You are on page 1of 174

1

Weste rn Kentuck y Universit y ~


4~.

i rur r~

AN APPLICATION OF OPTiMAL CONTROL


THEORY TO AN ANTI -TANK WEAPON SYSTEM

r ~ 6f ~\Ll ~ f l

u-I

PrirLiple Researchers : Dr. Randy J. York


Dr. Daniel C. St.Clair
Department of Mathematics
and Computer Science
-

Research Assistant Peter Sisler

I
a

~~~~~~~~~

S ECU MITY C L A S S I F I C A T I O N OF T H I S P A G E (When Data Enterer!)

REPORT DOCUMENTATION PAGE


T~~~~EPO RT NUMBER

.
~

BEFORE

2. G OVT ACCESSION NO.

T I T L E f e n d .SuA.iI1~ )

--

- .
_ _ _ _ __
.
~~~~~~ ~~~~~ ~

An App lication of Op t imal Control Theoc~~ to


an AntiTank Weapon System s

1
~R a n dy J / Y o r k

3.

14.

Commander ; U.S. Army Missile Research and


Developm en t Comm and
ATTN : DRDMITCN , Redstone Ar senal AL ~~c~gpg
MONITOR
& ADD RESS(I1 differ ent from ~Cont rol ling Office)

a CONr *e
w u
~ ~~~~~ ~~ ~zr

15.

ISa .

thu

PR
ECT . TASK
PRO
EN
EL
~~~
A R E A & WOR K~~~
UNIT~~~
NUMBERS

ne d78~
13. ~~~~~~

i
~~ J.... / J-f .

DISTRIBUTION S T A T E M E N T (of

CONTROL LING O F F I C E NAME AND A D D R E S S

16.

Final re

~~~~ .~~~~~~~~~~~~

n~~~~~~~~~~~~~~~ P~~~ T NUMBER

10.

II.

.1
f .1

NUMBER

.4~~~~~~~~~~ F n B renT ~~~~~r Q r r n c py ERED_

~~~~~~~~ FORMIN G ORG ~~~~ ZA T I O N NAME AND A D D R E S S

Western Kentucky Un iversity


Bowling Green , Kentuck y 42101

FORM

R E C I PI E N T S C A T A L O G

165

S E C U R I T Y CLASS. (of thia repo rt)

Unc lassif ied


~ I~ Ck~~~~

IIIC A T IO N / D O W N G RA D I N G

Report)

Approved for publi c release; distribution unlimited.


~ I
.
I
l

DISTRIB UTION S T A T E M E N T (of the abetted entered Ir. Block 20, if differen t from Report)

tO .

S U P P L E M E N T A R Y NOT ES

19.

K E Y WORDS (Continue err rever e. aide if nec ...ary end Identify by block number)

20.

Opt imal control law formulation


Optimal control law implementation
\\Terrninal
guidance
knci_ Tank weapon system

(Camtthu.
ABST~~~i
C1

DD

~~ ~

,,v.r .. at

~~ ft n.c...atv

~~ d fd.nhify by bloc k number)

Th e pr obl em of imp lementing a threestate optimal control law in ai

an~~. t ank weapon system was examined. The weapon system was represented
by a detailed six degree of freedom simulation .
The der ivation of the control law is included along with a four
sta te simulation that was used to model tt1e 6DOF simulation . The procedure
used to implement the contro l law is given in Chapter III , and t he me t hod
of ob taining optimal control parameters appears in A ppendix C. The ques tioi
of sens itivity is also examined using both simulations.

~~~

1473

EOPT ~ON OF I N O V 65 IS ODSOLE TE


i
,
/
/

~ i

/ /~6

SECUP I V C LA S

U n c l a s sj ..~~ ed
c f o~ cj I THIS

PAGE: (WIr.n Del. Ent.r. ~t)


4
L~~~1

June 1 , 1978

FINAL REPORT
for
Wes tern Kentuck y University Contract # DA.AK 4077C0052
(Reds tone Arsenal Con tract # N- TB-3-3)
~~~ ~

I
AN APPLICATION OF OPTIMAT CONTROL
~
THEORY TO AN ANTI-TANK WEAPON SYSTEM

71

to

Redstone Arsenal , Alabama

~ Mt. i

Harold Pas trick , Con tra ct Supervisor


G ui dance and Con trol Direc torate
Research and Developmen t Labora tory
U. S. Army Missile Command
35809

~~~~t

Dr. Rand y J. York and Dr. Daniel C. St. Clair .


Depar tment of Mathematics and Computer Science
Western Kentuck y University
Bowling Green , Ken tucky 42101

II

_ ~~~~~

from

P e t e r S i sler , r e s e a r c h a ss i s t a n t

~~~~~~~~~

01

~~~~ i~. w

IWA .

ABSTRACT
The problem of implementing a threestate optimal control
law in a antitank weapon system was examined . The weapon system
was represen ted by a detailed six degree of freedom simulation .
The der ivation of the control law is included along with
a fours tate simulation that was used to model the 6DOF simulation .
The procedure used to implement the control law is given in Chapter III ,
and the method of obtaining optimal control parameters appears in
Appendix C. The question of sensitivity is also examined using both
simulations. Final conclusions and recommendations appear in Chapter V.

ACKNOWLEDGEMENTS
Dr. Harold Pastrick , con t rac t supervisor , of the U .S . Army
Missile Command , has been most help ful w ith his su gges tions and
demands. We would also like to acknowledge the assistance of
Dr. Thomas Madron and his Research and Compu ter Services group at
Wes tern who have patiently answered our many questions and requests.
This contract , wi th its security requirement , has posed special
problems , and we would like to thank Dr. Glenn Crumb and his assistants
at the Office of Grants and Contracts for their aid.
Th is report could not have een prepared without the hel p of
Debra Wheeler , who not only was most care ful in the typing of this
manuscript , but who was able to spot and correct oversights on our
part.

Someone has to be las t , bu t we hope that no affront is taken .


We would like to especiall y thank our wives for their patience and
understanding they showed during this contractual period and the
prepara tion of this final report.

IdiL J

--

~~~~~

- - .~~~-~~~ --~

.-------

~~~~~~~

. ----- .
.--

______________________________________________________________________________________

- -

.- _- .

CONTENTS
Chapter
I.

Page
Introduction.
1.1
1.2

II.

2.6

IV .

1
2

Imp lemen tation of the Six Degree of Freedom Simulation


on the IBM 370 System.
2.1
2.2
2.3
2.4
2.5

III.

Backgr ound
Imp lementation of the Control Law

Incompatability of IBM and CDC FORTRAN


Disk Storage and CJS
Selection of Comp iler
Simulation Simplification for Developmental Work .
Add itional Capability Added
2.5.1 R o ll , Yaw Control
2.5.2 Plo tting Capability
General Review of the 6DO F Simulation
2 .6.1 Airframe Model
2.6.2 RotatIon
2.6.3 Transla tion
2.6.4 Aerodynam ic Forces and Moments
2.6.5 Targe t Model
2.6.6 Guidance and Control Models
2.6 .7 Gyroscope Model
2.6.8 Guidance Filter Model
2.6.9 Ac tuator Model

Con trol Law Imp lem en ta tion Ideal Case


3.1 Br ief Summary of Control Law
3.2 General Outline of Approach
3.3 De termination of Contro l Parameters y and 8
3.4 6DOF Simulation Implementation
3.4.1 Modifica tion of Control Module Cl
3.4.la Imp lem en t a ti on of t he Con t rol Law.
3.4.lb Con trol Module Cl Listing
3.4.2 Evaluation of Control Gains GPIT , CYAW
3.4.3 Control Parameter Improvement
3.5 Sensitivity Anal ysis
3.5.1 Sensitivity for the FourState Simulat ion.
3.5.2 Sensitivity with the 6DOF Simulation
Control Law Imp lemen t at ion Prac ti cal Case
4.1 Disadvan tages of the Ti~reeS tate Controller
4.2 A Con troller Using A , X
4.3 Imp lemen tation of the A , A , 0 Con t roller
4.4 Est imatIon of the Timetogo

..

3
3
4
4
4
4
5
5
5
7
7
8
9
10

10
11
11

. .

12
14
16
17
18

18

18

27

27
27
27
29
36

36
37

38

-.-

-_

Page
V.

39

Conc\~usions and Recommendations for Future Study

40

VI. Simplified 6DOF Simulation


References

124

Appendix
A.
B.
C.
D.
E.

LL.~t.I

--

Optimal Terminal Guidance With Constraints


at Final Time
FourS tate Simulation Listing
Determination of Optimal Control Parameters:
An App lication of Mathematical Programming .
HookeJeeves Algorithm Listing
Derivation of the ThreeState Control Law

iv

--

A l
Bl
.

Cl
Dl
El

..
~~~~ ---~~~~~ -

___________________________

LIST OF ILLUSTRATIONS
Figure
2.1
2.2
2.3
3.1
3.2
3.3
3.4
3.5
3.6
3.7
4.1

Page
Modularization of the 6DOF Simulation
Block Diagram of Pitch or Yaw Guidance Channel
Ideal Seeker Characteristics
Geometry of Tactical Missile Target Position
Hooke Jeeves Algorithm for n = 2
Original Feedback Control Syst em
Missile Roll Stabilized Position
Initial Implementation of the 3State Controller
RZE vs. RXE
THETA vs. TIME
Missile Target Geometry

6
7
10
12
17

19

. . . 28

18
34
35
36

--

LIST OF TABLES

Table
3.1
3.2
3.3
3.4
3.5
3.6
3 7

Page
Definition of Variables
New Variables for Subroutine Cl
Performance for Var ious Initial Conditions (Baseline I)
Performance for Various Initial Conditions (Baseline II)
~
Performance Dependence on GPIT
Performance Dependence on y and 8
6DOF Simulation Runs for Various Initial Conditions

12
20

. . 31
30
.

32
32
33

,
I,

_
_

_
_

- ~~~~ .__

Chapte r 1
INTRODUCTION
1.1

Back ground

Recent intelli gen ce suggests that the impenetrable nature


of heavy armor nay be susceptible to missile attacks at a relativel y
hi gh angle of i mpa ct , with respect to the horizon. In many modes of
direct encounter , the torget may not be reachable with a bod y p itch
attitude ang le of the proper magnitude. There are several poss ible
reasons fur this condition , including lack of energy (fuel), lack of
t ime to ni.inel1v ~r int o the more desirable attitude , or lack of control
in f o r m a tion b y appropriate sensurs to command tile response. This
condition has been recognized for some t ime at the Missile Research
and Developm ent Command , and consequentl y ther e have been attempts
to modif y trajectory shapes by a variety of pred etermined control
laws . ilowover , therL has been a certain lack of robustness in the
solutions obtained over the entire range of conditions anticipated.
This situation motivated a search for optimal solutions to the guidance
problem and a study of tiadeoffs among the suboptimal candidates that
were deemed feasible.
Terminal guidance schemes for tactical missiles may be based
on a c l a s s ical a p p r o a c h , such as a p r o p or ti onal nav iga tion and gu idance
law [3 ,4 ] , or on a modern control theoretic approach [1 ,5 ,6]. In the
latter , a control. law is derived in terms of t imevary ing feedback
gains when formulated as a linear quadratic control problem . A sub
optimal terminal guidance system for reentry vehicles , der ived us ing
the modern approach , was the basis for the initial work of this
p r o b l em .

f r.
~

Kim and Grider ~2] studied a suboptimal terminal gu idance


system tor a reentry vehicle by p lac ing a constraint on the bod y
attitude ang le at impact. Their problem was oriented to a longrange ,
highalt itude mi ssion . Their scenario was formulated as a linear
quadratic control problem with certain key assumptions. The ang le of
at tack of the reentry vehicle was assumed to be small and thus was
neg lected . Furthermore , the autop ilot response was assumed to be
instantaneous , i.e., ~ ith no lag t ime attributed to the transfer of
input commands to output reaction.

Thes e con di ti ons have been stu died in an ex t ension of t heir


earlier work where a formulation is given for a system that has finite
time delay. In fact , t he increase a n d decre ase in tim e delay has
interesting ramificaLions on the solution. The angleofattack
assumption is investigated , and , altho ugh not solved analyticall y in
closed form , the system is derived [9 1.

There is more than just a passing academic interest in this


pr oblem . As suggested previousl y, the ant iarmor role of several Arm y
weapon systems very well may be enhanced by this techn ique. The
reduc ti on to a prac t i c al imp lementation or mechanization is the aim
of this contract .

1.2

--,.
-.-- ~~~~~~~
~~~~

~~~~~~~~~~~~~~~~~~

~~~~~~~~~~~~~~~~~~~~

Imp lementation of the Control Law

In order to investigate the problems arising from t he prac ti cal


implementation of such an optimal control law on a real system , a de t a iled
six degree of freed om (6DOF) digital simulation of such a system was
needed. Such simulations are so complex that to implement the control
law from just a programming point of view would require that the simulation
would have to be in modular form; otherwise , the programming d i f f ic u l t ies
would obscure the implementation difficulties . Such a simulation of a
semiact ive laserguided airtoground missile was provided to the authors
by Dr. Harold L. Pastrick of the Army Missile Command , the contract
supervisor. Unfortunatel y , n o n e was ava ilable t ha t would work on the
IBM37O system at the Un iversity of Kentuck y that would be used see
sec t ion 2 .1 for a d iscussion of the considerable difficulties encountered .
The deta iled 6DOF simulation would give an accurate represen
tation of the real system and how it would respond with the new controller.
Al though a s imp ler fourstate simulation would be used as a guide in
impleme nt ing t he con t r o l law , it assumed only a point mass representation
of the missile. The 6DOF simulation had wind tunnel test data used in
the cal culation of aerodynamic forces and moments , and co u ld be u sed to
examine performance under more realistic conditions.

The original work [2] assumed no lag in the autop ilo t , ano ther
poss ible source of trouble. A more complex control law based on a first
order a uto p ilot lag model had been derived [9] and was available for
implementation if needed. Perhaps the mo st crucial assumption of all
t o be t es t ed , was that of no hard constrain t on the controller as is
presen t on the missile in the tail fin stops. The onl y cons train t in
the formulation of the control problem was a sof t cons train t presen t
in the cost function J to be minimized , Equation (3.3).

The appro ach u sed was t o im p lemen t t he con t rol law w it h the
set of state variables g iven in [2]: missile attitude ang le e , p r o j e c ted
missiletotarget distance on ground Yd. and ra te id . Perfec t knowledge
of the states were assumed . Since i
T cannot be measured , the set
d~ d
of state variables was then altered to include the line of sight
ang le A and rate ) .
In this setting, all var iables were to be ass umed
available at first , and t hen r ea lis ti c m e a sureme nt s from t he seeker
f o r A ,~ was to be used , along with a gyro measured attitude ang le
See section 1 .2 for a more complete outline of the approaches used , along
wi th other sections of Chapters III and IV .
Chapter V.

Conclusions and recommendations for future stud y appear in

2
I-- -

~~~~~~~~

_ .

..

~~~~~~~~~~~~~~~~ _

~~~~~~~~~

;IuuIuuI
~~

- ~--

-~~~~~~~~~~~~~~ u~~~

-.-

- .--

- .- . - ~~ ~

- -- --- -

Chapter II

IMPLEMENTATION OF THE 6DOF SIMULATION


ON THE IBM37O SYSTEM

2.1

Incompatability of the IBM , CDC FORTRAN s

One might be inclined to think that all FORTRAN s are


compa tabl e , but such is certainl y not the case. Each computing
sys tem , whe ther it be that of IBM or CDC , has cer t ain uni que
features of which a programmer can take advantage. The code of
the 6DOF simulation took extensive advantage of such features
and made imp lementation on the IBM37O system quite difficult.
All incoinpatabilities between the two FORTRAN s had t o be de t ec t ed
and resolveda process which required several months of effort .
The f inal goal of the altering of the FORTRAN code was to produce
a simulation that woul.d run on the IBM37O as well as the CDC6600 ,
that is , the code was not to take advantage of the special features
of either system.
Some of the programming inconsistencies were easy to
spo t and correct. Among theue were the use of * for titles in
FORMAT statements , variable names that were too long, restrictions
on handl ing al phanumeric character data , res t ric ti ons on com bin ing
the initialization of data for arrays and specifying their length ,
the number of lines on wh ich a st at ement can be con tin ued , res t ric tions
on the indexi ng of parameters in DO loops.
Other programming incompatabilities were more difficult to
de tect and correct. The p lotting capability in the simula ti on made
extensive use of in house p lotting subroutines and had to be bypassed.
A very troublesome programming practice was that of not initializing
variables when their initial value was to be 0 . The CDC machine
will zero out its memory with the start of any new job . The IBM
computer does not. The g a r b a g e in those par t icu lar memory loca t ions
will he used in the executation phase without any warning to the user.
The way th is p roblem was f inally solved wa s to run the program with
a WATFIV comp iler which will flag uninitialized variables .
Some subroutines such as those involvad with linear inter-

p o la t ion ~~t une or two vari ables , mad e such extensive use of special
CDC array m an i pulation features that they had to be rewr itt en and ,
consequent ly, re tested for accuracy .
2.2

Disk Storage and CJS

The simulation was of such length that disk storage was


essential. Once the simulation was p laced on disk , t he problem was
still how to implement the numerous corrections decided upon . To
t h i s end , it. was decided to use a new IBM editing creation called CJS ,
Conversa tional Job System which was a subset of a larger package
called CMS , Conversational Monitor System . EspecIall y desirable
feature s Included were the ability to search for ph r a s e s , t o dele t e
or add lines , rear range wh o le sec t ions of c ode , t o make global
c h a n g e s a s w e ll a s lo c a l ones , and to create executive programs to
s e r v e a wide variet y of needs .

~~~~~~~~~~

--

-- -.- - -.~~~~~~
~
~ ~~
~

To reduce execution time , all subroutines which were to


remain unchanged were stored on d i s k in obj ec t f orm to lessen
com p ilation time.

Sel ection of Comp iler

2.3

The %.~ATFIV FORTRAN com p iler was cho sen for t he ini tial
phase of the work because of its rather extensive collection of
syntax error messages. Once the syntax errors were removed , the
Glevel comp ile r was u sed becau se of it s speed in comp iling. When
finally the po in t was reached were code changes were few , the
Hlevel comp iler was incorporated because it optimizes the execution
of the code. ln order to do t his , it does take longer to compile.
Simulation Simp lif icat ion for Dev elopm ent al Work

2.4

For t he develop m ent al phase , a shor t er , less comp lex


simulation was desired , one that would be less expensive to run.
The or iginal 6DOF simulation provided capabilities that were neither
needed nor d esired. A variety of seeker modules , ac tua tors , and
autop ilots were available. The simulation was set up to run Monte
Carlo sets and to give various statistical analyses for the results .
There was in add it ion a p lotting capability which would be quite
desirable but was unusable due to the utilization of subroutines
found only on the CDC system .
To reduce comp ilation and run time , it was decided to
pare down tile simulation so that the resulting shorter form of the
simulation would contain onl y t he mod u les an d su brou ti nes necessary
for the development work . To this end , some 27 subroutines were
deleted . They were C2 , C2I , NORMAL , RANNUM , MCARLO , AERROR , TERROR ,
CEPAS , CEPP , NORM , KSTEST , TABLE , PPLOT , XLOC , G2 , G2I , S4 , S41, CS ,
C5I , RESET , PLOT4 , S2 , S3 , PLOT2 , PLOTN , and SUBL1.
A lis ting of the shortened form of
appears in Chap ter VI .
2.5

the 6DOF simulation

Add itional Capability Added

2.5.1

Roll , Yaw Con t r o l


Since the initial work in [2] assumed motion in one plane

only, the capability of controlling roll and yaw mo tion was added .
The simula ti on alread y had a roll control feature , bu t this had t o
be al tered somewhat when the yaw control was added. The user sets
switch OPTN3 equal to 0 or 1 , depend ing on whe ther he wishes to
allow the missile to roll or not , respec t ivel y. If no roll is
selected , then by setting another switch OPTNYW to 0 or 1 , he can

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

~~~~

-- ~~~~~~~~~~~~

- _aU~~ ~~~
J

allow the missile to leave the pitch plane or f o r c e it to f l y in it.

The coding of this al t erat ion of subrou tine D2 fo llows :


IF (OPTN3.LE .O.) GO TO 45
IF (OPTNYW.LE.0 ) GO TO 55
GO TO 65

CRAD*FNXBA/FMIX

45

WPD

55

WRD

65

W QD

2.5.2

(CRAD *FMBBA+(FMIX_FMIY)*
WP*WQ/CRAD)/FMIg
(CRAD*FMYBA+ ( FM I~~~FMIX) *
WP* WR/CR A D )/ FNI Y

Plotting Capability

The orig inal plotting capability was rep laced with one
prov ided by Dr. St. Clair which would generate line printer plots.
Any variable in the Carray could be plotted against time , and any
two variables in the Carray could be plotted against each other.
Since the grap hs are done on the line printer , they are ra ther
crude in appearance , but they do provide some insi gh t into
how variables are chang ing with respect to t ime .

The variables to be p lotted are stored on disk. Although


the plotting subroutines that are used on the IBN37O cannot be used
on the CDC6600 , any program using a package compatable with the CDC
6600 could be used to read off these stored variables.
2.6

General Review of the 6DOF Simulation

The modularization of the 6DOF di git al sim ula t ion of a


sem iactive laserguided airtoground missile is given in Figure 2.1.
A block d i a g r a m f o r one channel (p i t c h or yaw ) is shown in Fi gure 2 . 2 ,
and a b r i e f d e s c r i p t i o n of the various components follows . For a more
detailed examination , refer to [111 and [12].
2.6 .1 A irframe Model
The a i r f r a m e model included t r a n s l a t i o n a l and r o t a t i o n a l
dynami cs and generation of aerodynamic forces and moments. Airframe
dynam ics were simulated in 6DOF with the missile orientation represen ted by three Eu ler ang les. Maneuvering control was achieved by
means by four surfaces in cruciform configuration hinged about fixed
stabilizing fins.

- -- - ~~~~~~~~~~~~~~

0
UI-

~-

~
U

.--l

U
-

I~r

I_ J

.4

X I - OZ

., 4
~

4-

~o

ha.

~fl

fl1
t
~~ ~~~ i

_ ii
~

if

-I

I z

O.
~

~~~0u. w

t~ o

<U .

icc ~~~
ix i~~~~
>

I>

,,,x5~~~
i

>
9I - z
~~

cli

I>

00

~~~~~~
I-

~ ~1

_,

~~~~U.

,
U

iII>

I
..

I-

0 0

,
.
,
~~~~~~~~~

r4

i,
UJ

0 0

IH
Z

6
~~~~ _ -

- _

_ -

~~~~~~~~~~

- -

--~~~

~-

.-

-~~~~~~~~~~

--_ -~ ~~-

SAMPLE

L J OE TECT
~~

GUWANCE

J_a.I

._J

t~~~~~
LOS

GEOMETR Y

Figure 2.2.
2.6.2

ACTUATOR

~~~~[AMP

_H

0li

- -_-_- -- ~

TURNING
RATE

AIRFRAME
AERO

Block diagram of pi tch or yaw guidance channel.

Rotation

1
X~ B
I~ Q~
I~

M~~

(2.1)

M
+
YA
M

I )P R
X B B

(2.2)

(2.3)
X~~ B~ B
~~Y
where ~~~
R are rotational rate components about body axes X
8, ~~R
~~~ B;
Z respectively
x and 1y are moments of inertia about XB and T B ana
B
~
M~~ , MYA, M ZA are componen ts of aerodynamic momen t about the body~ axes.

~~~~~~

ZA

By resolving bod y rotational rates into the inertial and


intermediate axis systems , the following equations for Euler ang les are
ob tained :
=

(Q Bcos I

~E

~~~~~~~~
=

2.6.3

R~ sinq ~~)/ costp ~

(2.4 )

+ R cos
B
~~E

~B

ESIfl*E

(2.5)

(2.6)

Translation
It is advantageous to express the translation equations of

motio n in t erms of i n e r t i a l axes giving

~~~~

- - _-~- ~

n
r

(2.8)

ZA

YA

mg

(2.9)

- - - - - - -

_ _ - -

- - - ------ -

mZ

(2.7)

mY

F~~

mX

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

-~-~~

---- .
~ -- ~~
~~~~~~~~~~

__ ~~

~~~~~~~~~~~~~~~~~~~~~~

where m is the mass of the missile ; F y~ , F YA, FZA are the aerodynamic
forces in the inertial axes ; and g is acceleration due to grav ity .
Aerodynamic forces are ob tained in inertial axis form by transforming
body axis components according to the following matrix equation :

Fx

F~~
F YA

[TI

(2.10)

F~

F ZA

Fz

where F x, F~ , F z are the body axis aerodynamic force components.


bodyearth transformation matrix has elements defined in [8].

The

2.6.4 Aerodynamic Forces and Moments


For both missile models , aerodynamic forces and moments are
expressed in terms of body axes. The missile trajectory is divided
into two sections; unguided flight before the target laser spot is
visible (the preacquisition phase) and the guided portion of the
trajectory after the seeker has acquired the target spot (the post
acquisition phase).
Aerodynamic force and moment components for both flight phases
are as follows :

Fx
F~

F
=

YA

ZA

=
=

qSD(C~

=
=

=
+

q SC~~

(2.11)

qSC~

(2.12)

~ SC~

(2.13)

DP C /2V)
B P

(2.14)

C l2 7)
qSD(C~~~ +
~~B MQ~ ~
qSD(_C
~~ 0 + DRBC NR /2V)

(2.15)
(2.16)

where q is the dynamic pressure , S is a reference area (crosssectional


area of the missile), D is a reference length (diameter of the missile),
and V is the total speed of the missile relative to the atmosphere.
Dynamic pressure is given by
q

2
pV /2

(2.17)

where p is atmospheric density and is a function of altitude.


Aerodynamic coefficients and derivatives referenced in Equations
(2.11) through (2.16) are obtained from wind tunnel tests and other
estimates are expressed in graphical form as functions of angle of attack ,
Mach number and , during the postacquisition phase , control vane angles .
Coefficients and derivatives are defined for both models by differing
functions for the two flight phases . Details of the coefficients for
both models are presented in [8].

--

_.~~

Angle of attack components (m and ) may be defined by the body


~
axis components of velocity. However , for digital
computation , there is
some advantage in expressing a and ~ in terms of integral equations using
wind axis coordinates . Thus , using equations for & and
~ derived in [8]
and making small angle approximations , the equations for angle of attack
components are :
=

(F
ZB
(F

YB

aF

XB

)/ mV

I XB) 1mV
~~

8P

uP

(2.18)

~B
R

(2.19)

where the body fo rce components (F XB, F YB, F ) include aerody namic
ZB
f o r ces a n d gravity force resolved into body axis components.
2.6.5

Target Model

The t a r g e t model included target position as a f u n ct i o n of


time and the calculation of missiletarget miss distance at intercept.
Ta rget motion was specified deterininistically as eithe r constan t acceler
ation or cons tant velocity from an initial starting point . However , the
effect of jitter in the designating laser beam direction was intro
duced by applying a random disturbance to the calculated target position .

Target to missile disp lacements are


AX

X~

where X~ , Y~ , Z~
missile position
two Euler angles
axes relative to
displacements in

(2.20)
(2.21)

~m

Z
Z
(2.22)
~
are target position coordinates and X , T
Z are
coordinates . The gyro platform is charac~ eriied by
P ) which define the orientation of the platform
~~~~ 51 axes parallel to X , Y , Z. Target to missile
inertial
digital seeker axes are given by
AZ

AX
AT

AX

5
= [5 ]

AZ
5

(2.23)

~~

AZ

where the transformation matrix [S] has elements as shown in [8 ] .


From the target to missile disp lacements in seeker coordinates ,
the boresi ght error ang les are defined as:
BEPSZ
BEPSY

tan

tan 1 (A Z /AX )
5
5

(AY / ~~~
5

-~~~ - .
~~~~~~~~

(
2.24)

A Z~ )

- . -

--

(2.25)

-_ - ~~~~~~~~~~ ~~~~~~~~~~~ -

-_

~~~~~~~~~~~~~~~

~~~~~~~

~~~~~~~~~

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

- -

~~~~~

2.6.6 Guidance and Contro l Models


The guidance and contro l models included the seeker d e t e c t o r
and gyro , guidance filters , and vane actuators . The laser detector is
mounted on a 2DOF gyrostabilized platform in the nose of the missile ,
and the laser beam is viewed through a f ixed lens . After the gyro has
been uncaged during the terminal homing phase of the missile flight ,
gimbal torquing signals are generated to null the spot displacemen t
from the center of the detector and the resulting gimbal angles and
rates relative to the missile are used to generate guidance and stabilizing signals for input to the guidance filters . Detector output
Is from a sampleandhold uni t which operates at a fixed period .
A proportional navigation guidance schem e is employed; the guidance filters are leadlag networks in each of the p itch and yaw channels
which serve to decoup le the missile natural frequency in p itch and yaw
from the control system.
The ideal characteristics of the seeker transfer function are
shown in Figure 2.3.

I
2

FOV LIMIT

I
2

j:

FOV LIMIT

LOS ANGL E

Figure 2.3.
2.6.7

Ideal seeker characteristics.

Gyroscope Model

For man y of the studies using the di g ital simulation , it is


desirable to model a gyro which has no dynamic time constants. The
simplest model possible is one represen ted by a perfect integration ,
i.e., 1/s in the Lap lace notation.
As is well known [8], for example , the differential equation
relating output axis motion to input torque or input rate is given by
JO + BO + KO = H4c

10
~~~~

~~~~~

- .- -

--

- .

2
where J ~ ftlb/(rad/sec ), B
H -s- ftlb/(rad/sec).

ftlb/(rad/sec), K

- ~

ft/rad , and

2.6.8 Guidance Filter Model


The guidance filter or compensation network is of the lead
lag type wi th transfer function (T3s + l)/(T4s + 1), where T is the
3
lead time constant and
is the lag time constant. The input to the
guidance filter comes from two sources , the sampleandhold output of
the seeker and the damping network output.
2.6.9

Actuator Model

A firstorder actuator was used in the simulation with transfer


function AA/ (s + AA ) where AA is the actuator model constant. The input
to the actuator is from the guidance filter , and the output drives the
canards . The yaw control canards are on a common shaft , as are the
pitch canards.

-...
~~

, -- ~~~~ ~
~~~~~~~~~~

~~~~~

-:1

_
~~~~~~ ~~

--

- -- ~
~
~ ~

_ _ - _

~~~~~~~~~~~~~~~~~~

~~~~~~~

Chap ter III


CONTROL LAW IMPLEMENTATION
3.1

IDEAL CASE

Brief Summary of Control Law

In the paper Terminal Gu idance for Impact Attitude Angle


Cons train ed Fli ght Trajectories [2], a simp l i f i e d model of a m issile
ta rget scenario is used to produce a three s t a t e control law t h a t will
result in the missile impacting v er t i c a l l y on the target , Figure 3.1
deplicts the geometry of the terminal guidance phase , and Table 3.1
defines the variables introduced .

~~~~~~~~ A CT ICAL MISSI LE


A L COS O
I.E

~~~~~~

i../

VER TICAL
REFERENCE !

Figure 3.1.

__________

Geometry of Tactical Missile

TARGE T

Target Positions.

Table 3.1. Definition of Variables


Variable

Definition
Missile position variable projected on the ground (ft)
Target position variable (ft)
Position variable from missile to target projected on
=
the ground
~~d

_. uIUcIUI
~
~~

~ - ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

____________________

V. -

__
~

_ _ _ __- -~~
~

_-

Table 3.1. Continued


Variable

Definition
Time derivative of

AL

(ft/sec)
~d
2
Late ral acceleration of the missile (ft/sec )

Body attitude angle of the missile (deg)

The following set of state variables was chosen for the


modeling process.
Yd
-

(3 .1)

AL

0
Under the following assumptions :
1)

the ang le of a t t a c k is small and can thus be n e g l e c t e d , and

2)

t h e a u t o p i l o t has zero lag ,


C~~~( t )~~~~

U =

C2 (t ) ? d

c3 (t ) 0

(3.2)

a c o n t r o l law of the form can be o b t a i n e d w h i c h w i l l minimize the


performance index
J

Y~~~( t ~~)

t 2
t)dt
Y0 (t f )+ ~f f u (

(3.3)

where the s t a t e v a r i a b l e s are s u b j e c t to t h e f o l l o w i n g d y n a m i c s :


=
=

AL

(3.4)

~d

(3.5)

A L cos 0
w 1A 1

Ka u

(3.6)

K 1u

(3.7)

P e r f o rmance is considered to he acceptable when the following constra ints


are satisfied:
IY d (t f )i

<
<

5 feet
5 d eg r ees

~~~~~~~~

(3.8)
.

(3.9)

The time varying coefficients appearing in the control


law are g iven b y:
c1

[~ g(tft)

[ ~ g(t ft)

=
C3

[
K
~~~~ a

2
2
gyK ~ (t f t ) / 2 1/ A

gyK ~ (t f t )

yK~ g ( t f t )

(3.10)

/2]/A

/6 ]/A

(3.11)
(3.12)

where
g
and

-hKi

(b

cosO in the linearized form)


-

2
2
3
Y~ K a (t f t ) + ~g (tf t) /3
2 2
4
yg Ka (tf t ) / 12 ,
t

time , t f

(3.13)

(3.14)

time of impact.

A derivation of this control law is given in Appendix E.


The assumption of no lag in the autop
li ot has been removed
by Pas t r i ck and York , O p t imal Term inal Gui dance wi th Cons t rain t s a t
Final Time s [9]. This work discusses a control law more general in
na ture which allows the control law of [2] as a limiting case.
Appendix A contains a detailed examination of this more realistic ,
and con seq uen t l
y, more complex con trol law.
3.2

Gene ral Outline of Approach

A simp l i f i e d s i m u l a t i o n u s i n g the dy n a m i c s as modeled in


section 3.2 was developed (Appendix ! ) . Whenever f ea s a b l e , new ideas
~
were tried out on the simplified simulation first to tes t their merit.
This approach is not onl y p h i l o s o p h i c a l ly sound , b u t there was a very
prac tical reason for it as wellmoney , or rather the lack thereof.
The large scale simulation received from Redstone Arsenal required
large amounts ol memory and was r a t h e r e x p e n s i v e to run (@ $20 f o r a
comp l e t e fli ght).

The smaller four state simulation was made to resemble the


large 6DOF one as closely as possible. The original constant velocity
was replaced with a t ime varying velocity profile taken from a 6DOF
simulation run. Thc parameters K 1, K a w 1that appear in the four
~
state dynamL.s were reexamined and evaluated
to be sure that they
reflected the dynamics of the actua l weapon system . It was found that
the parameter values given in [2] of a typical tactical missile did
reflect the dynamics in our case as well. One discrepancy was that
our autopilot lag was more accurately represented by
w

9.8

in stead of
w = :
~
-

(3.15)

________

_____________________

- - ~

- - -

but it was decided to use the smaller value in the developmental


work as this represents more lag in autopilot. Since our small
scale simulation ignores aerodynamic forces and moments , it was
fel t tha t this increased lag mi gh t make o u r s imp le simulation
even more represen tative of the actual system.
The 6DOF simulation was altered as well. Since the
initial work in [2] assumed fligh t in one p lane onl y, it was
decided to firs t try to imp lement just the p itch control , and
then later a yaw control. Consequently , the roll and yaw c on t rol
parameters were set so that the missile was allowed to neither
roll or yaw . Once the control law was properly implemented in the
pitch channel , then symmetry would suggest that it could be implemented
in the same manner in the yaw channel.
Once the contro l law was performing properly with the four
state simulation , then it would be used on the 6DOF simulation . The
approach used in implementing the contro l law was to assume that all
of th e states T
d Td and 0 were known and could be measured exactly,
and that it was possible to generate the time vary ing coefficie,. -s
c 1, C 2 , c~ without error. It was felt that for the initial implementation that it was better to leave the choice of states as given in
[2], alth oug h there is in fact no way to measure the range and range
rate states of
and 1d For this reason , this imp lementation is
referred to as the ideal case. For the practical case , an alternative
set of state variables was used.
The control parameter values appearing in [2] were op timized
f o r a ver ti cal impac t an g le , but the initial geome t ry was t ha t of a
for tyfive degree triangle (10 ,000 f rom the target 10 ,000 up).
When us ing their control p arameter values of
y

3823

and

13

6.94E 04

(3.17)

with our low profile trajectory (200 x 2600), the m issile in the
simp le simulation would overfly the target. The same type of trajectory
was obtained on the 6DOF simulation . Basically , the t rouble seemed
to be that the missile could not turn in the time allowed. The presence
of aerod ynamic forces and moments made this problem even more acute.
I t was decided to f i r s t t ry to achieve an a t t i t u d e angle t h a t
would be less demanding on the control system , such as
0(tf)

45 degrees.

(3.18)

A ppend ix A show s how it is possible to use the same control law to


ach ieve an arbitrary ang le. These initial runs also suggested that
the geometry for the terminal gui dance pha se needed t o be a lt ered
to allow more hei gh t with which to work. Thus , it was decided tha t
the trajector y would need to consist of two phases : a preprogranmied
p i tch maneuvi~r to gain altitude followed b y a t erminal guidance phase.

i-

rr

~~~~~~~~~~~~~~~~

_____________________

Initial condition s for the terminal gui dance phase we re

chosen t o be:

1000 , Td

5000, 0

85 .

(3.19)

The same assumptions were made as in [2] of no ang le of att ack and no
lag in the autop ilot. It was decided to imp lement the 3state controller
because of its relatively simple form. Should lag become a problem ,
the 4state controller discussed in Appendix A could be tried. The
basic problem then was t o f ind the op t imal values of the two con t rol
parameters r and 8 that will minimize the cost functional J given as
Equation (3.3). This problem was solved using the Hooke Jeeves
Al gorithm as discussed in section 3.3.
With the following parameter values ,
y

5525.51

and

.19E06

(3.20)

a miss distance of 3.3 was achieved with an attitude angle at impact


of 45.270. With these control parameter values , the control law was
used with the 6DOF simulation . Section 3.4 discusses how the control
law was implemented. In summary , thoug h , a m iss dis tance of 18 fee t
and an attitude angle of 37 degrees was obtained before efforts were
made to impr ove on the con t rol parame ter val u es , see sec tion 3.4.
Finally, perf orma nce sens iti vi ty t o in it ial condi t ions is disc u ssed
in section 3.5.

~~~

3.3

mm

i. ni z e d :

Determination of Control Parameters y and 13


Two c o n t r o l p a r a m e t e r s appear in the cost functional to be
J

2
Y~ ( t ~~)

2
i [0(t~ )_45]

t 2
~J~ u (t)d t

The problem was to determine optimal values for


following success criteria could be achieved :

hd

(t

f)I

<

(3.21)

and 13 so that the

y-

5 feet
<

(3.22)

5 degrees

(3.23)

The appr oach used to solve this problem was to view it as


the mathem atical pr o gramm ing problem of m in imizing
=

Y 2 (t )+ [ 0(t ) _ 4 5 ] 2
d
f
f

(3 . 2 4 )

where the t i n ~~ t Ion evaluatio n was achieved through a computer run of


the fourstate simulation program with specified values for y and 8.
(see Appendix C for a discussion of mathematical programming in
general and this problem in particular ) .

~~~~~~~~~ -~~ -~~~~~_-

---

--

- - -

---- -

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

To make t h e m a t h e m a t i c a l programming problem m a t h e m a t i c a l ly


f e a s i b l e , it was assumed that F(y,13) varied in a continuous manner with
respec t to its two arguments. Since F(y ,13) cou ld no t be expressed
in any cl o sed f orm , there was no der ivative information available .
Af ter some thought it was decided to use the Hooke Jeeves Algorithm
to tackle this problem as it is rather straight forward , has almos t
guaranteed convergence , and was readil y available.

Th is par t icular algori thm will make two exp lora tory search
moves , one in each coordina t e direc t ion , to reduce F(y,13) , and then
will follow this with a pattern search (in the direction of the
diagonal).
0

Figure 3.2.

Hooke

Jeeves Algorithm for n

Using this algori thm , the optimal control parai et er values


~
were found to be
y

5525.508

13

(3.25)

l9OEO6

which yielded the following results:


Yd ( t f )
3.4

3.3

and

0 (t )= 45.27
f

(3.26)

6DOF S i m u l a t i o n Imp l e m e n t a t i o n

The f irst task in implementing the threestate optimal control


This modification is given
in section 3.4.1.

law was to m odif y the control module Cl.

The second task was to evaluate the two gains GPIT , GYAW
that appear in the control module , see section 3.4.2.

~~~ .

~~~~~~~~~~~~~~~

~~~~~~~~~~~~~~~~~~~~~

The l a s t t a s k was to improve upon t h e values of - -~ and 3


given by t he Hooke Jeeves Al g o r i t h m w o r k i n g w i t h t h e f o u r s t a t e
s i m u l a t i o n w h i c h was b e i n g used t o a p p r o x i m a t e t h e 6DOF s i m u l a t i o n .
This improvement is discussed in section 3.4.3.
3.4.1

Mod ification of Control Module Cl

3. 4 . l a .

Imp l e m e n t a t i o n of the C o n t r o l Law

Figure 3.3 gives the original feedback con trol system

based on propor t ional navi gat ion . As discussed in section 3.1 ,


it was decided to initiall y con t rol p it ch mo ti on only. The roll
stabiliza t ion channel was lef t in t act , as was t he cross coup ling

of the pitch and yaw control channels. This cross coupling is


necessary since this particular missile when it is roll stabilized
flies with the following tail fin confi gu ra t ion :

NO. 1

~~~~~~~~~~

.:: 3
~~~~~~~~~~ NO.2
LOOKING FORWA RD . -O

Figure 3.4.

Missile Roll S t a b i l i z e d P o s i t i o n

All f o u r t a i l f i n s are r e q u i r e d to e x e c u t e a p i t c h maneuver , as


would also be true for a yaw or roll maneuver.
The initial command signal t o t he t ail fin occurs at the
first summation in the p itch (
upward) and yaw (lower) control
channels. Th is signal is the difference between the commanded
attitude ang le and the sensed attitude angle (the feedback signal).
The new commanded attitude angle was given by the control law
u

cl(
t )Yd

C 2 ( )c d +
~ ~

c3 (t)0

(3.27)

The gui dance f i l t ers in t he p it ch an d yaw channels were replaced by


gains GPIT and GYAW for the initial imp lemen tation.
is given in Fi gure 3.5.

The new autop ilot

3.4.lb. Control Module Cl Listing


Table 3.2

The new variables that have been introduced are given in

2~~~

______

_______________

tci

~~~~

-c :
~

~~~~~~

~~~~~~~~~~~~~~~~

~~~

3,

.
,
~~~~~~

_.._

.~;

~1
u

~~~~~~

~-

________________________________

______________

- -

~~~~~~~~~~~~~~~~~~~~~~~~~

-~~~

.~~~~~~~~~~ -- ~- -- - -~ .- - ~~~~~~~~~~~~~~~~~~~~~~~

~~~~~~~~~~~~~~~~~~~

Table 3.2 . New Var iables for Subroutine Cl


Name

Symbol

BETA

13

BDELTC( I )

cj

C Array
2917

Def inition
Control parameter

856

Commanded t a i l fi n p o s i t i o n , i

BJJ

881

P i t c h si gnal a f t e r coupling

BJS

537

PI t ch signal bef or e coup ling

BKK

882

Yaw s i g n a l a f t e r coupling

BKS

545

Yaw signal bef ore co up ling

BJJSSS

987

P itc h signal before limi t er

BKKSSS

988

Yaw si gnal b e f o r e l i m i t e r

BTHT

350

Attitude pitch ang le (measured from horizon)

BXX
989

1, 2 , 3 , 4

Roll signal af ter limi t er

CTHTA(T)

c (t )

C o n t r o l law c o e f f i c i e n t f o r 0

CXREL(T)

c (
t)

C o n t r o l law c o e f f i c i e n t f o r Y

CXRELD(T)

Control law coefficient for

DELTA(T)

Roll signal before ne twork

BXXSSS

(t)

d
d

Denomina tor for control coefficients

-bK /w
1 1

GAMMA

2916

CPIT

2921

Gain in p i t c h channel

CYAW

2922

Gain in yaw channel

2918

Propor t ionali ty cons t an t appearing in

2919

Cons tant appearing in autopilot lag model

2000

Time

KA

K
K

11k

Loca t ion in

a
1

Control parame ter

dynam ics

TGOX

E s t i m a t e of t i m e t o g o

TGOY

Estimate of t imetogo in z direction

~~~

.--- ---

in x d i r e c t i o n

--

,,
~~-, -

.= =

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

L.~~~ _ 2 .

Table 3.2.
Name

Sym bol

Wl

XRE~.
XRELD
YREL

YRELD

_ _ _ _ -~

--

- - __

2906

Estimate of time of impact

2920

Au top ilo t lag cons t an t

~d

2904

Relative xdirection missile t a r g e t


posi tion (projected on the ground)

~d

2905

Time derivative of XREL

2902

Relative ydirection missile target


position (projected on the ground)

2903

Time deriva t ive of

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

-~~~~~~~~~~~~~~~ - -

i~ . . _ ~~~~ _

(TGOX , TCOY)

YREL

-_ r _

Descri ption
Mm

IF

~~~~~~~~~~~

Continued

L o c a t i o n in
C Array

TGO

~~~~~~~~~~~~~~

~~~~~~~~~~~~~~~~~~~~~~~~~~~~

-~ ~~~~

~~~~

.-

SU B RO U f1N E C l
COMM ON C ( 3 8 3 0 ) , G R A P H ( 3 O O ,~ 4 )
DIM ENS ION BDELTC (~4) ,V A R ( 1O l )

C**INPUT

DATA

EQUIVALENCE

EQUIVALENCE
E QU I VA L EN CtL
EQUIVALENCE

(CC

(C (
(C(
(C (

860 )
d6 -l )
~62)
361 3 )

, TDY

( C (

869)

,GBIA S
, GN
,W N2

)
)
)

)
EQ Jl V A L ~ NCE (C( d6L4 ),WN1
)
EQUIVALENCE ( C ( 865) ,WL
( C ( 866 ),WLXX 1 )
E QUIVALENCE
EQ UIVALENCE (CC ~6T) ,WLXX 2 )
E Q U I V A L E N C E (C( d68) , WLJK 1
E Q U I V A L E N C E

EQUIVA LENCE

--

WLJK2

(C( 870) , H JK

~ (l) , WXX
812) ,DXX
~37$ ) , WJK
17N ) ,DJK
875) ,GXX
8!6), GJK
d77) ,RES
E Q U I - .A L ENC E ( C ( 878 ),QDN

EQUIVALEN CE
EQ U IVAL ENC i.
EQ U IVAL ~ NCE
EQ U IV i ~L :J4CE
EQ U IV A L EN CE
EQ U IV ~ LEN. E
EQU IV i L ENC E

(C(
(C(
(C (
(C(
(C(
(C(
(C(
(C(
(C(

EQU IV O LEN C E
EQUIV A LEN CE
E Q U i V A L E . CE ( C (
~
EQU IV A LE NC E. ( C (
E Q U I V A L E N C E ( CC
E Q U iV A L E N C E ( C (
C

EQUIV ALENCE
EQUIV A LENCE

E Q C 1 V A L~JC t .

E C U IV A L E N C o
EQ UI 1iA L E N C E
EQ U V A L E N C E
EQ U I VA L E N C L
EQUIVALENC E
L Q U IV A L NC E
~~
EQUIV A LENCE
EQU IV I - L E N C E
EQUIVALENCE
E Q U IV A L L -C E
~
EQUIVALE NCE

89 2 ) , Q B I A S

d93) , RB IAS
~ 9 9) , OPTC 1

9L17) ,GNS

(C( 9~48) ,~4S1


(C( ~ 49) ,WS2

C * * I N P U T S F R O M OT H ER
E Q U I V A . EN C E ( C (

~
EQU I V A L E N C E
EQUIVALE N CE

879) , Q U P
3 9 0 ) ,HXX

( CC
C(
~

C~~~~~

S T A TE

V A R I A BL E

EQ UIVALENCE
EQUIVALENCE
EQU iV A L E N C E

E Q U I V A L E N C E

EQUIVALENCE

-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

)
)
)
)
)

)
)
)
)
)

)
)

i 7 ) , SPII I

STHT
9 7 ) ,SPS I
8 7)

~ C ( 353) , BPH 1

( C ( 35 4 )
~
(C ( j 5 5)
( C C 4O 3 )
~
( C C L4 O 1)
6 l)
( CC
C( 1~46 2 )
( CC 4 6 3 )
(C( 1233)
(C (l7 47)
~
(C( 17 14 3 )

,BTF12
,.B PS1
,~~LAMQ
,~~LAMR
,CAG E
, TKRZ
, TKRY
,BDR
,WH
,WQ
( C ( 1 7 3 9 ) , WP

N1 - J D U L E S

C
C * * I N P U T ~, F R C ; ~ M A I N P 8 O G R A N
EQUIVA lE NC E ( C ( 2 u 0 0 ) , T

)
)

)
)
)
)
)
)
)

J U TP U T S

(C (
(C(
(CC
(C(
(CC

~OO) , nLQSDD)
803) ,~ LQSP )
)
~3O~ ),~~LQ3D )
807),~~LQS
dO S ) , WLQSSD)

~~~~~~~~~~~~~~~~~~~~

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

~~~

--

~~~~~~~~~~~~~~~~~~~~~

E Q U I V A L E N C E (CC
EQUIVALENCE ( CC
E QUIVALENCE
(CC
(
EQ U I V A L E N C E
E Q U I V A L E N C E .C C
E Q U I V A L E N C E (C(
EQUIVALENCE (CC
EQUIVALENCE (C(
EQUIVALENCE CCC
EQ U I V A L E N C E (CC
EQ U I V A L E N C E (CC
E Q U I V A L E N C E (C(
E Q U I V A L E N C E CC C
E Q U I V A L E N C E (CC
EQ U I V A L E N C E (CC
E Q U I V A L E N C E CCC
E Q U I V A L E N C E C C .
EQUIV ALENCE (CC
E QUIVALEN CE ( CC
EQUIVALENCE
(CC
EQUIVALENCE
EQUIVALENCE

EQUIVALENCE

E Q U IV ALENC E

EQUIVAL ENCE
EQ U I V A L E N C E
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQ U I V A L E N C E
EQU iV A L E N C E
EQUIVALENCE
EQ UIVAL ENCE
EQ U I V A L E N C E
EQUIVAL ENCE
E Q U IV A L E U C L
E QU I V A L E N C E
EQUIVALENCE
EQ U I V A L E N C E

L.EN CL
~Q 2i
E QU I V A L E N C E

8 1 1 ) ,WL Q SS )
8 12 ) , WL RSDD )
)
)
~3 l 5 , WL R SP
8 l6 ) ,WL RSD )

8 1 9 ) ,WLRS

83 6) , BJJSD

)
)

820), WL R SSD )
823),WLRSS )
32~4) , BLQSSD )
827) , BLQSS )
828) , BLRSSD)
831 ) , BLRSS
832 ) , B J J S D D )
835 ) , B J J S P )
BJJS

8 3 9)

8L 0)

, BKKSDD)

, BKKSP
8~~4 ) , BKKSD )
)
~~~7 ) , BKKS
8~-~8), BXXSDD )
843 )

(C( 851 ) , I3XXSP

CC ( 852) ,BXXSD )
)
(CC 855 ), BXXS
C C C 9~~1) , BJSSD ) , (C( 93~4 ) , BJ SS
(CC 935 ) , BKSSD ) , (C( 938),BKSS
( C ( 953) ,SNP 1
(CC 950) ,SN P2
) ,
(CC 957),SN Q2
), (C( 960) ,SNQ1
) , (C( 967),SN R1
(CC 96L4 ),SNR2
(C( ~~ 7 l ) , B P C 2
(C(
) ,
9P 4 ) , B P C 1
(C(903) , H1 3P) , (C(904 ),t-11 3M)

) ,(c (

956) ,SNPO

), (C( 963) ,SN ~~O


),(C( 970) ,SNRO
) , (C(
977),BPCO

)
)

)
)

( C ( 9 0 5 ) , H 2 i 4 P ) , ( C ( 9 0 6 ) , H2 L I M)

(CC 9O7) , CDR FT 1) , (C(908) ,CDRFT2 )


( C ( 9 0 9)

CDRFTY )

(C (9814 ), CDR FTX)


(CC 1 6 ib ) , ANGX )
( C ( 7 B ) , B D R F T D ) , ( C ( 9 8 l ) , BDR F T )
~~
(C(965) , NLM T 1 ) , (C(986) ,NLMT2)
( C ( 9 8 7 ) , B J J S S S ) , ( C ( 9 8 8 ) , BKKSSS )
(C (989) , BXXSSS )
( C ( 9 9 0 ) , B J J S S L ) , ( C ( 9 9 1 ) , BKKSSL )
C ( 5 ~~O) , BJSDD) , (C(531 4) , B J S D ) , ( C ( 5 3 7 ) , B J S )

C C C ~~ja) ,BKSDD) , (C(5 L~2 ) ,BKSD) , (C(5 145) ,BKS)


E Q U I V A L E N C E (C(546) , BXSDD) , (C(550) ,BXSD ) , (C(553) ,BXS )
EQ U I V A L E N C E ( C ( 5 j 3 ) , C J S D ) , (C(5 L 11) ,CKSD ) ,(C(549) , CXSD)
E Q U I V A L E N E E (C (9142) ,WL2 )
E Q U IV A L E N L:~ . (C(943) , D J 2 )
EQUIVAL ENCE ( C ( 9 4 4 ) , wJ2 )

EQ U IVALE N E CC ( 9 L4 5 ) , DX2 )
~
ft ) ,wX2)
E Q U l .A L i WC~ ( C (
~~

C~

EQUI VALENCE
~OU T

P Uj A

E Q U I V A L L ~~CL

CMO T HER UUTPU ~3


EQ U I V A L L N C E
EQU i V A L E N C E
EQU I V A L E N C E
EQUIVAL ENCE
EQUIVALENCE

L
_

(C(2965) , V A R ( l ) )

( CC

856 ) ,N D E L T C ( l ) )

( C ( 8 8 0 ) ,BPHIS
( C ( 5 1 8 ) ,B 13 S S )
( C ( 5 H ) , 02 14SS )
( C C 8 d 1) ,BJJ
( C C d 82 ) ,i3K K

---

- - ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

EQUIVALENCE

~~~~~~~~~~~~~~~~~~~~~~~~~

- --~

8 3 3) , I3XXSS )

CC (

E Q U I V A L E N C E (CC 884 ),BJJSS )


EQ U I V A L E N C E (C( 8~~5) , BK KSS )
E Q U I V A L E N C E (CC 886),BTHTS )
EQUIVALENCE CCC d o 7 ) ,BPSIS )
C * *A D D I T I O N A L INPUT FOR NEW CONTROL LAW
)
EQUIVALENCE ( C ( 2 0 0 0 ) ,T
E Q U I V A L E N C E ( C C 3 5 0 ) ,BTHT
E Q J V A L E N C E ( C C j ) 1) , B P SI
)
)
E Q U I V A L E N C E ( C ( 1 6 1 5 ) ,RXE
)
E Q U I V A L E N C E ( C ( 1 6 0 3 ) , VX E
)
EQU IVALENCE ( C ( 1 6 1 9 ) ,RYE
)
(
C
(
16
2
3
)
EQUIVALENCE
,RZE
EQUIVALENCE ( C ( l l l ) ,VZE
E Q U iV A L E N C E C C C 1 6 O 7 ) ,VYE
)
E Q U I V A L E N C E ( C C 1 6 5 1 ) ,RTXE
(
C
(
1
6
6
0
)
EQUIVALENCE
,VTXE
)
E Q U I V A L E N C E ( C ( 1 6 5 5 ) , R TYE
)
E Q U I V A L E N C E ( C ( l 6 6 1 ) ,VTYE
E Q U I V A L E N C E ( C ( 2 9 l 6 ) ,GA M M A )
E Q U I V A L E N C E C C ( 2 9 1 7 ) ,B E T A )
EQUIVALENCE ( C ( 2 ~j l 8 ) , K A )
E Q U IV A L E NC E ( C ( 2 9 1 9 ) ,K 1)
E Q U I V A L E N C E ( C ( 9 2 O ) , l)
~~
~~
E Q U I V A L E N C E (C (2921) ,GPIT)
EQ U IV A L E N C E (C (2922) ,G YA W )
C * *A U D IT I O N A L OUTPUTS
EQ U I V A L E N C E CC(29014 ),X REL
)
EQ U IVAL E NC E. (C(2905) ,XRELD )
)
E .AJ IV A L ENCE (CC2;02) ,Y R EL
EQUI VALENC E. (C(2- ~O3 ) , YR ELD )
E Q U I V A L E N C E (C (2906 ), TF)
C
N E A L K A ,~~ 1
DA T A B / . 7 O 1 l /
C

DATA

P I D A/ 1 . 5 / 0 / U b/
B E i A 4 * 2 ~~~~~ A M M A * B E T A* K A * * 2 * ( T F _ T ) + B E T A * G * * 2 * ( T F _ T ) * * 3 / 3 . +

U E [. TA ( f )

(GA ~ N1~ * * * 2 * KA * * 2 * C rr~~T)~~)/l2 .


C r H r~~~r ) ~~~_ B E T A * ~~\ M M A * K A + ( G A M M A * K A * G * * 2 * ( T F _ T ) * * 3 ) / 6 . ) / D E L T A ( T )

C X r ~r . L ( T ) (_ B E T A * 3 * ( T F _ T ) _ ( G * G A M M A * K A * * 2 * ( T F _ T ) * * 2 ) / 2 . ) / D E L T A ( T )
_
.3
CX R: -.Lu(T )
~ ~. 3 E fA * * (TF_ T )**2_ C *GAMMA* KA* *2 * (TF_ T) **3/2.)/DELTA (T )

C JUIDA~-.:E CI~ ~AL Si4 P 1NG


C 1G U I A .z. S IT C r L N i
~~ ~~
~~
~

QSL)
f)
Lk
~ ~~

4N 2 *(4NU* (WLAMQ
3N C~~LQ 3 U / W L

IF (Ti< R Z.

T.O.

0.

W RC

WLRS )

QDN

~
.~ . 0.~ .AND.
T.GT.
.GT . 0 . ) GO TO 5
+

+ QBIAS

2 . *W L R S D
+ GBIAS

+ RBIAS
T.GT.TDY) GO TO 4

.A N A .

L3.L 1 S

WLQS )

I F ( CA G E
4 IF ( T K R Y
LHSD D
~

WLQS ) 2. *WLQSD)
)

TDY) WQC

0.

RBIA 3

I F ( C A G E . CT .
5 C UN r IN UE

0 .)

WRC

WLAMR

24

RBIAS

WLAMQ

+ QBIAS

+ GBIAS

--~~~~~

~ LQSSD

WN1 (WQC

WLQSS )

WL R SS )
WN1 * ( WR C
I F ( W N 1 .GT . 0 . ) GO TO 3
WL Q SS = W Q C
4LRSS
WRC
~
BL QSSD
W LQSS
WL RSSD

BLRSS D

C RA T t

Y RO

WLR SS

D Y NA M

ICS

AND

LI
MI TI NG

BDRFTD = ( C D R F T 1 *A N G X + C D R F T 2 )
BTH T S=B TH 2+B D R F T

BPSI S=_ B PS 1 +CDR FTY*I3DRFT


BPHIS =_ BP H 1 +CDRFT X * BD R FT
( W Q C O S D ( B P H 1)
B P H I S D = WP

SNR2

BP H S

W S 1*WS2*(GNS*SPSI_ SNRO)

BPHS S W P 1
BTHT S - SN Q 1

BTHTS
BP S IS
BPSIS
8 CONTINUE
BXX
BP H S
B T SS = BTHTS

BPS S

BPSIS

C * * * * * S P E C IA L CASE
I F( O PT C 1

. L E.

BL QSSD = Q B I A S
BLRSSD = 0 .
BDC
0.

C
C

_ W R * S I N D (_ B P H l ) )

*SIND (.8p 31)/COSD (..Bp 31)


BPi-I S = BPHISD / ~4 L X X 2 + BPHIS
I FC G N S .LE. 0 .) GO TO 8
SNP2
W SlWS2(GNS SPHI~~SN P0)
SN Q2
~ Sl*W S2* (GN S*STHT_ SNQO )

(WS 1 +WS2)SNP1
(WS 1 WS2)*SNQ 1
(WS1+WS2)*SN R1

SN R 1

PROGRAMMED
0 . )

GO

TO

FLIGHT

BT :O .
BP=O.
IF(T .UE .1. 0000 .A N D . T .LE .3 .2000)BDC= 5.
.AND. T.LE.6. -4000)BT= 5.
IF (T.GE.4 .2
IFCT.GE .7.6 000 .AND. T .LE .9 .9000)BDC=5 .
IF (T.GE. 1 1 . 0 00 .AND. T.LE. 13 . 100)BP :5.
IF(T .ur. .15.200 .AND . T .LE .17 .300 )BDC
~.
IF(T .GE .18 .320 .AND. T.LE .20.560)BT = 5 .
IF(T.GE.2 1 .545 .AND . T .LE .23 .675 )BDC= 5 .
IF (T .GE .24.((O .AND. T.LE .26 .910 )BP=5 .
C(520)~~ B PCO s -bPB T
b PC2 18 . 18*20 . o(C BDC BPCO ) ( 18. 18+20.57 )*BPC 1
BXX
UXX + BPCO
13 rS3~ BTSS-BT
BP SS~ BPSS B P

9 CO NTINUE
C ***NE~ O P T I M A L C ONI R O L FOR P I T C H , YAW
1/W i
X REL = R TX E RXE
Y R E L = R T Y E R ~~E
X REL D zV T X E V X E
Y R ELD ~~V T Y E V Y E
C ****E 3TI ~1ATE OF TIME- ID-GO
T3 x = A 8 S ( X R E L / x R E L D )
71 TGUZ ~~A B S ( R i E / V Z E )
88 C O N T I N U E
rn

.--

SIGNALS

_ _ ._ _ _ __ t _._ __ _~ _
~
~

ri

- --- - - - - ~~

~~~~~~~~~~~~~~~~~~~~~~~

. -~~~ _ L~~~~~~~~~~~~ rT ~~~r--

~~~~~~~~~~

- ~~~~~~~~

- - ~~~~~~~~ - -

- -

~~ ~~ ~

TGO : A MI N 1 (TGOX ,TGOZ )

TF = T + TGO

BTHTR=BTHT/57 .29578

BJS=CTHTA(T) (P102+BTHTR)+CXREL (T) XREL+CXRELD (T)XRELD


BJS=-BJS
B K S :0 .

BJJ = BKS -E3 JS

B K K= B K S B J S

B J J S S S = G P IT * B J J

BKKSSS=G YAW BKK

C
C G UIDA NCE SIGNAL SHAPING AND LIMITING

BXXSD

BXXSP

BXXSDD = WXX* (WXX* (BXX


BXSD=CXSD

BXXS )

2.DXXBXXSD )

BXXSS = GXX * ( ( BXX S DD + ( WLXX 1+WLXX2 ) * BXXSD )/ (WLXX 1* WLXX2 ) + BXXS )


BXXSSS = BXS
CS IGNAL LIMITING
10

BJJSSL=BJJSSS
B KK S S L B~~KSSS

I F ( B J J S S L . G T . H i 3P ) B J J S S L: H

3P

1F (BJJSSL. LT . Hi 3M ) BJJSSL :H 1 3M

IF ( B K K S S L . G T . H 2 4 P ) B K K S S L = H 2 14 P
BKKSSL. LT . H2 ~4M) B K K S S L = H 2 4 M
B 13SS= BJJSSL
B24SS :BKKSSL

IF(

C
CCOM MANDS TO ACTUATORS

BD E LTC ( 1 )=B13S .. +BXXSSS


13DELTC(2 )zB2A 4SS+BXXSSS
BD E L T C ( 3 ) z B 1 3 S S B X X S S S
B D E L T C ( 4 ) :B 2 14 3 3 - I 3 X X S S S
RETURN

END

- ~~~~~~~~~~~

~~ UII~~

3 . 4 .2

___________________

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

E v a l u a t i o n of C o n t r o l Gains ~t lT , GY AL

Two new c o n s t a n t s , GPI T and GYAW , tru gains in the p itch


and yaw con tr o l channel s , were introduced in the control module Ci
to r e g u l a t e the amp litude of the tail fin command signal. As ran
be seen f r o m Figure 3. 5 , the signal is fed thr ough a l i m i t e r a f t e r
the multi p l ication by the g a i n .
By working w i t h f o u r s t a t e simulation , optimal values
were o b t a i n e d f o r t h e c o n t r o l p a r a m e t e r s , ~ and ~ (see s e c t i o n 3 . 3 ) .
Under the assumption that the fourstate simulation approximated the
6DOF s i m u l a t i o n , it should be possible to obtain a reasonable miss
d i s t a n c e and impact ang le w i t h the more comp lex simulation by using
t h e p r e d e t e r m i n e d o p t i m a l v a l u e s of t h e c o n t r ol p a r a m e t e r s . Several
r u n s w e r e made w i t h v a ry i n g values of the gains GPIT and CYAW . The
amplitude of the control si gnal had to be such t h a t the t a i l f i n s
were no t at the stops throug hout most of the flight , but would be
hard over at the end to achieve the desired attitude ang le at impact.
A m i s s d i s t a n c e of 18 f e e t and an a t t i t ud e impact ang le of 37 degrees
was achieved w i t h t h e f o l l o w i ng gain v a l u e s :
GPIT = GYAW = .000243
3. 4 .3

C o n t r o l P a r a m e t e r Improvement

The v a l u e s of the contro l param eters y and


~ wer e altered
(
using the Hooke Jeeves Al gor ithm) to improve performance in the
600F s i m u l a t i o n . R at h e r t h a n a t t e m p t to incorporate the 6DOF simu
latlon as a function subprogram to evaluate F(Y ,E), it was decided
to b y p a s s t h e a n t i c ipated programming d i f f i c u l t y by j u s t r u n n i n g t h e
simul at ion any time y or B was changed and F(y,B) needed to be reevaluated for the HJ Al gorithm. Al though the iterations thus
perforr -ied were slow , overall i t seemed to b e t h e mo s t s t raigh t forward
way to proceed .

3.5

S e n s i t i v i ty Anal y s i s

3.5.1

S e n s i t i v i t y f o r the FourState Simulation

Wh en the 3state control law was used on the 6DOF simulat ion ,
a small change in l o w e r i ng t h e missile v e l o c i t y f r o m 1095 f t/ s e c to
1090 f t/ s e c r e s u l t e d in t h e miss d i s t a n c e j umping to 264 . Clearl y ,
t h e q u e s t i o n of s e ns i t i v i t y had to be i n v e s t i g a t e d f i r s t w i t h t h e
s i m p l e r f o u r s t a t e simulation and then with the 6DOF simulation .
I t was n o t i c e d t h a t t h e i n i t i a l c o n d i t i o n s f o r t h e t e r m i n a l
g u i d a n c e p h a s e g iven b y E q u a t i o n (3.19) had been poorl y chosen in t h e
sense that the target was out of the field of view , a n d so an al t erna t e
set was used :

1000

5000

89

(3.29)

The v e l o c i ty was left the same at 1095 ft/sec and the initial attitude
0
ang le was increased to 89 since the missile would be flying approximately
horizontal after the initial prepro grammed p itch maneuver.

_
_
_
_

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

_ _ _ _ _ _ _ _ _ _ _ _ _ _

- z-

_ _ _ _ _

rs
0

~~
0

(J) &)

(1

(fl (f)

>

cs~~~~

--)

1)
~

l..i

C
0

0
4J

V 1
~

:
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

LL1
L
L LJ
L~]~
~ ~

_ ___ __ :L_ _ ._ _
~~ ~
__

- :

:~~~~-~~~~ -

:~~

~~~~~~~~~~~~~~~~~~~~~~~

~~~~~

--

~~~~~~ r~~~~~~~~~ :

V a r y ing each of t h e i n i t i a l c o n d i t i o n s
H , 0 , and
~~
~ d -generated
V~ (t ) separately, Table s 3 .3a , 3.3b , 3.3c , 3.3d were1 0
using the fourstate simulat ion. As can be seen from them , decreases
in t he acqu is iti on ra n ge are t olerable , but when it is increased to
6500 the simulation becomes unstable. The heig h t can be increased
w i t h o u t d e t r i m e n t , but when lowered to 800 instability occurs. The
ini t ial a t t i t u d e ang le can onl y be decreas ed b y two degrees. Performance is most sensitive to decreases in velocitya decrease of
5 ft/sec of velocity from 1095 ft/sec to 1090 ft/sec causes instability .
Recall that the performance obtained with the 6DOF simulation consid
erabl y worsened as well when the velocity dropped 5 ft/sec.
The above information suggests that a worst case scenario

abou t which t o o p t i m i z e the c o n t r o l p a r a m e t e r s mi gh t be:


=

1000

~d

5000

, 0 =

87.5 ,

1080 ft/sec

(3.30)

An initial attitude angle of 87.5 would hopefull y allow a larger


window for acceptable initial attitude angles. A lower velocity was
chosen as it would be more demanding on the control system . Using the
Hooke Jeeves Al g o r i t h m , t h e o p t i m a l control parameters turned out to
be
y

5 4 6 0.9 9 2

.l8929lE06

(3.31)

w i t h a r e s u l t i n g p e r f o rmance of
Y (t )
d f

.915

e (t )
~

45 .012 .

(3.32)

The f o u r s t a t e s i m u l a t i o n was ran in d o u b l e p r e c i s i o n to m i n i m iz e t h e


e f f e c t of r o u n d o f f e r r o r on the r e s u l t s . V a r y i n g i n i t i a l c o n d i t i o n s
one at a t i m e , Tables 3. 4a , 3. 4h , 3 .4 c , 3.4d were produced .

Th e chang es in p erforman c e when acquisi ti o n range , heigh t , or


initial attitude angle are altered agree qualitativel y w ith the previous
results. Y c an be decreased and H can b e increased wi th p e r f o r m ance
t
d
remaining acceptable. The ingerval of acceptable initial attitude
0
ang les would be from about 86 to 89.5 . Tl~e one very curious coqiplete
r e v e r s a l of s e n s i t i v i ty is in t he var iable .Yd . Previousl y, whe n
lessened , pe rformance worsened; now , when
in c reases , performance
~ d fourstate simulation
wor sens. With these control parame ters , the
pr edicts that the missile should not acquire the target until i t s v e l o c i t y
decreases below 1080 ft/sec.
3.5.2

Sensitivity with the 6DOF Simulation

The c h o i c e of GPIT , GYAW for the 6DOF simulation imp lemen t a ti on


turned out to be a very delica te one for the new set of initial condi
tions . Two comp le t el y different types of trajectories were observed
t o r v a r i o u s v a l u e s of GPIT , depending on w h e t h e r it was above or below
some value close to .0004975. When GP1T was less , the missile would
ove r f l y the t a r g e t and the s i m u l a t i o n would t e r m i n a t e when the maximum
fli ght time parameter was exceeded. For GPIT greater than this number ,

Table 3.3.

Y d (t f)
(ft)

0(t f)45
(deg)

I.C.
Var iable

I.C.
Value

3.3a

Y d (tO)
(ft)

3000.
3500.

1.61
4.63

5.31
0.24

4500.

3.09

0.48

4000.
5000.
5500.
6000.

6500.
H t ( t 0)
( ft )

800.
900.
l000.~

1100.
1200.

1300.
1400.
1500.
1600.

3.3c

( t 0 )
(deg)

8 5.

86.
87.
88.

89.*
90.

3.3d

V M (tO )

(ft/sec)

T!
~ III~

Performance for Var ious Initial Conditions (Baseline I)

Table

3.3b

1090.
1095.

1100.
1105.

5.03

2.83

2.80
17.85
4.83

0.00
l27 .tO

Uns t able

Uns t able

3.05
2.80

5.13

2.39

3.35
0.00

1.04

2.29

6.67

2.79

11.78

4.77
4.37
4.97

0.35
0.58
4.12

Unstable
5.95
58.70
1.49
2.99
4.07
3.48

2.80
2.47

0.00
2.64

Unstable
2.80
0.00

4.14
5.65

0.32
3.79

Baseline I
(denoted by * )
H (t 0) = 1000
~

0(t0)= 89
Y d (tO ) = 5000
=

1095 f t / sec

-y

5535.43

.18939E 6

V M (t O )

_ _ _ _ _

_ _ _ _ _

-~

---

~~~~

-. -

--- ~~-~~~~~

Table 3.4. Performance for V a r i o u s I n i t i a l C o n d i t i o n s


(Baseline II)

Table

I.C.
Variable

i.C.
Value

Y d (tf)

3.4a

Y (t 0)
~ f t)

3000.
3500.

2.41
4 .13

3.4b

H t (t0)
( ft )

4000.
4500.
5000. *
55 00.
6000.
700.
800.
900.
1000.
~
1100.
1200.

1300 .
1400.
1500.

1600.

3 .4 c

0 ( t 0)
(deg)

85.5
8 6 .5
k
87.5
88.5
89.5
9 0 . 5

3 .4d

VM ( t O )
(ft/sec)

1065.
1070.

1075.
1080. *

1085.

Cft)

4 .96
12 .9 8
0 .92
16 .24

U( t f)45
(deg)

4.24

Unstable

Uns table

-0.07
0.69
6.57
0.01
0.5 15

17 .45
124 .35
0.92
4 .3 3

33.40
10.97
0.01
0.78

5.57

4.18

3.84
3.30
3 .9
1.7 2

0.81
3.86
0.08
5.12

Unstable
5 .22
0.54
0 .92
0.01
1.23
9.79
5.49
0.17

Uns t able
4 .82
5.84

0 .46
0 .92

Unstable

0.05
0.87

2.05
0.01

Baseline II
(denoted by 5 * )

H (t 0)= 1000
~
9(t 0) = 87.5
Y d (t O )

5000

V M ( t O)

1080 ft/sec

-y

5460.99

.l8929E 6

~~~~~~~~~~~~~~~~~

_____________

~~~- ~~ ~~~~~~~~~~~~~

~~~~~~~~~~~~~

~~~~~~~~~

--

~~~

~~~~~~~~~~~~~~~~~~~~~~~~~~

--

the missile would impact considerably shor t of the target. Given the
seven di git accuracy of the computer , it was not possible to obtain
reasonable performance. See Table 4.1 for a closer examination of
this behavior. (The control param eters y and B were set at 5460.99
and .l8929lE 6 , respect ively.)
Table 3.5 . P e r f o r m a n c e Dependence on GPIT
RXE(ft)

GPIT

.000 5
.000 499
.000498

.000497
.000495

RZE(ft)

0 (t

1.1
.8
.4

1210.
1144.
945.

) (deg)

-28.6
27.5
46.1

5 67.
591.

530.
529.

t>t

t>t

The control parameters were then altered in the hope that


performance could be improved and sensitivity reduced. With

GPIT

(3.33)

.497843

and the missile impacting 510.6 sho r t of the t a r g e t , several


runs we r e made with various values of the control parameters y, 8.
As can be seen from Table 4.2 , small changes can resul t in a
t o t a l ly d i f f e r e n t t r a j e c t o r y .

Table 3.6

.P e r f o r m a n c e

y
5460.996
5461.
5460.996
5460.996

.l895E6
.l895E

. l9 l OE 6
.1920E 6

Dependence on y and B

RXE(ft)

RZE(ft )

510.6
536.
118.
536.

380.
178.
408.

8 ( t ) (deg)
f

.5

62

t tmax
~
27. 9
t tmax
~

Desp it e the small m iss dis t ance and excellen t impac t angle
achieved w i t h t h e f o u r s t a t e s i m u l a t i o n , it proved to be impossible
to achieve reasonable performance through an appropriate choice of
t h e g a i n p laced in t h e p i t c h and yaw channels. Since vary ing t he
control para m e te rs was uns uc cessf u l as well , it was decided t o re t urn
t o the alternate set of initial conditions for which the missile was
14 above cro wd w i t h an attitude ang le of 38.1 0 below the horizon .

--

----

- - - -- ---

~~--

----- ~~~~~~~ -

32

- - -- -

- --- ~~ - ,- -

---

~~-

______________________________

_____

--

----- .=-
~~
~~~~~ -.~~~~~~

=-~~~~ _ ~~~~~~~~

=--

~~~~~~~

Initially , the control parameters were


y

5525.508

, B

.l9E 6

CPIT

.243E3

(3.34)

Using the Hooke Jeeves Algorithm , the miss distance was reduced
to 10.4 with an attitude ang le of 38 .2 whe n t he con t rol parameters were 5525.45 and .l6OE 6 , respec tivel y.
Different runs were made on the 6DOF simulation with various
initial conditions. Table 3.7 contains the results~ The f i r s t line
represents the baseline case.
Table 3.7.
Y (t )
d O

H (t )
0

5000.

1500.

4500 .
5500.

1500.
1 500.

5000.
5000.
5000.
5000.

1400.
1600.
1500.
1500.

5000.

1500.

5000.

6DOF S imulation Runs for Various Initial Conditions


0(t )
0

V (t )
M O

5.
5.

1090.856
1090.856
1090.856
1090.856
1090.856

.3
1.6
1065.3
690.7
116, 7

1090.856

495.1

5 ,

5.
5.

4.

1090.856

5.
5.

1085.856
1095.856

6.

1500.

RXE

RZE

10.4
68 .9
1.2
.5
1.5

.7

150.4

.9
448.7

127.1
1.2

0( t )
f
38.1
44.
77.5
4 7 . 2
89.7

32.4

.9

41.8

35.5
43.4

~~~

As can be seen from Table 3.7 , performance is very sensitive


to any change in the i n i t i a l condition . Such sensitivity was not
observed with the fourstate simulation . This is probabl y due t o the
fact that in the simpler simulation there was no hard constraint
8n
t h e c o n t r o l l e r u ; whereas , in t he 6 00F s im u la t ion a bound of 10
existed for tail fin disp lacemen t.
Figures 3.6 and 3.7 represent the baseline trajectory.

A--

__________________

~ =s-~ ~ ~~~~~

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

- -~-:

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

--~ ~=~-------

~~~~~~~~~~~~~~~~~~~~~~~~

I
I
I

* **

* **

* *S
*

*
*

* **
* **

* 5
U

* **
*

a..
a-.

*
*

* *5

0
*
*

S
*

+0

rl
~

S
3
5 *

*
*

0~

(4
0

0
I
~~
+ (V~

c,J

*
*

- ---i --

Y ~

*
*

l:r~
+0

* 5

*
*
5 *

I:

~~~~~~~~~~~~~

_
_

-i - -

~~~ r~~~~~~

-~~- -~~

~~~~~

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

5 *

5 - 5

5 *

* *

5 * 5

-4
4-

5 *
* 5
5 *

* US
* 5
0*

5 *
*
U

5 *
*
5 *

* 5 *

5 *

* **

5
*
*

*
*

*
*
*

I
I
1
I
I
I
I

0
0

*
* 5

5 5
* 5
*

5 * 5

*
*
5 *

**
**

S .

I
I
I

~~
I +0

.
-

0
0

O
0

U.
0

N-

.
-

,
-

.
-

Lz3

a..

~~

3
~ ~

5 *
I

* 5
* 5 *
N-

+ 0
0

~
I
I
0
I
0
I
I
~~
1 + 0
I
0

I
I
0
I
I
I
0
I
0
I
I
~~
1 + 0
(\J
I
I
I
-0

I
I
I
0
I
0
I
(..
I
~
1 + 0
I
0
I
I
Lt 1
I
I
I
0

*
*
*

** *

0
0

U_ I
5 *
5 *
* 5
5 *
5 *
*
* 5 *
5 *

a-.

0
I
0
I
I
L.~
1 + 0

*
S
*
*

~~
+0
C\J

a-

r1
~

0
0

I
I
~~
1 + 0
I
0
I

5*

Li ;:

~~
+ 0
~~

*
*
*

0*
*
*

--

1
I
4
I
I
I

I
1
I
I
I
I
I
I

5 *
*

--

+~~~0

I
I
I
I

I
I
I
I
I
I
I
I

** * *
** *

---

N-

-, ~ r

~~~~

r
~~~~~~~ -~~c

--

Chap ter IV
CONTROL LAW IMPLEMENTATION
4.1

PRACTiCAL CASE

Disadvantages of the ThreeState Controller

Only one of the three states used to control the missile


Bo th range
and range
~d
ra te
are not available as there is no radar or other
such measur ing
devices present. The laser guided missile can detect the reflected
laser beam and would have A (the time derivative of the lineof sight
angle A ) approx imated b y the seeker. This alternate set of state
variables (A ,A ,e) would seem to be more attractive than the set
because of the possibility of using actual physical measurements.

can be me as u red , the attitude ang le 0.

4.2

A Controller Using A , A

Figure 4.1 gives t h e rela ti onshi p be tween the various state


variables .

_II

MISSILE

~~~~~~~~~~~~~~~~~ T

~~MU SSI U

TARGET

F ig u re 4 .1.

Missile

Target Geometry

Using F ig u re 4 ,1 , the follow ing equation is obtained.


=

H t t an (8+ X )

(4.1)

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

T~~~~~~~ ~---

- - _-

~~~~

~~~~~~~~~~

~~~~~~~~~~~~

---

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

~~~~~~~~~~~~

-~~~~

T)ifferentiating,
=

2
.
H sec ( + ) ( + ) + t a n ( 9 + A ) H
~
~
~ ~ ~

(4.2)

Substituting , the control law becomes of the form

f ( ~~,A

u=

,G ,8 h~~,H~ )

4 .3)

where
f ( A ,~~,0 ,~~,H ,H )

2
L H t a n (+ ) + c (H sec ( + ) (O+A )
1 ~
2 ~
~
+

4.3

t a n ( +~ ) H )

(4.4)

+ c 0

the A , \ , 0 C o n t r o l l e r

i m p l e me n t a t i o n o t

i n i t i a l l y, all s t a t es were assumed to be known . Under


this assumption , the following performance was obtained:
=

****

P (t )=
f

****

(4.5)

As f o r i m p l e m e n t i n g t h e two p a i r s of v a r i a b l e s (A , and
one of ea ch pair will be measured phys i call y , and the other member
will be appr ox im a t ed by so m e appr o pr ia te s ig n a l pro cessingeither
an integration or differentiation network.
~

A s for the last pair , neither H~ nor H t can be meas ur ed


directly.
rt
possible, Oowever , to approximate H , if the a t t i t u d e
~
and v e l o ci t \ of t h e m i s s ile are k n own , by us ing

_
%~

V ~~~cos 0

( 4 . 6)

If the initia l hei ght is known , Equation (4.6) could be used to


u p d a t e t h e , s t i m a t e of H . The o n l y way t h a t t h i s m i g h t be done
~~
would be in the osign (If tne preprogrammed pitch maneuver.
Th ere seems to he no way to avoid the need for range and
ra:~ t~ rate information. Equation (4.6) seems to come as close as
possible to drawing upon variables that can be measured. The velocity
V~1 is not rne, o~i red but does behave nicel y thr oug hout the flight , almos t
a l inear function of t . Perhaps a profile of V,1 co u ld be s t ored f o r
o r generated during the flight.

N ot e :

LL

~~~

app earing in E q u a t i o n ( 4 . 5 ) have been i n s e r t e d


m- m n - o the eode h a d some errors in it , These numbers should
be close to those previousl y obtained since the two formu
lations are m athematicall y equ ivalent. This approach was
not followed up due to the sensitivit y problem discussed
in Chapt er III . It was felt that it was importan t to
ou tl ine the approach to be used .

1h

~~~~~ *~~~

~~~~~~~
_

~~~~~~
-

~~~~~~
--

~~~~~~~~~
-- -

~~~~~~~~
-

4.4

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

~~~~~

fl.__

__I

~~

Estimation of Timetogo

The problem of e s t im a t i n g timetogo that is needed for


the control law remains even if all other problem s are resolved . A
fixed estimate could be used , but it would have to vary depending on
the initial conditions.
A second approach m ig h t be t o u se H and t b ut since these
~
two variables are being estimated , this me tho~ probabl
y wo u ld no t f a r e
too well.
A third alternate , perhaps , is to use the seeker, If the
intensity of the reflected laser beam can be measured , alo n g wi t h
the rate at which the intensity is varying , then an estimate of
t imetogo can be obtained ,

Intensity I is inversel y proportional to the square of the


d i s t a n c e f r o m the m i s s i l e to the target , the slan t range R. Thus

I =

k/R 2

(4.7)

Differentiating with r e s p e c t to t ime y ields ,


dI/d t

-2kR

3
dR/dt

(4.8)

He n ce ,
timetogo

d R/ dt

dI/ d t

(4.9)

- --~
~~~~~~

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

~~

--i --

~~

--

Chapter V
CONCLUSIONS ANt) RE C0 ~~1ENDATI0NS FOR FUTURE STUDY
Even assuming p e r f e c t knowled ge of t h e s t a t e s d d ~~
~
t h e best p e r f o r m a n c e that could be o b t a i n e d w i t h the 6DOF~ simulation
was

RZE

10.4

0 (t~~) =

(5.1)

38.1 .

When one considers tha t the center of mass of the t a r g e t is a b o u t


5 above gro und , the performance is very close to satisf y ing t h e
c r i t e r i a f o r success . When actual measurements are used , one can
only expect perform ance to deteriorate.
It must be L pt in mind that the above was assuming t~at
all states could be measured exactly. When actual measurements are
used for some of the variables and estimates for others , t h e performance
If the threestate controller is to work , these e r r o r s
~iili d e g r a d e.
should be q u i t e small.
The o t h e r o b s e r v a t i o n t h a t needs t o be made is that t h e
p e r f o r m a n c e is e x t r e m e l y s e n s i t i v e to changes in the initial
conditions when the target is acquired. Even if t h e miss d i s t a n c e
and attitude ang le at impact were zero , this sensitivity would d e f e a t
onv practical imp lementation of the control law. Small changes in
the control parameters also resulted in large changes in performance ;
that is , the assumption of continuity that was made in determ ining
th e control parameters is not valid for the actual system as
r e p r e s e n t e d by t h e 6DOF s i m u l a t i o n .
The o n l y c o n c l u s i o n t h a t can be r e a c h e d is that the three
s t a t e c o n t r o l l e r c a n n o t be s a t i s f a c t o r i ly imp lemented in t h e 6DOF
s i m u l a t i o n . The main exp l a n a t i o n t h e a u t h o r s f e e l is t h a t t h e r e is
a hard c o n s t r a i n t on the controller in the DOF simulation , but not
in t h e f o u r s t a t e s i m u l a t i o n . This , more t h a n t h e a s s u m p t i o n s of
neg 1ec ~ing the ang l e of a t t a c k , or no l a g in t h e a u t o p ilot , d i c t a t e s
that the imp lernLrit ation will not prove to be s a t i s f a c t o r y . The
c o n t r o l p a r a m e t e r s can be a d j u s t e d so that a successful flight is
a c h i e v e d f o r m o s t seti- of i n i t i a l c o n d i t i o n s , but the extreme sensitivity
makes t h i s imp l e m e n t a t i o n of no v a l u e .
The auth ors feel that it would be worthwhile to reformulate
and solve t h e c o n t r o l p r o b l e m so t h a t a h a r d c o n s t r a i n t is p r e s e n t on
the controLie r.
t-u ch a controller (which will be of the bangbang
v a r i e t y ) s hou i h be imp l e m e n t e d , assum ing again that all 3tates can
be ncasured , to verif y wh ether or not the presence of the hard constrain t on the t a i l fin deflection is the main cul pr it for the
sen sitivity problem. If successfully implemented , then the problem
of pra cti ca l imp lement ation of .;uchi a control law could be examined .

--

_
_
~~ ~~~ :L~~~

~~~~~~~~~~~~~~~~~

.-

~~~~~~~~~~~~~~~~~~~~

- --

~~~~~~~~ -

~~~~~~~~~

Chap ter VI

SIMPLIFIED 6DOF SIMULATION


This listing is comp lete except for the omission of sub

ro ut ines COSD , SIND , ATAND which are the same as the FORTRAN library

subroutines but work with degrees instead of radians.

The old subroutine DUMMY has been relabeled TIMEV . TIMEV


was placed as an en t ry poin t in DIJMMY bu t the presence of argumen t s
caused a syntax error which was most easil y correc t ed by renam ing
subro ut ine DUMMY .
Since the altered simulation does not have the random
error disturbance capability, the main program could be shortened
b y omi tt in g t he sec ond half which is th e s t a t is tic al analys is par t
for Monte Carlo sets.

__________________________

- ~~~~~~ _

~~~~~~ - -

~~~~~~~~~

- -~~~~~~~ ~~~

~~

: i
:i-:;
:
-- -~~~~~~~~~~~ ._ ~_~~~~~~
~~

C*****DIMODS TO BE USED W ITH F O R T R A N


C
COMMON

REAL

AMRK

~~~~~~~~~~~

.-

--- - - - - - - - - - - - -

~~~ -

~~

ROUTI NE

INTEGRATION

C(3830) ,G R A P H (3O O ,~ 4)

KA ,K 1

C OM M O N / C E PASS/X (1 0 0) , Y(

10 0 )

) , ( C ( 2 b 6 ~4 ) , D E R ( 1 ) ) ,
( C ( 2 6 6 3 ) ,HMAX
) ,
(C(2662) , H:-~IN
C
(C(2561 ),N
), ( C ( 2 5 6 2 ) , I P L ( 1 ) ) , (C(2965) , VA R (1)) ,
) , ( C ( 2 0 1 0 ) ,STEP
),
( C ( 2 (JJ ) , T
), ( C ( 2 0 1 1 ) , KSTEP
C
( C ( 2 0 1 2 ) , LSTEP ), (C(2008) ,PLOTNO ) , ( C ( 2 0 0 9) , NO PLOT ),
C
C
(C(2023) ,O POINT) , (C(2025) ,TIME (1)) , ( C ( 2 3 2 5 ) , VLA I3 LE ( 1 , 1 ) ) ,
C
(C(3167) ,N OO U T ), ( C ( 2 0 2 2 ) , O P T N 1 O ) , ( C ( 2 0 0 6 ) , R E P P L T ) ,
C
(C(2d65) , EU (1)) , (C(2765) ,E L ( 1 ) ) , (C(2007) ,PTLESS )
(C(1971 ) ,RITE
), (C(1972) ,RKUTTA )
E QUIVA LENC E
) , (C (1975) ,NPT
) , (C(197 ~4 ) , NJ
E Q U I V A L E N C E ( C ( 1 j73) ,KASE
EQUIVALENCE (C (3512) , I S G C T ) , (C(3721) ,I T C T) , (C(3511) ,RNSTRT)
E Q U I V A L E N C E (C( 21) ,IbVNSW )
Ec ~j IVALENC E (C( 22) , IPLOT)
EQUIVALENCE ( C ( 2 3 ) ,X L A M B D )
E Q U I V A L E N C E (C( 2~4 ) , K S S I G )
EQ U I V A L E N C E (C( 25) , CEPS IG (1))
E Q U I V A L E N C E (C (3 00) , R M I S S )
U IMLNS ION RM ISST( lOO)
E Q U I V A L E N C E (C(1000) , RMI SST(l) )
L)
~.Q U IVALE NC E (C(301 ) ,
E~~U I V A L E N C E ( C (3 0 2) , RYE)
EQ U I\1ALE N CE(C( 1 9 ) , P S I Z E )
EQ U I V A L E N C E (C (303) , RZF)
( C ( ~~1) , L C E P )
E QUIVALENCE
IE3L)
EQ- ~ I V A L E N C E ( C ( 3 3 2 5 ) , N C A S E ) , ( C ( 6 2 5 ) ,
D
E
R
S
V
)
EQ U I V A L E N C E ( C ( 2 6 6 2 )
,

EQUIVALENCE

E Q U IV A L E N C E

E. QU IVA L ENCE
r

L-

IJ IV A Lt N C t .

~J IV A L E N C E

~EAL KSSIG

(C(3000) , V S D

( 1) )

(C(3 010) , V MtI AN (1))


(C(302 0), I M V N D X ( I ) )
(C(3 030), IM VCT )

L~~F EG ER C E P S I U
0ll4LN ~~lO N CE P S I G ( 6 )

D I M E N S I O N TIME(300 )
V L A D L E ( 2 , 15 )
DIME N SION
IPL (lOO)
,
DIM ENS iON
VA R( 1O 1 )
, EL (100)
DIME NSIO N i N V ~ DX ( 1 O) , V M E A N ( 1 0 ) , V S D ( l O )
)
E Q u I v A L ~ ;NcE ( C ( 1 b 3 0 ) , hN
)
E U U IV A L E N C E (C(1981 ) , R N T
,
PLOTN
I4)
IU
IV
ALIJ
CE
(C(
1982)
,
~
~
t Q U I V A L E N C E (C(1983) ,PLCTU2)
EL U I V A L E N C E (C( 19c ~4) , N PLOT )

~ , , Jtv AL E .~ CE

D E R (1 0 1 )

, EU (100)

(C(2020) ,L C O N V )

C ( 1 6 5 1 ) , R T X E ) , (C(1655) ,RTYE) , (C (1659) ,RTZE) ,


EQUIVA L E NC t
. ( C ( h t , 1 5) , lc X E ) , ( C ( l 6 1 9 ) ,R Y E ) , ( C ( 1 6 2 3 ) , RZ E )
~
I N T : - ; E R OP O I NT
~;~ rE~~h~ R - JPT
~ X t E k :~ A L A U X i U~3

~~~~~~~~~~~ * * * * d $ * * * 0 0 0 0t KU CHANGE* ****** *******


~
1)0 1 1 1 1 I
1 , 3d30
1 1 1 1 C(I )
0.0
C * * * * * h $ * * * * * * I * * * * * * * $ * * $ * * * W KU C I IA N G E * * * * * * * * * * * * * *
I o C 1 :0

__ i

~ ~~

41
-

-- --- _ ---- - -~~~~

- ---

-- --

--

-- -

---

- ----

--

-----

- - -- -

______

~~~~~~~~

---~~~~~

-- -
~~~~
~~

~~~~~~~~~

~~~~~~~

-~~~~~~--

- ~ ~~~~

~~~
___________
-

~~~~~~ ~~~~~~~

ISGCT~~1
ITCT :O

ITCT ~

MO JL~~-1

I TSNDX

THIS CALL TO SUBROUTINE


RANNUM IS TO PERMIT USE OF
DIFFERENT RANDOM NUMBER GENERATOR STARTERS ( I R N S T ) .
IF SUBROUTINE NORMAL .13 CALLED WITHOUT FIRST CALLING
RANNUM , THE RANDOM NUMBER SEQUENCE WILL ALWAYS
BE STARTED WITH THE SAME NUMBER ( ENTERED AS A DATA
STATEMENT IN SUBROUTINE NORMAL ) , WHICH WILL RESULT
IN THE SAME SEQUENCE ALWAYS BEING GENERATED.

C
C
C
C
C

L~~~~1
NP
0

CALL COUNTV

C
C
C
C

1000 CAL i Z E R O
~
1001 I F ( P L O T N O . L E . O . ) G O T O 7
I F ( R E P P L T . G T . 0. )GOTO(
REPPLT

IF ( R E P P L T . GT . 1 . 0 )

N PLOT :O
CALL OINPT 1

NOOUT

CO N TINUE
CALL R A N N U M ( 1 . ,RNSTRT ,D U M )
I F ( IS G C T .G T . 0 )

C (1976)
K A SE :
0

1.

CALL MCARLO

0.
1.
1.

USE NEW NO. 11 ,7 ( DISCARD OLD )


USE OLD PLUS THOSE ADDED
USE NEW NO. 7 ( DISCARD OLD )

(DUM , 1 , IDUb-1 )

LS T EP
STEP
N PLOT11:PLOTNLI

N P L O T2 ~~PL O TN2

NO PLOT PLOTNO
NO P LOT :N PL OT
1002 C A L L
SUBL 1
AUXI
1003 C A L L
10011

CALL

SUBL2

1005 CmJ ~4TINUE


D E R ( 1 0 1 ) DER ( 1)
~~
IF(DEH( 1 ) . G T . D E R S V ) D E R ( 1 ) :DERSV
C ( 1 -~ 7 6 Y

1006 C A L L
1007 N J ~~N 1

1.
AUXS U3

CALL A M R K ( A U X S U B )

C
C
C
C
C
_

1 008 CONTINUE
1009 CONTINUE
H DE LX ~ RTXE R XE
RDELY :R TYERYE
RDELZ R T Z E R Z E
I F( R D E L Z . L T . 0. . O R . RDELX .LT . 0 . )
CALL SUBL3
IF ( KSTEP .EQ. 1 ) GO TO 1007

_
_

_
_

LCONV :2

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

~~ L-~~~

~~~~~~~~~~~~~~~~~~~~~~~~~~~

T~~T~~

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

~~~~~~~~~~~~~~~~~~~~

C
C**************** *SAVE MISS DISTANCE FOR EACH RUN OF THE MONTE CARLO
C
R U N SET********************************************
C
C
C
I F ( L C E P . N E . 1) GO TO 20
LCEP:O
NP
NP + 1
X ( N P ) z RY F
Y ( I J P ) : HZF
R 1IS S T ( NP) : R M I S S
~
GO TO 21
20 CONTINUE
WRITE (6 ,805)
805 FOr MA T ( 1 H 0 ,/ / /)
~
W R I T E ( 6 ,22 ) L
22 F O R M A T ( 1 H O , 1 1 X ,2 b H* $ * $ * $ WARNING RUN NUMBER ,I3,
371-1 DID NOT I N T E R S E C T T A R G E T PLANE *$*$*$ ,/ ,
/ / , 26X ,38H$*$*THIS RUN DROPPED FROM DATA SET* $*$)

W RITE( , 805)
21 CONTINUE
L :L + 1

C
C
C~~

~~ C O N T R O L

PARAMETER

OUTPUT

P R I N T -466 , C ( 2 9 1 6 ) , C ( 2 9 1 7 ) , C (2918) ,C(2 919) ,C(2920) ,C(2921)


1166 F O R M A T ( 1 UO , G A M M A : , E15 .6 , 11X ,
BETA : , E15 .6 , / ,
KA : ,F 1 0 ,6 , 4X ,
Ki
,F1O .6 ,4X , W i : ,F10.6 ,/ ,
1

~
2
GP I T
,F 1O .6)
THOF F
C (350)115

~4RITE( 6 , 12 1 2) C (3 5 O ) , THOFF
1 ~~~ F O H M A T ( O T H E T A MISSILE IN DEGREES

C
C

C
C
C

,/ ,

VARIATION

******** *W KU
UML L

: ,E 1 5 .7
DESIRED 145 DEGREE IMPACT: ,E 15 . 7 ,/ / )

FROM

PLOTTING

RO U T I N E* ** * * * * * *

KPLOT

* * * * * * * * * E N D WKU

PLOTTING

ROUTINE*****

C A L L PR O C E A
i~ (O PTN1 O .GT . 0.) CALL

CALL

RESET

1F (L$TEP. EQ . 5. OR

C A L L T I M E V ( DELT )
WHITE (6 , 96 ) D E L T

DUMPO

.LSTEP .EL .7 .O R . N O P L O T . E Q . O ) G O T O S

96 F O X MA T ( 1 H
LE

-~~

~~

, 17iISTA HT PLOTTING AT , F111 .7)


PT :PT LESS

O PJ I NT :O P O IN i L E S S P T

CHLL P L O T 1 1 ( G R A P H ,O PO INT ,V LAB L E ,T IME , NPLOT L4 , NPLOT2 ,NOPLOT )


U, L PLOT2
~~
UAL. L PLOTN
CALL TIM EV(D ELT)

W R ITE ( 6 , 97

_ _

)DELT

_ _

_ _

~~~~~~~~~

- ~~~~~~~

- ---- ------~

~~~~

-~~~- -~~~~~~~~ -

-~~~~~~~~~~~ ~~~
-

~
-

~~~~~~~ ~~~~~~~~~~~~~~~~~~~~~~~

- m~~ ~~

97 FO~ MAT( 1H , 18HPLOTTING ENDED AT ,F i4.7)


5 GO TO (1000 , 1001 , 1 002 , 1003, 10 011 , 1005 , 1006 , 1007, 1008 , 1009, 1 0 1 0 ) ,
1 L~~TEP

10 10 C O N T I N U E

C
C
C~~* * * MEAN
C
C
C

VARIANCE

AND STANDARD

DEVIATION

W R I T E ( , 10 0 )
100 FORMAT( 1H1 , 13X ,36HMEAN V A R I A N C E AND STANDARD DEVIATION! ,
1 7X , 1OHCL OCAT ION ,9X ,8H M EAN VA R , 19X ,7HSTD DEV/)
DO

120 I:1 , IMVCT

ILOC

: IMVNDX ( I)
WR ITE(6 , 1 0 2) ILOC , VM EAN (I) , VSD( I )
102 F O R M A T ( 1 O X , I5 , 8X , E 1 5 . 8 , 1 1 X , E 1 5 . 8 )
120 CONTINUE

C
C*** *********MONTE C ARLO AND CEP LOGIC F O L L OW S * * * * * * * * * * * * * * * * * * * ** * * * * *
C
C
C

IF ( N P . LT. NCASE. AND. L . L E . ( NCA sE +5 ) ) W R I T E ( 6 ,8 0 7 )


TO 8
807 FORMAT(1 H1 ,3 (/) , 39H
THIS RUN ADDED DUE TO BREAKLOCK ,3(/))
J :0
.4HITE(6 , 800)
800 F O R M A T ( i H l , 96X , 1O1-I YM I SS
, 1OHZMISS
, 1OHMISS DIST I ,
1
6X , i 2 ~~
I F ( N P . L T . N C A S E .A N D . L . L E . ( N C A S E 5 ) ) G 0

3 )

00 801 I:1 , NP
J
J +1
WRI TE( 6 ,8O2) X( I) , Y ( I ) , RMI SST (I)
802 FORMAT( 6X , 1OH 1
, 10l-i 1
1 , 101-1 1
, 1OH 1
, 1OEI 1
2
, 11-l 1 ,F 9 . 5 ,2H 1 ,F 9 . 5 , 2H 1 ,F8.5 ,2H
WRIT E (6 , 8 0 3)
803 FORMAT( X , 123H
2

IF (J .GT.30 ) W R I T E

IF (J .GT .jO) J :0

, 1OH 1
, 1OI-l 1

, 1OH 1
, 1OH 1

1)

(
6 , 8 0 0)

801 CONTINUE
IF (IBL .LE. 0)GO TO 8011
L :L 1
X I d L~~IB L
XL :L
~A T 10: X I B L / XL

8- ~u

L.}

W R I T E ( 6 , 8 O 6 ) I B L ,L , R A T IO
111 1 , 1 5(/ ) , 1X , 1O ( 1 1 H* B R EA K L O C K * )

FORMAT(

/ , 1X , 1 1I1*B R EAKLOC K* , 3 X , 16HTHIS RUN SET H A D , 111, 2 S H B R E A K L O C K


.FLIGHTS OUT OF , 111,2 3 FIG IVING A PROPORTION O F . , F6 . 11,11X , 11H BREAKLOC
.K
/ , 1X , 9HBR EAK LOCK ,3X , 17HTHIS RUN SET HAD , I4 ,214HBREAKLOCK FL

b
-

--.

-- -

--

- --- ~

44

-- ~~---

- -~~

-
~ .

--- - - -

----

- -

--

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

.IGIITS OUT OF , I11 ,2 3 H G I VI N G A PROPORTION OF ,F 6 .4 ,L4X ,9HBREAKLOCK


/ , 1X , 11H *BR EAKLOCK* ,88X , 11H*BREAKLO C K* ,
/ , 1X , l i H * BR E A K L O C K * ,88X , 11H*BREAKLOCK* ,
/1X , iO( 1 1H *BR EA KLO C K* ) )
8u11
CONTINUE
CALL CEPAS (NP , IBV N SW ,IPL OT , XLAMBD ,KSS IG ,CEPSIG , PSIZE)
11668 COLI T INUE
CALL EXIT
STOP

END

ri:.
I-

-~~~ *
-~~ ,
~~~~~ -----~~~~~~~~~~ --

.. --~~~~ -.-~ ,- ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ -~ --~~~~~ -

--- - i
- -- ~~~
~~~ ~~
~~~~~

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

SUBROUTINE AMRK (A UXSUB )


COMMON C(3830)
DIMENSiON CSAV( iOO ) , IPL (iOO)
REAL K1 (100 ) , K2 (iOO ) , 1(3 (100) , K Ll (iO0)
T)
EQUIVALENCE (C(2000),
EQUIVALENCE (C (26614) , DELT)
NJ )
EQUIVALENCE (C(19711 ),
IPL
(1))
EQUIVALENCE (C(2562) ,
EQUIVALENCE (C(1976) , XNDRK)
XNDRK : 1 .
DO 1 I : 1 ,NJ
J : IPL (I)

C
C****STORE INITIAL VALUES
CSAV ( I)
C(J+3)
C
C*** COMPUTE K 1
K 1( I)
DELT*C(J )
1 C(J-s- 3) : C S A V ( I ) + . * K 1 ( I )
T :T + . 5 * D E L T

C ALL ~ UXSUB

C*** C~ .MPUTE

DO 2 I

K2

: 1 , NJ
IPL (I )
J
K2(I )
DELT*C(J)
CSAV( I ) + .5*K2 (I)
2 C (J~~3)
C A L L A U X SU B

.-

C
C*** COMPUTE K3
1 , NJ
DO 3 I
IPL (I)
J
DELT*C(J)
1K 3 (I)
C(J+3 ) : CSAV(I) + K3 (I)
T
T + 5*DELT
CALL AUXSUB
C
C*** COMPUTE 1 (11
DO 11 I : 1 , N J
J z I P L( I)
K4 (I)
DELT*C(J)
L4 C ( J 3 )
(.3AV(I) + ( K 1 ( I )
XNDRK : 1 .
CA LL AU X S U B

2. *( K 2 ( I )

RET U R N

END

-.

--- .

---.

K3(I) )

KLI(I))/6.

---

--

~~~~~~~~~~~~~~~

~~~~~~~~~~~~~~~~~~

~~~~~~~~~

______________

_________

S U B R O U T I N E AUXI
CO;IMO N C ( 3 8 3 0 )
EQUIVALENCE ( C ( 2 3 6 1 ) ,NOMUD ) , ( C ( 2 3 6 2 ) ,X M O D N O ( 1) ) ,( C ( 2 5 6 1 ) , N)
XM O D NO ( 99 )
DIMENSION
N : 1
DO 1 I:1 ,NOMO D
L :XMODNO(I)
GO TO (1 ,2 ,3,~4 ,5 ,6 , 7 , 8, 9 , 10 , 11 , 12 , 13 , 14 , i5 , 16 , 17 , 18 ,19, 2O ,2 1 ,22 ,23
1
,214 ,25 , 26 ,27 ,28 ,29,30 ,31 ,32 ,33,34 ,35 ,3 6 ,37 ),L
2 CALL A l l
GO TO 1

CALL A2 I
GO TO 1

14 CALL A 31
GO TO 1
5 CALL A 111
GO TO 1
6 CALL A5 I

GO TO 1

7 CALL CU

GO TO 1
8 CALL C21
GO TO 1
9

CALL C3 1
GO TO 1

10 CALL C141
Ii

GO TO 1

CALL C5 1

GO TO 1
12 CALL C61
GO TO 1
13 CALL C71
GO T O 1
111 CALL C81

GO TO 1

15 CALL C91
GO TO 1
16 CALL C 1O I
GO TO 1
1T CA LL Dli
GO TO 1
18 CALL D21
GO TO 1
19 CALL D3I
U0 TO 1
20 CALL DII I
GO TO 1
21 CALL D5 1
GO TO 1
22 CALL G i l
GO TO 1
23 CALL G2I
GO TO 1
211 CALL G31
GO TO 1
25 CALL G~4 I
GO TO 1
26 CALL G5 1
GO TO 1
_ ----

--

-- -r -

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

27 CALL G6I
GO TO 1
28 CALL 811
GO TO 1
29 CALL S2I
GO TO 1
30 CALL S3I
GO TO 1
1
3 CALL SIII
GO TO 1
32 CALL S5I
GO TO 1
33 CALL S I
GO TO 1
314 CALL S7I
GO TO 1

35 CALL S8I
GO TO 1

36 CALL S9I
GO TO 1

37 CALL S J O I
1 CONTINUE
RETURN

END

.4-!

~~-~~~--~~

_ _ _

_
__.
~~~~~ .. ~~~~~~~~~~~

~~~~~~~~~~~~~~~~~~

______

_______________________________

--

~~~~~~~~~~~ -~~~

SUBROUTINE AUXSU L3
COMMON C (3830)
)
EQUIVALENCE (C(2003 ),T) , (C(2361) ,NOMOD ) , (C(2362) ,XMODNO (1 )
) , (C(2562) ,IPL(1)) , (C(2664 ),DER(l ))
EQUIVALENCE (C (256l) ,N
EQUIVALENCE (C(2965) ,V A R ( 1 )
EQUIVALENCE (C(2020) ,LCONV )
DIMENSION
DE R (1 0 1 )
V A R ( 1O 1 )
IPL (100)
,
,
DIMENSION
XMODNO (99)
EXTERNAL A l
EXTERNAL A2
EXTERNAL A .~
DC 1 I:1 ,NO MO D
I F ( L C O NV . EQ. 2 ) R E T U R N
L :XMODNO (I)
GO TO (1 ,2,3, 14 ,5 , 6 ,7,8 , 9 , 10 , 1l , 12 , 13, 114 , 15 , 16 , 17, 18 , 19, 20 ,21 ,22 ,
123, 2L1 , 25 , 26 , 27 ,28 ,29,30 ,31 ,32 ,33 ,3 Z4 ,35 , 36 ,37),L
)

2 CALL Al

GO TO 1

3 CALL A2

GO TO 1
4 CALL A 3
GO TO 1

5 CALL AII

GO T O 1
6 CA LL AS
GO TO 1
7 CALL C l
GO TO 1
8 C4~LL C2
GO TO 1
9 CALL C3
GO TO 1
10 CALL C11
GO TO 1
1 1 CALL C5
UU TO 1
12 C A L L C 6
GO TO 1
13 CALL C !
GO TO 1
111 CALL Cd
G~ : TO 1
15 CALL C9
GO TO 1
16 C A L L C1 0
GO TO 1
17 CALL Dl
GO TO 1
18 CALL D2
GO TU 1
19 CALL D3
G O TO 1
20 CALL D 11
GO TO 1
21 CALL D5
GO TO i
22 C A L L o i
GO TO 1
23 CALL G2

~~~ L .

~~~~~~~~~~~~~~~~~~~~~~~~

----- -- - - ~~ - -

___________

GO TO 1

CALL G3
GO TO 1
25 CALL G11
GO TO 1
26 CALL G5
00 10 1
27 CALL 06
2 11

GO TO 1
28 CALL S i
GO TO 1
29 CALL S2
GO TO I
30 CALL S3
GO TO 1
.11 CA LL SLI
GO TO 1
32 CALL S5
GO TO 1

33

314
35
36
37
1

CELL

S6

0 0 10 1
CALL S7
GO TO 1
CALL S8
GO TO 1
CALL S9
GO T O 1
CALL S1O
CONTINUE

RETURN
END

~ ~~~~~~~~~~~~~~ - --

-~~~- -

--

~~~~~~~~~~~~~~~~~~~~~~~~~~~

rr

----

- - -

~~~

Bl ~O U T I N E Al
COMMON C (3830

C
C**TABLE LOOKUP FOR A E R O COEF

COMMON

/ N C 5 Z / N C 5 ( 14 )
*/ NC 1Z / N C 1( 2 )
/NC2Z / NC2 ( 14 )
/ NC3Z/ NC3(11 )
* / CA 1 /V M 1 (l 5 )
/CA2/BA2(6) , VM2(6) /CA3/L3A 3 (7 ) , VM3 (5)
* / C A 5 / L 3 A 5 ( 7 ) , VNI5 (3 )
/CZPFZ/CZPF (- ~5) /CZ2FZ/CZ2F (35) /CM PFZ/CMPF(35 ) /CM 2FZ /CM2F(35)
*/CYL4FL /CYL 4F(36) /CNL IFZ/CN 4F(3 6) /CLLIFZ/CLIIF(21) / C L 2 F Z / C L 2 F ( 2 1)
*/CZDFZ/ CZD F( 3 5 )
/CMDFZ /CMDF(35 )
*/CM QFZ ./CMQ F (36) /CLPFZ/CLPF (36) /CLDFZ/CLDF(21 )
*/CXOFZ/CXOF ~ l i )
/CMOFZ/ CV iOF (6) /CAII Z/CAI4 (6)
COMMON
COMMON /NC6Z/NC6(4) /CA 6/6A6( 6) , VM (4) /DXCPFZ/DXCPF(214)
/CA7/BA7 (9) ,VM7 (3 ) /DCMFZ/DCMF(27 )
COMMON / N C T Z I N C 7 ( 1 1 )
COMMON /NCB Z/NC8(2 ) /CA B Z/CA8 (3 ) /DCMOFZ /DCMOF (3 )

C
C** INPUT DATA
E Q U I V A L E N C E (C (1200) , OPTA 1
E Q U I V A L E N C E (C(12140) ,CL 1
E Q U I V A L E N C E (C( 12L 11 ) , CL2
E Q U I V A L E N C E (C (12142) ,CL3
E Q U I V A L E N C E (C( 12~43) , DCM iF(1 ))
E Q U I V A L E N C E (C(i214b ) , CLO
E Q U I V A L E N C E (C(13O7 ) , RFLGTH)
C
C * *I N P U T S FROM OTHER -IODULES
E Q U I V A L E N C E (C (020L1 ), VMA CH
E Q U I V A L E N C E (C(0367) , BALPHA )
E Q U I V A L E N C E ( C ( 0 3 6 8 ) ,BALPHY )
E Q U I V A L E N C E (C(0369) ,B A L P H P)
E Q U I V A L E N C E (C(037O) ,B P H I P
EQUIVALENC E (C(1196) ,BDELT 1 )
E Q U I V A L E N C E (CU 197 ),BDELT2)
EQ U I V A L E N C E (C (il 9d ) ,BDELT3 )
E Q U I V A L E N C E (C(1l99) , BDELTLI )
E Q U I V A L E N C E ( C ( 1 5 5 1 ) ,OP T M
E Q U I V A L E N C E ( C ( 1 5 5 5 ) ,U D L 1
E Q U I V A L E N C E ( C ( 1 5 5 6 ) , UDL2
)
EQUIV A L E N C E ( C ( 1 5 5 7 ) ,UDL3
(
C
(
1
5
5
8
)
EQUIVALENCE
,UDL4
C
C * * INPUT5 FROM M~~I;-J PROGRAM
)
E~~U I V A L E N C E ( C ( 2 0 0 0 ) , T
E Q U IV A L E N C E (C(2O20) , LCONV )
C
C**OUTPUT iu MODULES
E Q U I V A L E N C E ( C ( i 2 0 0 ) ,OPT UNG )
( C ( 1 2 0 3 ) , CX
E QUIVALENCE
EQ U I V A L E N C E (C(12O4 ) ,CY
EQ U I V A L E N C E (C(i205) , CZ
)
)
EQ U I V A L E N C E (C (12O6), CLP
E Q U I V A L E N C E (C(12u7 ) , CMQ
)
)
)

)
)

E Q U I V A L E N C E (C (1208) , CN R
E Q U I V A L E N C E (C(1209), CL

EQ U I V A L E N C E
E QUIVALENCE

~~~~~~~

(C(1210 ) , CM
(C(1 2i 1) , CN

~~~~~~~~~~~~~~~~~~~~~~~

)
)

)
)

=--

EQU I V A L E N C E
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQU IVALENCE
C* * OT H ER OUTPUTS
EQUIVALENCE
EQUIVALENCE
EQUIVALENC E
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
E QUIVALENC E
E Q U IV A L E N C E

- -

~~~~~~~~~~~~~~~~~~~~~ -- - - - -- ~~

( C ( 12 3 7 ) ,CCQ )
( C ( 1 2 3 8 ) ,CC R )
( C ( l 2 3 9 ) ,CC RTOQ )
( C ( i 2 5 7 ) ,SDP)
( C ( 12 6 7 ) ,SD Q)
( C ( 1 2 7 7 ) , SDR )
( C ( 1 2 1 2 ) , CXO
( C ( 12 1 3 ) ,C ZO
( C ( 12 l 4 ) ,DCZ2
~
( C ( 1 2 i 5 ) ,CZDQ
( C ( 1 2 1 6 ) ,CZDR
( C ( 12 i 7 ) ,DCYII
( C ( 12 1 8 ) ,CMO
( C ( 12 1 9 ) , DCM2
( C ( 12 2 O ) ,CMDQ
( C ( l 2 2 l ) ,C M DR
( C ( i 2 2 2 ) ,DCNII
( C ( 1 2 2 3 ) ,DC L1
(C ( 12 2 L l ) ,DCLII
( C ( 12 2 5 ) ,CLDP
( C ( 1 2 2 b ) ,VM
(C(1227) ,BAP
(C(1228) , BDL
(C(1229), BDM
(C(123O) ,BDN
( C ( 1 2 3 i ) ,BDP
( C ( 1 2 3 2 ) , BDQ
(C(1233 ) ,BDR

)
)
)

)
)
)
)
)
)

)
)
)

)
)

)
E Q U I V A L E N C E (C(1236) , CH 1
E Q U IV A L E N C E (C (1237) ,CH2
)
E Q U I V A L E N C E (C(1238) , CH 3
E Q U I V A L E N C E (C(l239) , CH11
)
E Q U I V A L E N C E (C(12L10) , C Hl i
E Q U I V A L E N C E (C (1241 ) ,CH2 1
)
EQ U I V A L E N C E (C (1242) ,CH 31
E Q U I V A L E N C E (C(12L13) , CHLI1
DIMENSION NDCLD(2) , D C L D V M ( 3 ) ,
DIMENSION DCLDFF(3 ) , DCM 1F (3 )
EQ U iV A L E N C E (C (1234) , DCLDO )
E Q U I V A L E N C E (C(1235) , DCLDA)
)

DCLDOF(3 ), DCLDAF (3 )

~~~~~~~ YA W P A H A N E T E R I N P U T
EQ U I V A L E N C E (C(290l) , O PTN YW)
C
L)A TA ~ A T L B 1 / 4 H C X O / , WATLB2/ L IH CMO / ,W AT L B3/ L IH CZO / ,WA TLB II
# / LIHCMO /
DA1 A
A T L B D/ 1 1 t i D C M2 / ,WAT LB6/11 HCZDQ/ ,WATLB7/ 11HCM DQ/ ,WATL B8/
~
114I-I DC Z 2/
DAT A ~ AT LB 9/4RDCY 11/ ,W A T L l O /4HDCN II/ ,WATL 11/4HCLP / ,WATL 12/
//14HCM.~ /
DATA ~~A T L 1 3 /L l H C N R / ,WATL14/IIHCLDP/ ,WA T L 1 5 /4 HD C LI I/
DA TA W A T L 16 / L4 H D C L 2 / ,W A T L 1 7/ 4 H
/ ,W A T L 1 8 / 4 H D X C P/
D A T A WA TL 1 9/ 4 H D C M /
DA TA NDCLD / 3 , 0 /
DATA DCLDV M / .500 , .9 5 0 , 1 . 25 0 /
DA TA DCLDO F / . 0 0 0 ,- .0 0 13 , .00 10/
UAT A UL L L A E / . 00 1 6 , .00 23 , . O O 10/
DATA DCLDFF / .06 ,. 10 ,.OLI /
DATA SDPCC ,SCCR Q/O.0 , O.0/
C
52

__________________________________________________________________________________________

- - - - - -

- -

~~~~~~~~~

MULTIPLE
USPHI

UC P I-1 I
US 2PHI
USLI PHI
US2PH2

~~~~

_-

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

ANGLE F O RMULAE
: SIND (
BPHIP )

AND ABSOLUTE

~~~

.-- .
-~ ~ - - ~~~

-.~ .- --~~~.-~~~- ~~~~~~~~~~~~~

-~~~~

VALUES OF ANGLE OF ATTACK

C O S D ( B P I I I P)
S I N D ( 2 . *BP1-I IP)
S IND(11 .* BP H I P)
: US2PHI**2
:

C
C* * LIM IT TABLE ARGUMENTS
B D Pz ( B D E L T 1 - B D E L T 2 + B D E L T 3 + B D E L T 1 4 )/ 1 1 + S D P + S D P C C
BDQ :(+BDELT 1 +BDELT2+BDELT3+BDELT4 )/14+SDQ+SCCRQ
BDR :(BDELT 1 +BDELT2 BDELT3-,-BDELT4)/4+SDR
SDPCC :CCQ*BD Q+CCR *BDR
SCC R Q :C C H TOQ B DR
BAP
BA LP H P
UAL
A B S (BALPHA)
UBT
ABS (BA LP IIY )
VN1
V M ACH
IF (BAP .GT .2O .)B ~ P:2O .
IF ( U A L . G T . 2 0 . ) U A L z 2 O .
ip (UBT .GT .20.)Ut3T :2O ,
C
C**TABLE LOOR UP FOR .AERO COEF
IF~T .GT . 0. .AND. C ( 1 9 7 6 ) . L T . O . ) GO TO 1000
X F:O.
CXO :FINTP 1( VNI , V M1 ,CXOF ,N C 1 (1 ),XF ,WATLB1 )
X FIO.
CMO :FINIP1 (
VN ,CA L 4 , CM OF ,N C 2 ( 1 ),XF ,WATL B2)
0.
XF :
C Z O : F I N T P 2 ( B A P , VM , B A 3 , V M 3 , CZPF , N C 3 ( 1 ) , N C 3 ( 2 ) , N C 3 ( 3 ) , XF , WATL ~33)
CMO: F I N T P 2 ( B A P ,VM ,B A3 , VM 3 , CMPF ,N C 3 ( l ) ,N C 3 ( 2 ) ,N C 3 ( 3 ) ,X F ,WATL BI4 )
D C M2 : F I NT P 2 ( BA P , VM ,BA 3 , VM 3, CM2F ,NC3 ( l) , NC3( 2 ) ,N C 3 ( 3 ) ,XF ,WATLB5 )
CZD Q:FINTP2(BAP , VM , BA3 , VM 3 , CZDF ,NC 3 (1), NC 3 (2) ,NC3 (3 ) ,XF , WATLB )
CZDR : CZD Q
CMD Q : F I N T P 2 ( B A P , VNI , B A 3 , V M 3 , CM DF , N C 3 ( l ) , N C 3 ( 2 ) , N C 3 ( 3 ) , XF , W A T L B 7 )
C~~DN
CMD Q
D C Z 2 : F I N T P 2 ( B A P , VM , B A 3 , V M 3 , CZ2F , N C 3 ( 1 ) , N C 3 ( 2 ) , N C 3 ( 3 ) , XF , W A T L B 8 )
XF Q .
~
DCY11: FINT P2(BAP , VM ,BA2 , VM2 ,CY4 F ,NC2(1) , N C 2 ( 2 ) , NC2(3) ,XF ,WATLB 9)
DCNL I zF INTP2 (BA P , VM , BA 2 , VM2 ,CN4F ,NC2( l ) , NC2 (2) ,NC2 (3) ,XF , WATL 1O)
C L P : F I N T P 2 ( BA P , VM ,BA2 , V M2 ,CLPF ,N C 2 ( 1 ) ,N C 2 ( 2 ) , NC 2 ( 3 ) ,XF ,W A T L 1 1 )
XV :O .
CMQ : F I N T P 2 ( U A L , VM ,BA2 , V M2 ,CMQF ,N C 2 ( l ) , N C 2 ( 2 ) ,N C 2 ( 3 ) ,XF ,W A T L 1 d)
XF :0.
C N Rz F I N T P 2 ( U O T ,VM , BA2 , VM2 ,CMQ F ,N C 2 ( 1) ,N C 2 ( 2 ) ,N C 2 ( 3 ) ,XF ,W A T L 13 )
XF :0.

XF ,W A T L 1 4)

C L D P :FI N TP 2 ( BA P , VM , BA5 ,

VM 5 , CLDF ,N C 5 ( l )

OCL2 :FINTP2(BA P , VM , BA5,

VM5 ,CL2F , N C 5 ( 1 ) ,NC5(2) ,NCS (3),XF ,W A T L 1b )

,N C

5(2),

NC5(3 ) ,

DC L 1 1 : F I N T P 2 ( B A P ,VM ,BA 5 , VM 5 ,CLIIF ,N C 5 ( i ) , N C 5 ( 2 ) ,N C 5 ( 3 ) ,XF ,W A T L 1 5 )

X E:O .
D C L D O :V I N T P 1 (
V M ,DCLDV M ,DCLDOF ,N D C L D ( 1 ) , XF ,W A T L 1 7 )
D C L DA : F I NT P 1(
V M ,DCLDVM ,DCLDAF ,N D C L D ( 1 ) ,XF ,W A T L 1 7 )
DCL. F F I N T P 1 (
VM ,DCLDV M ,DCLDFF ,N D C L D ( 1) ,XF ,W A T L 1 7 )
~~
X F :O .
PX C P : F 1 N F P 2 ( u A P , VM ,BA6 ,VM ,DXCPF ,N C 6 ( 1) ,N C6 ( 2 ) ,N C 6 ( 3 ) , XF ,W A T L I B )
XF u.
DI -1 :FINTP ( I3AP , VM ,B A 7 , V M7 ,DCM F , N C 7 ( l ) ,NC7 ( 2 ) , N C 7 ( 3 ) ,XF ,W A T L 19 )
~
~~
X l :O.
DC~~O :FINTP 1 (
VM , CA 8 ,DCMOF ,NC8 (i) ,XF ,WATL 17 )
D;:1i: F INTP 1 (
VR ,CA B , DCM 1F ,NC8( 1 ) ,XF ,WA TL 17 )
C MO

CMO

~ --

_________________

CM O : CMO
1000 CONTINUE

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

~~~~~~~~~~ - --- --

--~.---

: - ~~~~~~~

DXCP*CZO *lO.8

C
C * * A EHO COEF WIND A X I S
DCMO :O .O
CZDR USPHI BDR
CZP z CZO + D C Z 2 * U S 2 P H 2 + C Z D Q * U C P H I * B D Q
CMP : CMO + D C M 2 * U S 2 P H 2 + CMDQ* UCPH I* BDQ
CMDR* USPIII* BDR
DCNII*USIIPHI + CMDQ* USP1~JI* BDQ + CMDR* UCPHI* BDR
CNP
CY P :
DCYII US4PHI + CZDQ * USPHI* BDQ + CZDR UCPH I BDR
C
C~ T RANSFORMATION FROM WIND TO BODY AXIS
CX
CXO
CL : DCL2* US2PH I+ DCL I4* US 11PUI + CLD P* BDP
CY : CY P ~ UC PHI
CZP* USPH I
IF(OPTNYW.GT.O. )CY:0.
_ CYP* USPH I
CZ
CU
CNP*UCPH I
CM P*USPH I
CM
CNP*USP H I + CM P*UCPUI + CMO + DCMO
DCLDR : DCLDO+ DCLDA*SIN( 6.2832*DCLDF*BALPHP )
I F O P TA 1 .LE. 2 . ) GO TO 1
CL
CL + DCLDR * BD R
RETURN
1 CL : CL + CLO + (CL 1*USPHI+CL2*US2PHI+CL3*SIND(3 . *BPHIP))*BAP/8.

RETURN
END

.L
.
~~

C ZP * U C P H I

- -~~

~~~

~~~

--

.
~
-

-- -~~~~ ~

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

SUBROUTINE A2
BODY AXES
C * * A E R O FORCE AND MOMENT MODULE
COMMON C ( 3 8 3 0 )
101 FOR MAT ( 1I- I O ,LIX ,2 1HFRONT LUG CLEARS RAIL ,5X ,3HT :
, 1PE 1O.2 ,5X ,
1P E 1O .2 )
*
9HREL VEL :, 1P E 1O . 2 ,5X ,1I4 HPITCH MOMENT
~,
C
C** INPUT DATA
EQUIVALENCE ( C ( l 3 O 6 ) ,RFAREA )
EQUIVALENCE ( C ( 1 3 0 7 ) ,RFLGTH )
EQUIVALENCE ( C ( 1 3 1 6 ) ,RLUG
EQUIVALENCE ( C ( 1 3 1 7 ) ,RAIL
Q U IV A L E N C E (C(133O) ,AGV
E Q U I V A L E N C E (C(171 12) ,
AMP2) , (C(17 ~46) ,
AMP 1 )
EQUIVALENCE (C(1332) ,CP H A S
E Q U I V A L E N C E (C(1333 ) ,AGH
EQUIVALENCE (C(133 11) ,RLGZ
E Q U I V A L E N C E (C(11 4O5) ,QI 3 U RN
E Q U I V A L E N C E (C(1627) ,AGRAV
C
C * * I N P U T S FROM OTHER MODULES
)
)

)
)

DIMENSION

IS N D X ( 1 1O )

EQ U I V A L E N C E
E QUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQ U I V A L E N C E
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
E Q UIVALENCE
E QU IVALENCE
EQUIVALENCE
E QUI VALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIV A LENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE

(C(3634), ISNDX (1)) ,


(C(0203) , PDYNM C)
(C( 2011),V MACH
(C(0207 ),VAIRSP)
(C( 350),BTHT
(C( 38O ) ,HANGO

(C(3512) ,

13512)

C C ( 1 2 0 3 ) , CX

( C ( 12 0 1 1 ) ,CY
(C(1205) ,CZ
)
(C(12O6) ,CLP
(C(1207), CMQ
(C(12O8) ,CNR
(C(1209) ,CL
(C(1210) ,CM
(C(1211 ) ,CN
(C(1236) ,CH1
(C(i237 ) ,CH2
(C(l238 ) ,CH3
( C ( 12 3 9 ) , CH14
( C ( 1 3 2 0 ) , FMXTH
(C(1321) ,FMYTH
( C ( l 3 2 2 ) , FMZ T H )
(C(l L 4ll ) ,FTHX
( C ( 1 1 4 1 2 ) ,FTHY
( C ( 1 L 1 1 3 ) ,FTHZ
( C ( 1 14 2 2 ) ,RLCG
( C ( 1 7 2 3 ) ,CFA 2 3
( C ( i 7 3 5 ) ,CFA 33
( C ( 1 7 3 9 ) ,WP
)
( C ( l 7 4 3 ) ,WQ
(C ( 1737 ) ,
FMX) ,
)
( C ( 1 7 4 7 ) ,WR
(C ( 17 118 ) ,
FMIX)
(C ( 17 3 8) ,
WPTO)
CHAD )
( C ( l7 5 1) ,
(C ( 6 2 6 ) ,
VIB )
(C (2 000) , T
(C ( 1972 ) ,R K U T T A )
)

)
)

)
)
)

(C ( 174 1 ) ,

FMY) ,

( C ( 1 7 4 5 ) , FMZ)

E QUIVALENCE

~~~

- - ---

~~~~~~~

--

~~

- ---- -

( C ( 1 9 7 5 ) ,NPT )

C* *0UT PUTS
)
EQUIVALENCE (C(1300) ,FXBA
EQUIVALENCE ( C ( l 3 O l ) ,FYBA
EQUIVALENCE ( C ( 1 3 O 2 ) ,FZBA
EQUIVALENCE ( C ( 1 3 0 3 ) ,FMXBA
EQUIVALENCE (C(13014 ),FMYBA
EQUIVALENCE ( C ( 13 0 5 ) ,FMZBA
EQUIVALENCE ( C ( 1 3 0 8 ) ,RDELCG)
EQUIVALENCE ( C ( 1 6 2 8 ) ,DMASS )
C
EQUIVALENCE ( C ( l 7 1 1 8 ) ,FMIX
EQUIVALENCE ( C ( 17 L 19 ) , FMIY
EQUIVALENCE ( C ( 17 5 0 ) ,FMIZ
C
C**OTHER OUTPUTS
)
EQUIVALENCE ( C ( 1 3 0 9 ) ,FMH1
EQUIVALENCE ( C ( 1 3 1 0 ) ,FMH2
)
EQUIVALENCE (C (13 11 ), FMH3
)
EQUIVALENCE ( C ( 1 3 12 ) ,FMI-14
EQUIVALENCE ( C ( 1 3 2 3 ) ,FMXLUG )
EQUIVALENCE (C ( 132 I ) ,FMYLUG )
EQUIVALENCE ( C ( 1 3 2 5 ) ,FMZLUG )
EQUIVALENCE ( C ( 3 5 0 4 ) ,OPTNL4 )
DATA FLG2 ,FLG1/O., 0./
C
C* * FO R C E VECTOR COMPONENTS
UQS : PDYNMC * R FA REA
UQSL :UQS* RFLGTH
)

)
)

F X B A :UQ S ( ~~C X ) + F T H X
FYBA ~ UQS * CY + FT HY
FZBA :UQS*CZ +FTHZ

C~~

C
C* *

AERO MOMENTS

UL2V
0.
IF (VAIRSP
(CL
FMXBA
(CM
FMY BA
(CN
FMZBA

CALCULATE
FMH 1 :
FMI-12
FMH3 :
FMH4 :

( NOTE FACTOR

OF 2 . 0 IN DAMPING COEFFICIENT)

.GT . 0.) UL2V :


+ C L P * U L 2 V *W P )
+ CMQ * UL2V *WQ )
+ CNR * UL2V *WR )

R FLGT H / (2 . *VAIRSP )
+ FMXTH
* UQSL
* UQSL + FZBA RDELCG + FMYTH
FYBA * RDELCG + FMZT H
* UQSL

HINGE MOMENTS
C H 1* UQSL
CH2* UQSL
CH3*UQSL
CH ~4*UQSL

C** M0ME Nr s AND FORCES DUE TO LUGS

IF(QBURN .LE . . 0. .AND. RANGO


UFZL2: FZLUG
FYLUG : 0 .
FZLUG : 0 .
FMXLUG
0.
FMYLUG : 0 .
FMZLUG
0.
IF ( F L G 2 . G T . O . ) GO TO 714
FMX :O.
FMY:0 .
FMZ :O .
DO 6 1:1 ,13512

_
~~~~~ ~~~~~~~~

-_

--

--

---

.LE. RAIL+RLUG) GO TO 70

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

----- - --:
~ ~~~~ ~~~~~~~~~~~~~~~~~~~~~~~~~~

- ~~ ~~

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

~~~~~

~~ ~~~~~~ ~~~~ ~~i

IDO :I
IF(ISNDX(I).EQ.1743 )CALL MCA RLO(DUM , 1 , IDO)
IF(ISNDX (I).EQ.1747 )CALL MCA RLO( DUM , 1 ,IDO )

CONTINUE

1.
C (13 )
WRITE (6 , 10I)WP ,W Q , WR
FO R M A T ( 1 H ,5OX ,2 1HT IPOFF RATES--ROLL
1011
711 YAW
,F 6.l )
C( 2 6 6 4 ) : C ( 2 7 6 1 4 )
FLG2
1.
W R I T E(6 , 102 ) T ,V A I R S P ,UFZL2
WRITE (6 , l03 ) R A N G O
: ,F6 .II)
103 FORMAT(32X , 9 H R A N G O
102 FORMAT (1HO ,36H
REAR LUG CLEARS
RAIL FORCE
*
1OH R EL VEL : ,F8 .3, 1 H

GO TO 74

70 IF (RANGO

RZDD

:AGV

,F 6 .1 ,9H PITCH

RAIL
:

,F 8 . 2)

,F 6 . l ,

,F8 .II,

.LE . RAIL) GO TO 72

RYDD
AGH
FYLUG :_ . (FYBA+DMASS*AGRAV*RYDD _ DMASS* (FMZBA*RLCG /FMIZ
_ FMXBA*RLGZ/FM IX)) / (1. + DMASS*( RLGZ**2/FMIX +RLCG**2 /FM IZ))
*
FZLUG : - (FZBA + DMA SS*AGRAV*(CFA 33_ RZDD ) + FMYBA*
R L C G * D MA S S / F M I Y )/ ( 1 . + DMASS * RLCG * RLCG/FM IY )
*
FMXLUG z FYLUG*RLGZ
FM Y LUG
FZLUG * RLCG

FMZLUG

!~~-

- -

FYLUG * RLCG

IF (FLG 1 .GT . 0.) GO TO 714


FLG 1
1.
W R I T E ( 6 , 101) T , VA IR SP , FMYLUG
W R I T E ( 6 , 1O 3 ) RANGO

C
C
C
C
C
C

W PTO- MONTE CARLO VALUE OF TIPOFF ROLL RATE


DTLUG T IME INCREMENT BETWEEN FIRST LUG AND
LAST LUG DROP OFF TIME (TIPOFF ROLL RATE

FMX- -ROLL MOMENT THAT GENERATES

OCCURS AT LAST LUG )

DTLUG :.026
FMX ~ WPT0 * FM I X / C R A D / D T L U G

TIPOFF ROLL RATE

CONTINUE

C
C

GO TO 74

72 C O N T I N U E

RZDD :0.
RYDD

:0.

RZDD

AGV
AGH

IF(RANG O

73

RYDD :
FYL UG

.LT. RAIL .3 ) GO TO

73

( F Y B A + D MA S S * A G R A V * ( C F A 2 3 + R Y D D ) )
- ( F Z B A + D M A S S * A G R A V * ( C FA 3 3 _ R Z D D ) )
FMXLUG : F NIXBA
FMYBA
EMYLUG :
FMZBA
FMZLUG :
FLG1 : 0 .

FZLUG

57
- - -

--- --- - --

_ _ _ _ _ _ _ _ _ _ _ _

FLG2 :O .
74 CONTINUE
C**TOTAL FORCE AND MOMENTS

FYBA
FZBA

FMXBA
FMYBA
FMZBA

FYBA

FZBA

FMXBA
FMY BA

:FMZBA

FYLUG
FZLUG
+
+
+

FMXLUG
EMYLUG
FMZLUG

C**LAUNCH TRANSIENTS MOMENTS (1 YAW ,2 PITCU ,3ROLL MOMENTS)

75

I F( F L G 2 . G T . 0 . ) G O TO 75

I F (V I B . L E . O . ) G O TO 75
CALL LTRAN ( T ,DELT ,A M P 2 ,FMY ,WQO ,1 ,2)
CALL LTRAN( T ,DELT ,A MP1 ,FMZ ,WRO , 1 , 1)
CONTINUE

FMXBA : FMXBA + FMX


F M YBA :FM YB A+ FMY
FMZ BA :FMZBA + FMZ

RETURN

END

L~~~~~~~~

_ _ _ _

~~~~~~~~~~~~~~~~~~~~

~~~~

--- _

~~~

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

- -- -

SUBROUTINE A 3
C** ENGINE MODULE
COMMON C ( 3 8 3 0 )
C

C
C** INPUT DATA

(C(1313 ) ,RFXCG )
(C(13114 ),RFY C G )
E Q U I V A L E N C E (C( 1315 ) , RFZC G )

EQUIVALENCE

EQUIVALENCE

EQUIVALENCE
EQUIVALENCE

( C ( 1 14 O 1 ) ,BALPHT )
( C ( l 4 0 2 ) ,BPHIT )

EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE

(C(i416 ) ,DWP
(C(1111 7) ,RDCGO
(C(1l 418) ,RDCG F
(C(141 9 ) ,FMIXF
(C(11 42O) ,FMIYF
(C(1421) ,RLCGO
(C(11423 ) ,FM IXO )
( C ( i 1 4 2 1 4 ),FM IYO)

EQUIVALENCE

(C(11403),QNALGN)
(C(114014 ),PC F TH )
(C(14O5) , Q BURN )
)
(C(1414) ,CI S P
)
(C(11415) , DWT

C
C** INPUTS FROM OTHER MODULES
EQUIVALENCE (C (2000) ,T
C

)
)
)

)
)

C** OUTPUTS
EQUIVALENCE
EQ U I V A L E N C E
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE

(C (13O8) , RDELCG)
(C(1320 ) , FMXTH )
(C(1321 ), FMYTH )
(C(1322) , FMZTH )
)
(C(1409) ,UDWP
(C( 1410) , FTHRST)
(C (1411 ), FTI-I X
)
(C(14 12) ,FTHY
)

EQUIVALENCE
EQUIVALENCE
EQUIVALENC E
EQUIVALENCE
EQUIVALENCE

(C(11 422) ,RLCG


)
(C(1628) , DMAS S )
)
(C(l748) ,FMIX
)
(C(17149) ,FMIY
(C(175O) ,FMIZ
)

EQUIVALENCE

( C ( 1 14 1 3 ) , FTUZ

C**STATE V A R I A B L E S AND THEIR DERIVATIVES


EQU iV A L E N C E (C(11496) ,UIMPD )
)
EQ U I V A L E N C E (C(1499) ,UIMP
C**LOOK UP TABLE FOR THRUST
DIMENSION NTH(2) , T H A ( i l ) , T H F ( l l )
DATA WATLAB/ 4HHRS T /
DATA NTH/ l1 , 0/
DATA THAI
0.,.025, .125 , .250 , .750 , 1.500 , 1.625 , 1.750 , 2.00 , 3 . ,
*100. !
DATA T r i F / . 1 , 1800. , 17 50 . , 1650. , 1600. , 1L400 ., 1250., 600., 300., 0.,
*3 /
C
IF (QBUHN.GT.O .) RETURN
CALL T A B L E ( T ,T I-I A ,THF , N T H ( 1 ) , XF ,WATLAB ,FT HRST )

LL

~~~~~~~~~~~~~~~~

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

-~~ -

~~~~~~~~~~~~~~~~~~~~~~

IF (QNALGN) 20 , 20 , 10
10 USINA :SIND(BALPHT)
FTHX :FTHRST*COSD (BALPHT)
FT H Y :_ FT H RS T * U SI NA * S IN D( B P H IT )
FTHZ :FTHRST*USINA*COSD (BPHIT)
_ FTUY *RFZ CG + FTHZ R FYCG
FMXTH

EMYTH

FMZT H

GO TO 30

FT HX * RFZCG
_ FTHX * RFYCG

FTHZ*RFXCG
FTHY*RFXCG

20 FTF!X ZFTHRST
FTHY :0.
FTHZ:O .

FMXT H O.
FMYT H:O .
FMZTU :O .

C
C
C

30 CONTINUE
UIMPD
UDWP
DMASS
RDELCG
FMIX
EMI l

FMIZ
C

--

FTHRST
U IM P / C IS P
(DW T-,-DWPU DWP )/32 .174
REIC GF )* UDWP/DWP
( RDCGO
RDCGO

FMIXO
FMIYO

FMIY

(FMIXO_ FMIXF )*UDWP/DWP


(FMIYO_ FMIYF )*UDWP/DWP

R LCG
RLCGO + RDELCG
IF (FTHRST .GT . 0.) RETURN

WRITE (6 , 100) T
100 FORMAT (/ / 1 1 4 9 BURNOUT T IME: ,F8 .14 ,511 S E C . )
QB U RN= 1.0
RETURN
END

_
_

_
_

_
_

---.
~~~

________

_________

- i
~~~~~

~~ : ~~~~~~

- --

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

~~~~~~~~~~

-:

SUBROUTINE A 3I
COMMON C(383O )
DIMENSION IPL(100) , ISNDX(40)
EQUIVALENCE (C(3634) , ISNDX(1)) , (C(3512) , 13512)
EQUIVALENCE (C( 367),BALP HA )
EQUIVALENCE (C( 368),BALPHY )
EQUIVALENCE (C( 370),BPHIP )
EQUIVALENCE (C(1308) ,RDELCG )
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE

EQUIVALENCE
EQU iVALENCE
EQUIVALENCE

)
EQ U I V A L E N C E
)
EQUIVALENCE
EQUIVALENCE
)
EQUIVALENCE
)
EQUIVALENCE (C(1739) ,WP
EQUIVALENCE (C(17L 13) ,WQ
)
EQUIVALENCE (C (1747) , WR
)
EQUIVALENCE (C(171 48) ,FM IX
)
EQUIVALENCE (C(17149) ,FMIY
)
EQ U I V A L E N C E (C(1750) ,FMIZ
T)
EQUIVALENCE (C(2000) ,
)
EQUIVALENCE (C(2561) ,N
EQUIVALENCE (C(2562) ,IPL (1))
CHAD )
E QUIVALENCE (C(175l ) ,
EQUIVALENCE (C( 626),
VIB)
FMX ) ,
E Q U I V A L E N C E (C(1737) ,
DATA IFLG 1 ,IFLG2/ O ,0/
) : 14 96
IPL( N
N :N -i- i
C(1499) = 0.

(C (1741 ) ,

FMY) , (C(17145) , FMZ )

IF (QBURN .GT. 0.) GO TO 10


CRAD=57 .2 95778
FM X _ O
FMY :O.
FM Z=O ,
.
0.
~1P
0.
WR

)
)
)

: 0.

BALP HA
0.
BALPI-I Y = 0.
BPH IP : 0.

C MONTECARLO THRUST DIRECTION


C
DO 5 I = 1 , 13512

ERRORS

( C ( 14 12 ) ,FTHY
( C ( 1 4 l 3) ,FTHZ
( C ( 1 1 4 1 5) ,DWT
(C(1418) ,RDCGF
(C (114 19) ,F MI XF
(C(1142O ) ,FMIYF
(C (1628) ,DMAS S

( C ( 13 2 0 ) ,FMXTH )
( C ( 13 2 1 ) ,FMYTH )
(C(1322) ,FMZTH )
(C( 1405) ,QBURN )
)
(C(141l ) ,ETHX

100
I
IF(ISN DX( I ) .EQ .1313 ) CALL MCARLO (DUM , 1 ,
IF( ISNDX( I ) .EQ .131L 1 ) CALL MCARLO (DUM , 1 ,
F(ISNDX( I ) .EQ. l315 ) CALL MCARLO (DUM , 1 ,
IF(ISNDX(I) .EQ. 1140 1 ) CALL MCARLO (DUM , 1 ,
IF(ISNDX( I).EQ. 1402) CALL MCAR LO (DUM , 1 ,
C~ M ONTE CARLO TIPOFF ROLL , PIT C H AND YAW RATES
61

IDO)

IDO )

IDO )
IDO )
IDO )
-

~~~~

______

~~~~

~~~~~~~~~~~~~~~~~~

I F ( I . 5 N DX ( i ) . E Q . 1 7 3 8 ) C A L L MCARLO ( DUM ,1 ,IDO)


I F ( 1 S N C X ( I ) . E Q . 17 146) IFLG2:0
IE( ISN DX ( I) EQ. 1742 )IFLG1= 0
CONTINUE
.

IF (VIB .LE. 0.)GO TO 6


CALL L T RA N ( T ,DELT ,C ( 17 4 6 ) , DUM ,W RO ,IFLG2 , 1)
CALL LTRAN( T ,DELT ,C( 17142) ,DUM ,WQO , IFLG 1 ,2 )
W Q:WQO/FM IYF
*CRAD
W R =W R O / F M I Y F
* C RA D

CONTINUE

IF LG 1 1
IFLG2= 1

RETURN

10 C ONTINUE

FTHRST = O.
FTHX:0 .
FTHY=O .
FT HZ =0 .
FMXTI-1:O.
FMYTH:0 .
FMZTH=0 .

DMAS S

DWT/ 32 .17 14
RDELCG : RDCGF
FMIX
FMIXF
FMIY = FMIYF
FMIZ = FMIYF
RETURN

END

62
-

~~~~~~~~~~~~

BLOCK

DATA

COMMON
* /NC1Z /NC 1 (2)
*/ CA 1 /V M 1 (1 5 )

/NC 2Z/NC2 (4 )
/NC3Z/NC3 (4)
/CA2 /BA2 (6), VM2 (6)
/CA5/6A 5 (7) ,VM5( 3)
*/ C A 3 / B A 3 ( 7) , V M 3 ( 5 )
COMMON
/CMOFZ /CM O (6) /CA 4Z/VM 4(6)

/NC5Z/NC5 (LI )

/CM2FZ/CM2(35)
/CMPFZ/CMP( 35 )
/CZ2FZ/CZ2 (35)
*/CZPFZ/CZp (35 )
/CL4FZ/CL14(21)
/CL2FZ/CL2(21 )
*/CYI4Fz/CYI4(36)
/CNI4FZ/CNI4 (36)
/CMDFZ/CMD 1(35 )
*/CZDFz/CZD1 (35)
/CLDFZ/CLD1 (21)
*/CMQFZ/CMQ(36)
/CLPFZ/CLP (36)
*/CXOFZ/CXO (15)
COMMON /NC 6Z/NC (4 ) /CA6/BA6 (6) , VM6 (14) /DXCPFZ/DXCPF (24 )
COMMON /NC7Z/NC7( 4) /CA7/BA7 (9) ,VM7( 3) /DCMFZ/DCMF (27 )

COMMON /NC8Z/NC8 (2) /CA8Z /VM 8 (3 ) /DCMOFZ/DCMO(3 )


DATA N C 1/ 1 5 ,O/
DATA NC2/ ,6 ,36 ,O/
DATA NC3/7 ,5 ,35 ,0/
DATA NC5/i , -5, 2 1 ,0/
DATA VM1/O. 0 ,0.44 , O .6 ,0 .7 ,0.8 ,0.85, 0.9 , 1 .0 , 1.1 , 1 .2
*
1. S , 1. LI , i.6 , 1.8 ,2. 4 /
~
~
DATA VM2/O .0 ,0.7 ,U .9 , 1.1 , 1. 11 ,2.0/
DAT A VM 3/ .5 , . 8 5 , .9 5 , 1.05 , 1 . 2 5 /
DATA VM 14/ .5 ,
.95 , 1.05 , 1.25 , 2.!
DATA VMS! .5 , .95 , 1.25/
DATA BA2/ 0. , 4 ., 8., 12 ., 16., 20./
DATA BA 3 / O. , 2., LI . , 8 . , 12 ., 16 .,20./
DATA BA5/ 0., 2., 4., 8., 12., 16 ., 20./
DATA CXO/. 3O 8 , .293 ,.295 , .299 ,.313, .33O ,.352 ,.1492,.616 ,.696 ,
*
.7D2 , .79 1 ,.850 , .859, .832/
DATA CMO /
*Q~~10 ,+0.25 ,+0 .OO , O .1 5 , O .O5 , .00 /
DATA CZP /
* 0.00 , 0.20 , 0 . 5 0 ,
1.30 , 2.25 , 3 .2 5 , 4.30
~ 0.00 , 0.20 , 0.50 , 1.35 , 2.35 , 3.40 , 11.45
* 0.00 , 0.25 , 0 . 6 0
1.~45 , 2 . 4 5 , 3 .55 , 4.70
,
* 0.00
0. 30 , 0.65 , 1.5 0 , 2.50 , 3 . 7 0 , 14.90
,
* 0 .00 , 0 . 30
0. 65 , 1 .5 0 , 2.50 , 3.70 , 5.00 /
,
DATA CN1P /
~+0 1 O , +U.40 ,+O .140 ,+O .OO , 0 .65 , 1 . 1 5 , 1 .50
*+O.25 ,-s-O. 145 ,Q~ L43 ,-s-O .10 , 0 .50 , 0.80 , 0.85
~ +O .OO , O .2 5 ,+O .20 , O .30 , O .90 , 1 . O 0 ,O. 70
* ...O 5 , +O .0O ,+0 .O0 ,0.35 ,O.90 , 1 . 1 0 ,O .50
~
*_ Q U) , +~).1C ,+0.10 ,-O .2O , O .55 ,O .45 ,O .50 /
DA TA C L U 1 /
.049 , .050 , .055 , .056 , .066 , .069
* . 05 3 ,
*
.0D ~4 , .0 5 1 , .050 , .053 , .057 , .062 , .061
*
~~~~ , .0 46 , .053 , .055 , .062 , .065 , .066
*
.1)55 , .053 , .053 , .054 , .056 , .060 , .061
*
.048 , .046 , .0115 , . 0 45 , .047 , .050 , .051 /
DATA CMD 1/
1~~
185
190 , .195 , .205 , .23 0 , .2L1 5
* .2O5
190
195 , .215 , .230 , .240
1*~
*...,2 (J5 , - .2 00 , - .21 0 , .225 , .2115 , .255 ,- .27O
* ... .220 , .21 0 , . 2 10 , . 220 , .225 , .211O ,- .255 ,
180
190 ,- . 185
180
185 , .205 , .220 /
DATA CL D1/
* .0235 , .0240 , . 0 2 5 5,
.0285 , .0 32 5 , .0320 , .034 5 ,
0 , .0280 , .0295, .0325 , .0355, .0370 , .0370 ,
* .02 (
*
.0285 , .0300 , .0315, .0335 , .0370 , .0375 , .0345!
.~~~~~~~~ ,

~~~ .

- .

,~~~~~

~~~ .

--~

------ -

-~~~~~

--~~~~-

_---~~~~~~~~~~ --- - - -

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

IuUI
~~

DATA CY4/
0.00 , 0.05 , 0.1 4 , 0.30 , 0.55 , -1.05 ,
* 0.00 , 0.06 , 0.16 , 0.32 , 0.57, 1.07,
* 0 . 0 0 , 0 . 0 7 , 0.18 , 0.35, 0.63, 1 . 1 0 ,
* 0 . 0 0 , 0.10 , 0 . 2 3 ,
0.45, 0.80 , 1 . 140 ,
* 0.00 , 0.08 , 0.20 , 0.40 , -0.70 , 1.27,
* 0.00 , 0.06 , 0.15 , 0.30 , 0.60 , 1 . 1 3 /
DATA CN14/
*
0.00 , 0.05 , 0.15 , 0.43, 0.97 , 1.85 ,
*
0 . 0 0 , 0.05 , 0.12 , 0.~48 , 1.05, 1.92 ,
2.12 ,
*
0.00 , 0.07 , 0 . 2 3 ,
0.55 , 1.1 7,
1.65,
3.05,
*
0.00 , 0.10 , 0.30 , 0.80 ,
*
0.00 , 0.06 , 0.25 , 0.70 , 1.55 , 2.95 ,
*
0 . 0 0 , 0.0 14 , 0 . 2 0 ,
0 . 6 5 , 1.50 , 2 . 8 5 /
DATA CL14/
.005 , .016 , .036 , .067 , .110
* .000 , .002 ,
.001 , .002 , .006 , .018 , .042 , .020
* .0 00 ,
*
. 0 0 0 , .0 00 , .000 , .003 , .013 , . 0 0 9 , .0 1 3 /
DATA CL2!
*
.000 , . 005 , .013 , .0314 , . 0 6 6 , .062 , .010
*
. 0 0 0 , .0014 , .013 , . 03 4 , . 0 8 0 , . O3b , .055
*
. 0 0 0 , .003 , . 0 0 7 , .023 , .044 , .030 , .0 1 O /
DATA C Z 2 /
.1)6 ,
*
.0 2 ,
.00 ,
.25 ,
.55 , 1.0 0 , 1.58
*
.00 ,
.02 , .06 , .25 , .58 , 1.04 , 1.63 ,
*
.00 ,
.25 , .55 , 1.00 , 1 .58 ,
.02 , .06 ,
*
.00 ,
.02 , .06 , .25 , .55 , 1.00 , 1.54 ,
*
.02 ,
.06 ,
.52 ,
.83 , 1 . 1 2 /
.00 ,
.25 ,
DATA CM2/
* ..O~ 0O , O. O5 , O . 1 0 , 0 .~40 , 1 .05 , 2 .05 ,3 . -4O
* ...0 00 ,~~ 0 . 05 , u . 1 0 , _ 0 .140 ,1. 05 , 2 .20 ,3 .75
* _ Q . tJO ,~~0.00 ,O .O5 , O .3O ,O.9 5 , 2 .10 , 3 .5O
2.45 ,3 .90
* ...Q.OO ,.
~0.05 , 0 .10 , 0.45 , 1 . 1 5 ,
*_ O~~OO , 0 .05 , O . 1O , O .55 , 1 . 3 5 , 2 .~40 , 3. 6 /
DATA CM Q/
* 1. 75 , 2.7 4 , 3.06 , 3 .07 , 2.88 , 2.55,
* 1.75 , 2.73,
3. 02 , 3. 03, 2.87 , 2.57,
* 1.75 , 2.68 , 2.98 , 3 . 0 0 , 2.88 , 2.63,
* 1.75 , 2 . 7 1 4 , 3. 14 , 3.3 0 , 3 .22, 3.26 ,
* 1.75 , 2.7 14, 3 . 1 1 4 , 3 . 3 0 , 3 .25, 3 .3 14,
* 1.75 , 2.79 , 3 . 2 0 , 3 . 3 8 , 3 .50, 3.59!
DATA CLP/
* .03 , .057 , .072 , .079,
.081 , .076 ,
~~
* . 038 , .057 , .072 , .079, .081 , .077,
*
- .038 , - .057 , .072 , .079, .082 , .079,
106 , .117 ,
* . 042 ,
.06 1 , .078 , .093,
*
.04 1 , .059, .07 6 , .090 , .103, . 1 1 3 ,
*
- .038 , .053, .068 , .083, .099, .114!
DATA NC 6/6 , LI , 2L1 , 0/
DATA BAb/0., 2. , 4 .,8. , 12 ., 16./
DATA V 19!.85 , .95 , 1. 05 , 1.25/
D A T A LX C P F /
* .OOu O ,. 0000 , .000 0 , . 0 000 , . 0 0 0 0 ,. 0000 ,
* .0000 , .0000 , .0000 , .0055 , .0075 , .0085 ,
* .O1 L IO , .0 135 , .0125, .0120 ,.0 1 15 , .O 1 1O ,
* . U 4 1 5 , . ~ ) , 41O , . O ~~~Q , . 0 2 8 O , . 0 1 9O , . Q 1 5 0 /
i~A TA NC7/~~,J , 27 , O/
DATA BA (/O.C , 2.0 , 14 .1 , 6.1 , 8.1 , 10.2 , 13 .2 , 16.3, 19 .3 /
*

UAT A

L.
~~

DAIA
-

V: ~~( / . 5 , . 9 5 , 1. 25 /

DCMF/

64

_ _ _ _

_ _ _ _ _ _ _
_ _ _ _ _ _ _ _ _ _ _ _
___

* .0139, .0 4 94 , .1 190 , .2259, .285b , .34 7 9 , .4 59 8 , .4 968 , .657 1,


* .05 1O , . 0 8 4 8 , . 1798 ,.2 990 , .3 7 6 4 ,. 4 8 3 3 , . 5 5 1 9 , .6 1 1( , .7 5 00 ,
* .0 276 , .0826 , .1 630 , .2792 , .3 3 8 1 ,. 11157 , .4 9 9 5 , .5 14 36 , .5379 !
DATA NC8 /3, 0!
DATA VMB / .5 , .95 , 1.25/
DATA DCMO / .O1 39, .0510 , .0276/

END

65
_

~~~~~~

- -_ -

-- -

SUBROUT INE C l
COMMON C ( 3 83 0 )
Dl ME N ~~iUN BDELTC (4) ,VAR (1O 1 )

C
C** INPUT DATA
)
EQ U I V A L E N C E (C( 86O ) ,TDY
EQUIVALENCE (CC 861) ,GBIAS )
)
EQ U I V A L E N C E (C( 862) , GN
)
E Q U I V A L E N C E (C( 863),WN2
)
E Q U I V A L E N C E ( C ( 8614 ) ,WN 1
)
EQUIVALENC E (C ( 8 6 5 ) ,W L
E Q U I V A L E N C E (C( 866),W LXX 1 )
EQUIVALENCE (CC 867),WLXX2 )
EQUIVALENCE (CC 868) ,WLJK 1 )
E Q U I V A L E N C E (C( 869 ), WLJK2 )
)
E Q U I V A L E N C E (CC 870),HJK
)
E Q U I V A L E N C E (CC 871 ) ,WXX
)
E Q U I V A L E N C E (C( 872) ,DXX
)
(CC
EQUIVALENCE
873),WJK
)
E Q U I V A L E N C E (C( 874 ),DJ
)
EQUIVALENCE (CC 875) ,GXX
)
E Q U I V A L E N C E CCC 876) ,GJK
)
EQ U I V A L E N C E (CC 877) ,RES
)
E Q U I V A L E N C E (CC 878),QDN
)
E Q U I V A L E N C E (CC 879),QUP
)
EQU IVAL [NC E (CC 890),HXX
(CC
EQUIVALE N CE
892),QBIAS )
EQUIVALENCE (CC 893),RBIAS )
E Q U I V A L E N C E (CC 899),OPTC 1 )
)
EQUIVALENCE. (CC 947) , GNS
)
E Q U I V A L E N C E (CC 948),WS1
)
E Q U I V A L E N C E (CC 9149), WS2
C
C* *INPUTS FROM OTHER MODULES
)
E.~U I V A L E N C E (CC
77),SPH I
)
EQUIVALENCE (CC
87),STI-IT
)
E Q U I V A L E N C E (CC
97),SPSI
E Q U I V A L E N C E CC( 353) , BPH 1
)
)
E Q U I V A L E N C E (CC 3514 ), BTH2
)
J
I
V
A
L
E
r
E~~
~CE (CC 355),BPS 1
E Q U I V A L E N C E ( CC 14 O 3 ) ,WLAMQ )
E~~ !~~V~ L ENCE ( CC 4 0 7 ) ,WLAMR )
)
E Q U I V A L E N C E (CC 1461) ,CAGE
)
EQUIVALENCE (C ( 4 6 2 ) ,TKRZ
)
E Q U I V A L E N C E ( CC 4 6 3 ) ,T K RY
)
E Q U I V A L E N C E ( C ( 1 2 3 3 ) ,BDR
)
E Q U I V A L E N C E (C(171 47) ,WR
EQUIVALENCE (C (171 43) ,WQ
)
E Q U I V A L E N C E (C(1739) ,WP
C
C * * I N p U r s FRJi 1 MAIN PROGRAM
E Q U I V A L E N C E (C(2000), T
C
C* * STATE V A R I A B L E OUTPUTS
E Q U I V A L E N C E (CC 800),WLQSDD )
E Q U I V A L E N C E (CC 8 0 3 ) ,W LQSP )
E Q U I V A L E N C E CC C 804), WL QSD )
)
E Q U I V A L E N C E (CC 807) ,WL QS
E Q U I V A L E N C E (CC 808),WLQSSD )

--

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

-- - -

- -

EQUIVALENCE
EQUIVALE NCE
EQUIVALENCE
EQUIVALENCE
EQ U I V A L E N C E
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQ U I V A L E N C E
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQ U I V A L E N C E
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
E Q U I V A L E NCE
E Q U IV A L E N C L
EQUIVALENCE
EQUIVALENC E.
EQu IVALENCE
EQUIVALE N CE
EQ U I V A L E N C E
EQUIVAL ENCE
EQUIVAL L N \ E
EQ U I V A L E N C E
EQ U I V A L E N C E
EQ U I V A L E N C E
EQ U I V A L E N C E
EQUIVALENCE
E Q U iV A L E N C E
EQUIVALENCE
E~ U I V A L E N C E
E.O J I V A L L N C E
f-L c J IV LEN C E
~
E~~u IV A L E N C E

~~~~~~

(CC 8 1 1 ) ,WL Q SS
(CC 812) , WLRSDD )
(C( 815) ,WL R SP )
C C C 816) , WLRSD )
)
(CC 819) ,WLRS
(CC 820) ,WLRSS D )
(CC 823) ,WLRSS )
(CC 82L1) ,BLQSSD)
(CC 827) , BL Q SS )
(CC 828) ,BLRSS D)
C C C 831 ) , BLRSS )
(C ( 832) ,BJ .JSDD)
(C ( 835) ,BJJS P )
(C( 836) , BJJSD )
)
(C( 339),BJJS
(C( 840) ,BKKSDD )
(CC 843) , BKKSP )
(C( 8LILI ),BKKSD )
)
(CC 8147) , BKKS
(CC 8448) , BXXSDD )
(CC 851) , LxxSP )
(C( 852) ,BXXSD )
)
(CC 855), BXXS
(C( 931 ) , BJ SSD ) , CCC 934) , BJSS
(CC 935),BKSSD ), (CC 938),BKSS
), (CC 953),SNP1
(CC 950),SNP2
), (CC 96O) ,SNQ1
(CC 957),SNQ2
) , (C( 967),SN R 1
(CC 9614 ), SNR2
) , (CC 974) ,BPC 1
(CC 971 ) ,BPC2
(
C
CC( 903 ) ,H 13P) , (9 04 ) ,H13M )
(CC905) , H214P) ,(C (906) ,H2 14 M )
(C(907) , CDRFT 1 ) , CC (908) ,CDRFT2)
(C(909) ,C D R F T Y )
(C(98L1 ), C D R F T X )
(CC 1676) ,ANUX )
(CC978 ) , B D R F T D ) , (C(981) ,B D R F T )
(C(985) ,N LMT 1 ) , (C(986) ,NLMT2 )
(C(987) ,BJJSSS) , (C (988) ,BKKSSS)

- ~

-- -_
~

~~~~~~~~~~~~~~~~~~~~

)
),(C(
),(C(
) ,(C(
) ,(C(

956),SNPO
963),SNQO
97O) ,SNRO
977),BPCO

E Q U I V A L E N C E CCC 989 ) ,BXXSSS)


EQUIVALENCE. (C(990) , BJJ SSL ),(C (991) ,BKK SSL )
EQUIVALENC E CCC 5 3O ) ,BJSDD) , (C(5314) ,BJSD ),CC (537), BJS)
E Q U I V A L E N C E (C(538) ,BKSDD) , (C(5L12), BKSD), (C(5145) ,BKS )
E Q U i V A L E N C E (C(5~46) ,BXSDD ) , C C C 5 5 O ) ,BXSD) ,CC (553 ) ,BXS )
E..- U I V A L E N C E (C(533) ,CJSD) ,CC (5 141 ) ,CKSD ) ,(C(549), CXSD )
EQU i V A L E N C E (CC942) ,WL2 )
EQ U I V A L u N C E
EQU iVALENCE
EQUJVALENCE
EQ U I V A L E N C E

EQ U IV ~ LENC E (C(2965) ,VA R( 1 ))

C * * OUT PU T S
C

CC( 943 ) ,DJ2)


CCC9 L I4 ) ,W J 2 )
(C(945) , DX2 )
CCC 94 6) ,WX2 )

EQ U i V A L E N C E

CM*OT HE. R OUTPUTS

EQUIVAL E NCE

EQ U I V A L E N C E
E QU I V A L E N C E
EQUIVALENCE
EQUIVALENCE

(C( 856) ,BDELTC ( 1 ) )

(CC 88O) , BPFiIS

(C(518), B13SS)

CC (5 1 9 ) ,B211SS)
)
CC( 881 ) ,BJJ
CCC

882), BKK

~~~~~~~~~~~~~

~~~~~~~~~~~~~~

-:

T i

EQUIVALENCE

EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE

~~~~~~~~~~~~~~~~~~

__ _ _

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

(CC 883),BXXSS )
(C( 884),BJJ S S )
CCC 885) , BKKSS )
886) ,BTHTS
8 8 7) ,BPSIS

(CC
( CC

)
)

C** GUIDANCE SIGNAL SHAPING


C**GUIDANCE SWITCHING
WLQSD = WL Q SP
W LRSD z WL R SP

W N 2 * ( W N 2* ( W L A M Q
WLQS )
WLRSDD = W N 2 * ( W N 2* ( W L AM R
WLRS)
G N * ( W L Q S D / W L + WLQS ) + QBIAS
WQC
WRC = G N * (W L R S D / W L + W L R S ) + RBIAS
WLQSDD

IF

2. W L Q S D )
2. *WLRSD

GB IAS

.AND. T.GT .TDY) GO TO 4

C T K R Z . G T .0 .

W LQS DD = 0.
WQC = QB IAS + GBIAS + QDN
IF (CAGE .GT . 0. .AND. T.GT . TDY) WQC
14 IF (TLKRY .GT . 0.) GO TO 5
WL RSDD
0.

W RC

RBIAS

I F( C A G E .G T . 0 .) WRC = WLAMR
5 CONTINUE
W N1 M ( WQC
WLQSS )
WLQS S D
WLRSS )
WLRS SD = W N 1 * ( W R C
IF ( W N 1 .GT . 0.) 50 T O 3
WLQSS = WQC

WLAMQ + QBIAS + GBIAS

RBIAS

WLRSS
3 BLQSSD
BLRSSD

WRC

WLQSS

WLRSS
C
C**RATE GYRO DYNAMICS AND LIMITING
BDRFTD= (CDRFT1*ANGXCDRFT2)

-~~~~~~~

BTHTS= BTH2BDR FT
BPSIS=_ BPS 1 +CDRFTY *B DRFT
B PHI S=_ BP H1C DR FTX*BDRFT
(WQ*C OSD(_ BPH1 ) _WR*SIND( _ BPH 1 ))
BPHISD = W P
S IND( ...BPS1)/C OSD ( ...BPS1)
*
BPH S = BPHISD / WLXX2 + BPHIS
6 IF(GNS .LE . 0.) GO TO 8
SNP2
WS 1*W32*(GNS*SP1~
(W S 1 +W S 2 ) S N P l
SNQ2
(WS1+WS2)SNQl
~I S1 *~~S2*(GNS*STHT_ SNQO )
S N R 2 = W S 1* ~~S 2 * C G N S * S P S I_ S N R O )
-

1 I_ S N P 0 )

B PH S

BP1IS - S N P 1
B i H T S = BTHTS
SNQ 1
BPSIS = BPLIS
SNR 1

( W S 1 + WS 2 ) S N Rl

8 CONTINUE

BXX
BPHS
BTI- ITS
BTSS
BPSS
BPSIS
C *****3PEC IAL CASE

P R O G R A M M E D FLIGHT
IF (OPTC 1 .LE . 0.) GO TO 9
BLQSSD = QB IAS

BLR SSD
BDC

BT=O.
BP~ 0.

0.

0.

IF(T .GE .1.0000 .AND. T.LE .3.2000)BDC :5.


.A N D .
T . L E . 6 . 14000)BT : 5.
IF (T.GE .7.6000 .AND . T . LE .9 .9 0 0 0)B D C= 5 .
I F ( i. GE . 4 .2

~~~~~~~~

~~~

~~~~

~~~~~~~~~-~~~~~~-

-~~~~~~~ ---

--

-----~~

-~~--

_ : T i
~

ITT ~~~~~~~~~~~~~~~~~~

--- -~~~~ r,- - - - ,-==

~~~~~~~~~~~

~~

==~~~~~~:-

- -~~~~~~~~~~ = ~

IF (T. GE. 11.000 .AND. T.LE. 13 . 100)BP= 5 .


IF (T.GE.15.200 .AND. T .LE .17.300)BDC : 5.
I F (T . G E . 1 8 . 320 .AND. T . L E .2 0 .56 0 ) B T : 5 .
IFCT.GE.2 1.5L 1 5 .AND. T.LE .23 .675)BDC :5 .
I F( T .G E . 2 1 1 . 770 .AND. T.LE.26. 910)BP:5.

C C52 0 ) =BPC O+B PBT


BPC2 = 1 8 . 1 8 * 2 0 . 5 7 * ( B D C_ B P C O )_ ( 1 8 . 1 8 + 2 0 . 5 7 ) * B P C 1
BXX
BXX + BP CO
BTSS=BTSS-BT

BPSS = BPSSBP

C**SUMMA TI ON OF RATE DAMPING AND GUIDANCE SIGNALS AND THEIR DERIVATIVES


(BLR SS - BPS S) (BLQSS
BTSS)
9 BJJ

BKK

( BLRSS BPSS )

C
C* * G U I D A N C E SIGNAL SH A PING
BXXSD = BXXSP

BJJSD
BKKSD

( BLQSS BTSS)

AND LIMITIN G

B JJS P
BKKSP
BXXSDD
WXX*( WXX*(BXX - BXXS ) - 2 . *DXX*BXXSD )
BJS D=CJSD
BKSD =C KSD
B XSD C XSD
HSATz23 .
IF(ABSCBJJS).LT.HSAT)G0 TO 20
BJJS :SIGN(HSA T ,BJJS )
VAR(NLMT 1+ l )= BJJS
IF( BJJS*BJJSD GT. 0.O)BJJSD=O .0
=

20

25

BJJSDD=W JK *(WJK* (GJX *BJJ_ BJJS)_ 2 .*DJK*BJJSD)


IF (ABS(BKKS).LT .HSAT)G0 TO 25
BKKS :SIGN (HSAT ,B KK S )
V A R (N L M T 1 + 3 )= BKKS
IF (BKKS*BKKSD .GT .0 .O)BKKSD:0 .O

BKKSD D=WJK* (WJ K *(GJK *BKK_ BKKS)_ 2. *DJK*BKKSD)


B X X S S = G X X * ( ( B X X S D D + ( W L X X 1+ W L X X 2 ) * B X X S D ) / ( W L X X 1 * W L X X 2 ) + B X X S )
BJJSS:BJJSD !W LJK 1 +BJJS
IF(ABS(BJJSS) .GT. H S A T )B J J S S = SI G N ( H S A T , BJJSS)
BKKSS=BKKSD !WLJK 1 +BKKS

**

30

35

10

L~

IF (A 83 (BKKSS).GT.HSAT)BKKS S:SIGN (HSAT , BKK SS)


tI IGH FREQ UENCY SHAPINF OPTION

IF(1 .X2 *WJ2 .LE .0 .)GO TO 10


IE (ABS(BKS).LT.HSAT)GO TO 30
BiC$=S IGNCHSAT ,B K S)

V A R ( N LA T 2 + 3 )= B K S
IFCBKS*BKSD .GT .O.O )BKSD :0.O
z3KSDD W J 2 * C W J 2 * C B K K S S_ B K S ) _ 2.*DJ2*BKSD )
IE (ABS (BJS) .LT .I-iSAT)GO TO 35
BJS = SIGN(HSAT , B J S )
VA R ( N L M T 2+ 1 )=BJS

IF (BJ3*BJSD. GT .0.0)BJSD=0 .O
BJSDD ~~ J 2* (WJ2* (BJJS S_ BJS )_
2.*DJ 2*BJSD )
BX SDD ~ WX2 * (WX2 *(BXXSS_ BXS)_ 2. *DX 2*BXSD)
BKK SSS=BK SD/wLJK2 +BK S
IF (A B S (BKKS3G ) .GT .HSAT)BKKS SS=SIGN (HSAT ,BKKS SS )
BJJ SSS=BJSD/WLJK2+BJS
IF(ABS (BJJ3SS) .GT.HSA T)BJJSSS :SIGN (HSAT ,BJJ SSS )
BXXS 3S=BXS
BJJSSL :BJJSSS
BKKSSL :BKKSSS
IF (BJJSSL.GT .H13P)BJJSSL :H13P
69
_

- -

~~~~ r ~~

-=

~~~~~~

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~

IF(BJJSSL.LT .U13M)BJJSSL :H 13M


IF( BKKSSL. GT. H2 4 P) BKKSSL = H24P
IF ( BKKSSL. LT. H24 M) BKKSSL:H2 14M
B13SS :BJJSSL
B24SS= BKKSSL

C
C**COMMAND S TO ACTUATORS
BJSSD : 20 . * ( B 13 S S_ B J S S )
BKSSD = 20 . * ( B2 ~4SS _ B KSS )
B13SS = BJSS D!12 5 .+BJSS
B24 SS
B K S S D / 12 5 . + B K S S
BDELTC ( 1 ):B13SS+BXXSSS
BDELTC (2 ):B24S S+BXXSSS
BDELTC( 3 )=B 1 3SS BXXSSS
B D E L T C (4 ) = B 24 S S B X X S S S
RETURN

END

~~:

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

~~~~
~~~
- -- -

-------r--

------

- ---

IPL(N
) 800
I P L( N + 1) = 8 0 4
IPL( N+2 )
808
I PL( N +3 ) = 812
IPL (N+ 4) : 8 1 6
I PL( N + 5) : 820
I P L( N+6 )
82 11
I P LC N + 7 ) : 828
IP L( N+8) = 8 3 2
NLMT 1= N+9
IP L( N+ 9 ) = 836
I P L( N+ 1O )
8 14 0
IP L ( N + l 1 ) = 844
IPL (N -12) :848

I P L( N+ 1 3 )
N : N +14
C( 80 3)

CCC

852

0.

C(8O 7)
0.
C (811 )
0.
C (8 15) = 0 .
C(819) = 3.
C(8 23)
0.
C (827) = 0.
C(831 ) = 0.
CC8 35 ) = 0.
C (839) = 0.
C(8143)
0.
0.
CC847 )
CC85 1 )
0.
CC855 )
0.
CC 8 1 1 ) = CC 378)
CC 82 3 )
CC 8 7 9 )
C ( 11 13 )
CC 831)
CC 8147) =C( 831)
CC 831 )
CC 839)

PROGRAMMER

IF(CC899) .LE.O.) G0
I P L (N) 971
I P L ( N 1) - = 9 7 4
N N +2
CC 97 14 ):0.

C (861)

TO 9

CC 977 ):0.

I F C C ( 9 4 9 ) .LE. 0 .) GO TO 10
9
C * * G Y R O POT NOISE
) : 950
IPLCN
I P L( N+ 1 ) = 953
I P L( N2 ) : 957
IPL(N+ 3 ) : 960

IPL (N+11 )

964

IPL(N+5 ) 967
N : N + 6
C( 953) : 0.
-

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

SUBROUTINE C l I
COMMON C (3830)
DIMENSION IPL (100)
)
EQUIVALENCE ( C ( 2 5 6 1 ) ,N
)
EQUIVALENCE ( C C 2 5 6 2 ) ,IPL( 1)
EQUIVALENCE ( C ( 9 8 5 ) ,NLMT 1) ,( C ( 9 86 ) ,NLMT2)

L.

~~~~~~~~~~~~~

~~~~~~~~~~~~~

-~~~~

~~~~~~~~~~~~~~~~~~~~~~

T~~

~~II~~~~

~ ~~~~

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

:
T
T
T ~~~~~ -

C( 956) : 0.
C( 960)
0.
C( 9 6 3 ) = 0.
C( 967)
0.
C( 9 7 0 ) = 0.
10 CONTINUE
I F ( C ( 9 4 4 ) * C ( 9 14 6 ) . L E . O . O ) G O TO 20
I P L( N) = 5 3 0
IPL (N +1 ):5314
NLMT2 = N+ 1
I P L( N+2 ) = 5 3 8
I P L( N+ 3 ) = 54 2
IPL( N+ LI)=5 46
IP L( N+5 ) :5 50
W = N+6
C( 533) : 0 .O
C ( 5 14 1 ):0., 0
C (549)=O .0
C(537)=0 .0
C (5L15):0.0
CC 553 ) :0 .0
20
CONTINUE
)
IPLCN
931
IPL(N +1 )

935

N =N +2
C( 934) : 0.
0.
CC 938)
IPLCN)=97 8
N= N +1
CC981):0 .0
RETURN

END

7
_ __ _ _ ~ __

_ _

_ _ _ ____

__

___________

~~~~~~~~~

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

--

~~~~~~~~~

- .
~~~~~~~~~~~~ : --

SUBROUTINE CLI
C
C** HELFIRE SIMPLIFIED ACTUATOR MODEL
C****** NON
LINEAR MODEL ~~~~~~
C
COMMON C ( 3 8 3 O )
DIMENSION BDSS(14 ),BDSC 4 ) ,BDSD ( ~I )
DIMENSION BDELTD(LI ),BDELT (4),BDELTC (4 ),V A R (1O 1 )
DIMENSION BDLT(4) , BDT(4)
14 )
WDS2D (LI ), W D S 2 ( 4 ) , WDS1D(14 ), WDS 1 (
DIMENSION
DIMENSION IPL (101 )
DIMENSION NC2(2) , CB2C ) , CHAF(6)
DIMENSION
G1 (14),G2C 14) ,G3 (4),W 1 ( 4 ) ,ZN(4),WN (14)
DIMENSION H1 (4) , H2C 4 ) , BH (14 )
DIMENSION C B 1C 6 ) , CHDU(6) , CHDL(6)
DIMENSION AIH( 14 )
EQUIVALENCE (C(521) , A IH (1 ))
EQUIVALENCE ( C C 1 1 1 b ) , G3 ( 1) )
E Q U I V A L E N C E C C ( 1 1 2 U ) ,G 1 C 1 ) )
EQUIVALENCE (C (112 ~4) ,Z N C l ) )
EQUIVALENCE ( C ( 1 1 2 8 ) ,W N ( 1 ) )
EQUIVALENCE (CC 113 2) ,I 1 ( 1 ) )
~
EQUIVALENCE ( C ( 1 1 3 6 ) ,G 2 ( 1 ) )
E QUIVALENCE (C(11L 18 ) , H 1 C 1 ) )
E Q U I V A L E N C E CC C 1152) , H2C 1 ) )
E Q U I V A L E N C E (C (1 156) , B H C 1 ) )
C

C**INPUT DA TA

EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE

(C (11L I0) , OPTACT )


( C C 1 1 I 4 1 ) ,BDP
CC( 11 42) ,BDQ
)
(C(11 L 13 ) ,BDR
(C( 1 14 1 4) ,EFF
)
( C C 1 1 ~45) ,HX
(C (1146) ,BDB
(CC1 1L I 7 ) , FM 8S
)
(C(13O 6) , R F A R E A )
(CC1307 ) , RFLGTH )

C
C**INPUTS FROM OTHER MJDULES
EQUIVALENCE CC (0203 ) ,PDYNMC)
E Q U I V A L E N C E (C( 204),VMACH
EQUIVALENCE CC( 367),B A L P H A )
EQ U I V A L E N C E (CC 368), BALPHY )
E Q U I V A L E N C E (CC 377), BALPD )
EQ U I V A L E N C E (CC 378),BALYD )
EQUIVALENCE (C( 856) , BDELTC(1))
EQUIVALENCE (CC 12 SLI ) , DELTB)
EQ U I V A L E N C E (C(1192) ,BDTC1 ))
E Q U I V A L E N C E (C(1 196) ,BDLT(1 ) )
EQUIVALENCE CC (525) , BDSSC 1 ))
E Q U I V A L E N C E (C(1310 ), FfrIH2 )
)
EQUIVALENC E ( CC 1 3 1 1 ) , FMH3
)
E Q U I V A L E N C E (C(1312 ) ,FMH I4
C
C * * I N P U T S FROM MAIN PROGRAM
)
EQUIVALENCE ( CC 2 0 0 0 ) ,T
)
(C(
2013)
EQUIVALENCE
,DOC
)
EQ U I V A L E N C E (C 2561 ) ,N
~~--------- -~~~--- -~~---

- - -

---

--

__________________________________________

EQUIVALENCE

DATA
DATA
DATA
DATA
DATA

- ~~~~~~

--

--- -

-- ~

(C(2562) ,IPLC 1 ))

WA TLB 1 /LI HC H1LI/ ,WATLB2 /LIHCH23 / ,WATLB 3 /4HFMHA/


NC2/6 ,0/
CB2/
.00 ,
.50 , . 8 5, 1.05 , 1. 40 , 1.60/
.4 , 1.6 , 3 .1 , 3.6 /
CHAF! 0. ,
.1 ,
C B 1 / . 0 0 0 0 , .8500 , .95 00 , 1.050 , 1.2 5O , 1. 6 0 0 /

DATA CHDU !.0007, .0007 ,.0007 ,.OO2O ,.OO 2O ,.O020/


DATA CHDL/.0007 , .0007, .0018 , .0018 , .0018 , .0018/

C~ STATE VARIABLE OUTPUTS


BDELTUJ)
B D E L T C C 1) - BDP + BDQ BDR
BDELTC (2 ) = BDELTC(2)
BDP + BDQ + BDR
BDR
BDELTC(3)
BDELTC(3 ) + BDP + BDQ
BDELTCC 14) = BDELTCC 4 ) + BDP + BDQ + BDR
BDELT( l )=C (11O 3 )
-

BDELT (2):C(1107)

BDELT(3 )=C( 1 1 1 1 )
BDELT(14 ):C(1 1 1 5 )
BD S( 1 )~~C (1O87 )
BDSC2)=C( 10 91)
BDSC 3 )= C C 1095)
BDS(14):C(1099)
WDS2 (1)
CC 11b 3 )
WDS2 (2 ) = C ( 1 16 7 )
WDS2 (3) = CC 1 1 7 1 )
WDS2 (4) : C C 1 1 ~~5)
W DS1 (1) = C( 1179)
WDS 1 (2)
CC 11 83 )
WDS 1 ( 3 ) = CC 1187)
WDS 1 (4) = C ( 1 1 9 1 )

C
C**ACTUAT OR DYNAMICS
X K :O .
VM : A M I N 1 ( V M A C H , 1. LI)
CALL TABLE(VM , CB1 ,CH DU ,NC2( 1) ,XK ,WATLB 1 ,CH114 )
CALL TABLE(VM ,CB1 ,CHDL , N C 2 C 1 ) ,XK ,WATLB2 ,CH23 )
CALL TABLE(V M ,CB2 ,CUAF ,N C 2 ( 1 ) ,XK ,WATLB 3, FM HA )
FM14 : CH1 )4*PDYNMC *RFLGTH*RFA REA* 12 .
CH2 3 *PDYNMC*RFLGTU*RFAREA*12.
FM23
DO 30 1:1 , 14
(I+1)!2
J

FMHD

FM 1LI

I F C IA B S ( ( 2 * I_ 5 ) !2 ) .LE. 0) FMHD : FM23


F M H= F M H D * B D E LT ( I ) + F M H A * ( B A L P H A + B A L P H Y * (_ 1) * * I )
BDE :WDS 1 (I)-FMH !G2 (I)
W D S 2 D ( i) :
~4 N ( I )* ( G3 ( I) *B D E_ W D S 2 ( I) )

BDELTD(I)=WDS2 DCI)/W 1 (I)+WDS2(I)

B D S D( I) = 6 5 . * ( B D E L T D ( I) ! 12 5 . + B D E L T( I)_ B D S( I) )
BDSS(I )=BDSD (I )/125. +BDSCI)

BDH=G1 (I)* (BDELTC (l )_BDSS (I))

A I U ( I) : B D H
IF( BDI- I.LT. H 2 ( I ) ) B D H : H 2 ( I )
I F( B D H . G T . I- I1( I)) BDH= 1- 11(I)
WDS 1D( I ) : W 1 CI ) * ( B D H _ W D S 1 C I ) )
I F( B DT C I ) .LT . B D E L T ( I) B H ( I) ) B D T ( I) : B D E L T ( I) B H( I)
IF(BDT (I).GT.BDELT(I)+BHCI))BDT (I) :BDELT(I)+BH (I)

BDLT(I ):BDT (I)

C* * S U R F A C E POSITION LIMITER
IF((ABS (BDELT(I)).GT . 19.).AND. (BDELTD (I)BDELT (I).GT.O.))BDELTD( I)

L~~~ L.

-~~~~-

- -

-~~~~~

- -

_
_

_
_

30 CONTINUE
C ( 1 0 84 ) = B D S D ( 1)
C ( 108 8 ) = BDS DC2 )
CC 1 092 ) = B D S D ( 3 )
C( 1 096 ) : B D S D(4 )
C ( 1 10 3 ) = BDELT(1)
C ( 1 1 07 ) = B D E L T (2 )
C( 1111)
BDELTC 3)
: B D E L T (4 )
C ( 1 1 15 )

C
C** OUTPUT D E R I V A T I V E S OF STATE
C ( l l O O ) = B D E L T D ( 1)
C( 11O4 )
BDELTD(2 )
C ( 1 10 8 ) : B D E L T D ( 3 )
C ( 1 1 1 2 ) z BDELTD(LI )
C ( 1 l 6 O ) : W D S 2 D ( 1)
C C 1 1 6 1 4 ) = ID S 2D ( 2 )
~
C ( 11 68) = WDS2D (3 )
W DS2 D(4 )
C ( 1 17 2 )
=
C ( 1 1 7 6 ) = wDS 1DC 1 )
C(1 180) = W D S 1D ( 2 )
C ( 1 1 8 4 ) = WDS 1D(3 )
r ( 1 1 88 ) = WDS 1D(4 )
C
-

..

VARIABLES TO INTEGRATION

C * * * * ** * * * * * * * * * * * * * * * * ** * * * * * * * * **

RETURN

END

--

IF (T.GT.0. .OR . DOC .GE .6.) RETURN


W R IT E ( 6 , 150) T
DO 100 I=2 ,N
I P L( I 1 )
J
1 00 WR ITE(6 ,2 00 ) J , C(J) , C(J+3)
CALL DUMPO
150 F O R MA T C F 1 O . 14)
200 F O R M A T ( I 1 0 , 1 P 2 E 1 5 . 7 )

--

-~~~

SUBROUTINE

- ---- -

-.

-- ---- -

- -- -

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

C4 I

COM MON C ( 3 8 3 0 )
DIMENSION I P L C 1 0 0) , IS N D X ( L I O )
EQUIVALENCE C C ( 3 6 3 14 ) , IS N DX C 1 ) ) , ( C ( 3 5 12 ) , 135 12)
DIMENSION BDLT(4)
G 1 C 4 ) ,G 2(14 ) ,G3( ~I) ,W 1(L l ) ,Z N ( 4 ) ,W N ( 4 )
D IMENSION
D IMENSION H 1(L I ) , H 2 ( 4 ) , B H ( 4 )
EQUIVALENCE ( C ( 1 12 0 ) ,G 1 C 1 ) )
EQUIVALENCE C C C 1 1 2 1 4 ) ,Z N ( 1 ) )
EQUIVALENCE ( C C 1 1 2 8 ) , W N ( 1 ) )
EQUIVALENCE ( C C 1 1 3 2 ) ,W 1 C 1 ) )
EQUIVALENCE ( C ( 1 1 3 6 ) ,G 2 C 1 ) )
EQUIVALENCE ( C ( 1 1 1 6 ) ,G 3 ( 1 ) )
EQUIVALENCE ( C C 1 1 14 8 ) ,H 1 C 1 ) )
EQU iVALENCE (C(1152 ) ,H2(1 ) )
EQUIVALENCE (C(1156) ,BH(1 ) )
EQUIVALENCE (C(11O 3 ) , BDELT 1)
EQUIVALENCE C C ( 1 1 0 7 ) , BDELT2)
EQUIVALENCE ( C (1 1 1 1 ) , BDELT3 )
EQUIVALENCE ( C C1 1 1 5 ) ,BDELT14)
E Q U I V A L E N C E (C(12147) , FELECB )
EQUIVALENCE (C(1248) ,FELECQ)
EQUIVALENCE (CC1 2 14 9 ),F E L E C R )
EQUIVALENCE C C C 1 2 5 0 ) ,FMECHB )
E Q U I V A L E N C E (C(1251 ) ,FMECHQ)
E Q U I V A L E N C E CC( 1252), F M E C H R )
E Q U I V A L E N C E (C(1140) , OPTACT)
)
EQUIVALENCE C C ( 1 1 4 1 ) , BDP
EQUIVALENCE (C C 1 1 1 4 2 ) , BDQ
EQUIVALENCE (C(1143 ) , BDR
EQUIVALENCE (C(1196), BDLTC 1 ))
)
E Q U I V A L E N C E CCC256 1 ) ,N
E Q U I V A L E N C E CCC2562) ,IPLC 1 ))
CC 37 7 )=0 .O
C(378):O .0
CC 1 192 )=O .O

C( 1 193 )=O.0

CC 1 1 9 4 ) = 0 . O
C( 1 19 5 ) = 0 . 0
IP L C N ) = 1 1 0 0
IPL(N+1 )
1 1 04
I P L (N + 2 )

IPL (N 3 )
N = N+ 14

1108
1112

) = 116 0
IPL (N+1) : 11 64
IPL(N+2) = 1168
IPL CN + 3 ) = 1172
I P L( N+ 4 ) = 1176
I P L( N +5 )
1180
I P L( N6 ) : 1 1 8 4
I P L( N+ 7 ) = 1188
N : N -s- 8
IP L( N) : 1 084
I P L( N+ 1 ) : 1 088
I PL(N+2) :1092
I P L ( N+ 3 ) = 1096
I PL (N

N = N + 14

C ( 1 08 7 ) : 0 . 0

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 76

~~
-

--

--

_____________________________

_ _ _ _ _

~
~ ~~~~~~~~~
- --

- -

---

~~~~~~~~~~~~~~~~~~~~~~~~

C ( 1 0 9 1) = 0 . 0
C (1095)=O .O
C (1099)=O .0
C C 1 1 6 3 ) = 0.
C ( 1 1 6 7 ) = 0.
C ( 1 1 7 1) = 0.
C ( 1 17 5) = 0.
C( 1 17 9 )
0.
C( 1183 ) = 0.
C C 1 1 8 7 ) = 0.
C 119i ) = 0.

RETURN

ENTRY

Al l

C
C MONTE CARLO FIN MISALIGNMENT ERRORS
C
FELECB
=
0.
FELECQ = 0.
FELECR = 0.
FMECHB
0.
FMECH Q = 0.
FMECHR = 0.
DO 1 0 I = 1 , 135 12
11)0
I
I F( I S N D X( I) . E Q . 1250) CALL MCARLO (DUM , 1 , IDO )
IF (ISNDX(I).EQ. 1251 ) CALL MCA RLO CDUM , 1 , IDO )
IF(ISNDX( I).EQ.1252) CALL MCA RLO (DUll , 1 , IDO )
C MONTE CARLO FIN OFFSET (MODULE C I4I AND C4 )
IF (ISNDX(I ).EQ. 12147) CALL MCARLO (DUM , 1 , IDO )
I F ( IS N D X ( I) . EQ . 1 2 4 8) CALL M CARLO ( DUM , 1 , IDO )
I F ( I S N D X ( I ) . E Q . 1 2 4 9) CALL MCARL O (DUll , 1 , IDO )
DELTB
=
FELECB
+ FMECHB
DELTQB = FELECQ + FMECHQ
DELTRB = FELECR + FMEC HQ
10 C ON T I UE
~~
- I F ( OPTA CT .LE. 0 .) GO TO 20
OPTACT = 0.
DO 5 I= l , 14
READ(5 , 200 )G1 (I) ,ZN(I),W N ( I ) ,W 1 (I),G 3C I) ,G2(I) ,
) H2 (
I) , BH ( I ) , 18
* f-l i (
~~ ,
5
W R I T E (6 ,300)I8 , I ,G 1(I ) ,ZN(I) ,WN (I ),W 1 (I),G3 (I) ,G2 (I) ,
*
H 1 (1) , H2 (I ),BH(I )
200 FO RM A T (9F8 .3, I8)
300 F O RM A T C I 6 , I2 ,9F8.3)
20 CONTINUE
BDELT 1 = B DP +
BDELT2 = -B DP + BD Q + BDR
E3 DELT3 = BDP + BDQ
BDR
BDELT4 = BDP + BDQ + BDR
BDLT (1) = BDELT 1
BDLT(2) = BDELT2
BDLT (3 ) = BDELT3
BDLT (14) = BDELT 4
BDQ

BDR

RETURN
END

Jjj

SU BR OU T I N E
DTA (XIN , J)
COMMON C (3830)
DIMENSION REF(Lf ),BIT(4)

EQUIVALENCE C C ( 3 4 5 3 ) ,R E F ( l ) )
EQUIVALENCE ( C C 3 4 5 7 ) ,B I T ( l ) )
R E F ( 1 THRU 4 ) IS BDELTCC1 THRU 14)
SIG = 1.
IF(XIN .LT.0 .)SIG = l .

X IN =AB S(XIN)

14

10

100

IF(XIN. LT.REFC
J~))GO ~ TO ~1O ~
RITE(6
14
)CC
2OO
0
) , X IN ,J
,
~
FORMAT (6H DTOA ,2E20. l0 , 15)
XIN = R E F C J )

~~~~

~~

GO TO 100

X IN :X I N A M O D (X I N ,B ITCJ))

X IN= S IG *XIN

RETURN

END

.
~
~~~~~~ ~~~~~~~~

78
~~~

_ _ _ _ _

_ _ _ _ _ _ _ _ _ _ _

_ _ _ _ _ _ _ _

_ _ _ _ _ _ _ _

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

SUBROUTINE DUMPO
CO MMON C ( 3 8 3 0 )
DO 100 1=1 , 1 8 00 , 7
N = 0
DO 200 J = 1 , 7

K : I + J 1

200 IF ( A B S ( C ( K ) )
100 IF (N .GT . 0)
-

.GT . i . E b ) N :
W R IT E ( 6 ,30 0)
*
I ,C(I),C(I+1 ) ,C( I+ 2 ) ,C( I+3) ,C( I+4 ) , C ( I+ 5 ) ,C ( I+ 6)
300 FORMA. T ( 1H - ,I5 , 1P7E.1S.7)
.
-

RETURN

END

_ _ _ _

~~ ~

--

~~~~~~~~~~~~ r

----

SUBROUTINE Dl
C* * T R A N S L A T I O N A L DYNAMICS MODULE
COMMON C (383 0)
C
C* * INPUT DATA
EQUIVALENCE (C(1627 ),AGRAV )
EQUIVALENCE ( C C 1 6 2 8 ) ,DMAS S
EQUIVALENCE ( C ( 1 6 2 9 ) ,AT H RST )
EQUIVALENCE (C (1630) , ATURNT)
EQUIVALENCE C C C 1 6 3 1 ) ,B GA M T )
E Q U I V A L E N C E CC(16 39 ), OPTARG )
EQUIVALENCE (C (1681) , AD IVE )
)
EQUIVALENCE C C ( 1 7 5 1 ) ,C R A D
C
C**INPUTS FROM OTHER MODULES
)
EQUIVALENCE (C(1300), FXBA
)
EQUIVALENCE (C(1301 ) ,FYBA
)
E Q U I V A L E N C E (C (1302) ,FZ BA
E Q U I V A L E N C E (C (17O3 ) ,C F A 1 1 )
EQ U I V A L E N C E . (C (1707) , CFA 1 2 )
E Q U I V A L E N C E C C ( 17 1 1 ) ,CFA 1 3 )
(C(1715 ) ,CFA 21 )
E QUIVALENCE
E Q U I V A L E N C E (C(17 19) ,CFA22 )
E Q U I V A L E N C E CC( 1723 ) ,CFA2 3 )
EQUIVALENCE ( C ( l 7 2 7) ,CF A3 i
E Q U I V A L E N C E (C(1731 ) ,CFA 32 )
E Q U I V A L E N C E (C(1735) , C FA 33 )
)
EQUIVALENCE (C(2000) , T
C
C* * S T A T E V A R I A B L E OUTPUTS
EQUIVALENCE ( C ( 1 O O ) ,VXE D
)
E Q U I V A L E N C E ( C C 1 b O 3 ) ,VXE
)
EQUIVALENCE ( C ( 1 6 0 1 4 ) ,VYED
E Q U I V A L E N C E (C(16O7) , VYE
)
)
E Q U I V A L E N C E (CC 16O S ) ,VZED
EQUIVALENCE (C(1611) ,VZE
)
)
E Q U I V A L E N C E C C C 1 6 1 2 ) , RXED
)
EQUIVALENCE ( C C 1 6 1 5 ) , RXE
)
EQUIVALENCE ( C C 1 6 1 o ) ,RY ED
EQUIVALENCE ( C C 1 6 1 9 ) ,RYE
)
)
E Q U I V A L E N C E ( C C 1 6 2 O ) ,RZED
)
EQ U I V A L E N C E (C(1623) , RZE
EQUIVALENCE (C( 1640), VTARGD )
E Q U I V A L E N C E (C(1643) ,VTARG )
EQ U I V A L E N C E (C(1644), BPSITD)
EQ U I V A L E N C E (CC 1647) ,BPSIT )
EQUIVALENCE ( C ( 16 4 8 ) ,RTXED )
)
EQ U I V A L E N C E C C C 1 5 1 ) , RTXE
EQ U I V A L E N C E (CC 1652) , RTYED )
EQUIVALENC E CC( 1655) , RTYE
)
E QUIVA L.. NCE ( C ( 1 6 5 6 ) , RTZED )
)
E Q U I V A L E N C E CC (165 9 ), RTZE
C
COTIIER OUTPUTS
)
EQUIVALENCE CC( 1 624), AXBA
)
E Q U I V A L E N C E C C ( 1 6 2 5 ) ,A Y B A
EQUIVALENCE

(CC1626) ,AZBA

EQUIVALENCE

(C(1633) ,V DELY

EQUIVALENCE

(C(1632) ,V DELX )
80
~~~~~~~~~~~~~~~~~~~~~~~~

-, .
~~~ c~ _ =

~~~~~~~~~~~~~~~~~

_ -~~~~~~~~~


~~~~~~~~~~~~ I Z I ~~~~~~~ -

EQUIVALENCE
EQUIVALENCE

(C(1634) ,VDELZ )
(C(1635) ,RDELX )

EQUIVALENCE
EQUIVALENCE

(C(1637) ,RDELZ )
(C(1638) ,VCL SNG )

EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVA LENCE

(C(1636) ,RDELY

( C ( 16 6 0 ) ,VTXE
( C ( 16 6 1 ) ,VTYE
( C ( 16 6 2 ) ,YTZE
(C(1663) ,VDXB
(C(1664) ,VDYB
(C(1665) ,VDZ B

)
)
)
)
)
)
)
)
)

(C (1676) ,AN GX
CC (1677 ) ,ANG Y
(C (1678) ,AN G Z
(CC 37b ) ,RAN G E

EQUIVALENCE

EQUIVALENCE

EQUIVALENCE
EQUIVALENCE

C
C* *ADD AE RO AND THRUST FORCES TO GET TOTAL ACCELERATION
AXBA = F X BA / D M A S S
AYBA
F Y BA / D M A S S
AZ BA = FZBA/DMAS S
C
C * * R E S O L V E FROM BODY TO EARTH AXES
AXE = C F A 1 1 * A X B A + C F A 2 1* A Y B A + C F A 3 1* A Z B A
AYE = CFA12*AXB A+ CFA22*AYBA +C FA 32*AZBA
A Z E = CFA 13*AXBA +CFA2 3 *AYBA+ C FA 33 *AZBA
C
C* * I N T E G R A T E A C C E L E R A T I O N S
VXED = AXE

VYED

VZED

AYE

AZE

IN BODY AXES

AGRAV

C
C** CALCULATE TOTAL MISSILE ACCELERATION IN BODY AXES
VDXB
=
C F A 1 1 * V X E D + C F A 12 * V Y E D + C F A 1 3 V Z E D
VDYB
=
C F A 2 1 * V X E D + C F A 2 2 V Y E D + C F A 2 3 V Z E D
VDZB = CFA 3 1*VXED + C FA 3 2 V Y E D + C F A 3 3 * V Z E D
ANGX = V D X B / 3 2 . 1 7 4
AN G Y = VDYB /32 .i74
ANGZ = VDZB /32 . 174
C
C* * I N T E G R A T E VELOCITIES TO EARTH AXES
POSITION
10 RXE D = VXE
RYED = VYE

RZED

VTYE

VZE

C
C* *TARGET MOTION
IF (OPTARG .LE . 0.) RETURN
VTA RG D = A T H R S I * A G R A V
BPSITD= 0.
IF (VTARG.GT . 0.) BPSITD= A T U R N T * A G R A V * C R A D / V T A R G
C
VTX E = VTARG*CUSD (BGAMT) *COSD(BPSIT)
C

VTZE

R TXED

RTYED

RTZED

VTARG*COSD(BGAMT)*SINDCB PS IT)
VTA RG *SIND (BGAM T)

=
=
:

VDELX

VDELY

V I XE

VTYE
V TZE
VTX EVXE

VTYE -VYE

~~~~~~~~~~~~~~~~

- -

- -.

--

--

~1
C

V DELZ
VCLSNG

RETURN

V TZEVZE

( R DE LX*VDELX + RDELY *VD ELy + RD ELZ *V D ELZ )/RA NcIE

END

---

-~~~~~~ -_ _- ---

- .

- .-

..

T_ I

C
~~

SUBROUTINE

~~~~~~~~~~~

~~~~~~~~~~~ -

~~~~~~~~~~~~~~~~~~~~~~~

~--~-----.
~-

- -

-~--~.

1) 11

T RA N S L A T I O N A L DYNAMICS INITIALIZATION MODULE FOR Dl


CO MMON C C 3 8 3 0 )
)
EQUIVALENCE ( C C 2 5 6 1 ) , N
IPL(l
))
E Q U I V A L E N C E CCC2562 ) ,
DIMENSION IPL (100) , ISNDX(40) , ITNDX (1O)
EQUIVALENCE CC (3634) , ISNDX (1 )) , (C(35b2), 13512)

C** INPUT DATA


. )
EQU IVAL E~N C E ~~~C( Too).;vwxE
)
EQUIVALE NCE (CC 1O 1 ) ,V WYE
)
EQUIVALENCE (CC 102) ,VWZE
EQUIVALENCE (C( 2011),VMAC H )
EQUIVALENCE CCC 367),BALPHA)
EQUIVALENCE (CC 368),BALPH Y )
EQUIVALENCE (CC 427),BTHTG )
EQUIVALENCE (CC 1431 ) , BPS I G )
EQUIVALENCE (C(1639) ,OPTAR G)
EQUIVALENCE (C(1666) , BLOSV )
EQUIVALENCE (C(166fl ,RSLAN T )
EQUIVALENCE CC (1674 ) , VMWTE )
)
E Q U I V A L E N C E (CC 175 1 ) ,CRAD
EQUIVALENCE CC (35O2) ,OPTN2 )
E Q U I V A L E N C E (C (35 014 ), OPTN4 )
EQUIVALENCE (CC3506) ,OPTN6 )
C
C** OUTPUT TO MODULES
)
EQUIVALENCE (C(1615 ) , RXE
)
E Q U I V A L E N C E (C(1619 ) ,RYE
)
(C(1623)
EQUIVALENCE
,RZE
)
EQUIVALENCE ( C ( 1 6 O 3 ) ,VXE
)
E Q U I V A L E N C E CC (1607 ) ,VYE
)
E Q U I V A L E N C E C C C 1 6 1 1 ) ,VZE
)
E Q U I V A L E N C E (C(165l ) , RTXE
)
E Q U I V A L E N C E (C(1655) ,RTYE
)
E Q U I V A L E N C E CCC 16 59), RTZE
)
EQUIVALENCE (C (l668) ,RXO
)
E Q U I V A L E N C E CC (1 66 9 ) , RYO
)
EQ U I V A L E N C E (C(l670) ,RZO
)
c. .U i VA L E N C E (C(1671 ) ,VXO
~
)
E Q U I V A L E N C E (C (1672) ,VYO
)
EQ U I V A L E N C E (CC 1673) , VZO
EQUIVALENCE (C(1752) , B PHIO )
EQ U I V A L E N C E (C(1753) , BTHTO )
E Q U I V A L E N C E CC (1754 ) , BPS I O )
)
EQUIVALENCE ( C ( 1 6 6 5 ) ,RHZRO
E Q U I V A L E N C E (C(1635) , R D E L X )
E Q U I V A L E N C E (C(1636) , RDELY)
E Q U I V A L E N C E (CC163 7 ), R D E L Z )
E Q U I V A L E N C E (C(168O ) , RSJY M C )
E Q U I V A L E N C E C C C 1 6 8 1 ) ,RSJZMC)
E Q U I V A L E N C E (C(3753) ,ITNDX(1) ) , (C(3721) ,ITCT)
EQUIVALENCE CCC 1 76 1) , AO 11 )
E Q U I V A L E N C E (C (1762) , A 0 1 2 )
EQ U I V A L E N C E C C C 1 7 6 3 ) , A O 1 3 )
EQUIVALENCE ( C ( 1 7 5 5 ) , A 0 2 1)
E Q U I V A L E N C E CC ( 1756), A 0 2 2 )
E Q U I V A L E N C E (C (1757 ),
A02 3)
EQ U I V A L E N C E (CC1 75 8 ),
A 031 )
.

. .

(C (l759) ,
CCC 176O ) ,
( CC 1 7 6 14 ) ,
CC (17 65 ) ,
(C(l766) ,
(CC1767) ,

EQUIVALENCE
E QUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE

C
C

--

- -

A0 32)
A0 33 )
P1)
Qi )

(C(1768) ,
(CC1769) ,

EQUIVALENCE
EQUIVALENCE

P2 )

R2)

XB O1 )
YBO 1 )
Z B O 1)
XB O 2 )
YBO2 )
ZBO2 )

E Q U I V A L E N C E (C(1770) ,
EQUIVALENCE (C(1771 ) ,
E Q U I V A L E N C E CC( 1772) ,
EQUIVALENCE C C C 1 7 7 3 ) ,
EQUIVALENC E (CC 36O) ,BPL-I1ER )
EQUIVALENCE (CC 361) ,B T H 2 E R )
EQUIVALENCE (CC 362),BPS 1ER)
EQUIVALENCE ( C C 1 5 6 2 ) ,GSPOT Y )
EQUIVALENCE C C C 1 5 7 2 ) ,GSPOTZ )
EQUIVALENCE (C(1581 ) , SIGP i)i)
EQUIVALENCE (CC 1579) ,
ZF i-~)
WO )
EQUIVALENCE (CC158O) ,

C* ZERO OUT SPOT JITTER MAX/MIN STORAGE LOCATIONS THAT ARE SAVED IN OUTP
C(1567) = 0.
C (1568) = 0.
C (l577 ) = 0.
C (1578)
0.
C P R I N T E D FROM MODULE GLI
WO
3 .911
ZETA = .7145
C

C SPOT J I T T E R

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

_ _ _ _ _ _ _ _

MONTE

CARLO

INITIAL

VALUES

RSJYMC
0.
RSJZMC = 0.
DO 500 IOL = 1 , ITCT

IT SNDX

I OL

IF (ITNDX(IOL) .NE. 1680) GO TO 502


IPL(N)= 1560
IPL (N+l )= 1563
N =N-s- 2

IFCSI GPOT.NE .0.) GSPOTY

.7O7*SIGPOT/SQRT(WO /4 ./ZETA
=
CALL M C A R L O ( R N S T R T , 14 ,ITSNDX )
502 IF (ITNDX(IOL) .NE .1681 ) GO TO 500
I P L( N ) = 15 7 0
I P L C N + 1 ):1573
N =N + 2
IFCSIGPOT .NE . 0. )

G SP OTZ

.707*SIGPOT/SQRT (WO/ 14./ZETA

CALL M C A R L O C R N S T R T ,14 , ITSNDX )


500 CONTINUE
IP L( N)

IPL (N-s .1)

C (26614))

C(266L1))

1600

IPL(N+2 ) =
IPL(N+3 ) =
IPL(N 14 ) :
IPL (N+5) =
IPL(N+6 ) :

16014

1608
1612
1616

1620
164 0

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

~~~~~~~~~~~~ ----

~~~~~~~

---

IPLCN+7 ) = 1644

IPL(N+8)

16118

IPL (N+9) = 1652


IPL (N+iO) = 1 656

N = N+ 11

CC 36 3 ) :0.

CC 3614) = 0 .
CHAD = 57 .29577t~

IF. (-OPTN2 .LE O .)- .G 0 TO 1002


IF (OPTA RG .LE. 0.) N = N 5

C** CALCULATE MISSILE PARAMETER


RYE = 0.

RTZ E

0.

XBO2

0.

INITIAL CONDITIONS

RTYE = 0.
RTXE
0.
BPHI O = 0 .
XB O 1 = 0 .
YBOl = 1.
ZBO1 = 0 .
H

C
C***

1 8 0 2= 0 .
ZBO 2= 1.

MONTE C A R L O AUTOPILOT GYRO DRIFT


P1
0.
Q l = 0.
P2 = 0.
R2=O .
DO 503 I = 1 ,13512
11)0 = I
IF(ISNDX(I).EQ. 1764) CALL MCARLO
IF(ISNDX(I).EQ. 1765) CALL MCARLO
IF(ISNDX(I) .EQ.1766) CALL MCARLO
IF (ISNDX(I).EQ . 1767) CALL MCARLO

RATES

(DUll ,
(DUll ,
(DUll ,
(DUll ,

5 03 CONTINUE
C
C
C*** AUTOPILOT GYRO BIAS E R R O R S
C
1002 BPH1ER = 0.
BTH2ER

IDO )
IDO )
100 )
IDO )

0.

0.
DO 11 I = 1 , 135 12
IDO = I
IF(ISNDX( I).EQ.360) CALL MCARLO
IF(ISNDX(I ).EQ .361) CALL M CARL O
IF (ISNDXCI).EQ. 362) CALL MCARLO
CONTINUE
B P S 1E R

1,
1,
1,
i,

(DUll , 1 , 100 )
(DUll , 1 , IDO )
(DUll , 1 , IDO )

11
C
C** INITIALIZE M A T R I X COEF FOR AUTOPILOT GYRO MODELS
C
USP H I1 = SIND (BPHIO + B P H 1E R )
UCPI- II1 : COSD ( B PH IO + BPH 1 ER )
USTHT2 = S INDCBT H TO + BT (-I2ER)
UCT HT 2 = COSD ( BTHTO + BTH 2ER )

U3PSI1
U C P S I1

A0 11

SIND(BPSIO + BPS 1ER)


COSD (BPSIO + BFSlER )
UC PSI 1*UCTHT2
:
=

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

~~~~~~~~~~~~~~~~

_____________

____________________

A 0l2

-USTHT2

A022

A 021

USPSI1*UCTHT2

A 01 3

_ USP S I1*UCPHI 1 + UCPSI 1*USTHT2*USP HI1


U C PSI1 * UC PHI 1 + USP SI1 *USTHT2 *USPH I1

A02 3

UCTHT2 USPHIl
U C P S I 1* U S T H T 2 * U C P H I 1

A 03 1
A0 32

USPSI1*USTHT2*UCP HI1

A 033

UCTHT2 * UC P1~1I1
IF (OPTN2 .LE . 0) RETURN

C
C MISSILE

INITIAL

ATTITUDE

USPSI 1*USPH I1

UCPSI 1* US PH I 1

ERRO R ~

DO 5 I = 1 , 13512
11)0 = I

IF(ISNDX (I) .EQ . 1752) CALL MCARLO


IF(ISNDX (I).EQ .1753 ) CALL MCARLO
IF (ISNDX(I).EQ .1754) CALL MCARLO
5 CO NTINUE

(DUll , 1 , IDO )
(DUll , 1 , IDO )
(DUll , 1 , IDO )

IF (OPTNLI .GT. 0.) GO TO 30

CRAD *ARS IN(SIND(B PSIG)*RSLA NT/RXE)


COSD(I3PSIO)
TTI-iTG
= SINDCBT I-ITG )/COSD ( BTHTG )
BTHTO = A T A N D ( C -R Z E / R X E
TTHTG*CPSIO ) , ( C P S I O
=

GO TO 40
30 CONTINUE

~~~~

IF (OPTN2 .GT .1. 0) GO TO 10


R X E =_ R S L A N T * C O S D ( b L O S V )
R Z E = R S L A NT * S I N D ( B L O S V )
GO TO 20
10 RSLANT = SQRTCRZE **2 + RXE ** 2)
20 RH = R H L R O R ZE

BPSIO
CPSIO

IF ( O P T N 4 .GT . 1.)
UST = S I N D ( B T HT U)
USP

UCP
UCT

=
=
:

TTHTG*RZE/RXE ))

GO TO 40

SIND(BPSIO)
COSD( BPSIO)
COS D(BT HTO )

UCP H = C O S D ( B P H I O )
SIND (BPUIO )
USPH
RXBA = _ R X E * U C P * U C T + RZE * UST
RYBA = _ R X E* ( U C P * U S T * U S P H USP*UCP H ) RZE* UCT * USPH
RZBA = _ RXE* (UCP *UST *UCPH + USP*USPH )
RZE * UCT * UCP H
BTHTG
A T A N D ( R Z B A , RXBA )
BPSIG = ATA ND( RY BA , (RXBA*COSD (BTI .ITG)_RZBA*SIND (BTHTG) ))
40 CONTINUE

24 VSOUND = 1117. 3
.OO392*R }~I
IF (OPTN6 .LE. 0.) VMWTE : VMACH*VSOUND
VMWXY = VMWTE *COSD( BALP HA
BTHTO)

VXE

VIE
VZE
RX O

RYO
RZO
VXO

VYO

-- ---

.-

=
=
=
=
:
=
=
:

VMWXY

V MWXY
VMWTE

COSD(BALPHY + BPSIO)
* S IND (BA LPHY + BPSIO )
* SIND ( BA LPHA
BTHTO )

RXE
RYE
RZE
VXE

VYE
- --

- - --

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

-- -- ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ --
-

~~~~~~~~~~ T~~~

-:-----

VZ O = VZE
RETURN
END

87

---:
- ---- - ~~ ~~ ~~

~~~~~~~

T~~ = - ~

.
~~~~~

~~~~~~~~~~~~~~~~~~~~ ~-~

~~ --:

SUBROUTINE 1)2
ROTATIONAL DYNAMICS MODULE
C (MMO N C ( 3 8 3 0)
~

C~~

C
C * * D A T A INPUTS
EQUIVALENCE ( C ( 1 7 4 8 ) ,FMIX
)
EQUIVALENCE (C(1749) , FMIY
EQUIVALENCE ( C ( 1 7 5 0 ) ,FMIZ )
)
EQUIVALENCE (C (1751 ) , CRAD
E Q U I V A L E N C E (C(3503) , OPTN3 )
C
C**INPUTS FROM OTHER MODULES

EQUIVALENCE
EQUIVALENCE
EQUIVALENCE

L
..

(C(1303) ,FMXBA )
(C(13O4 ) ,FMYBA )
(CC1305 ) ,FMZBA )

C
C**STATE V A R iA B L E OUTPUTS
EQ U I V A L E N C E (C(1700) ,CFA 1 1D )
E Q U I V A L E N C E (C( 1703) , CF A 11 )
EQUIVALENCE (C(17014 ), C F A 12 D )
EQUIVALENCE (
~~(17O7 ) , CF A 12 )
E Q U I V A L E N C E (C(17O 8) ,CFA 13 D)
EQ U I V A L E N C E ( C (1 7 l l ) ,C FA 1 3
EQ U I V A L E N C E (C(1712) , C F A 2 1 D )
E Q U I V A L E N C E (C (17 15) ,CFA21 )
EQ U I V A L E N C E (C(1716 ) , CFA22D )
E Q U I V A L E N C E (C(1719 ) , CFA22
E Q U I V A L E N C E (C(172 0),CFA23D)
EQ U I V A L E N C E (C(1723) ,CFA23 )
E Q U I V A L E N C E (C(1724) , C F A 3 1D )
EQUIVALENCE (CC 1727) , CFA 3 1 )
E Q U I V A L E N C E ( C ( 1 ( 2 8 ) ,C F A 3 2 D )
E Q U I V A L E N C E (C( 173 1 ) , CFA 32 )
EQUIVALENCE (C(1732) , CFA 33D)
E Q U I V A L E N C E (C( 1735 ) ,C FA33 )
)
E Q U I V A L E N C E (C(1736) , WPD
E Q U I V A L E N C E (C( 1739) , WP
)
EQ U I V A L E N C E (C(17140) , WQD
)
E~~U IV A L E N C E (C( 17143 ) , WQ
)
EQ U I V A L E N C E (C(1744 ), W RD
)
EQ UI V A L iLN C E. ( C ( 1 7 14 7 ) ,WR
C
~~~~~~ Y A W PI~ RA 1 E TE R I N P U T
E~~U IV A L L N C E C C ( 2 9 O 1 ) ,O PTNY W )
C
C**INTE. GR~~TE BODY A N G U L A R RATES
IF (OPTN 3 .LE.O .) GO TO 45
IF (OPTNYW.LE.U.) GO TO 55

GO TO 65
45 WPD = C RA D * F M X BA / F M I X
55 -.4hD
65 W QD

(CRA D *FM ZBA+(FMLX_ FM IY )*WP*WQ /C P.AD)/FM IZ


(CRAD*FMYBA+ (FMI Z_ FM IX) *WP* W R /CRAD)/FM IY

C
C**INTEGRATE ATT i TUDE DIRECTION COSINES
149 C F A 1 1D= ( CFA 2 1 * ,.1 R_ C F A 3 1 *W Q ) / CRA D
C F A 12 D = C C FA 2 2 * r~H_ C FA 32*W Q )/CRAD
CFA 13 0= CCFA2 3 * ~4R CFA 33 WQ)/ C RAD
CFA2 1 D = (CFA 3 1* ~ft_ C F A 1 1 * W R )/C R A 1 )
CFA22D = (CFA 32 *W P_ CFA12*WR) /CRAD

L L _

~~~~~~~~ _~

TT
~~~~T T . I

CFA23D
CFA 31D
CFA32D
CFA33D

RETURN
END

=
=

(CFA33*Wp_ C FA 13*WR ),CRAD


(CFA 11*W Q_ CFA2 1*Wp ),CRA D
(CFA12*WQ_ CFA22*Wp )/CRAD
(CFA 13*wQ_ cFA23*wp ),CRAD

SUBROUTINE D 21
C**ROTA TIONAL DYNAMICS INITIALIZATION MODULE D2IEUL
COMMON C(3830)
DIMENSION
IPL (100)
CINPUT DATA
EQUIVALENCE (C(1752) ,BPHIO )
EQUIVALEN CE (C(1753), BTHTO )
E Q UIVALENCE (C(1 75~ ),BPSIO )
CINPUTS FROM MAIN PROGRAM
)
EQUIVALENCE (C(2561) ,N
)
EQUIVALENCE (C(2562) ,IPL
C**STATE VARIABLE OUTPUTS
EQUIVALENCE (C(1703) ,CFA 11 )
EQUIVALENCE (C(1707), CFA 12 )
EQUIVALENCE (C(1711 ) ,CFA 1 3 )
EQUIVALENCE (C(1715), CFA21 )
EQUIVALENCE (C(1719) ,CFA22 )
EQUIVALENC E (C(1723) ,CFA23 )
EQUIVALENCE (C(1727), CFA3 1 )
EQUIVALENCE (C(1731) ,CFA32 )
EQUIVALENCE (C(1735) ,CFA33 )
C**OTHER OUTPUTS
EQUIVALENCE (C(1755) ,A 021 )
EQUIVALENCE (C(1756) ,A022 )
EQUIVALENCE (C(1757) ,A023 )
EQUIVALENCE (C(175 ~~),A0 31
EQUIVALENCE (C(1759) ,A 032 )
EQUIVALENCE (C(17b0) ,A0 33 )
C**INIT IAL CALCULATION OF EULER ANGLE MATRIX OF DIRECTION COSINES (CFA)
USPHI
SIND (BPHI O)
UCPFI I C O SD (BPHIO )
USTHT = S IND (BTFIT O )
UCTHT
COSD(BTHTO )
USP SI SIND (BP S IO )
UCPSI
COSD(BPSIO)
CFA 11
UCPSI*UCT 1~1T
CFA 12
USPSI*UCTHT
CFA 1 3
USTUT
_ USPSI*UCPHI+UCPSI*USTRT*USPH I
CFA2 1
CFA22
UCPSI*UCPHIUSPSI*USTHT*USPHI
CFA 23 = UCTHT*USPHI
CFA 31
UCPSI*USTHT*UCPUI+USPS I*USPHI
CFA32 = USPSI*USTHT*UCPH I_UCPSI*USPH I
U CTHT*UCPH I
CFA33
C
C* *INITIALIZE MATRIX COEF FOR FREE GYRO MODEL (S)
C
C **INTEGRATED PA RA MATER LIST (IPL) FOR WPD ,WQD ,WRD ,AND CFAD
IPL(N)
1700
IPL(N+1 )
17O ~
IPL (N+2)
1708
IPL(N+3 )
1712
IPL (N ~ ) 1716
IPL(N+5 ) = 1 120
IPL(N+6) = 172k
IPL(N7) = 1728
IPL(N+8) = 1732
IPL (N+9) = 1736
IPL(N+ 1O ) = 1 PlO
_

IPL(N+11 )
N = N+ 12

RETURN
END

17~4~4

FUN C TI O N FINTP 1 (X ,X I ,YI ,N ,x ,XL)


DIMENSI O N XI (N ), YI (N)

IF (F .GT . 0.)G0 TO 30
DO 10 1=2 , U
IF (X .LE. XI(I)) GO TO 20
10 C O NTINUE
I = N
20 PCT = (XX I (I 1 ))/ (X I (I )XI( I 1 ))
F = 1.
30 FINTP1 = YI (I 1 ) + PCT* (YI (I)_YI (I_ 1))

RETURN
END

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

FUN C TI O N FJNTP 2(X ,Y ,XI ,YI ,ZI ,NX ,NY ,NXY ,F ,XL )
DIMEN SION X I (NX ),YI (NY), ZI (NXY ), T(2), COL(1O)

IF(F .GT . 0.) GO TO 30


DO 10 1=2 , NY
IF(Y .LE. YI (I)) GO TO 20
10 C O NTINUE

I = NY

(Y YI( I 1))/ (YI(I ) YI( I 1 ))


20 PCT
30 DO
W J= 1 ,2
L = I + J
2
(L_ 1 )*NX
L 1NX
L 1NX1
L 1NX + 1

LNX

NX

DO 50 IR = L 1NX 1 , LNX
50 COL(I RL 1NX) = ZI (IR )
40 T (J ) = FINTP 1 (X ,X I ,COL ,NX ,F ,XL)
T(1 ) + PCT* (T (2)_T(1))
FINTP2

RETURN
END

V ..

SUBR O UTINE 03
CAIR DATA M ODULE 03
COMMON C(3830)
C**INPUT DATA
EQUIVALENCE (C(0208), RHZRO )
C**INPUT S FROM OTHER MODULES
)
EQUIVALENCE (C(O100) , VWXE
)
EQUIVALENCE (C(01O1 ) ,VW Y E
)
EQUIVALENCE (C(0102) ,VWZE
)
EQUIVALENCE (C(1603) ,VXE
)
E Q UIV ~~ ENCE (C(16O7) ,VYE
)
EQUIVALENCE (C(1611 ) ,VZE
)
EQUIVALENCE (C(1623) ,RZE
C**INPUT S FROM MAIN PROGRAM
C**STATE VARIABLE OUTPUTS
C ** N O NE
C**OTHER OUTPUTS
EQUIVALENCE (C(0200) ,VMWXE )
EQUIVALENCE (C(0201) , VMWYE )
EQUIVALENCE (C(0202) ,VMWZE )
EQUIVALENCE (C(0203) ,PDYNM C)
EQUIVALENCE (C(O2OL ~),VMAC H )
EQUIVALENCE (C(02O5) ,DRHO )
EQUIVALENCE (C(O2O ) ,VSOUND)
EQUIVALENCE (C(0207 ),VA I RSP )
)
EQUIVALENCE (C(0209) ,RH
CCALCULATE PRESENT ALTITUDE
RH= R ZE+R HZR O
C**CALCULAT E MISSILE VELOCITY W RT AIR MASS IN EARTH AXES
VMWX E
VXEV ~1XE
V~1WYE
VMWZE = VZ t~-VWZE
SQRi (VMWXE*V MWXE +VMWYE *VMW YE+VMWZE *VMWZE)
VAIR SP
C **A IR DEN SIU , SPEED OF SOUND , DYNAMI C PRESSURE , AND MACH
DRHO: (.O ~~ 75 )/(1 .+ .3325E_OL ~*RH+RH*RH*RH* .O2315E_ 12)
_ .00392*RH+111 7.3
YSOUN D
(DR }iO*VA IRSP*VA IRSP )/6 ~4.3~4L~
PDYNM C
VMA CH
VAI HSP/VSOUND
RETURN
END

~~~~~~ - -~~~~~~ , .

.-~~~~~-

rr

--

,
.

SUBR O UTINE QL~

* 1* * * * *
* ** * * * * * **** * * ** * ** * * * * * * * * ** ** ** * * ** ** * *

C~~ THIS IS A SUBROUTINE (NOT A MODULE ) CALLED BY STAGE 3


C** STOPS PROGRAM AND COMPUTES MISS DISTANCE

**
**

***********
**** ** * * ***** * *** * **** * ***** ** ***** * **** * ** ** **
COMMON C(3830)

COMMON /XMA /XMAX (1~,7)


100 FORMAT( 1HO , 1 7H HISS DISTAN C E = ,IPE 15.7/
*
1HO , 17H FLI G HT TIME
= , 1PE15.7 )
200 FORMAT( 1l-I O , 9X , 8HRDELX
, 1PE15.7 , 8X ,8HRDELY

= , 1PE15.7 ,
BX ,&IRDELZ = , 1PE15.7)
300 FO RMAT (1HO , 140X ,8HRY F P = , 1PE15.7, 8X ,8HRZF P
, 1PE15.7)
EQUIVALENCE (C( 357),BGAMH )
(C( 358),BGAMV )
(CC 371) , RANGE )
(C(1635) ,RDELX )
(C(1636), RDELY )
(C(1637), RDELZ )
)
EQUIVALENCE (C(2000) ,T
E QU iVALENCE CC( 156~4) ,YMC )
EQUIVALENCE (C( 1565) ,YMC2)
EQU iVALENCE (C(15P4 ),ZMC)
EQUIVALENCE (C(1575) ,ZMC2 )
(C(202O) , LCONV )
EQUIVALENCE (C( 300),RMISS )
( C (3 0 1 ) , L
)
(C( 302) ,RYF
)
(C( 303),RZF
)
EQUIVALENCE (C( 31) , LCEP)
E QUIVALENCE (C(3721),
ITCT )
(1))
E Q U I V A L E N C E (C(300 0),VSD
EQUIVALENCE (C(3O10 ), VM EAN (1))
E Q U I V A L E N C E (C(302O) , IMV N DX( 1 ))
EQUIVALENCE (C(3J3 0), IMVC T )
E Q U I V A L E N C E (C (1651 ) , RTXE) ,(C(1655) ,RTYE) , (C(1659), RTZE), (C(1615 )
,RXE ), (C(16 19) , RYE ) , (C(1623) ,RZE)
REA L* o PITCh , YA ~
/
DATA PITCH ,YAW/ 8H PITC H
, BHYAW
DIMENSION 1M~~~DX(1O), V M EA N ( 1O ) , VSD( 1O)
DIMENSION V SUtI (1O ), V S 2 ( 1 O )
LCEP
0
RDELX ZI~ ~X E R X ~
R D E L I= R T Y E R Y E
RDELZ = R T Z E R Z L
IF(RL ~~LZ .LT . 0. .OR. RDELX LT . 0.) LCONV = 2
IF (R~~~~E .GT . 530.) GO TO 20
*

UCU = S IND(BGAMV)

UC 33 = COSD (I3GAMV)
UC2 1 =- .SIND(BGAMH )
UC22
COSD(BGAM iI )
UC 1I
UC22 *UC 33
UC 12
_ UC2 1*UC 33
_ UC22 *UC 1 3
UC 3 1

UC32
RXF P
RY FP
RZFP

UC21*U 01 3
UC 11 *RL )ELX + UC12* R DELY + UC 130RDELZ
U C2 1*RD ELX + UC22~ RDELY
1JC 3 1 *RDELX + UC32*RDELY + UC33 RDELZ

IF (RXFP
IF(IMVCT

.GT. 0.) GO 10 10
.LE . 0) GO TO 50
-

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

--

~~~~~~~~~~~ -

~~ -

~~-

.
~

DO 5 I= 1 , IMVCT
IDO = IMVNDX( I )
VALUE = C(IDO)
VSUM( I) z VSUII (I) + VALUE
VS2 (I)
VS2(I) + VALUE* *2

TCASE

TCASE 1
L
1
V M EAN (I) = VSUM (I)/TCASE
IF(L .NE. 1) GO TO 2

(VS2(I)
(VSUM(I)**2)/TCASE)/TCASE
GO TO 3
2 S2
(VS2(i)
(VSUM (I)**2)/TCASE)/TCASE1
SQRT(32)
3 VSD(I
32

5 CONTINuE
50 CON TINUE
PCT
UXFP/(RXF P
UXF P )
RDX
UDELX - PCT*(RDELX - UDELX)
R O Y = UDELY PCT*(RDELY UDELY )
RDZ = UDELZ PCT* (RDELZ UDELZ )
RYE = UY FP PCT*(RYE P UYFP)
RZF
UZFP PCT*(HZFP UZFP )
TZERO
UT - PCT*(T - UT)
SQRT (RYF**2 + RZF**2 )
RMISS
W RITE (6 ,600)C (630), PITCH
WR ITE( 6 ,600)C(63 1 ),YAW
600
F O R M A T ( 1H O ,60X , 214H+MAX BREAKLOCK VALU E F10.5 ,5H
W RITE (6 ,~4OO) L
4OO
FORMAT(
1 110 , 13HR UN NUMBER = , 12)
~
IF(ITCT.LE .O)GO TO 30
CALL M C A R LX ( DUM ,2,RNSTRT)
WRITE (6 ,50 0) C(1567 ) , C(1568), C(1577), C(157~~)
XMCSPT
SQRT(YMC2*YMC2 + ZMC2*ZMC2)
WRITE(6 ,2555 )YMC ,YMC2
WHITE (6 , -~~56)ZMC ,ZMC2 ,XMC SPT
CONTINUE
30
500 F0H~4AT( 1H0 , 11 X , 13HMA X SPOT Y
,F6.2 , 14H MIN SPOT Y
1
12K , 1 3 H M A X SPOT Z = ,F6.2, 1L ~H MIN SPOT Z

2 )

IN ,A8 )

=
=

F6.2/

,F6.2//

2555 F O RM A T ( 1H O , 11 X ,26HSAMPLE SPOT JITTER Y M E A N = ,F1O. 5 , 6X , 12HMEAN SQUA


1RE= ,F 1 0.5)
2556 FORMA T ( 1HO , 1 1 X ,26HSAM PLE SPOT JITTER Z M E A N = ,F1O.5 , 6X , 12HMEAN SQUA
1R E = ,v l J.5 , 6X , 1ddSPOT RAD IAL RMS :,F1O.5 )
WHITE (6, 1 00) RM ISS , TZERO
WRITE( 6 , 17 )XA i~X
E OR MAT (7E1 7 .8 )
17
2OO) ROX , RDY , RDZ
~RITE (6 ,
WRITE(6 ,330) RYE , RZE

LCOUV

LCEP
RETURN

10 UT
I
UDELX =
UDELY = RO ELY

UDELZ

UX i P
U Y FP
ULF P
RETURN

uXEP

H YFP
RZEP

20 C O N T I N U E

RETURN

___________________________

~~~~~~~

~~~~~

.
~~~ --

SUBROUTINE 05
C**COORD INATE CONVE R SION MODULE
COMMON C (3~330)

C
CINPUTS FROM OTHER MODULES
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENC E

(C (0200) ,VHW XE )
(C(0201) ,V MWYE )
(C(0202) ,VMWZE )
(C(0207),VAI RSP)
)
(C(1317) ,RAIL
(C (1405),QBURN )
(C (1603) ,VXE
)
)
(C(1607) ,VYE

EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQU IVALENCL
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVA LENCE
EQUIVALENCE
EQUI VA LE N CE
E QUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE

)
(C(1615 ) ,RXE
)
(C(1619) ,RYE
)
(C(1623) ,RZE
(C(1635) ,RDELX )
(C(1636) ,RDEL Y )
(C(1637 ) ,RDELZ )
(C (1651 ) ,RTXE
(C (1655) , RTYE
)
)
(C (1659) ,RTZE
(C(16b ~3) , RXO
(C( 1669) , RYO
)
(C(1 !O) ,RZO
)
~~
(C(1671 ) , VXO
)
(C(16i ),VYO
)
~~
)
(C(1673) , VZO

E Q U IV A L E N C E
EQU IVALEN CE
EQUIVALENCE
EQU IVA L~.NCE
EQUIVALENCE

(C(16~3 1) ,RSJZMC)
(C(1682) ,RSPOTX)
(C(1683) ,RSPOTY)
(C( 16814 ),RSPOTZ)
(C(3753) , ITNDX(1 ))

EQU IVALENCE

EQUIVALENCE

EQUIVALENCE

(C(1b11 ) ,VZE

(C(16 ~ 0),R SJYMC )

(C(3721) ,

ITCT)

EQUIVALE NCE (C(1560) , SXPDD)


EQU IV \L EH CE (C(1561 ) ,
RX)
E Q U I V A L E N C E (CC 1562) ,GSPOTY)
E Q U I V A L E N C E (C(1563) , SXPD)
SXP)
E Q U 1V A L E N ~~E (C( 1566) ,
E Q U I V A L E N C E (C(157O ) , SYPDD)
E Q U I V A L E N U E (C(1571) ,
RY)
E Q U I V A L E N C E (C(1572) ,GSPOTZ)
EQ U I V A L E N C E (C(1573),
SYPD )
E Q u I V A L E N C E (CC1576),
SYP)
ZETA )
E Q U I V A L E N C E (C(1579 ),
E Q U I V A L E N CE (C(1SBO ) ,
WO)
DIMENSION ITN DX (1O)
EQUI tIA LENC E (C(17 03),C F A 1 1 )
E Q U I V A L E N C E (C (1707 ),CFA 12 )
EQU iV A L E N C E (C (17 11 ), C F A 13 )
EQU i V A L E N C E (C(1715 ) , CFA2 1 )
E Q U I V A L E N C E (C(1719) ,CFA22 )
EQ U I V A L E N C E (C (1723), CFA23 )
EQUIVALE N C E (C(1727) ,CFA 3 1 )
E Q U I V A L E N C E (C(1731 ), CFA 32 )
E Q U I V A L E N C E (C(1735 ) ,CFA 33 )
)
EQUIVALENCE (C(1751) , CRA V

XBO 1 )
E Q U I V A L E N C E (C (17b 8),
EQUIVALENCE (C (1769) , YBO 1 )
EQUIVALENCE (C(177 0), ZBO1 )
EQUIVALENCE (C(1771 ) , XBO2)
E Q UIVALENCE (C(1772), YBO2)
EQUIVALENCE (C(1 773) , ZBO2)
P1)
EQUIVALENCE (C (17b ~4),
Q1)
EQUIVALENCE (C(1765) ,
EQUIVALENCE (C ( 1 76b ) ,
P2)
R2)
EQUIVALENCE (C (1767 ),
EQUIVALENCE (C(1761) , A O 1 1 )
E Q U I V A L E N C E (C(1762) , A0 12)
EQUIVALENCE (C (1763 ) , A 013 )
)
EQUIVALENCE (C (1755) ,A 021
EQUIVALENCE (C(1756) ,A022
)
)
EQUIVALENCE (C(1757), A02 3
)
EQUIVALENCE (C(1758), A 031
E Q U I V A L E N C E (C(1759) ,A 032 )
)
E Q U I V A L E N C E (C(1 760) , A 033
)
(C(2
000)
EQUIVALENCE
,T

C
C**OTIIER OUTPUTS
EQ U I V A L E N C E (C(0350) ,BTHT
)
EQ U I V A L E N C E (C(0351) , BPS1
)
E QUIVALENCE (C(0352), BPHI
EQUIVALENCE (C( 353) ,BPH1
)
E Q U I V A L E N C E (CC 3514) ,BTH2
E Q U I V A L E N C E (CC 355),BPS1
)
EQUIVALENCE (C(0356) ,VTOTE )
E Q U I V A L E N C E (C( 0357),BGAMH )
EQUIVALENCE (C(0358) , BGAM V )
E Q U I V A L E N C E (C(0363) ,BTHLV )
E Q U I V A L E N C E (C(03614) ,BPSLV
E Q U I V A L E N C E (C(O365) , BLA MV )
E Q U IVA L E N ~~ (C(0366) , BLAMH )
E Q U I V A L E N C E ( C ( 0 3 6 7 ) ,BALPI-IA)
E Q U I V A L E N C E (C(0368), BAL PHY )
E Q U I V A L E N C E (C(0369) , BALPHP)
E Q U I V A L E N C E (C(0370) ,BPHIP )
EQUIV A L E N C E (C(0371) ,RANG E )
)
E Q U I V A L E N C E (C(0372) ,RXBA
)
EQ U I J A L E N C E (C(037 ,~ ), RYBA
)
E QU 1VAL E:~CE (C(0374 ), RZBA
)
E Q u I V A L E N C E (C (1663) ,VDXB
)
E Q U I V A L F N C E (C(1661 4) ,V DYB
)
E Q U I V A L E N C E (C(1665) , VDZB
E Q U iV A L E N C E (CC 377),BALPD )
E Q U I V A L E N C E (CC 378 ), BALYD )
E Q U I V A L E N CE (C( 380) ,RAN GO
)
E Q U I V A L E N u L (C( 390), RXL
)
E Q U I V A L E N C E (CC 391 ) ,RYL
E Q U IV A L F N C E (CC 392) , RZL
)
)
EQUIV A L E N C E (C( 393) ,BPH2
C
C CA LCULAT ION OF H E A V I N G , PITCH , ROLL EULER ANGLES IN DEGREES
[3PH I = A TA N D(C FA2 3 , CFA33 )
BTHr = A T A N D ( _ C F A 1 3 , SQRT(C FA 11 *CFA 11+ CFA 12 *CFA 12 ))
BP S I = A i A N D ( C F A 1 2 ,C FA 1 1 )
C

C**F REE G Y R ) MODELS (IN ITIA L GIMBAL ANGLES ARE ZERO)

97

C** AUT O PILOT DRifT RATES

_ Q 1 * Y B O 1 / CRA D

DXBO 1
DYBO 1
DZBO1
DXBO2
DYBO2
DZBO2

XBO 1
YBO 1
ZBO1
XBO2

B12

DXBO1*T
1. + DYBO1* T

DZBO1 *T

YBO2

ZBO2
Bli

C Q1*XB0 1
P1* ZBO 1 )/CRAD
P1*YBO 1 /CRAD
R2*ZBO2/CRAD
_ P2*ZBO2 /CRAD
(P2*YBO2
R2*XBO2)/Ch1.D

=
=
=

=
=

DXBO2*T
DYBO2*T
1. + DZBO2*T

A 0 1 1 * C F A 1 1 + A 01 2*C FA 12 + A 013*CFA 13
A O 1 1 * C EA 2 1 + AO 12*CFA22 + A 013*CFA23

B 13
A O 1 1 * C F A 3 1 + A0 12*CFA 32 + A 013 *CFA33
B21 = A 0 2 1 * C F A 1 1 + A 022*CFA 12 + A 023*CFA 13
B22
A 02 1*CFA21 + A0 22*CFA22 + A02 3*CFA23
B23 = A0 2 1*CFA 3 1 + A 022*CFA 32 + A02 3*CFA33
A O 3 1 * C F A 1 1 + A 032*CFA 12 + A 033 *CFA13
B31
B32 = A 0 3 1 *CFA2 1 + A0 32*CFA 22 + A0 33*CFA23
A 0 3 1 * C F A 3 1 + A 032*CFA32 + A 033*CFA33
B33
XB 1 = B 1 1*X B 01 + B2 1*YBO1 + B 31*ZBO1
YB 1
B 12*XBO 1 + B22YBOl + B32*ZBO1
ZB1 = B1 3 *XBO 1 + B23*YBO1 + B33*ZBO1
XB2 = B 11 *XBO 2 + B2 1*YBO 2 + B31*2B02
YB2
B 12*XBO2 + B22*YBO2 + B32*ZB02
ZB2 = B 13*XB02 + B23*YBO2 + B33*ZB02
BPH 1 = A TAND (ZB1 ,YB1 )
BPS 1 = ATAND ( XB 1 , YB 1 /COSD (BPH1 ))
BTH2
ATAND (X82 ,ZB2)
BPH2
A T A N D (Y B2 , ZB2/COSDCBTH2 ))

C
C **CALCULAT ION OF TOTAL VELOCITY
VTOTE = SQRTCVXE*VXE +VYE*VYEVZE*VZE)
R DELX
R T X E R X E
RDELY
R T Y E R Y E
RDELZ
R T Z E R Z E
C
IF(C (1976) .LE.0.) GO TO 20

RXL
RYL

RXE - RXO VXO*T


RYE RYO VYO*T

VXL

VXE

RZL
RZE RZO VZO*T
RANGO
SQRT(RXL**2 + RYL**2

VYL
VZL

=
=

VIE
VZE

20 CONTINUE

VX O

RZL**2)

yb
VZO

CTRA N SFORM MISSILE LOS FROM

EARTH TO BODY AXES

C
C LINE OF SIGHT OF LASER SPOT WITH MONTE CARLO SPOT JITTER INCLUD ~ D
C
DO 500 1
1 , ITCT
IDO = I
IF (ITNDXC I) .NE. 1680) GO TO 501

RSJ YMC

GS P O TY * SXP

CALL M C A R L O

(DUM ,2 , IDO )

___________________________________

______

SXPDD

~~~~~~~~~~

WO *W0* (RX_ 2. *ZETA* SXPD/WO

501 IF(ITNDX (I).NE. 1681 ) GO TO 500


RSJZMC

SY PO D
500 CONTINUE

RXBA

RYBA
RZBA

UVP 1
UVP 2

=
=

.-.

~~~~~
~~~

~~~~~~~~~

SXP )

GSPOTZ*SYP
(DUM ,2 , IDO )
W O*WO* (RY
2.*ZETA *SYPD /WO

CALL MCARLO
RSPOTX
RSPOTY
RSPOTZ

~~~~~~~~~~~~

=
=

RDELX
RDELY
RDELZ

+
+

SYP)

RSJYMC
RSJZMC

R SPOTX*CF A11

R SPOTX*CFA2 1

+ RSPOTY*CFA 12 + RSPOTZ *CFA 13


+

RSPOTY~ CFA 22

RSPOTZ*CFA2 3

RSPOTX*CFA 31 + RSPOTY*CFA 32 + RSPOTZ 1CFA33

VXE*RDELX +V YE *RDELY
R DELX*RDELX +RDEL Y *R DELY
UVP 3 = VZE *RDELZ
UVP ~ z S Q RT (UVP2)
RANGE
SQRT(UV P2+RDELZ **2)
C **VERT ICAL A N D HORIZONTAL LINE OF SIGHT ANGLES (EARTH AXES )
C
BLAM H
A TAND(RDELY , RDELX )
I3LAMV
ATA N D( RDELZ ,UVN4)
C
C* *VEBT ICAL AND HORIZONTAL PROPORTIONAL NAVIGATION ANGLES
IF (VTOTE .LE.1O .) GO TO 30
v X P = (UVP 1 +UVP 3 )/RANGE
( V YE*R DELX_ VXE*RDEL Y )/UVPL~
V~ P
VZP = (VZE*UVP2_ RDELZ*UVP 1 )/ (RANGE*UVPI4 )
BTHLV
ATAND (VZP , VXP)
BPSLV
ATAND (VYP , VXP)
C
BGAMV
A T A N D (_VZE , SQRT(VXE*VXE +VYE*VYE ))
BGAMI -I
A T A N V(VYE , VXE)
C
C**VELOCITY WRT AIR IN BODY AXES
VMW U
CFA 1 1*VMWXE +CFA 12*VMWYE +CFA 13 *VMWZE
VMWV = CFA2 1 *VMWXE +CFA22*VMWYE +CFA2 3 *VMWZE
VMWW
CFA 3 1*VMW XE +CFA 32*VMWYE +CFA 33 *VMWZE
C
C**VE R T ICAL A N D HORIZONTAL ANGLES OF ATTACK
IF (QBURN .LE.O. .AND . RANGO .LE.RAI L) GO TO 30
A T A N D (V M W W ,VMW U)
BALPI-I A
BALP HY = A TA N D CVMW V , VMWu)
US Q =V MW U * * 2
BAL,PD :CV IIWU *VDZB _ V MWW * VDXB) /(USQ+VMWW**2)*CRAD
BALYD = (VM ~4U*VDYB_ VMWV *VDXB )/(USQ+VMWV **2)*CRAD
C
C **ALPHA PRIME A N D PHI PRIME (WIND TUNNEL AXES)
IF ((BALPHABALP HY).EQ.O.) GO TO 30
BPUI P = A T A N D ( B A L P H Y ,BALP IIA )
30 ~3ALPHP :SQRT(BALP HA **2+BA L PHY **2 )

RETURN

END

.- * ~~~~~~~~~~~~ ~~~~

99
,

~~~~~~~~~~~~~~

- . , .--

-.

~~~

- .

.
~~~~~

~~~~~~~~~

-.

yr

~~~~~

SUBROUTINE LTRAN (T ,DELT ,AMP , Y ,YC ,IFLG ,K)


DIMENSION A (5 ,3), PHI (5 ,3), W(5 ,3)
DATA IMAX ,AE/ Z4 , _ 1 ./

DATA A/ l .
, 12. ,26. ,O. 1. ,4., 12. ,26.,0. 1. ,4. ,12. ,26.,0./
IF (IFLG .GT.O)GO TO 17
ZC=O.
W 1= 6 .28*1 1.
DO 1 I= 1 ,IMAX
CALL RANNUM(O. ,RNSTRT , RN )
PHI (I,K )=3. 14*RN
W(I ,K)= I*W1
C Z C IS INTEGRATI O N C ONSTANT FOR Z
B=W (I ,K )*T+PHI (I ,K)
ZC=ZC+ A( I ,K )*(AE*SIN(B)_W (I ,K)*COS(B ))/(AE**2+W(I ,K)**2)
C ONTINUE
YC=AMP *EXP(AE*T)*ZC
17
CONTINUE
,
~~~~~ .

Z=O.
DO 2 I= 1 ,IMAX

Z= Z+A (I ,K )*SIN(W(I ,K )*T+PHI (I ,K))

C O NTINUE

Y=AMP*EX P (AE*T)*Z

RETURN
END

Pr

~~

SUBROUTINE OINPT 1
BASIC INPUT SUBROUTINE OINPT 1

COMMON C (3830)

COM M ON/WKU l / O N A M E O ( 5 O )
EQUIVALENCE (C(3218) ,ONAME 1 (1 )) , (C(3268) ,ONAME2( 1 )) , (C(3318) ,ONA

ME3 (1)) ,

(1 )),

C
C
C
C
C

), (C(2361) ,NOMOD
(C(3328) ,ONAME Z~(I )

) ,

(C(2362), MODNO

CC(3168) ,OUTNO(1)), (C(2~161 ), N0SUB ), (C(2~ 62),SUBN O


(1)) ,
(C(3066) ,NO LIST ), (C(3167),NOOUT) ,
(C(3067) ,LISTNO(1)), (C(3117) ,VALUE( 1 ) ), (C(2008) ,PLOTNO ),
(C(2009) ,NOPLOT ), (C(2325), VLABLE( 1 , 1)) , (C(1) ,K(1))

EQUIVALENCE (C(2 010) , STEP)


EQUIVALENCE (C(198 ~4 ) ,NPLOT )
E Q U I V A L E N C E ( C ( 19 8 5 ) , O U T P L T ( 1 ) )
E Q UIVALENCE
( C ( 3 5 12 ) , I S G C T ) , ( C ( 3 5 1 1 4 ) , S I G M A ( 1 ) ) , ( C ( 3 5 5 ~4 ) , SIGLB
(1))
C
, (C(35914 ),S IGUBC 1 )) ,
* (c(363Q ) , IsN Dx(l )) , (C(367 )
~~ ,IDIST (1)),(C(3511) ,RNSTRT )

EQU IVALE N Ct (CC 372 1 ) ,ITCT ),(C(3723),TSGMA(1)) ,(C(3733) ,TLB (1)) ,


*(c (37L ~3),TuB(l)) ,CC (3753 ),ITNDX (1)) ,(C(3763) ,ITDIST (fl),(C(3773) ,
C
TSPER(1 )) ,
*(C (3783),TYPPER(1 )) ,(C (3793) ,TPSIG (1)),(C(3803) ,TNXST (1)),(C(3813)

, ITNDX2 (1))
E Q U I V A L E N C E (C( 21) , IBVNSW)
EQUIVALENCE (C( 22) , IPLOT)
EQUIVALENCE (C(19) ,PSIZE)
EQUIVALENCE (CC 23) ,XLAMBD)
EQUIVALENCE (CC 2~), KSSIG)
EQUIVALENCE CC ( 25),CEPSIG (1))
EQUIVALENCE CC (3825), NCASE)
E Q U I V A L E N C E (C(3020) ,IMVNDX( 1 ))
E Q U I V A L E N C E CC (3030 ), IMVC T )
REA L 8 CPERTY ,SSS
DIMENSION ONAM E 3 (1O) , ONAME 14 (10)
DIMENSION LISTNO(50 ) , VALUE(50)
DIMENSION SUBN0( 99) , IR(~4) ,VR(~4 )

DIMEN S IO N ALPHA ~ N ) ,ONAM E 1 (50 ) ,ONAME2(5O) ,OUTN0(50) ,MODNO (99)

DIMENSION K(351O)
DIMENSIO N VLA BLE (2 , 15)
DIMENSION JUTPLT( 15)
DIMENSION SIGMA( 140) ,SIGLI3 (40),SIGUB(40),ISNDX(L~0),IDIST (LtO )
DIMENSION TSGMA (1O) ,TLB (10) ,TUB (1O ) ,ITNDX (1O ) ,ITDIST (1O ),
*TSPER ( 10) ,TYP PEK (lO ) ,TPSIG( 10) ,ITNDX2( 10) ,TNXST (10)
DIMENSION IMV N DX (1O)
DIMENSION CEPSIG(6)

INTEGER CEPS IG

REAL K 3A1~
R E A L MOD N O
INTEGER OUTNO
INTLUE R OUTPLT
DIMENSION COEFF(28) ,REF(28) ,BREF(28),ATD(5) ,BATD (5),DTA(4),BDTA( 4)

EQUIVALENC E (C(3359), COEFF(1))


EQU~~VALENC L

CC (3387 ),REF(1 ))

EQUIVALENCE

(C(3 ~53),DTA(1 ))

EQUIVALE N CE (C(3~4l5 ) ,BREF ( 1))


E Q U I V A L E N C E (C (3 J 4L ~3) , A T D ( 1) )
EQUIVALENCE (C(3 ~ L~8),BATD(1 ))
L
EQUIVALENCE

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

(C(3457), BDTA(1 ))

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Pr
NAMELIST / D A P / C O E F F , REF ,~~flEF , ATD , BATD , DTA , BDTA
/
DATA CPERTY /8EI R
/
DATA SSS/8FIS
JAR

~ RITE (6 ,31)

3 1 FORMA T( 11H1 INPUT DATA/ )


1 R E A D ( 5 , 2 , E N D = 5 0 ) I R ( l ) , ( A L P H A ( J C ) , JC = 1 ,~4 ) , I R ( 2 ) , I R ( 3 ) , TPER , TP SGM A ,
*V R( 1 ) ,VR(2) ,VR( 3 ) ,IR (~4) ,VR(~4)
55 C O NTINUE
WR ITE (6 ,3O )IR(1 ), (AL PHA(JC ) ,JC= 1 ,1~),IR(2) ,IR (3),TPER ,TkSUM A ,
*V R ( 1 ) , V R ( 2 ) ,VR( 3)
* I R ( L ~) , V R ( ~4 )
30 FORMAT( 1X , I2 ,14 A 44 ,I5 ,I2 ,A1 ,F5.2 ,2E 15 .7, F 1O .4 ,15 ,F7.14)
2 F O RMAT (1 2 , L~A 4 , IS , I1 ,A 1 ,F3.2,2E15 .7,F 1 O. 4 , I5 ,F5.2)
7 I F ( I R ( 1 ) . N E . 1 ) GO TO 3
N O SUB
N OS UB + 1
SUBNO(NOSUB )
IR(2)
GO TO 1
3 IF( IR (l ) .NE. 2 ) GO TO ~
NOMOD
NOMOD + 1
MODNO (NOIIOD ) = IR (2)
GO TO 1
4
~ IF(IR(l ) .NE. 3) GO TO 5
L = IR (2)
C(L) = VR (1 )
IF (VR(2) .EQ. 0.) GO TO 1
NOLIST = NOLIST + 1
LISTNO (NOLIST ) = L
VA LUE (NOLIST)

VR( 1 )

GO TO 1
5 IF (IR (l) .NE. 14 )GO TO 6
N OOUT
NOOUT + 1
IF (NOOUT.GT.50) GO TO 1
U N A M E O ( N O O U T ) = A L P H A ( 1)

ONA M E1 CNOOUT) :ALPHA (2)


ONA M E2 CN O OUT ) = ALPHA( 3 )
O U T N O ( N O O ( J T ) = IR (2)
GO TO 1
6 IF CI R(l) .NE. 5) GO TO 16
READC5 ,DAP)

W RIT E (6 ,V A ? )

GO TO 1
16 IF (IRC1 ) .NE.7) GO TO 19
WPL O T =NPLOT +l

IF (NPLOT .GT.15 ) GO TO 1
DO 20 1= 1 ,2
20 VLABL E (I,NPLOT )=A LPHA(I+ l )
OUTPLT (NPLOT)=IR(2)
GO TO 1
19 IF (IR(1).NE.8) GO TO 18
IF (TPER.EQ.SSS ) GO TO 19~
IF (VR (L ~).GT .O .) GO TO 192
IFCI R (3).NE .O .AND. IR(3 ).NE. 1) GO 10 193
ISGCT= ISGCT+1

SIGMA( ISGC 1 )=VR( l )


SIGLB(ISGCT )=VR(2)
SIGUB (ISGCT ):VR(3)
ISNDX(I SGCT )= IR (2)
IDIST(ISGCI ):IR(3 )
GO TO 1
102

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

18 IF(IR(1) .NE. 9) GO TO 100


STEP = 11 .
READ (5 , 8)NP , IBVN SW ,IPLOT ,XLA MBD ,KSSIG ,(CEPSIG (I),I=1 ,5),PSIZE
8 EORMAT( 31Z4 ,2F 10. 3,512 ,E15 .7)
GO TO

100 IF(IR(1) .NE. 10) GO TO 191


IMVCT
IMVCT + 1
IMVNDX (IMVCT ) IR (2)
GO TO 1

192 IE (IR (3).GT.5) GO TO 193


ITCT= ITCT1
TSGM A (ITCT)=VR( 1)
TL B ( I T C T ) = V R ( 2 )

TUB (ITCT)=VR (3)


ITNDX 2(ITCT)=IR(2)
IF(IR(4) .GT.O) ITNDX2 (ITCT)=IR(L1 )
ITNDX(ITCT)=IR(2)
ITDI ST(ITCT )=IR(3)
TSPER ( I TCT ) = VR ( 4 )

TPSIG (ITCT)=TPSGMA
TYPPER (ITCT)=O.
IF (TPER.EQ.CPERTY) TYPPER(ITCT)= 1.
GO TO 1

19 ~4 R N S T R T = V R ( 1 )

GO TO 1
6 ,5518)
193 ~4RITE (
55 18 FOR M AT C1 X ,58HUNDEF INED DI STRIBUTION TYPE NUMBER ENTERED
CARD REJ
*ECTED)
WR ITE (6 ,3O)IR (1) ,(ALPHA (JC) ,J C 1 ,~4),IR (2),IR (3) ,T p ER ,T p SGMA ,
*V R (1 ),VR(2) ,VR(3 ) ,
*IR (4) ,VR (4)

191

GO TO 1

CONTINUE

NCASE =NCASE +1
RETURN

50 STOP
END

~~~~~~~~~ -~~~~~ -- -

-~~~-~~~~~~~~~ -~~~-

-~~~~~~~~~~ --

- -----

.
~

SUBR O UTINE O UPT 2

- . _

OUTPUT INITIALIZITION SUBROUTINE OUPT2


COMMON C(3830),G RAP H
EQUIVALE NC E (C (2017) ,DTCNT ) , (C(3167) ,NOOUT ) , (C(2016), PGCNT ) ,
),
( C ( 2 0 1 U , I TCNT ) , (C(2003) ,PCNT ) , (C(2015), CPP
C
)
)
(C(20
19),
(C(2
013),
C
(C(2018), TAPE
TA PEND ),
DOC
,
,
C
(C(2000), T
), (C(2021), KCONV ), (C(2025) ,TIME(1 )) ,
( C ( 2 0 0 8 ) , PL O TNO ) , ( C ( 2 0 0 9 ) , NOPLOT ) , ( C ( 3 1 6 8) , O U T N O ( 1 ) ) ,
C
C
(C(2O0 ~4), PPNT
), (C(2023) ,O POINT )
DIMEN SION GRAP H (1 , 1) ,TIME (300) ,OUTNO (50)
INTE G ER
PG CNT , DTCNT , OUTN O , O POINT
EQUIVALENCE (C(19B5), OUTPLT(1))

INTEGER OUTPLT

-~~

DIMEN SION OUTPLT( 15)


KCONV =O
ITCNT = D OC + 1 . 0
P C NT = T 0 . 0 0 0 0 0 1
PG C NT = 1
D TCNT = ( N O O U T +
IF C ITCNT .GE . 7) GO TO 2
ITCNT = ITC N T + 1
C ALL DUMPO
2 TIME(l )=T
OPOINT = 1
I F ( N O P L O T . E Q .O ) GOTO
DO 10 J:1 , NOPL OT
K=OUTPLT (J)
10 GRAPH (1 ,J)=C(K)
11 CONTINUE

RETURN
END

11

SUBROUTINE OUPT3
OUTPUT S U B R O U T I N E OUPT 3
CO MMON C ( 3 8 3 0 ) , G R A P H ( 3 0 0 , 4)
COMMON/ WKU 1/ONAMEO(50 )
EQUIVALENCE (C(3168), OUTNO(1)), (C(3218) ,ONAME 1 (1 )) ,
C (C(3268) ,ONAME2( 1 )) ,
C
(C(2017), DTCNT ), (C(3167) ,NO O UT ), (C(2016) ,PGCNT ),
),
C
(C(2014), ITCNT ), (C(2003) ,PCNT ), (C(2015), CPP
),
), (C(2664 ),DER
), (C(2018) ,TAPE
C
CC (2000),T
C
(C(2019) ,TAPEND ), (C(2008) , PLOTN O), (C(2009), NOPL O T ),
),(C(2004), PPNT
),(C(2025),TIME( 1)) ,
C
(C(2005) ,PPP
(C(2023), OP O INT )
C
)
EQUIVALENCE (C(1985) ,OUTPLT (1 )
5 0)
DIMENSI ON B(50),OUTNO(50),ONAME 1 (50 ) ,ONAME2 (
DIMENSION TIME(300)
DIMEN S ION O UTPLT (1 5 )
INTE GER DTCNT ,PGCNT , OUTNO
INTE G ER O PO INT
INTE G ER O UTPLT
DATA DER 1/O.O/

C
C SAVE SPOT JITTER MAX/MIN VALUES
IF(CC1 6SO).GT.C C1567 )) C(1567 ) = C(1680)
IE (C(1680).LT.C(1568)) C(l568) = C (1680)
IF(C (168I).GT.C(1577)) C (1577) = C (1681 )
IF (C(1681).LT.C(1578)) C(1578) = C (1681 )
C
IF (ITCNT. GT. 6) GO TO 7
IT C NT = IT C NT + 1
CALL DUMPO
PGCNT = 1
C
7 IF (DER. EQ. DER 1 ) GO TO 8
DER 1 = DEN
W R ITE (6 , 20)T ,DER
20 FORMAT( 1H ,SHT IME= , F14.7 ,2X , 1OL-ISTEP SIZE= , 1PE19 .7)
8 IF CT .LT . PCNT )GOTO 15
9 P C NT = P C NT + C PP
IF (PGCNT. NE. 1) GO TO 3
IF(NOOUT.LE.1 ) GO TO 3
1 W RITE (
6 ,2) (ONAMEO( I) ,ONAME 1 (I ) ,ONAME2( I) , I= 1 ,NOOUT)
4,7X , 3A
4 ,7X ,
2 FORMAT (1H 1 ,3X , 4HTIME ,5X ,5 (7X ,3A4 )//(20X , 3A4 ,7X ,3A
,3A
4)/)
13A 4 ,7X
PGCNT = 2DTCNT + 4
3 IF (PGCNT .GE . 86) GO TO 1
00 4 I = 1 ,NOOUT
J = OUTNO(I )
4 B(I) = C (J)
IF(NOOUT .LE. 1 ) GO TO 15
WRITE (6 ,5) T , (BCI ) , I = l , NOOUT )

5 F O R~1AT (//f ,F14.7, 1P5E 19 .7/(14X , 1P5E1 9.7))

PGCNT
PGCNT + DTCNT + 14
15 IF (T.LT .PPNT .OR.NO PLOT .EQ .O )RETURN
PPNT= PP NT+PPP
KPO 1NT =OPO INT +1
IF (KPOINT 300) 16 , 13, 18
13 WRITE (6, 14)
iLl FORMAT (/11111 * * **
WARNINGPLOTTING
C O INTS PL O TTED ** ** ~~/ / )

ARRAY

FILLEDONLY FIRST 300 P

I
16 OPO INT=KPO INT
TIME (OPOINT)=T
IF(NOPLOT .EQ.0 )GOTO 11
DO 10 J= 1 ,NOPL O T
K= OUTPLT(J)
10 GRAPH (OPOINT
,J)=C (K )
11 CONTINUE
1 8 RETURN

END

I
I ~.
~

~~~~

SUBROUTINE STGE2

COMMON C(3830)
EQUIVALENCE (C(2011) ,KSTEP ) ,
KCONV = 0
LCON V = 0
KSTEP = 1

(C(2020) ,LCONV ), (C(2021) ,KCONV )

RETURN
END

~~~~~~~~~~~~~~~~~~~~~~ ----

_ _

-.- .-

--

.- - .

- .
~~~~~~~~~~~~~~~~~

-.-

SUBR O UTINE STGE3


COMMON C(3830)
), (C(2001) ,TF
),
EQUIVALENCE (C(2000) ,T
), (C(2011) ,K STE P ),
EQUIVALENCE (C(2010) ,STEP
),
EQUIVALENCE (C(202I) ,K C ONV ), (C(2561) ,N
), (C(2664),DER(l )) ,
EQUIVALENCE (C(2663) ,HMAX
), (C(2965) ,VAR(1))
E Q UIVALENCE (C (2865 ),EU
),
), (C(1974) ,NJ
EQUIVALENCE (C(1973) ,KASE
,
DIMENSION
DER (1O 1 )
, VAR( 101 )
EXTERNAL A UX S UB
CALL G 41
IF (ABS( T-TF) .LE. 0.01 ) GO TO 20
IF ( (TFT ) .LT. 0.) GO TO 10
IF (LCONV .EQ. 2) GO TO 20
IF (LCONV .EQ. 1) GO TO 10
IF(DER(1) .LT. 0.) DER (1 )=_ DER( 1 )* .5

RETURN

10 IF(DER( 1 ) .GT . 0.) DER (1)= ~ DER(1).5


K C ONV
KC O NV + 1
IF(KCONV .GE. 10) GO TO 20

RETURN

20 PCNT = 1.0
IF (STEP .EQ. 11 .)GOTO 140
PREDER = DER (1)
DER (1 ) = 0.
NJ = N 1
N PT = 0
C ALL AMRK (AUX SUB )
DER (1) = PREDER
140 CALL OUPT 3
KSTEP = 2

RETURN
END

(C(2003) ,P C NT )
(C(2020),LC ONV )
(C(2662), HMIN )
(C(2765) ,EL (1))
(C(1975) ,NPT
EL(100)

SUBROUTINE

3
14
5

6
7
8

9
I

SUBL 2

COMMON C (3830)
EQUIVALENCE (C(2~46i) ,N OSUB ), (C(2462), SUBNO(1) )
DIMENSI O N
SUBNO (
99 )
DO 1 I = 1 , N O SUB
J = SUBNO (I )
GO TO ( 1 , 2, 3, Ll , 5, 6 , 7, 8, 9 ), J
C ALL INPT 2
GO TO 1
CALL OUPT2
GO TO 1
CALL STGE2
GO TO 1
CALL CNTR2
GO TO 1
C ALL RNDM 2
GO TO 1
C ALL AUXA 2
GO TO 1
C ALL A UXB 2
GO TO 1
CALL AUXC2
CO N T I N U E

RETURN
END

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

SUBROUTINE

3
~4

5
6

SUBL3
COMMON C(383O)
EQUIVALENCE (C (24461) ,NOSUB ), (C(21462) ,SUBNO C1 ) )
QIMEN SION
SUBNO (99)
DO 1 I = 1 , NO SUB
J = SUBNO(I)
GO TO ( 1 , 2, 3, 4 , 5, 6, 7, 8, 9 ), J
C ALL INPT 3
GO TO 1
C ALL OUP T3
GO TO 1
CALL STGE3
GO TO 1
C ALL C N T R 3
GO TO 1
C ALL R N D M 3
GO TO 1
CALL A IJXA 3
GO TO 1

~ CALL AUXB 3
GO TO 1
9 C ALL A UX C 3
1 CONTINUE

RETURN
END

SUBROUTINE Si
CSEEKER MODULE
C
COMMON C(3830)
)
EQUIVALENCE (C(2000), T
101 FORMAT (3 0U 0
TARGET A CQ UISITION
*
1 0ii
EPS Z = , 1PE 11 .3, 1 OH
10 2 F O RMAT (3 0H 0
PIT C H PLANE TRA C K
*
1O ~1
EPS Z = , 1PE 11 .3, 1 OH
YAW
ILANE TRACK
103 FORMAT (30H 0
*
1O ~-i
EPS Z = , 1PE1 1 .3, 1OH
C
C * * I N P U T DATA
EQUIVALENCE (C( 445),RLOCK )
)
EQUIVALENCE (C( 446),DT
)
EQUIVALENCE (CC 447),BDB
E QUIVA LENCE
E QU IVALENCE
EQUIVALENCE

EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQ UIVALENCE
EQUIVALENCE
EQUIVALENCE
EQU IVALENCE

E QU IVALENCE

EQUIVALENCE

(C(
(C(
(CC

4 4 8) , CFOV Z
4~4 9 ) , CFOV ~
450),GSX

( C C O 4 5 8 ) , WSN

E QU IVALENCE
EQUIVALENCE
E QU IVALENCE

EQUIVALENCE
EQUIVALENC E
EQUIVALENCE
EQUIVALEN CE
EQU IVALENCE

C
COTHER OUTPUTS

, 1PE 11. 3 )

)
)

)
)

(C(0Ll59) ,WL2

(C( U64 ),TRKZ

)
)

C I N P U T S FROM OTHER M O D U L E S
E Q U I V A L E N C E ( C C 0 3 7 1 ) , R ANGE
E Q U I V A L E N C E C C ( O 3 7 2 ) , RXBA
E Q U I V A L E N C E ( C ( 0 3 7 3 ) , RY B A
E Q U I V ~tL ~~N Ct. C C ( 0 3 7 4 ) , R Z B A
E Q UIVALENCE
( C C 1 7 3 9 ) , WP
E Q UIVALENCE
C C ( 1 7 4 3 ) , WQ
( C ( 1 7 ~4 7 ) , WR
EQUIVALENCE
C

CSTATE VA R IABLE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE

E lS Y

(C( 451) ,SEPS )


)
(CC L152) ,SWP
)
(CC 453),RBK
)
CC( 454),GEO
(C( 455) ,OPTNSK)
)
(C(01456) ,GS
)
(C(O~457 ),WSL

E Q u I V A L E N C E (C( 460) ,ST


E QU I VA L E r~CE (CC ~461), CAGE
E Q U I V A L E N C E (CC 462) ,TKRZ
EQUIVALENCE (CC 1463),TKRY
EQUIVALENCE

T = , F 8 .4 ,
EP S Y = ,1PE11.3 )
T = , F8.~4 ,
EPS Y = , 1PE 11 .3 )
T =

OUTPUTS
(C (0408), WLQD
(C(O 411), WL Q
(C(O412) ,WLRD
(C (0~415 ), WLR

( C ( O 4 1 6 ) , WL Q SD
CC (O~419 ) ,WLQS
( C ( 0 14 2 0 ) , WLRSD

CC (01423 ),WL R S
(C(O4 24 ),BTHT GD
(C(0L327) , BTHTG
CC (O428) ,BP S IGD
(C(O431) ,BPSIG

)
)

)
)
)
)
)
)

)
)
)
)

)
)
)
)

)
)

)
)

~~

- 9

---- -

- ~~~~~~ a
~~~~~---- -

EQUIVALENCE
EQUIVALENCE
E Q UIVALENCE
E Q UIVALENCE
E QUIVALENCE
E Q UIVALENCE
EQUIVALENCE
E Q UIVALENC E
EQUIVALENCE
E Q UIVALENCE
EQUIVALENCE

--

(C(11 ) ,BY)
(CC12) ,BZ)
( C ( O 4 0 3 ) , WLAM Q )
C CCO 4 0 7 ) , WLAMR )
(C ( 0 14 35 ) , BEPSZ )
( C ( 0 1 43 6 ) , BEP SI )
(C(0 ~437),WZ
)
)
( C ( 0 4 3 8) , WY
(C(O439) ,B G DEFL )
CC ( 4 6 5) , SDY )
CCC

466) ,

IF ( C( 899 ) .GT.O.O )RETU RN

SDZ )

C
CDIRECTION COSINES FOR BODY TO PLATFORM TRANSFORMATION
BTACT = BTHTG
BRACT = BPSIG
UCT=COSD(BTA CT)

UST= SINDCBTA CT)


UCP=COSD(BPACT)

USP= S INDCBPACT)
UB 11

UCT UCP

UB 3 3

UCT

UB 12 = UCTUSP
U 913 =UST
UB21 = USP
U 822 = UCP
UB23 = 0.
U531 = UST*UCP
UB 3 2 = USTUSP

C
C CALCULATE TOTAL DEFLECTION OF GIMBALS
BGDEFL=SQRT(BTHTG2BPSIG2 )
C
C TRAN SFORM LOS FROM BODY TO GIMBAL AXES
RX G = UB 11 *RXBA +UB 12*RYBA +UB 13*RZBA
RY G = UB2 1RXBA+UB22 RYBA+UB2 3RZBA
RZG = UB31RXBA UB 32 RYBA+UB33 RZBA
C
CLOS ERRORS IN PLATFORM COORDINATES
BEPSZ = ATAN D(RZG , RXG )
BEP S Y
A T AN D ( RY G , RXG )
C
C SEE KER OUTPUT SIGNALS
IF(CC1976) .LE .O .) GO TO 82
IF (T.LT.(ST-.000001)) GO TO 82

IF ( C ( 13) .LE .0 .)
C(13 )
1.
ST = T

C(2664)

820 CON T INUE

GO TO B20

DT / AINT(DT / C(2764))

ST = ST + DT
CV IDICON T R A C K E R

I F ( O P T N S K .L E . 0. ) GO TO 85
WLA M Q = GEO * BEP SZ
WLAM R = CEO * BEP S Y
WQP

WRP

W LAM Q

WLAM R

GO TO 30
CQUAD RANT TRACKER

85 CONTINUE

- -- -

- ~~~~~~~

~~~~---

--~~~-~~~

a,

IF (RANGE .GT. RLOCK ) GO TO 81


CZ = 2.BEPSZ/CFOVZ
CY z 2.BEPSY/CFOVY
IF ( C Z 2 .GT. 1. C Y 2 ) GO TO 81
BZ = SI GN (1. , BEP ~ L)
BY = SIGN( 1., BEFbI)
TKDB = BDB/2.(RANGE/32810.)2
IF (.ABS(BEPSZ).LT .TKDB) BZ = 0.
IF (ABS(BEPSY) .LT .TKDB) BY = 0.
CALL QD
IF (CAGE .GT . 0.) GO TO 82

UZ

BZ

UY = BY
CAGE = 1.
WR I TE (6 , i01 ) T , BEPSZ , BEP SY
GO TO 82
81 87. = 0.
BY = 0.
CSEEKER COMPENSATION
82 IF (OPTNSK .GT . 0.) GO TO 30
WLAM Q = BZ GS
WL AM R = BY GS
WQP = WLAM Q

WR P = WLA MR
IF ~WSL .LE. 0.) GO TO 83
WLQD

WL R D
WL QD

W LRD

WLA M Q

WLAMR

SEPS
SEE S
WLQD/WSL + WL Q
WLRD/WSL + WLR
WL QD

WLRD

WQP =
WRP =
WLAM Q = WQP
WLA M R = WR P
IF ( WSN . L E . 0 . ) GO TO 83
WLQS )
WLQSD = WSN ( W Q P
WLRS )
WLRSD z WSN (WRP
WQP = WLQS O/WL2 + WLQS
WRP = W L R S D / W L 2 + WLR S

CSEEKER SWITCHING LOGIC

83 IF (CAGE
PITCH PLANE
10 IF CTKRZ

.LE. 0.) GO TO 30

IF (BZUZ

TK R Z

.GT.

0 . ) GO TO 20

.GE . 0.) GO TO 12

1.

W R I TE (6 , 102) T , BEP SZ , BEPSY

12

GO TO 20
WLAMQ
WQP =

WL Q SD

BZGSX

WLAMQ
WLQD= 0.
UZ

= 0.

BZ

YAW PLANE

20 IF (TKRY .GT. 0.) GO TO 30


IF (BY UY

.GE. 0.) GO TO 22

TKRY = 1 .
W R I TE (6 ,103) 1 , BEPSZ , BEPSY

GO TO 30
22 W LAM R = BYGSX

WHP = WLAMR
WL R D = 0.

~~~~~~~

.- - - --_

.
_ - -1__ .
___ _ _ _
~~~~~
~~~~~~
-

=
=
=

113

,
-

~~--

~~ - - - - ~~ -

--- ~~~~~~~~~

- -~~~~~~

WLRSD =
UY = BY
3 0 C O NTINUE

0.

C
CMISSILE BODY RATES IN GIMBAL AXES
WZ = UB 3 1~~WP+UB32~ WQ+UB33WR
WY = UB21WPUB22 WQ+UB23 WR
C
CGIMBAL COUPLING
UZK = SWP ( BTHTG
+ .1BPSIG )
UYK = S W P * ( _ B P S I G
.1*BTHTG )

UZK

UZK

SDZ

UYK = UYK + SDY


C
CGIMBAL ANGLE DERIVATIVES
BTI-ITGD = WQP + (JZK WY
BPSIGD = W RP + UYK
W Z/UB33
C
IF ( CAGE .GT . 0 . ) RETURN
WLAM Q = 0.

WLAMR

- -

= 0.

WL QD = 0.
WL R D = 0.
WLQSD
0.
WL R SD = 0.
I3 TH TG D = 0.
BPSIGD
0.
RETURN
END

- -

SU BROUTINE Sl I
CSEEKER INIT.MODULE
COMMON C(3830)
DIMENSI O N IZ (5 0), IY (50), ISNDX (Z4O)
EQUIVALENCE (C(3634) , ISNDX(1)) , (C(3512), 13512)
EQUIVALENCE (C( 470), BTGERR )
EQUIVALENCE (CC 471) , BPGERR )
E Q UIVALENCE ( C C 4 6 5 ) , SDY )
EQUIVALENCE (CC 466), SDZ )
))
1 FORMAT( 5X ,2HB Z ,6X ,4 (I13, 111 )/ (13X ,l(I 13, 111 )
2 F O RMAT (5X ,2HBY ,6X ,4C 113, Ii1 )/(13X ,4(I 13, 111)) )
EQUIVALENCE (C( 11) ,BY)
E Q U I V A L E N C E C C ( 12 ) , BZ )
EQU IVALENCE(C(20 11 ) ,KSTEP )
E QU IVALENCE ( C ( 6 0 0) , I Z ( 1) )
EQUIVPLLENCE(C( 650),IY(1))
DIMENSION I P L ( 1 0 0 )

EQUIVALENCE (CC452 ) ,SW P)


EQUIVALENCE(C(0 41 1) ,WL Q)

E QU I VALENCE
EQUIVALENCE
E Q UIVALENCE
EQUIVALENCE
EQUIVALENCE
EQUIVALENCE

( C ( 0 4 l S ) , WLR )
(C(01419) ,WLQS )
( C ( 0 4 2 3 ) , WLRS )
(C(O427),BTH TG )
CC( 0431 ) ,BPSIG)
C C C 2 5 6 1 ) ,N )

EQUIVALENCE (C(2562) ,IPL (1))


EQUIVALENCE (C(3504 ),OPTN4 )
EQUIVALENCE (C(26b2)
,DE R SV )
IPL (N )= 424
I P L C N + 1 ) = 42 8
~
I P L C N + 2 ) = 4 08
IPL (N+ 3) =41 2
IPL (N +4 ) = 1416
IPL(N+5)= 420

N=N +6
C(41 1 )=0 .
C (415)=O.
C(419)=0.
C(423)=O .
BY = 0.
BZ= 0.
SDY
0.
SDZ
0.
DO 10 I = 1 , 135 12
IDO = I

C
C MONTE CARLO SEEKER OUTPUT STARTING VALUES
C
1F(ISNDX(I ).EQ .11) CALL MCA RLO (DUll , 1 , IDO )
IF (ISNDX(I ).EQ.12 ) CALL M CARLO (DUll , 1 , IDO )
IF (ISNDX(I).EQ. ~460) CALL MCARLO (DIJM , 1 , IDO )
I F ( A B S ( B Y ) . G T . O . ) BY = SIGN(1 .,BY)
IF (ABS(BZ).GT.O. ) BZ = SIGN (1., BZ)
C
C MONTE CARLO SEEKER POINTING ERROR
C
IF (ISNDX(I).EQ .470) CALL MCARLO (DUll , 1 , IDO )
IF (ISNDX(I).EQ. 471) CALL M CARLO (DUll , 1 , IDO )
C
~~

115

C
C

**

M ONTE C ARLO SEEKER DRIFT

IF (ISNDX(I).EQ. -465 ) CALL MCARLO (DUM , 1 , IDO)


IF (ISNDX (I).EQ . 466) CALL M CARLO (DUM , 1 , IDO )
-

10 C ONTINUE
BTHTG = BTHTG
BPSIG = BPSIG

+
+

BTGERR
BPGERR

WL QS=SWP (BTHTG-BPSIG)
WLQ=SWP (BTHTG-BPSIG)
WLR=SWP (BTUTG+BPSIG )
WLRS=SWP(BTHTG+BPSIG)
C(13) = 1 .
DERSV= .002
C(L461):0.
C(462)=0.
4 63 ):0.
4 614 ) =0
IF (OPTN4.GT.l .) GO TO 30
C(461 )= 1 .
C(1462)=1 .
C(-L4 6 3) : 1.
C( 14 6 14 )= 1.
30 CONTINUE
N I= 1
MI = 1
SET=O.
DO 200 1= 1 ,50
IZ (I)=0
200 IY(I)=O

RETURN
ENTRY Q D

IF(SET .GT.0.) RETURN


IF (NI.GT.50) RETURN
IF(MI.LE. 1O) GO TO 100
NI=N I+ 1
M I= 1
100 IZ(NI)=IZ(NI)+INT(BZ+2.)*10** (1OMI)
IY (NI)=IY(HI)+INT(BY+2 .)lO (lOM I)
MI=MI+ 1

RETURN
ENTRY S8

IF (SET.GT.0. .OR .KSTEP. NE.2) RETURN


SET=1.
WR I TE (6 , 1) (IZ(I) ,I= 1 ,NI)
WR I TE (6 ,2) (IY(I), I 1 ,NI)

RETU R N
END

- - - - -~~~~~~~~~~~~ -- .-

~~~~~~~~~~~~ - -

--

- -- . - . - ~~~

.
- ~~~~~ ---

--

SUBROUTINE
DIMENSION

XK

Y =
RETURN
END

0.

TABLE (X ,XI ,YI ,NX ,XK ,XLABEL ,Y)


XI (NX ),YI(NX)

FINT P 1 (X ,XI ,YI ,NX ,XK ,XLABEL )

--

rr

SUBROUTINE TABL 2(X ,Y ,XI ,YI ,ZI ,NX ,NY ,NXY ,XINTER ,XLABEL ,Z)
DIMEN SION XI (NX ) ,YI (NY) ,ZI (NXY )
Z=FINTP 2(X ,Y ,XI ,YI ,ZI ,NX ,NY , NX Y ,X I NTER ,XL ABEL )
RETURN
END

4,

rr
~

~~~

~~~~~~~~~~~~~~~

- -

~~~~

SUBROUTINE TIMEV (XPD Q )


DIMEN SION I C (6 ),X(100),Y(100) ,Y1 (2,4),T1 (300)
DUMMY SUBR O UTINE
ENTRY A 2I
ENTRY A Ll
ENTRY A 41

ENTRY

A5

ENTRY A5I
ENTRY C 2
ENTRY C21
ENTRY C 6
ENTRY C6 1

ENTRY C?
ENTRY C 7I
ENTRY C8
ENTRY C81
ENTRY C 9

ENTRY C91

ENTRY

ENTRY
ENTRY
ENTRY
ENTRY

ENTRY

ENTRY

ENTRY

ENTRY

C lO
C 1OI
D3
D31
D14
D 41I
D5

DSI

Gi
Gi l
G3 1
ENTRY GL II

ENTRY
ENTRY

-.

ENTRY G5I
ENTRY G 6

ENTRY 06I

ENTRY

ENTRY

ENTRY
ENTRY

S5
S51
36

S6I

ENTRY S7
ENTRY S7I
ENTRY S81
ENTRY 59
ENTRY 39 1
ENTRY SlO
ENTRY 3101

ENTRY
ENTRY

AUXA 1
AUXA 2

ENTRY
ENTRY
ENTRY

AUXB 3
AUX C 1
AUXC 2

ENTRY AUXA3
ENTRY AUXB 1
ENTRY A UXB 2

ENTRY A UXC 3
C NTR 1
C NT R 2
CN T R 3
ENTRY INPT 1
ENTRY INPT 2

ENTRY
ENTRY
ENTRY
ENTRY

INPT3

. - -

ENTRY N ORMAL (RX ,XL ,XU ,XM ,SG , RN)


ENTRY O UPT 1
ENTRY PR OC E S
ENTRY RANNU M (R ,S,T)
ENTRY RNDM 1
ENTRY RND M 2

ENTRY RNDM3
ENTRY STGE 1
ENTRY KIKSET
ENTRY C O UNTV
ENTRY WRITE
GOTO 2

ENTRY MCARLO (R ,M ,I)

GOTO 2

ENTRY AE R R O R (XL )

GO TO 2

ENTRY
CALL
ENTRY
ENTRY
ENTRY
ENTRY
ENTRY
ENTRY
ENTRY

TERR O R (XL )
EXIT
CEPAS (N ,I,IP ,XL ,S,IC ,P)
CEP P (X ,1,N ,S,X 1 ,I ,IC ,IP ,P)
N ORM (R ,X 1 ,XU ,XM ,S, R 1 )
KSTEST (Y ,N ,S ,X 1 ,SX , NI)
ZTABLE (Z ,F ,N)
PPLOT (X ,Y ,N ,C , I , R ,T ,XB ,YB ,XL ,P)
XL O C (X 1 ,H ,I ,IN)

GOTO 2

ENTRY 02
ENTRY G 2I
ENTRY S14

ENTRY S4I
ENTRY C 5
ENTRY C 5 1

ENTRY

RESET

ENTRY

PLOT 4(G,N , Y1 ,T1 ,NP ,NL ,NO)

ENTRY
ENTRY

PLOT 2
PLOTN

GOTO 2

ENIRY S2
ENTRY S3

ENTRY MCARLX
ENTRY SUBL 1
2 CONTINUE
RETURN
END

Pr

C
C

C
C
C
C
C
C
C
C
C
C
C

SUBROUTINE

PRINTER PLOTS OF SPECIFIED VARIABLES AGAINST TIME AN D AGAINST

EA C H OTHER.

..MAXIMUM OF 14 DEPENDENT VARIABLES PLOTTED AGAINST TIME


..IN PAIRED PLOTS , FIR ST VARIABLE IN PAIR IS A SSUMrD TO BE
THE INDEPENDENT VARIABLE.

POINTS PLACED IN WKU.YORK ON DPCOO2.


NEXT JOB STEP

NOPLOT C(2009) #PLOTS


NPLOTC ( 1 98 4) // PLOTSCOMPUTED???
VLABLE (2, 15)C (2325)8 CHARACTER AXIS NAMES
OUTPLT (15) C(1985) C ARRAY INDICIES OF VARIABLES TO BE PLOTED
GRAPH(300 ,Ll ) POINTS TO BE PLOTTED
TIME (300)C(2025) PLOT TIME INTERVALS
OPO INTC (2023) #POINTS TO BE PLOTTED
PLOTN2C (1983).- 1/PAIRED PLOTS INPUT
PPPC (2005 )W IDTH OF TIME SAMPLE INTERV A LINPUT
PPNTC(2004)-TIME PLOTTING IS BEGUN INPUT

C
C
C
C
C
C

COMMON C(3830),GRAPH(300 , 4)

EQUIVALENCE (C(2009) ,NOPLOT ),(C(1984) ,NPLOT) ,(C(1983), PLOTN2) ,


.(C(2325) ,VLABLE (1 , 1)),(C(1985) ,00TPLT (1)) , (C(2025), TIME (1)) ,
.(C(2023) ,OPO IN T ),(C(2005) ,PPP ), (C(20014),PPNT )
INTEGER O POINT ,NOPLOT ,NPLOT ,OUTPLT (15 )
DIMEN SION VLABLE(2 , 15) ,TIME (300)

NAMEL IST/TEST/N OPLOT ,NPLOT ,PL O TN2 , VLABLE ,O UTPLT ,OPOINT , PPP ,PPNT ,
.GRA PH

C
C
C
C
C

C
C
C
C

ACTUAL PLOTTING DONE BY

VARIABLES:

WKPLOT

DEFINE

F I L E 8 ( 1 U 0 O , 8O , E , N P O I N T )

OUTPUT GRAPH

LABEL S

NOPLOT :N PLOT
NPL 2= PLOTN2

W R I T E ( 8 l , l O ) NOPLOT , OPO INT , NPL2


10 FORMAT (313 )
IF (NOPLOT .EQ. 0) RETURN

PLACE TIME ARRAY

IN F I L E

DO 14 J = 1 , OP O I NT
1-14 .~RITE (8NPO INT , 11 ) T IME(J)
11 FORMAT(E 15.7)

TIME VS. --- PLOTS


DO 4 I:1 , N O PLOT

IF(I .GT .4) WRI TE (6 ,6)

IF (I.GT . 4 ) RETU RN
6 FORMAT( OWA RNI NU ..I>L I ..YO U CANNOT HAVE MORE THAN 14 TIME PLOTS ,
.1 , , 8 0 * )
C PLACE DEPENDENT VARIABLE IN FILE
DO 5 J = 1 ,OPOINT
-

~~~

----

~~~~-

-- - - --

~~~~~ ~ -

~- - -

- ---

Pr

______________________
-~~-~~~~~~ -~~~~--- - ~ -.
- - -- - --- - -- - --- - - -.-- ~~~

ft

~~~~

J , I)
5 WRITE (BNPOINT ,l l ) GRAPH (
W R I T E ( 8 N P O I N T , 2) VLABLE ( 1 , I ) , VLABLE ( 2 , I )
2 FO R M A T (2A 14 )
4 CONTINUE
W R I T E ( 6 , 6)
8 FORMAT ( OPLOT
RETURN
END

- - .

POINTS

PLACED

IN W K U . Y O R K )

.
-4

- -

~~~

-,

-r~~~.----.------~~~~~--.=.- -

_._

~~~~~

~~~~~~~

~~~~~~~~~~

SUBROUTINE ZERO
COMMON C ( 3 83 0)
EQUIVALENCE (C(19814) ,NPLOT )
EQUIVALENCE (C(2023) ,OPOINT )
E Q UIVALENCE ( C ( 2 3 6 1 ) , NO M OD )
EQUIVALENCE (C(2461) ,NOSUB )
E QUIVALENC E ( C ( 3 0 6 6 ) , N OL I ST )
E Q UIVALENCE ( C ( 3 16 7 ) , NOOUT )
INTEGER OPOINT
NOSLJ B
0
NOMOD
0
N OOUT
0
NOLIST = 0
OPOINT :O
N PLOT =0
RETURN

END

0.

L~~~~.

1
---~~- - ~~~ -

- - .-

---

~~~~~~~

---

-- ~~

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

~~

-- r
~~ :~ ~

- ~~

~~~~~~~~~

- -~~

REFERENCES
1.

Bryson , A . E . J r . and H o , Y.C., A pp l ied Optima l C o n t r o l,


Wal t ham , Mass., 1969 , pp. 148176 .

2.

B l a i s d el l ,

Kim , M. and Gr ider , K.V ., Terminal Guidance for Impact Attitude

Ang le Cons t rained Fl ig ht Trajectories , IEEE Transac tions on Aerospac e


and Elec tronic Systems, Vol. AES 9 , Nov . 1973 , pp. 852859 .

3.

Adler , F.P., Miss ile Guidance by Three Dimensional Proportional


Nav igation , Jo u rnal of A pp l ied Ph ysics, Vol . 27 , May 1956 , pp .500507.

4.

Mur taugh , S.A. and Cr iel , H.E., Fund amentals of Proportional


Nav igation , IEEE Spec trum, Vol.3 , Dec. 1966 , pp. 7685.

5.

Leistikow , L ., McCork le , R.D., and Rishel , R . W . , Optimum Control of


Air to Surface Missiles , Rept. AFFDLTR6 6--64 , Air Force Fli ght
Dynamics L a b . , Wri ght P a t t e r s o n Air Force Base , Ohio , March 1967.

6. Stallard , D.V. Discreto Optimal Terminal Con trol with App lications

to Missile Guidance , IEEE Transac tions on Automatic Control, Vol. AC1 8 ,

Aug. 1973 , pp. 373376 .


7.

Breza , M.J., Terminal Guidance of an Air to Surface Missile Using


Optimal Control and F i l t e r i ng , Master s Thesis , Air Force I n s t i t u t e
of Techology , Wri ght P a t t e rson Air Force Base , Ohio , June 1967.

8.

Fifer , S . , Analogue Computation, Vol. IV , New York : McGrawHill Book

9.

York , R..1 . and Pastrick , I L L . ,

Company , 1961 pp. 10871176.

Op timal Term inal Gu idance w it h Cons t rain t s


at Final Time , Journal of Spacecraft and Rockets, Vol. 14 , No. 6 ,
June 1977, pp. 381383.

10.

Willems , G., ~p~ imal Controllers for Homing Missiles, U.S . Army Missile
Command , Reds tone Arsenal , Alabama , Sep tember 1968 , Repor t No. RE68l5.

11.

Pastrick , H.L., Y ork , R.J., and St. Clair , D.C ., On the Realization
of an Optimal Control Law for a Terminally Guided Missile , Proceeding
~
of the Tenth Annual Southeastern Symposium on System Theory, March , 1978

pp. IVA 38 to IVA48.

12.

__

Lewis , C.L., Hooker , W .R ., Lee , A .W., Jr , and Harr i son , J.S ., T HAD T7
Missile MonteCarlo Termtnal Homing Simulation U t i l i z in L LS, Digi tal/
~
L inear and TV Seekers, U.S. Army Missile Command , Redstone Arsenal ,
Alab ama , July , 1976 , Report No . RG 7T2 .

~~~~~~~~~~~~~ ,

---

~~~ ~~-

,--

-- ~

--

--

- -, .

_2- ,----.:
___ =_~..
~~

- -~

~~ --

.
~~~

13.

Lewis , C.L. , An Engineering and Programming Guide for a Six Degree of

14.

York , R.J. and Pastrick , H . L . , ~ p tima l Con trol App l i ca tions f o r Missile

.-- -

Freedom, Terminal Homing Simulation Program, Guidance and Control Directorate , U.S. Army Missile Command , Reds tone Arsenal , Alabama , Jul y , 1973.
Systems, (Conf idential) Guidance and Control Directorate , U.S. Army Missile
Command , Reds tone Arsenal , Alab ama , F ebr u ary , 1977 , Technical Repor t
TG- 77 2.

.-,
uIuuuuI
~~~~~~ ~
~~~

________________________

Appendix A .
OPTIMAL TERMINAL GUIDANCE WITH
CONSTRAINTS AT FINAL TIME
Randy J . York
Western Kentucky University
Bowling Green , Kentucky 42101

Harold L. Pastrick
US Army Missile Command
Redstone Arsena l., Alabama 35809
and

Robert Pervine
Ph. D Candidate

Kent Pavey
Ph. D Candidate

U n i v e r s i t y of Kentuck y

Lexington , Kentucky

U n i v e r s i t y of Colorado

40506

Boulder , Colorado

80302

A BSTRAC T
A suboptima l termina l guidance law for a t a c t i c a l guided missile

was derived using a system of state equations describing the geometric


and dynamic conditions of the missile-target closure. The problem was
f o r m u l a t e d to minimize a l inear quadratic p e r f o r m a n c e index with con-

straints at fina l t ime . Initially , the autop ilot implementation assumed

an ideal instantaneous response. This assumption was then removed ,


and the resulting configuration was investi ga ted for various lag t imes .
As wou ld be expec t ed , the control laws deriv ed with the s implif ying
assumptions removed were much more complex . However , it was noted that
in the li miting case , when the a u top ilo t time lag tended to zero , a
simp ler control law surfaced . Comparisons were made then of the miss
dis tar ~e an d at t i tu de ang le at impact with vary ing lag co n di t ions . It
was det ermined tha t the suboptim a l control law was extremel y sensi tive
to various approximations , especially linear subarcs or cons tants , for
the t ime vary ing gains.

I.

INTRODUCTION

Recen t in t elligence suggests that the impenetrable nature of


heavy ar mor may be sus cep tible to missile attacks at a relatively high
angle of impact , with respect to the hori zon . In many modes of direct
encoun t er , the target may not be reachable with a bod y pi tch attitude
angle of t he proper magni tu de . There are several possible reasons for
this condition includin g lack of energy (fuel), lack of t ime to maneuver
in t o t he m ore desirable a tt i t ude , or lack of con trol information b y
appropria te sensors to command the response. This condition has been
recognized for some time at the Missile Research and Dev elopment Command
and consequentl y th ere have been attempts to modif y t rajec t ory shapes
by a vari ety of prede termined control laws . However , there has been a
cer tain lack of robustness in the solutions obtained over the entire
range of condi tions antici pa ted.

:__, _ _ _ _

- - - -~~~~- -~~~~~~~ :
~r

This situation motivated a search for optimal solutions to the

guidance proble m and a stud y of tradeoffs among the sub opti caa i. candi-

dates which were deemed feasible .

Termina l guidance schemes for tactica l missiles may be based on a


classical approach , such as a propor t ional navi ga tion and g u idance law
(3 ,4], or on a modern contro l theoretic approach (5-8]. In the latter ,
a control law is derived in terms of time-vary ing feedback gains when
formula ted as a linear quadratic control problem. A suboptima l terminal
guidance system for reentry vehicles , derived using the modern approach ,
was the basis for the initia l work on this problem.

Kim and Grider (2] studied a subopti ma l. termina l guidance system


for a reentry vehicle b y p lacing a cons t rain t on the body attitude
angle a t impact. Their problem was oriented to a long range hi gh

al titude mission . Their scenario was formulated as a l inear quadratic

control problem with certain key assumptions . The angle of attack of


the reentry vehicle was assumed to be small and thus was neglected .
Furthermore , the autop ilo t response was ass u med t o be ins t an t aneo u s ,
i.e., with no lag time attributed to the transfer of input comma nds
to output reaction.

These conditio ns have been studied in an extens ion of their earlier


work (9]. A formula tion is g iven for a system that has finite t ime
delay . In fact , the increase and decrease in time delay has interesting
ramifications on the solution. The angle of attack assumption is
inves tigated , and though not solved analytically in closed form , the
sys tem is derived .

There is m ore t h a n j u s t a passing academic i n t e r e s t in this problem.


As suggested previousl y, the antiarmo r role of several Army weapon syste m s very w ell may be enhanced by this technique . The reduction to a
prac tical imp le mentation of mechanization will be studied and described
in a future paper. This paper , however , summarizes the feasibility of
the concep t.

II.

STATE REPRESENTATION AND PROBLE M FORMULATION

The geo metry of the tactical missile-target position is g iven


in Fig ure A -I. Assuming that the angle of attack is small and can thus
be neglec ted (this assumption will be considered later ) and choosing
the following set of state variables

V
x~~

d
d
L

0
~~~~ ,

Yt- Y m
t

m
A

~~~~

(AI)

~: -~~~~~~

~~~

_.
.
. ..-.

~~~~~~~~~~~~~

~~~~~~~~~~ CTICA L M L SILE

~~~~~~~~~~~~~~~~~~

VERTICAL

REFERENCE

/
7

,
,

TARGET

_ _ _ _ _ _ _ _ _ _

Figure A-l .
Geometry of tactical missile
target positions .
where
a the position variable from the missile to the target ,
projected on the ground
Y a the position variable of the target
t

the position variable of the missile projected on the ground


the derivative of Yd, the missile to the target velocity
projected on the ground
A

L a the lateral acceleration of the missile

o a the body attitude angle of the missile


a

the angle of attack of the missile ,

ti~~ sys t em d ynamics can be expressed as

-A

cos 6

(A2)

_W A + K u
~
L L
1

AL

O K u
a
No t e t ha t t he lag in t he a ut op ilo t has been represented by a first
order lag ne twork
A (s)
L
u(s)

(A-3)

_ _ _ _ _ _

s+w

where u represents the control.


response of the autop ilot.)

(Previous work (2] assumed immediate

Lineari zing abo ut an operating point (i.e., cos 6


the system in the standard canonical form ,

b) and viewing

Ax + Bu

t he resul t is

~d

-b

~d

L
6

0
0

0 0

-w

9
A

(A-4)

K
-a

This op timal control proble m will have a controller of the form


U

C *Y +
d ~~~~ *~~~~+ C * o + c
~~~ y

*~~~

(A-5)

where C,~ , C,~, C 0, C~~ are time varying coefficients chosen to minimize
t he cost func t ional
J

Y~ (t ) + 792 (t )
f
f

+
~~~~

dt

(A-6)

-- -

.--

Mere
a f ina l (impact) time
t

a initial time
a constant weighting factors.

The integral term in the performance index of Equation (A-6) is used to


constrain the total expenditure of u. The actua l constraints on miss
distance and a ttitude angle at impact are :

~ 5 ft

I9(t f )I
III.

5 deg

PROBLEM SOLUTION

Using the Euler-Langra ng e formulation , a closed form solu t ioa


of the controller , u , was obtained as:
2

gyK (t
a
2

(+
~
-

~~a)

~~~~~

gy g
+

~~~~~~~~

-~~~

- -

~~~

(1

~~

t )
f

~~

(t

I
[(s )

- (I-

e
-j -~ ~ ,)

(+ )
~~

a)

t )

(gyi~~\

w (t_ t )
1
f ]

(t

W (tt
1
~~
e
}
/
A
)

W (tt
1
~~
e
)

W (t~ t
f
l

(t -

t )
f

)](t - t )
f

~
/i

~-s __ ___ _

~~~~~~ ~
-

- -

- -

--

-r.

c~~

yg2K(t

2
g ~ K(t - t )
f
2

* [2~~~~ 7 K

where

[ +
~

2
w (
t=t )\ (t
7~ K (
1
1
\
l
_
e
W

2
yK g
3

(\ l _ e

W (t-t )~
1
f \

t )
f

- y K ~~

)~//

W (tt
1
~~
e

)
(A-7)

W (tt
~~
~ - e 1
~~)2
i
a)(
+(

*~~ i + o

~~~~~~~

~~~

~~~~~~~~~~

{ (+
~~

t )
f~

2w~

- . -_- -

- .-~~ ca~~~~~~~r

E ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

( t

_ t
f

)j

W
(
3e ~~~~~~

3
t )
_: f
~ 2

(t

2
t )
f

}
/
A

tatime

I.
and

2
A a ~

(t;

_( )
~
t f)

L
(
5

4e

t t
(e

~~~~~~~~~~

L
~

1
~~~~~~

(
I

~~

1
)

W t t
1
~~~
2e ~~
)

2 &Yt_t
~~
e ~

)]

1+

(1

2yK ~~

e~~~

t_ t

~~)2

----- -~~~~~~~

+ (t

t )
f

(t - t

)
f ~

(
~

(t ~~t
f)

2
[~ g ] +

e~~~~~~~~~)

----

2
7K 2]
[
8

The coefficien ts are shown in Figures A-2 , A-3 , A-4 , and A-S for
following parameter values :
5~ K
0 0005 1 y
5, K
3823 ,
=

0.0000694, t
f

7.68.

The jinks near impact time are c h a r a c t e r i s t i c

of the choice of state variables .

200

TIME (sic)

Fi gure A-2.

Time vary ing optima l his tory


of coefficien t C~

~~
- --

--

- ~~~--- -- - - ~~~~ -- -

~~~

-- - - --- -

~~~~~~~
-- - - - - - - -- -

-- - - - --- - - - - - - -

--- - - - - -

--

--

- ------ -

~~~~~

__________

:
~~~~

TIME (si c)

Figure A-3 .

Time varying optimal history

of coefficient C.1.

10.000

C
~

:
10 .000

20 .000

30.000

4
TIME (sac)

Fi g u r e A-4 .

T ime varyi ng optima l history

o f coefficient C .
9

--- -

1__ ,_ _ _ ___
~
~~ ~

~ -~~- - - - -

- ~
---

_._

__ ~~~~

~~

~~~~~~~~~~~
-~
~~

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

1000

CA

0
1000

2000

3000
4000

Figure A-S .

TIME

i
-

(sic)

Time varying optima l history of coefficient

EFFECT IVENESS OF ThE CONTROL LAW

IV .

In an earlier work [21 , a control law of t he form


u

C~

~d

G.~~ *

*9

was derived under the assumption of zero autop ilot lag (i.e. A
The coefficien ts obtained were
[_ ~ g (t

t)

f-

[-~~ K

gyK ~ (t

f-

t)2/2]

2
t) - gy K~ (t
f-

2
7K g (t

t)3/2]

3
t) /6]

where
-b K
1
ga - ;--

- -

- - -

--

_ _ _ 1_ ._

~~

______ ___ _

- - - -~~~~

u).

-~~~

-T
~~~~~~~

~~~~~~~~~~~~~~~~

and

a2

Ya K ~~(t

t)

Sg 2 (t~

t)

2
Yg K ~

t)

(A8)

No te the relationshi p be tween the two c o n t r o l laws . As lag in the autop ilo t tends to zero i .e., W~
in Equation (A-7), the control law of
Equa tion (A-8) surfaces as the limiting case of the new control law .

Even though acceptable perforrnace is obtained us ing the control


law described b y the coefficients in Equation (A-8) there is some question regardin g the sensitivity of performance due to lag. In particular ,
this question arises : at what point does performance degenerate to
j u s t if y t he more co m p lex con t rol law t o achieve t he given performance
co nstraints? This is partiall y an s wered by r e f e r r i n g t o Fi gure A-6 ,
which is a plo t of miss distance versus lag for the two control laws ,

and Figure A-7 which is attitude angle at impact versus lag .


40

68 ft MISS AT
30-

N EW CONTR OL LAW
O L D CONTROL LAW

UNST ABLE FOR

20

1 0

, sl
~ 1 .2

I
4
LAG CONS T ANT

F igure A-6 .

A comparison of miss dist ance performance.

_________

____________

- --

.T

=- -~~~~~ -

~~~~~~~~~~~~

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

0.100
NEW CONTROL LAW
OLO CONTROL LAW

~~~~ UNSTABLE FOR~~~~~~~

o.o2:~~~~~~~~~~~~~~~~~~~~~~~ ~~
/
/
/
,

,
LAG CONSTANT

Figure A-7.

A comparison of impact angle performance.

For lag cons t ra in t s of

1, 2, the simulation with the control

law of E q u a t i o n (A-8) becomes unstable . This is not true with the law
describe d by Equation (A - 7 ) .
For
3 , a miss distance of 68 ft was

ob tained while the new control law had a miss distance of 6.1 feet.

Acceptable performance for Equation (A-8) is not obtained until


> 4.
It should be noted that the parameters in that control law were chosen
for a nomina l lag v a l u e of

5.

Should the amount of lag in the

tactical missile autop ilot vary , and no t be approx ima ted closel y a
p r i o r i , t h e n the more comp l i c a t e d control law needs to be implemented .
In other words , the control law of Equation (A-8) performs adequately
onl y f o r lag nea r t h e assumed nominal value.
The c o n t r o l law imp l e m e n t a t i o n was i n v e s t igated for a d ap t i b i l i t y

to app roximation si g n
a ls f r om p hys icall y realizable sources. The coeffic i e n t s ~~ C,~ , we r e app r oximated by linear functions pieced together at
th ree break p o in t s .

Pe r f o r m a n c e turned out to be too sensitive to the

approximatio n error , and the performance constraint could not be met .


The number of linear segments used was increased , bu t acceptable performance still was not achieved. It was concluded that successfu l
imple mentation would require more than merel y linea r approxima tions to
the t ime varying coefficients.

A-li.

- ~~~~~~~~ ---- ~~~----

The a n g l e of a t t a c k probabl y cannot be ignored for the larger


t a c t i c a l m i s s i l e s . For such a m i s s i l e , t h e s y s t e m of equations should
include the angle of attack a. In addition , bec a use it is feasible to
a c h i e v e o nl y a small a ng l e of a ttack a t im pac t , a reasonable performance
index to be minimized would seers to be:

(t ~~~) +

C~~ Y~~~

2
2
c 0 (tf )+ C a (t )+ C f ~~ 2 ( t) dt
2
3
f
4

(k-9 )

To p r o d u c e a f o r m u l a t i o n of the p r o b l e m which incorporates the


angle of a t t a c k cx , it is assumed t h a t cx can be expressed as a linear
combination of the control u and the attitude rate ~~. Assu m ing mo t ion
onl y in the p i t ch plane , then

a =

(pitch rate)

= (TAB

TCB )/ 1
2
2

where TAB

is t h e p i t c h i n g moment c o e f f i c i e n t due to angle of a t t a c k and


2
p i t c h r a t e , TCB 2 , is the p itching moment c o e f f i c i e nt due to f i n
d e f l e c t i o n . Now ,
TAB

-q Sd (C

cx

or
TAB

L cx
1

where

_ q S d

m / 12

d ynamic

missile r e f e r e n c e area

p ressure

missile reference dimens ion.

A-12
-

--------- -

-- - ~~~ --- - - -~~-- - --

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __
__
__
__
__
_ _ _

-1

I
A l so , noting p itch fin deflectio n (e ) is equivalent

TCB

q S . d ~~~ C

to control (u)

Therefore ,
9

L a
1

L e
2

L u
3

where
L

q ~~S

d .c / I2

Figure A-l , yields the fol lowing system for the s t a t e variables

~~
~cI ~d ~L

~d

0 1cz:

cos ( 9

W A.
~ l b

AL

- a)

e=e
L a+ L + L u
1
2
3

o
b

K u
3

Th is system can be linearized abo ut an operating

cos (9

a).

point , i. e . ,
Even so , the control problem w i t h the cost functional

in Equation (A-9 ) does not lend itself readil y to a closed form solutio n. Future work is antici pated using computer augmented algorith ms .

13

- ~~---- -~~

--~~~ :=~~~~~~ --..---

--

--

~~~~~~~~~~~~~~~~~~~~~~

~~~

. - ~~~~~~~~~~~~~~~~~~~~~~~~~~~~

-r

~~~~~~~

:?-

~~~~~~~~~~~~~~~~~~~~~~~

REFERENCES
I.

Bryson , A. E. J r . , and Ito , V . C., Applied Opt ima l ControL ,


Blasdell Publishing Company , Waltham , Massac h u s et t s , 1969 ,

p. 66.

2.

Kim , M. and G r i d er , K. V . , Terminal Guidance for Impact A t t i t u d e


Angle Constrained Flight T r a j e c tories , IEEE Transaction on Aerospace
and E l e c t r o n i c Systems, Vol. AES -9 , No. 6 , November 1973.

3.

Adler , F . P., Missile Guidance by Three-Dimensional Proportional


Navigation , Journa l of Applied Physics, Vol. 27, pp. 500507 ,
May 1956 .

4.

Mu r t a u g h

5.

Leist ikow , L., M cCork le , R. D., Rishel , R. W ., Optimum Control of


A ir to Surface M i s s i l e s , Report No. AFFDL-TR-66-64 , Air Force Flight
Dynamics Laboratory , W r i ght Patterson Air Force Base , Ohio ,

S.

A. and Criel , H. E ., Fundamentals of Proportional

Navi g a t i o n , IEEE S p e c t r u m ,

V o l . 3 , pp. 75-85 , December 1966.

March 1967.

6.

Bre za , M . J . , T e r m i n a l G u i d an c e of an Air to S u r f a c e M i s s i l e Using

Optima l Control and Filterin ,~~ Master s Thesis , Air Force Institute
of Technology , Wri ght Patterson Air Force Base , Oh io , June 1967.

7. Willems , C., Opt ima l Controllers for Homing Missiles , Report No.
RE-TR-68-15 , US Ar my M i s s i l e Command , Redstone A r s e n a l , Alabama ,
Sep t e m ber 1968.

8.

St allard , D. V . , C l a s s i c a l and Modern Guidance of Homing Intercep tor


Missiles , Report No. P2 47 , Ray theon Co; Missile Systems Div.,
Bedford , Massachuse tts , Ap r i l 1968.

9.

Pastrick , H. L. and York , R . J., Optimal Attitude Control for a


Guided Missile Application , Procnedings i976 Conference on Information Sciences and Systems, John Hopkins University , Baltimore ,
Maryland , April 1976.

A1 4

!IUuI
~~

Appendix B:

FourState Simulat ion Listing

FUNCTION KGCTRL(GAMMA ,BETA ,NBASE )


IMPLICIT REAL *8 (A.- H ,0Z)
REAL *14 KGCTRL ,GAMMA ,BETA
REAL
8 K 1 ,KA

COMMON/CONSTR/CQUANT

CO 14MON/YORK 1/DTRK ,
C
CINTEGRALS FOR RUNGK
U
YD ,YV ,THETA ,HA
11
,AL
CDERIVATIVES FOR RUNG K

,DYD ,DYV ,DTUETA ,DHA

,DAL

I! ,NX ,IYORK ,KUTTA

SIN(X) :DSIN(X)
COS(X) :DCOS(X)

C
C** * * * K IM_ G R I D ER CONTROL

C1Y(OG) :(BETA*G*OG

LAW
-

GGAMMAKA2OG2/2.)/DEL

C 1Y D ( 0 G ) :( ~~B E T A G O G2 + G G AM M A K A 2 5 O G 3 / 2 . ) / D E L

C 1THET(OG)= (BETA GAMMAKA .-GAMMA KAG2OG3/6 .)/DEL

777

C998
C.

C 1A L ( O G ) : O .
VMI (T)
-3 9 . 14 * T

CONTINUE

+ 1080.

FORMAT( 3F10 .O)

CINPUT VAL UES FOR GAMMA AND BETA


C
REA D (
5 ,998 ) GAMMA ,BETA
C

IF(GAMMA.EQ .O.) STOP


PRINT 100 ,BETA ,G AM MA
C 100 FORMAT( 1H 1 , 1X , BETA: ,D114 .6,2X , GAMMA: ,D114 .6)
C

TF :7.O
TIME :O .

DT= 1 ./128 .
DTRK :DT

NX :5

1< 1 :
1 .00
W1:9.8O

(A :.0005
JPRI :1O

YD :5000.0

VM :VM I(TIME)
VT :O .
AL :20.

C THDEG

:LAUI4

C H ANGLE

THDEG :8,.5

DELO

C
C
C
C

DEL O

:DESIRED

IMPACT ANGLE

145 .0

DELR A D :D E L O / 5 7 . 29 6
COS (DEL O)
B
B
0 .7 0 7 1 0 7
T H ETA :( THDEGDEL O )/57 .29 6
P R I N T 105 , GAMMA , DE L O , THDEG , B
105 FORMAT ( G A M M A : ,F1O. 14 ,2X , DELO : ,F1O. L$ ,2X , TUETA: ,F10. -4 ,2X , B : ,
AF 1O .4)
YV :V T VM SIN (THETA+DELRAD )
I-IA :1000.O
_______________________________________________________________________

~~~~~~~~~~~~~~~

~~~~~~~~~~~~~~~~

~
______

_____________

_____________

~~~~~~~~~~~~~~~~~~~

~~~~~~~~~~~~~~~~~~~~~~~~~~~

J :2 0

IF (NBASE .EQ. 0 ) WRITE (6,1319) TF , Ki , Wi , YD ,


#H A , VM , B , THDEG , DELO , KA
131 9 FORMA T(O INIT IAL VALUES OF TF ,G i O .4 ,/ Ki ,G10 .5 ,
* YD
,G 1O. -4 , 1-IA ,G 1O. -14 , VM ,G 10.14, B , G10. 4 , I

W i ,GiO.S,
INITIAL THDEG

U , 010.5 , IMPA CT ANGLE DESIRED , GiO. 5 , / , KA ,G10.5/0)


2 CONTINUE
THDEG :THETA57 .296
DELTH :THDEGDELO

XM :1 .EkY D

DO 10 KUTTA :1 , 4
VM

VM I (TIME )

G :K 1~ B/W 1
TGO :HA / (VM COS(THETA+DELRAD ))
DEL :BETA** 2_ GAMMA *BETA*KA **2*TGO_ BETA *G**2*TGO* *3/3 .+GPLMMA* G**2*KA
12TGO14/12.
CY :C 1Y(TGO)
CYD :C1YD(TGO )
CT:C 1THET(TGO)
CAL C ~AL(TGO )
U CY*YD+CYD*YV +CT*THETA+CAL*AL
DYD : YV

DYV -AL COS(THETA+DELRAD)


DT H ETA :KA U

DHA :VM COS (THETA+ DELRAD)


DAL :W 1AL +KlU
GOTO(30 ,5O ,6O ,LIO ) ,KUTTA

30 C O N T I N U E
60 T I M E :T I M E + .5DT

--

C
C
C
C

140 CONTINUE
5 0 C ALL RUN G K
10 CONTINUE
IF (DABS(YD) .LT.50.) DTRK :i./256.
35 IF (J.LT.JPF ~I) GO TO 3
36 C ONTINUE
PRINT 1 0 1 ,T IME ,YD ,YV ,THETA
101 FORMAT (OTIME :: ,F1O. Ll , L4X , REL DIST. YD : ,D114 .6,REL. VEL. YV : ,
1D 14 .6 ,~4X , T HETA ~~ ,D 14. 6)
PRINT 102 ,HA ,AL
102 FORMAT( V ERT. HEIGHT : ,D14.6 ,UX , AL : ,D14 .6)
PRINT 103, CY ,CY D ,CT , CAL ,U
103 FORMAT (1X , Cli : ,D114 .6,14~., Cu D: ,D1L I .6 ,14X , C 1IHETA : ,D14 .6 , 14X ,
1C1AL :: ,D114 .6, 14X , U : ,D 1J4 .6)

C
C

J :O

IF(TIME.GT .TF+ .2)GOTO 8


IF (UA .LT.O.)GOTO7
GO TO 2
7 C O NTIN UE
T I-I DEG :TH E TA57 .296
3

SET F U N C T I O N

KG CTRL

AND

:
ID2

CONSTRAINT
+

THDEG2

VALUES

CQUANT :O.O
WR ITE( 6 , 13 13 ) KGCT R L ,YL , THDEG , GAMMA , BETA , C Q UANT
1 3 1 3 FORMAT( OFUN C T I UN VALUE ,G 15.7 ,/ MISS DISTANCE ,G15.7 , -ANGLE : ,
G A M M A , 015.7,
$G 15.7 , I
BETA , 015.7 , CQUA NT ,G15.7, / , O ,

)
.6O( )

C
C

RETURN

PRINT 1O1 , T IME , YD ,YV , THDEG


PRINT 102 ,HA ,AL
-

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

- -

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

8 PRINT 4 0 5
405 FORMAT ( TIME EXCEEDS TF)

WR ITE (6 ,1313 ) KGCTRL ,YD ,


W RITE (6 , 1314 )

T I-IDE G , GAMMA , BETA , CQUAN T

1314 FORMAT( ORUN TERMINATED ,/ ,-4 (

RETURN

,80(),f))

END

SUBROUTINE RUNGK

IMPLI CIT REAL 8 (A.- H ,OZ )

DIMENSION X(5) ,DX(5),XA(5) ,DXA (5 )

COMMON/YORK 1 /DT ,X ,DX

, NX , IY OR K , K (JTT A
GO TO ( 1 O , 3 0 , 50 , 7 0 ) , K IJ TTA
10 DO 20 I= 1 ,NX
XMl ) : X ( I )
DXA(I) :DT*DX(I)
20 X(I) :X (I)-i- .5DXA(I)
U

RETURN

30 TDT :2.DT
HDT :.55DT
DO 40 I:1 ,NX
DXA(I) :DXA(I)+TDT*DX(I)
40 X(I) :XA(I)+HDT*DX(I)

RETURN

50 DO 60 I:1 ,NX

VDT :DTDX (I)


DXA (I):DXA(I)+2. VDT

60 X(I) :XA(I)+VDT
RETU R N
70 DO 80 I:1 ,NX
80 X ( I ) : X A ( I ) + ( D X A ( I ) + D T D X ( I ) ) / 6 .

RETURN
END

--

-~~~~~~~~~~~~~~~~~~ -~~~~~ -

. --~~~-- - - -~~~~~ --

- -

--

--

_________________________________________

______

Appendix C:

U-- -

--

--

Determination of Optimal Control Parameters:

An Application of Mathematical Programming


by
Drs . Daniel C. St. Clair and Randy J. York
western Kentucky University
Int roduction
The imp l e m e n t a t i o n of a c o n t r o l law in a s i m u l a t i o n s t u d y r e q u i r e s t h a t
the r e s e a r c h e r d e t e r m i n e a p r i o r i the value of several control law parameters .

Many of these parameters such as height , weight , distance , velocity , etc. are
determined by particular test requirements.

However , there may also be a set

of parameters which act as constants in the simulation but whose values are
difficult to determine apriori.

This paper suggests several easy to implement

m a t h e m a t i c al programming algorithms which can be used to produce reliab le esti


mates of such p a r a m e t e r s .
Each algorithm presented attempts to solve the ma thematical programming
problem
mm

(C 1)

J (a)

where a represents the vector of parameters to be determined and J (a)


is the scalar valued function being minimized.

In the application at the end

of th e pape r ,
+

2
(0(a)-O f )

(c-2)

w h e r e Y d (cz) is the missile miss distance , 0(a) is the attitude angle at impact ,
and O I S the desired attitude ang le at impact.

In that example , each d e t e r m i n a t i o n

of J(a) required execution of the control law simulation program .


In the next section , the c o m p u t a t i o n a l approach to solving (
1) is given
along wi th the description of several nonlinear programming algorithms .
References to both orig inal works and works containing additional test results

L.-

are provided to assist the reader in easy implementation of the algorithms .


The third section describes the app lication of the Hooke and Jeeves algorithm
to a specific control problem.
Review of Mathematical Programming Algorithms
The general approach to solving the nonlinear programming problem
mm

J(c z )

(1 bis)

is to proceed from a set of parameter estimates a


to a better set of
11

pa rameter estimates a t by using the equation

t1

The scalar step length

t~~

Ac~~ .

(C3 )

as well as the direction v e c t o r

L\ ci~

are determined

by the s t r a t e g y used in devising the particular method under consideration .


Some algorithms choose

t~~

b y p e r f o r m i n g a onedimensional search , such as those

described at the end of this section , along the t a . direction.


~

One area of

current research favors omitting the step length search by making appropriate
choices for Aa 1, for example see [1,21 .

Other algorithms [3 ,4) select the

direction of search and the step length simultaneously . The point a will
be accepted as the desired result if either of the following convergence
criteria are met , [5] viz.

<

where

and

a.

are arbitrarily snail and

I j .H

j_

(c4)

<
~~

represents the Euclidian norm .

if this criteria is not met , the algorithm is repeated.


The development of many nonlinear programming algorithms rely heavily on the
quadratic approximation of the objective function

J(cx)

provided by the f i r s t

t h r e e te rms of the Taylor s e r i e s .


J ( t

11

--

~T .)

J(a.

)
1

VJ ~~~~L~a~ +

~~

1~~~H
11

~ cj .

(C 5)

whe r e VJ~~~ is the gr a d i e n t vector and H~~ is the real symmecric Hessian
1
1

Both Vj .

mat rix .

and IL

are evaluated at a .1 .

Algorithm s are c l a s s i f i e d as 1.) d i r e c t search m e th o d s , 2 . ) m e t h o d s u s i n g


f i r s t p a r t i a l d e r i v a t i v e s , and 3 . ) m e t h o d s u s i n g second p a r t i a l d e r i v a t i v e s

contingent on whether the algorithm in question uses no term s , the first two
terms , o r all three term s r e s p e c t i v e l y of e q u a t i o n ( 5 ) .

Since t h e c a l c ul a t i o n

of partial derivatives is not feasible for the type of problems being considered
here , the algorithms discussed will be limited to those of the direct search type.
The development of direct search al gorithm s has been guided by e x t e n s i v e
experience and by thinking of the problem as that of following valleys down the
side of a mountain.
vector

These intuitively developed algorithms obtain the direction

of equation (3) by observing function values at previous points.

A onedimensional search in the


length

direction is used to determine the step

(see equation (3)).


e2

~~l

e1
Figure:

Hooke

Jeeves A l g o r i t h m f o r n

One simp le , but very effective , algorithm is Hooke and Jeeves Pattern
Search [6]. This algorithm consists of a series of exp loratory moves
followed by a pattern move , see the figure above . Beginning at a , a one
d imensional search is made in each of the n coordinate directions.
C-3

~~~~~~.
.

~~~~~ - -

--

- - - -

Thus , for

t:hese exploratory moves , the

of e q u a t i o n (3) are the n c o o r d i n a t e d i re c t i o n s .

The s e l e c t i o n of ~~ comp l e t e s the f i r s t set of exp l o r a t o r y moves.

The o n e d i m e n s i o n a l

search done at each step insures that


J(a j ) ~
The n e x t move is c a l l e d a p a t t e r n move and is made by c h o o s i n g

= an

an+l

(C6)

~~~~~~~~

Since J(a +1)is not usually evaluated , it may be the case that
~
>

~~~ n+i)

A second set of exploratory moves then beg ins at


point

and proceeds to produce the

~2n+l~
~~~~~~~~~~~~~~~

~~

a pattern move is made from a2ii+l . Otherwise , one returns to a~ and beg ins a new
set of exp loratory moves.
The p a t t e r n moves take large steps along valleys while exploratory moves

lead back down to the valley f l o o r .


that

is selected as
~n+1
a

where

Some p r o g r a m m e r s modif y equation (6) so

n+l

+ t

(a

a)
o

is a ste~ 1engt h chosen so


~~

~~~~~~
E i t h e r v e r s i o n is easY to p r o g r a m and r e q u i r e s onl y f u n c t i o n v a l u e s .
~~~~~~~~~

Hooke and

Jeeves ~-i~ gorit hm is particularL y attractive for use in minimizing functions having
1cn ~~

curved valleys.
Rosenbrock s algorithm [7] subscribes to the same general valley following

p h i l o s o p hy as does Hooke and Jeeve s a l g o r i t h m , hut Rosenhrock updates the set of

search directions b y adding each successful new search direction to the set while
removing the oldest search direction from

the set.

~~~~~~~~~~~~ --

-~~~~~~

n
~~~~~~~~~~~~
~~~~~~~~~~ --- -~~~~~~~~~~~~~~

_ ---- --

-~ ----a ~~~~~~~~~~~~~~

--- --

Given a set of n mutually orthogonal directions

compu t es

U 1, U

2, ...

from an initial estimate

~~~~~~~

A simple linear search is u sed to de t ermine

- j ,

Rosenbrock s method

by usin g equation (3).

This sequence of steps is

.
1

repeated until nonzero values are found for all

t~~ ,

1, 2,

,..,

n.

A f t e r every n i ter a t i o n s , the set of n search di n-ctions Lia is r eplaced


~
by the set

___ j
~

t czk
~
where k

= 1, 2 ,

..
.
,

n.

t~~

~~~~~

~~~~~

This new set of d i r e c t i o n s is orthogonalized by the

familiar GramSchmidt orthonormaUzation process [8] and the process is repeated.

Hence , t h e method a l i g n s the first search direction along the valley which has
-

just produced a favorable reduction in the value of the objective function.

further development o~ this idea has been pursued b y Davies , Swann , and Campey
~

[9].

An al gori t hm by ~ .J.). Powell [10] is designed to search in such a way that


the directions of search

-- z~ are mutually conjugate to the Hessian approximation

of inequality (5), v iz.


11.

f or

and k ,

1.

When J() is quadratic , i.e. inequality (5) becomes an equality, it can be


shown [11] that t h e minimum will be found in n steps.

The iterative procedure for Powells algorithm is given below .


1.) For i

1, 2 ,

...,

n , compute
.
~~1

where
2.)

t ~~

= T_

ii ~~~~~~~~~~~
1
1

is chosen such that J(a1) < J ( a 1) .


~~

max (J ( ~~
ci~ )
)J (
)
.
11
1
l<i<n
produces this maximum .
Set ~

3.) Calculate J
4.) If either J

Let k be the value of i which

3
3

= J(2
>

%a ~ )and define J 1

J ( c ~~) a n d J

2
or (J 2J +J )(J J -L )
1
2 3
1 2 ~

use the old directions Au1, I

1, 2 ,

.
.
.,

> ~~~~

(J

j )
1
3

n and repeat t h e

C5
-

-- --

-- _ -

____________

a l g o r i t h m u s in g

a a s

--

the s t a r t i n g point.

5.) If neither condition in Step 4 holds , replace t ct wi th


Aci..

( a a ) .

Pic k the new s t a r t i n g point


_*
a
0

where

~ k

so that

k ~~k

has been calculated to insure


~~~~~~~~~~~

Steps 15 a r e r e p e a t e d u n t i l t h e termination criteria a r e met


new

(see ( 4 ) ) .

Each

c* produced in St ep 5 will be conjugate to those previously produced by


~ k

this step.

The initial direction vectors

~~~~~,

1, 2 ,

.
.
.,

n ar e us uall y

chosen as the coordinate directions .

In certain test cases , Powell observed that his algorithm failed to compute
a new Ac~ when such computation was needed to preserve the linear indepen
de rtce of the current search directions.

Such behavior occurs when the valley of

the function becomes quite narrow and elongated and

be up da ted by S tep 5.

~~

repeatedl y fails to

Zangwill [12] incorporated per iodic coordina t e searches

in Powell s algorithm to resolve this problem.


While t h e r e are many o t her good direc t search al gori th m s available , the
ones presen ted here represent methods which are easily programmed and wh ich

req u ire a minimum back ground in nonlinear optimization.

The reader is referred

to Himmelblau [5] and to Lessman [11] for a comp lete discussion of methods ,

programming details , and examp les.

We conclude this section with a comnxnent

about oned imensional search algorithms.


To d e t e r m i n e the s c a l a r t. in equation (3) any algorithm may be used which
produces a scalar

t.

such that
il

+ t .
~~~~~)
1

<

J( :
~i i

).

While any t~ satisfying this criteria is acceptable , the greater the reduction
in the function value , the better is the performance of the entire algorithm .

- -

--

--

Popular oned imensional search algorithms include the Fibonacci search and the
quadratic fit algorithms.

These algorithms as well as others are described in

}limmelblau (5], Lessman [ill, and Cooper and Steinberg [13].


Application
The example cited here arises from the authors stud y [14] of a control law
presented in 1973 by Kim and Grider [15].

This law was designed to minimize

both the miss distance and the body attitude angle at impact for an air to
ground m i s s i l e p u r s u i ng a g r o u n d t a r g e t moving at a constan t speed .

The s t a t e

equation used was:

_ K

u cos

Ku

where
the distance between the missile and the target ,

projected on the ground ,


0

t h e bod y a t t i t u d e ang le of the missile (measured

K ,K ,W
1 1

from the vertical)

= constants

and u is the controller.

The controller u was given as

u (t ) = [C , C . , C ]
y

0
wi th

C~

= (~~~ g~~

2 2
a ~~K ~~ / 2 ) /t.
1

--

(-cz ct K
1 2

2
2

a~~~K
1

+ cz~~g

23
~~

/3

~~g

2 ,2 4
K

/12

C7
---.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~~
~~~~~

~~~~~

- -- - - .
~~~ ~~

~~

~~~~~~~~~~~~~~~~~~~~~~~

~~-ere
g

~ K 1COS 0 f /W 1
t

= the v e r t i c a l h e i g h t of m i s e i l e

The c o n t r o l l e r was desi gned to n i n i i : z e

= time

. H/(V*cosB)
~

(t )+ a 0 (t )+ a 7j
f
1
f

Kim and Crider reported a - -iss


~
impact of 0. 115 degrees .

the velocity of missile

= time of impact

t~~ e c n c t

functional:

i (t)dt

d i s t- i n c e

o~ 0.045 f e t w i t h

ar

a t t i t u d e angle at

They used t h e f o l l o w i n g initial va)ue

o obtain these

results :
H

= 10000 f e e t
=

10000 f e e t

1
K

0 = 45 degrees
V =
W

2000

= 0.0005
= 3823

ft ./second

5 rad ./second

= 6 .94 X lO~~

The au t h o r s w i s h e d to produce similar results for miss distance and attitude

angle at impact while vary ing H ,

e~ and V. The control law was also modified


~d
so that the attitude angle a t impac t 0 could be specified . The constants

and K were to remain unchanged . The parameters a

to achieve the desired result.


To obtain accep table values for a

gramming problem was formulated ,


mm

w h ere values of Y (a) and


d
T
[a , ~ 1
r
2

- -

3(
~~)

and a , had to be estimated

and a2 , the fou r- ing nonlinear pro-

2 (a)
~d

(0(a)_O

2
)

(2 b is)

a) would be d e t e r m i n e d by running
~(

The Hoo l e and Jeoves algorithm was s e l e c t e d

--

__ .

--- ~~~~~~~~~~ - --- --~~-_

;ls

the simulation with


the method to be

~~~~ -- --

--
~ ~~~~

--

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

~~

used in solving the problem , since any algorithm relying on derivative information
for J ( c t) had to be excluded.

The Table gives the results obtained for the initial

conditions :
H = 1000 f e e t
=

5000 feet

1095 f t ./ s e c .

= 45

The values of W 1, K 1, and K were t h e same as those used by Kim and G r i de r .

Initial
Val u es

Final
Val u es

3823.598

5525.508

0.34OE4

0.l9OE 6

1520.600

10.970

36.950

3.301

57.462

45.270

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

a
a

J(cz)
miss distance (feet)
attitude impact angle (dearees)
Table :

Computati onal Results

The column labeled I n i t i a l Values gives the i n i t i a l choices of


a

and

as well as the resulting objective function value , the miss distance , and

the attitude impact ang le.

The Final Values column shows the improved values

of a1 and a obtained after repeated applications of the Hooke and Jeeves


2
al gorithm. Note that the final attitude impact angle was 45.2700, just 0.270
over t h e d e s i r e d i mp a c t ang le of

= 45

0~

Before leaving this example , a computational note is in order.

The Final

Values reported in the Table were achieved by e x e c u t i n g t h e Hooke and Jeeves


algorithm , evaluating the results , and reexecuting the algorithm using the
b e s t r e s u l t s from the previous run . This series of steps was repeated several

times before the Final Values were obtained.

This is not unusual behavior f o r

any a l gorithm to exhibit since the choice of search direction s and step lengths
used by an algorithm may cause criteria (4b) to be satisfied before a minimum
C 9
_

--- -

-.-

----

~~~~~~~

value of

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

the o b j e c t i ve f u n c t i o n is o b t a i n e d .

-~

~~~~~~~~~~

~-~~~~ - -

Restartin g the algorithm will

usually be successful in the further refinemen t of the parameters being


estimated.
Conclusions
N o n l i n e a r p r o g r a m m i n g al g o r i t h m s are a v a l u a b l e t o o l in the d e t e r m i n a t i o n
of many control law parameters .

This paper h as presented a review of several

direct search algorithms and an example of how these algorithms can be applied
to obtain parameter estimates.

As can be seen in the con t rol problem example , the problem of finding optimal
values for the control parameters

and a is replaced
2

with a mathematical

pr ogramming problem ; however , t h i s new problem s t i l l has hidden in i t two unknown

search parameters t

and t .

The advantage of this formulation is that there

are several techniques available f o r finding t

and t

as discussed in the

paragraph just above the examp le.


In applyin g the a lg o r i t h m s , the approach is essentially the same.

Specify

the parameters to be estimated and the o b j e c t i v e f u n c t i o n t o be o p t i m i z e d .


n o n l i n e a r programming a l g o r i t h m t h e n uses v a l u e s of

t h e obj ective f u n c t io n at

various points to produce improved parameter estimates.

The process is repeated

u n t i l no a d d i t i o n al i r p r o vem e n t can be made in p a r a m e t e r v a l u e s or u n t i l t h e

objective function reaches a specified value.

-~

-S

~~~~~~~~~~~~~~~~~~~~
~S

cb
-

The

-------- - --S-S-- ---- -- -

- - _-i

______

~~~~~~~ ~~~~~

~~~~~~~~~~~~~~~~~~~~~~

~~~~

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

RE FE RENCES
1.

Bass , R.

A Rank Two Algorithm for Unconstrained Minimization ,

M a t h e m a t i c s of Computation, Vol. 26 , 1972 , pp. 129143.

2.

Fletcher , R. A New Approach to Variab le Metric Algorithms ,


Computer Journal , Vol. 13 , 1970 , pp. 317322.

3.

A Method for the Solution of Certain NonLinear


Levenberg, K.
Problems in Leas t Squares , Q u a r t e r ly of App lied M a t h e m a t i c s,

Vol. 8 , 1944 , pp. 2 12221.


4.

5.

An Al gori thm for LeastSquares Estimation


M a r q u a r d t , Donald W.
of NonLi near Parameters , Journal of the Socie ty of Industrial
App lied Math -matics, Vol. 2 , 1963 , pp. 164168.

Himmelblau , D.M .

Hill, 1972.

Applied Nonlinear Programm ing .

New York:

McGraw

and Jeeves , T.A. D i r e c t S e a r c h S o l u t i o n of Numerical


and Statistical Problems , Journal of the Association for Computing
Machinery, Vol. 8, 1961, pp. 212229 .

6.

Hooke , R.

7.

Rosenbrock , H .H. An Automatic Method for Finding the Greatest or


Least Value of a Function , The Computer Journal, Vol. 3, 1960, pp.
175184 .

8. Rice , J.R.

Experiments of Gram Schmidt Orthogonalization ,

Ma thematics of Computation , Vol. 20 , 1966 , pp. 325328.

9.

10.

Box , M.J. , D. Dav ies and W .H. Swann . Nonlinear Optimization

Techniques: I. C. I. Monograp h No. 5.


and Boyd , Ltd., 1969.

Edinburg h , Scotland : Oliver

An Effic ient Method for Finding the Minimum of a


Powell , M . J . D .
Function of Several Variables Without Calculating Derivatives .

Computer Journal ,

Vol. 7 , p p . 155 162 , 1964.

11.

Lessman , R.E.

12.

Zangwill , W .I. Minimizing a Function Without Calculating Derivatives .


Compute r J o u r n a l , Vol. 10, pp. 293296 , 1967.

13.

Optimization.

Mixed Noridcrivative Algorithms for Unconstrained


R oll a :

U n i v e r s i t y of M i sso u r i R o l l a , 1975.

Cooper , L. and Steinberg, 1). Introduction to Methods of Optimization.

Philadelphia:

W.B. Saunders , 1970.

c-li
- -

--

- - 5 -

- S-~~~~~~~~~~~~~~ -

--

_a.__ __ ___ _ ~
~
~~

- -- - - - --

~---5--~-

--S------------ -

~~~~~~~~~~~~~~~

14. Pastrick , ~I.L., Yo rk , R . J . , and S t . C lair , D . C . On t h e


R e a l i z a t i o n of an Optimal Co nt r o l Law f o r a Terminally Guided
Missile , Proceedings of the Tenth Annual Southeastern Symposium
on System Theory, March , 1978, pp. IVA38
IVA48 .

15.

KIm , M . and Grider , K.V . Terminal Guidance for Impact Attitude


Ang le Constrained Flight Trajectories , IEEE Transaction on Aerospace
and Electronic Systems, Vol. AES9 , 1973 , pp. 852859.

r~~~~~~~ -

~~~~~~~

Appendix D:

.~~~~~

- - -_

________

Hooke Jee ves Algorithm L i s t i n g

R EA L 8 C Q U A N T

COMMON N BASE , X , N , LA , K , KK , N P R N T , NT R Y
C O M M O N / C O N S T R / CQUANT
E X T E R N A L COMPUT
DIMENSION X0 (10) ,X( 1O) ,XMAX( 1O) ,XM IN (lO)
NT RY =O
N BA SE =0
REA D( 5 , 10 000)TOL ,AL PH ,BETA ,DEL ,LIM ,LT ,LSN ,N ,NPRNT
10000 FOR M AT (~4 E 1 2 . 0 , I 1 O , 4 I L h )
R E A D ( 5 , 1 0 0 0 1) ( X 0 ( I ) , X IIA X ( I ) , X M I N ( I ) , 1= 1 , N)
10001 FORMAT(3E 15 .O)
C
C I N P U T THETAO FOR K G C T R L F U N C T I O N
READ(5 , 10001 ) CQUANT
C
5
DO 10 I= 1 ,N
X ( I ) :X O ( I )
10
0
LA :

15

LT ,LIM ,TOL ,LSN ,DEL ,

XMAX ,XMIN ,

CALL PATTRN(COM PUT ,


A L-P H , BETA )

READ (5 ,10002 ,END=9 99 )L ,ALPH ,BETA ,DEL ,LA

10002 FO RMAT( h E 12.O ,I1 )


IF (LA.EQ.9) Go TO 15
IF (LA .GT.0) GO TO 5

999 STOP

END

~~

SUBROUTINE PATTRN (COM PUT , XMAX ,XM IN ,


IDEL ,ALPH ,BETA )
COMM O N NBASE ,X ,N ,LA ,K ,KK ,NPRNT , NTRY

C
C
C
C

X IS
C IS
XMAX
XMIN

LT ,LIM ,TOL ,LSN ,

DIMENSION X( 1O) ,XMAX (1O) ,XMIN( 1O)


DIMENSION C (100),D(100)
LOGICAL TYPL

ARRAY OF INDEPENDENT VARIABLES


ARRAY OF INDEPENDENT VARIABLES AT LAST BASE POINT
IS ARRAY OF MAXIMUM LIMITS ON X

IS ARRAY

OF MINIMUM LIMITS ON X

C A) IS A R R A Y OF STEP SIZES FOR X


C LA IS P PIMA RY BRANCHING CONTROL
:1 FOR INITIALIZATION
C
C
= 2 FOR INITIAL MOVE OF X(K ) IN EXPLORATORY SEARCH
:3 FOR REVERSE MOVE OF X(K ) IN EXPLORATORY SEARCH
C
:
C
~ FOR PATTERN MOVE
:5 FOR INITIAL MOVE OF X (K) AFTER PATTERN MOVE
C
C
6 FOR REVERSE MOVE OF X (K) AFTER PATTERN MOVE
C
= 7 FOR BASE POINT
: 5 WHEN SEARCH IS FINISHED
C

C K IS INDEX OF X
C KK IS COUNTER FOR NO. X ,S TESTED SINCE LA ST BASE POINT
C N IS NO.

OF I N D E P E N D E N T

VARIABLES

C DEL IS INITIAL STEP SIZE CONTROL

= 0 . 0 1 FOR
C LT IS CONTROL ON MAX KK BEFORE BASE POINT

0 INPUT

N i FOR 0 INPUT

: N FOR I N P U T G R E A T E R THAN N
C
C LIM IS MAX NO OF MOVES : 1. 0 E 5 FOR 0 I N P U T
C TOL IS BASE POINT TEST TOLERANCE AND M DII LIMIT ON STEP SIZE
C
1 .OE5 FOR 0 INPUT

- - --5 .

- -

--

- -- - - - ---5-

5---5-

--5-

- - - ~~-

-5-

-- --- 5-

- --5-

_ _

- 5--- -

C LSN IS CONTROL ON FUNCTION

EVALUATION

C
EVALUATION IS DONE IF LSN.NE.O
C ALPH IS R A T I O TO I N C R E A S E D ( K )
C
= 2 . 5 FOR 0 I N P U T
C BETA IS RATIO TO D E C R E A S E D ( K )
C
= 1/3 FOR 0 I N P U T
TYPE : . FAL SE.
IF(LA .EQ.9)GO TO 800
LA = 1
100
NCT = O
DO 14 L= 1 ,N

2000

IF ( X M A X ( L ) . G E . X M I N ( L ) ) GO
W R IT E ( 6 ,2000)L

12
13

14

CONTINUE
IF ( N C T . L T . N ) GO

LA = 1

17

20

80 0
810

10

LA : 10
GO TO 500
IF (XMAX (L).EQ.XMIN(L)) NCT=NCT+ 1
IF ( X ( L ) X M A X ( L ) ) 12 , 14 , 11
X ( L ) =XMAX ( L )
GO TO 14
Jr ( X M I N ( L ) X ( L ) ) 14 , 14 , 13
, ( L ) = XMI N (L )

11

15
16

AT BASE POI r ~!

FORMAT( 1HO ,68H0


INPUT DATA ERROR . LIMITS REVERSED FOR INDEPENDE
1NT VARIABLE NO. ,I4)

10

2010

TO

- -

TO

15

CALL COMPUT(SN )
W R ITE (6 ,201O)
1

FORMA T (1HO , L47H0 EACH PAIR OF UPPER AND LOWER LIMITS IS EQUAL
26H
NO SEARCH IS PO SSIB LE
)
LA : 1 0
GO TO 500
IF ( L T ) 1 7 , 17 , 16
IF (LT.GT.N) LT=N

GO TO 20

LT =N 1
IF ( L T . L T . 1)

LT = 1

IF ( L I M . L E . 0 ) L I M :1E5

IF (TOL.LE.O) TOL :1.OE5


IF ( D E L . L E . 0) D E L :O . O 1
IF (ALPH.LE.O) ALPH :2 .5
IF ( B E T A . L E . 0) B E T A : 1 . O / 3 . O
DO 810 L :1 , N
D( L ) = ( X M A X ( L ) _ X M I N ( L ) ) * D E L
C(L )=X(L)

WRITE (6,3000) (D(L) , L= 1 ,N)

3000 FORMAT( O$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ /


U I N I T I A L STEP SIZES / ,8E15.7 /
NO FAIL = O
NPF = O
NCT = 1
K= 1

KK:1

M 1= 1
M2 = 1
CALL C O M P U T ( S N )
SP:SN
SC= SN
LA = 1
GO TO 500
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

5-

- - -

-~~~~~~ --

5
~-~~--

- -5 - -

110
120
130
14 0
150
160
200
201
210

22 0
222
223
224
230
231
232

23 4
--

235
280
282

283
281

28 4

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

320
33 0

332

--

_ - -_ _ - _ _ --

T Y P E : .F A L S E .
LA=2
GO TO 150
IF ( . N O T . T Y P E ) GO TO 120
LA= 5
IF ( D ( K ) . E Q . O ) GO TO 34 0
X (K ):X (K )+ D (K )
IF ( X ( K ) .G T . X M A X ( K ) . O R . X ( K ) .LT .X M I N ( K ) )
1 GO TO (500 , 3 1O ,33 0 , 500 , 3 1O ,33 0 , 500 , 50 0) , LA
G0 T0 850
IF (SNSP) 201 ,310 ,310
IF (TYPE) D(K) :D(K)ALPH
SP= SN
NPF = 0
M 1= 1
t42= 1
K :K + 1
IF (K.GT.N) K :1
IF ( L T . L E . 1 ) GO TO 230
IF ( K K . L T . LT ) GO TO 224
GO TO 230
K K :K K + 1
GO TO 130
IF (SP+TOL*ABS(SC).GE .. SC) GO TO 234
IF ( T Y P E . A N D . ( N P F . L T . 5 ) . O R . . N O T . T Y P E ) G0TO232
GOT0234
LA :?
NBASE :NBASE+ 1
Ml =1
IF (LSN) 850 , 520 ,850
K K :K K + 1
IF ( KK N ) 130 , 130 , 235
N O F A I L :N O F A I L + 1
IF ( N O F A I L 1 5 ) 2 8 O , 500 , 500
IF (.NOT.TYPE) GOTO281
N PF :O
D O 283 I : 1 , N

X(I)=C(I)

G O T O 284

IF (M1 .GT.N) GO TO 300

KK = 1
M 1:1
M2 :1
SP:SC

GO TOM 10
300 WRITE(6 ,2O20 ) SC
GO TO 500
IF (TYPE) 00TO312
31 0
3 11
LA= 3
GO TO 313
LA :6
3 12
X(K)=X(K) D (K) D(K)
313
GO TO 160
321

~~~~~~~~~

IF (SN.GE.SP) GOTO33O

D(K) : D(K)
GO TO 210
X (K )= X (K )+D (K )
D X : T O L A BS (X ( K ) )
IF ( D X . L T . 1 . O E 3 0 )
D (K ) = D ( K ) B ETA
_

DX :1 .OE 3 0

D-3
_

5 -

- --

IF (A B S ( D ( K ) ) . G E . D X ) G0T0340
D(K) :SI GN(DX , D ( K ))
M 1:M 1+ 1
IF ( .N O T . TYPE ) GO TO 220
M2 :M2+ 1
IF ( M 2 . L E . N ) GO TO 220
M2= 1
NPF =NPF+ 1
G0 T0 220

333
340
342

34 3

500
2020

2030

END MINIMUM

FORMAT

OUTPUT BLO C K FOR LA ERROR

507

IF ( L S N ) 511 , 520 ,51 1


SP= SN
SC= SP
LA :11
KK = 1
D O 526 L : 1 , N
P = C ( L)
C (L) :X ( L )
IF ((2*X(L)_ P).LE.XMAX(L)) G0T0523
X ( L ) :XMAX ( L )

522

GO TO

523
52 4

530
850

ENTRY AND LIM FINISH

GO TO 8SO

510
511
520

527

COUNT

(I3, 5E 19.8)

IF (LA .EQ.l) GO TO 110


IF (LA.GE.8) GOTO900
LA :8
DO 507 I :1 , N
X(I)=C(I)

505
506

526

FAILURE

W R I T E ( 6 , 2 0 2 0 ) S N , SP , SC , DEL , TOL , A LPH , BETA , TYP E , N , K , KK , LA , LT , LSN , M 1 , M


12 , NPF ,NCT ,LIM ,NBASE ,NOFA IL
DO 501 I :1 , N
W R I T E ( 6 , 2 030) I , X ( I ) , C ( I ) , D ( I ) , XM A X ( I ) , X M I N ( I )

501

525

STEP SIZE TESTS AND PATTERN

CONTINUE
SP: , E 1 5 .7 , 5H
DEL : ,
SN : , E 1 5. 7 , 5H
F O R M A T (/ / / 5 H
SC= , E 15 . 7 , H
1 E15 .7 ,/6H TOL= ,E15 .7 ,8H ALPHA :,E15.7 ,TH BETA :
,E15.7/
18X ,8H TYPE
,L1 ,3X ,3H N :,13,3X ,3H K :,13,3X ,4H KK :,14 ,3X ,-4H LA :,
4H
LT
: , I 3 , 3 X , 5H L SN= , I 3 , 3 X , 4H M l = , 13 , 3X , 4H M2 = , I 3 , 3 X ,
113, / BX ,
15H NPF :, I 3 , / BX , 5H NCT :,16 ,3X ,5H LIM :, 17 ,3X ,7H NBASE :,I4 ,3X ,
18H NOFAIL= ,13, // 3M NO ,9X , 1HX , 18X , 1HC , 18X , 1HD , 16X ,-4HXMAX , 15X ,
1J I HXM A X , / )

526

I F( (2 X ( L ) - P ) . G E . X M I N ( L ) )
X ( L ) :X MI N ( L )
GO TO 526
X ( L ) : 2 X ( L ) - P
CONTINUE
N OFA IL :0
G O TO 85 0
SP= SN
TYPE : .TRU E .
GO TO 140
NCT =NCT+ 1
IF ( N C T L I M ) 852 , 851 , 851
IF (LA 8 )
500 , 852 , 500
CALL C O M P U T ( S N )

GOTO52 5

851
852
C
IF MOVE S U C C E S S F U L , CHECK H A R D
C
C
G O TO 1 000
IF ( S F .LT . S N ) GO TO 1000
--

-~~~

55--

CONSTRAINTSELSE

-5- --

CONTINUE.

C
C
C

IF (
ABS(ABS(CQUANT) -45.) . L E .
5 . 0) GO TO 1000
WRITE (6, 1 0 0 2) C QU ANT , LA , NCT , ( X ( I 4 ) , 14: 1 , N )
1002 F O R M A T ( 0 , 5O ( # ) , / H A R D C O N S T R A I N T V I O L A T E D :
, I2 , N C T = ,15 ,/ X : ,12G9.4)
%G 9 . 4 , LA
WRITE (
6 ,1003) ( D (I - 4 ) , I4:1 ,N)
1003 FORMAT( ARRAY OF STEPLENGTHS: / , 1 OG 1 1 . 4 )

C QU ANT

~~~

IF CONSTRAINT VIOLATED , FAKE INCREASE IN FCTN. VALUE AND LET ORIGINAL


CODE SHORTEN STEP LENGTH.
SN
SP + 1.0
:

1000 GO TO (
100 ,200 ,320 ,530 ,200 ,320 ,510 ,900 ,800),LA

900

RETURN

END

SUBROUTINE COMPUT(Z)

C
C

REAL KGCTRL
COMMON NBA SE ,X ,N ,LA ,K ,KK ,NPRNT ,NTRY
DIMENSION X (10)

YY: 100.0* (X(2)_X(l)**2)**2+(1_X(1))**2


CALL KIM GRIDER CONTROL LAW
YY :ABS (KGCTRL(X(1 ),X(2),NBASE ))
NTRY=NTRY+ 1
IF (LA.EQ.7) NBASE =NBASE+ 1
IF (LA.EQ.. 1 .OR . LA .GT.6 .OR . NPRNT .NE.O)
1WR ITE(6 ,5)YY ,LA ,K ,KK , NTRY ,NBASE , (I ,X (I),I:1 ,N)
F ORMAT ( 4H 0 SN ,E15.7 ,3H LA ,12 ,2H K ,I2 ,3H KK ,I2 ,5H NTRY ,15 ,
X ( ,I1 , 1H) ,E15 .7)))
1 6H NBASE , I5/(4 (4H
Z=Y Y
R ET U R N

END

INC KGCTRL DBLF CTN

//GO .SYSIN
0.0000 1

6199.992

DD

. 0 0 0 0 23 5 1 59 2

87.5

1.5

6200.
.1

0.050

0.

0.01000

0500

0.

1*

I
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

-5----

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Appendix E:

---- -5

Derivation of the ThreeState Controller

Problem : Derive a control Iaw of the form


u = c

( t )Y

c (t)Y
2
d

c 3 (t)e

(E-1)

to minimize the cost funct ional


J

= Y~~~ (t~~ )

where the state variables


=

-~e 2 (t~ ) ~J~~ u2 (t)dt

~~~~~~

= (K /W )ub
O=K -

& are subject to the dynamics

~~

~d

(E-2)

(E 3)
,

cos O(t )
f

(E4)
(E5)

Solution: Using the technique of Lagranglan multipliers , define


the Haniiltonian to be
~~~~ + A Y

1 d

X K bu / W + X K
2 1
1
3

(E6)

Following the method of solution as outlined in [1], in order f o r


u to be the optimal controller ,
=

0 = 2~ u

X K b /W
2 1

+ X 2K

and so
K
R1
2 1~~ 1

A K ]/(2~3)
3

The con t r o l l e r u then is determined once the Lagrang ia n m u l t i p lie r s


are known .
Determ i nation of A 1,

~2 ~3

The condition t h a t t h e y must s a t i s f y is


=

~~~~~~

1,2 ,3

(E9 )

whe r e
X

~th state

_- -

1,2 ,3

--

- - _ --

-- -- -- - - - -

_ - - - -- -~~~~~~

_-

---~~~~~~~~~~~~~~ 5~~~~~~~~~~

Thus,

~H/BX 2

au / ax

= 0 =

H/- Y
~ ~ d

aH/ Yd =
~

3 =

~~~~~

(Ell)

~~ 2

= U

(E 1O)

(E12)

Boundary conditions f o r A , A , A are given by:


1 2
3

A ~(t )= aU/ ~ X 1(t f ) , i = 1,2,3 .


f

(El3)

Thus ,
A (t )
1 f

X (t )
2 f
~ 3 (t f

2 Y (t )
~ d f

(E14)

(El5)

) = 2 .Y )(t ) .
f

(E16)

Solving the differential equation (E10) and using condition


(
El4) , we get

A 1 (t)

= 2 .Y d

(E17)

(tf ).

Equation (Ell) combined with Equation (El7) yields

(E 18)

A (
t + c .
t) = _2 Y (t )
2
~ d f

Boundary Condition (E15) imp lies


A (t)
2

2 Y (t f ) t
d

2Y

(t )t .
f f

(El9)

For A 3, Equation (El2) combined with condition (El6) gives


A (t) = 2 y0 (t ).
1
3

(E20)

Hence , the control u given by Equation (E8) becomes


u(t) = Ig.Y~ (t )(t_t ) K
f
f

Y O ( t )J/8
f

(E21)

whe r e

K 1b/W 1 .

(E22)

The controller can now be determined if Yd (t )and O (t f )are


f
known in closed f o r m .

-_

~~ - --- 5--

~~~~~~~

Determination of Y (t ) and 6 ( t )
d

On ce the form of the controller has been determined as in


Equation (E21), we can return to the original system of differential
equations (E3)
(E5) and solve for Y (t), 0(t). Equation (E4)
a
becomes

gu

)(t_t ) K
~[
~ Y~~(t f
f

-rO (t ) ] / ~ .
f

(E-23)

Integrating ,
2

)(t_t )/2
= ~[
~ Y~ (t f
f

K Y8 (t )(t_t ) + d ]/ ~
f
f
1

(E 24 )

and again ,
Y

(t)

3 /6_K . 10 (t )(
2/2
(t_t )
t...t )
~[
~ Y~ (tf )
f
f
f

d (t_ .t )+ d ]/ ~ .
f
1
2

(E25)

A l so , fo r 8 ( t ) ,

K~ [g Y~ (t )(t_t f ) K~ y0 (t )J/~
f
f

so
2
(t_t )/ 2
0(t) = K~ [g Y~~(t )
f
f

)
~ 0(tf)(t-t f + d 3}/ 6

(E-26)

Now , letting t
t 0 we determine d 1, d 2, d 3 f rom Equations
(E24), (E25) , (E26). Substituting and simplifying yields:
d

= _ (t _ t

2
t
)~~.Y~~(

) / 2 + (t _t ) K a .Y 0 ( t )
f
f

3
2
= (t _ t ) g .Y ~~(t ) / 3 + (t -t ) (-K
f
f
0 f
+

(tQ _ t )(_ ~ Y~~(t )/g) + ~Y~~(t )/ g


f
f
0

2 (g.Y
= (t t )

o f

(
t ))/2 + (t t )K

d f

o f a

~Y~ (t .)/~

(E-27)

Y0 (t ))/2
f

(E-28)

yO (t ) + 88 (t )/K

Using d 1, d 2, d 3 and now letting t=t f , we can w r i t e do wn two


equationstwo unknowns: Y d ( t f ) an d 0 ( t f )
.
Y (t )= g d
d f

--- - 5 5

~~~~

(t

5-

2 /~

2 (_K
_ t ) 3Y (t )g 2 / ( 3 ) + (t -t )
t )g)/
~~ f
~
~ yO (
0 ~
f
f
(2~3) + (t _ t )(_Y d (to )+ Y (t )
O f
d o
(t

~ -3

(E29)

_______________________

and
0(t ) = K d IF

a3

2
2
t ))/(26) + (t -t )K y0 (t )/~ + 0 (t )
0(t ) = (t -t ) (-gK Y (
f
o f
ad
f
o f
a
f
o

Solving Two Equations


The system fot Y
Y (t ) [ l g 2 Q 3 / ( 3 8) J

Two Unknowns

(t

),

8( t

+ 0 (t )[K

d f

(E-31)

) is (with Q

ygf 2 I (21&)]

t _ t ).
0 f

Y (t )
d o

f2Y (t )
d o

(E32)

and

Y (t )[gK Q 2 I(2 j1
~
d f
a

+ O(t )[1K ~~~~~~

0(t ) .

(E33)

Using Crammer s Rule to solve , we calculate the determinant of the


coef f i c i e n t m a t r i x :
det A

(l/ ~ 2 ) [yg 2 K 2 c~4 / 12

8g 2 Q 3 / 3 +

+ ~~

(E-34)

Defining

I3

~~ E

2 det A

(E35)

we obtain the following:


Y

2
2
c [_ Y (t )
6K YY d (t ) 1
~ ~ d o
.
2
2
0(t )K yg8]/2}/~
~ [2 8Y~ (t)K ~

2
(t )= ( Y (
f
d
~~

0(t

t)

(E 36)

) = ( 2O (t )
~~

8gK Y(t )Q 2/2

[38~ K Y ~~(t)

ad

2
2g 88(t )]Q 3/6}/L~

(E37)

Therefore , returning to Equation (E2l),


u ( t ; t ) = gY (t )12/8
o
d f

K yO (t )/~3 ,
a
f

(E38)

and substituting Equations (E36) , (E37 ) and simplify ing, yields


u(t)

Y ( t ) [ g F . gK 2y1A 2 + gK 2yc~2 I2]I~


d
a
a
2
2yf23 + K 2 yg A 3 / 2 ] / t
+ Y (t)[g8c
gK
~
~
~

2 3
~(t )[K -,g ~ /2

K~y8

E-4

(E 39)

2 3
~K g ~l / 3 ]/ ~

--

-- :
-

~~~~~

- ---~:
i
~

- -::

r--

H ence , we conclude t h a t
c1 (t) =
c2 (t )

c3 (t)

[gf

~(tft )

gK ~ y ( t f t )

2
[g ~ (tft)

gK

2
3
[KaYE (tf_t) /6

/ 2]/i\

2
_t)/2]/
a y(t f
~

K Y ]/A
a~

(E40 )

(E41)
(
E-42)

where
=

2 2
4
yg K (t -t) / 12
f

2
(t -t ) I3
~g
f
~~

2
2
Y8K a (t f_t ) + 8

(E-43)

and

L L_

K b/W
1 1

--

~~~~~~~~~~~~~~

Fr

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

.--

~~~~~~~~~~~~~~~~~~~~

~~~~~~~~~~~~~~~~

- - . .,
~
~~
- _ ~~~~~~

._

- --.,.- _ - -

D1~--TRIBUTION
ARMY

No . of Cop ies

Defe nse Documentation (~e n t~~


Cameron Station
Alexandria , Virginia 22314

12

Commander
ASD
ATTh: DRCPH }iFF , Col. Dobbs

Wrigh tPatterson Air Force Base , Ohio

45433

Commander
P ica t inny Arsenal

ATTN : DRCPMCAWSPA , Col. R. Ph illips

Dover , New Jersey

07801

LOCAL
DRDMIT

DRDMITG

DRDMITCN
DRCPMHF
DRCPNHFE

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Dr. Kobler

Col . Fiert

Mr. H u f f
~h . Ga m b i l 1
Dr. Pa~ t r i c k

1
1
2

Mr . Coiner

- -5

-~~~~~~~~~

- ~~~~~~~~

1.

- -

-5- -

- ~~~~~~~

-55~~~~~~~~

- 5 - - --

--

L~~..