Scribd Logo
Upload

Welcome to Scribd!

  • NASA Derivation and Definition of Linear Aircraft Model 1988

    Uploaded by

    Mary-Jane Stevens
    17 views108 pages
    NASA Derivation and Definition of Linear Aircraft Model 1988

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    Report this Document
    NASA Derivation and Definition of Linear Aircraft Model 1988

    Copyright:

    © All Rights Reserved
    Download as PDF or read online from Scribd
    Flag for inappropriate content
    17 views108 pages

    NASA Derivation and Definition of Linear Aircraft Model 1988

    Uploaded by

    Mary-Jane Stevens
    NASA Derivation and Definition of Linear Aircraft Model 1988

    Copyright:

    © All Rights Reserved
    Download as PDF or read online from Scribd
    Flag for inappropriate content
    of 108
    NASA Reference Publication 1207 1988 NASA National Aeronautics and Space Administration Scientific and Technical Information Division (@hpssinirs.nasa.gowssarch jep7 Derivation and Definition of a Linear Aircraft Model Eugene L. Duke, Robert F. Antoniewicz, and Keith D. Krambeer Ames Research Center Dryden Flight Research Facility Edwards, California CONTENTS SUMMARY INTRODUCTION SYMBOLS Vectors Matrices... . semen teReegssages SUBEERPUS 0 oe coe ca sce eo ae een sees Superscript 1 NONLINEAR SYSTEM EQUATIONS 1.1 Definition of Reference Systems... « 1.2 Nonlinear State Equations... . « 1.2.1 Rotational acceleration. 6... 0s eee eee eee 1.2.2 Translational acceleration. 6... 1.2.3 Attitude rates, 1.2.4 Earth-relative velocity. « « 1.3. Nonlinear Observation Equations . 13.1 Accclerations... 6... seen ee 1.3.2 Air data parameter: 1.3.3 Plightpath-related parameters 1.34 Energy-related parameters. . 1.3.5 Force parameters. 1.3.6 Body axis rates and accelerations. sees 1.3.7 Instruments displaced from the vehicle center of gi 1.3.8 Miscellaneous observation paramictors. « 2 LINEAR SYSTEM EQUATIONS 2.1 Linearization of the State Equation - 2.2 Linearization of the Observation Equation . . 2.3 Definition of Matrices in Linearized System Equations 24 Elements of the Linearized System Matrices. . « 3 CONCLUDING REMARKS APPENDIXES. A—AERODYNAMIC FORCES AND MOMENTS B—DERIVATION OF THE WIND AXIS TRANSLATIONAL PARAMETERS V, 4, AND 6 BL Preliminary Definitions B.2_ Derivation of V Equation B.3. Derivation of & Equation Th Derivation of # Equation eer_//_mtemonanse mss E BLANK NOt FILMED 20 2 2L au 2 22 23 a 26 aT 30 31 at 35 35 35 36 37 C—GENERALIZED DERIVATIVES 39 C.L Generalized Detivatives of the Time Derivatives of State Variables... 0... eee es 39 C.2 Generalized Derivatives of the Observation Variables 13 D—EVALUATION OF DERIVATIVES 49 D.t Preliminary Evaluation : 49 49 DLL Rolling moment derivatives. . . D.1.2 Pitching moment derivatives. « : 50 D.l.3. Yawing moment derivatives... . catanesus (60 D.14 Drag force derivatives... «+ + amgsee 51 D.L.5 Sideforce derivatives... beeen eee . Sl D.1.6 Lift force derivatives... . « sEmaz 52 D.2 Evaluation of the Derivatives of the Time Derivat of the State Variables... +. « 53 Roll acceleration derivatives. eee 53 Pitch acccleration derivatives. . se 51 Yaw acceleration derivatives... 000s eee eee ee bene 88 Decoupled roll acceleration derivative ee sames 56 .2.5 Decoupled pitch acceleration derivatives. . 57 D.2.6 Decoupled yaw acceleration derivatives. 6-0 e vee e erence eee eres BPE 1.27 Total vehicle acceleration derivatives... . « « eee BS D.2.8 Angle-of-attack rate derivatives. abies be ute Bate 59 D.2.9 Angle-of-sideslip rate derivatives. 60 D.2.10 Roll attitude rate derivatives: siete ve ete 62 D.2.11 Piteh attitude rate derivatives. 62 D.2.12 Ieading rate derivatives. veces 68 1.2.13 Altitude rate derivatives. gym 16 D.2.14 North acceleration derivatives... 06. c vee eee e eee e eee eee nee 65 D.2.15 East acceleration derivatives. amass 65 D.3. Evaluation of the Derivatives of the Observation Variables 66 1.3.1 Longitudinal kinematic acceleration derivatives. 66 1.3.2 Lateral kinematic acceleration derivatives... . or D.3.3. Zbody axis kinematic acceleration derivatives. @ 68 D.3.4__x body axis accelerometer output derivatives. 69 D.3.5_y body axis accelerometer output derivatives. .. - 69 D.3.6 z body axis accelerometer output derivatives. « 70 D.3.7 Normal accelerometer output derivatives. n D.38 Derivatives of x body axis accelerometer output not at the vehicle center of gravity, 7 Stereo bia owe 5 2 D.3.9. Derivatives of y body axis accelerometer output not at vehicle center of gravity. . . errr ree 2 D.3.10 Derivatives of z body axis accelerometer output not a vehicle center of gravity... . aay eases 3 D.3.11 Derivatives of normal accelerometer output not at velile center of gravity... see eee : piamae ena: D.3.12 Load factor derivatives. cee ee see 5 1.3.13 Speed of sound derivatives. seveeee 6 D.3.14 Mach number derivatives. 66.66 e ev es eee 7 D.3.15 Reynolds number derivatives. . beet e eee eee ees TT D.3.16 Reynolds number per unit length derivatives. . . or swors 78 D.3.17 Dynamic pressure derivatives... 2... 5. ict e een eee 7 D.3.18 Impact pressure derivatives... 0... eve eee eee sg 80 D319 Mach meter calibration ratio derivatives. - . 8 1.3.20 Total temperature derivatives... - 82 D.3.21 Flightpath angle derivative 82 1.3.22 Plightpath acceleration deri 83 1.3.23 Vertical acceleration derivatives... . « sone st 1.3.24 Specific energy derivatives, ©. 00-0 eee eee eee eee ee ovee 85 D.3.25 Specific power derivatives. sees 86 D.3.26 Normal force derivatives. 87 D.3.27 Axial force derivatives. . . : 87 1.3.28 x body axis rate derivatives... .. . « e@anse ae aast 168) D.3.29 y body axis rate derivatives. . vee 89 D.3.30 z body axis rate derivatives. . : 90 D.3.31 x body axis acceleration derivatives ++ 90 D.3.32 y body axis acceleration derivatives. 6... s+. « a1 1.3.33 z body axis acceleration derivatives. . : eee eee 1.3.34 Angle-of-attack sensor output derivatives... 6. 5 * - 8 D.3.35 Angle-ofsideslip sensor output derivatives... cece 98 1.3.36 Altimeter output derivatives. Sinmaccamernwe came ecm 196 13.37 Altitude rate sensor output derivatives. 2... scimin 282535 95 D.3.38 Total angular momentum derivatives. . eas 4mm ee 965 1.3.39 Stability axis roll rate derivatives. 6... ceed D28mEE 97 D.3.40 Stability axis pitch rate derivatives... . Bre Ou ie Siete a Le. OT D.3.A1 Stability axis yaw rate derivatives. 66... eee ee es 98 REFERENCES 101 SUMMARY ‘This report documents the deri fa linear aircraft model for a rigid aircraft of constant mass flying over a flat, nonrotating earth. The derivation makes no assumptions of reference trajectory or vehicle symmetry. The lincar system equations are derived and evaluated along a general trajectory and include both aircraft dynamics and observation variables. INTRODUCTION ‘The need for linear models of aircraft for the analysis of vehicle dynamics and control law design is well known. Those models are widely used, not only for computer applications but also for quick approximations and desk caleulations. Whereas the use of these models is well understood and well documented, their derivation is not. The lack of documentation and, occasionally, understanding of the derivation of linear models is @ hindrance to communication, training, and application ‘This report details the development of the linear model of a rigid aircraft of constant mass, flying over a flat, nonrotating carth. ‘This model consists of a state equation and an observation (or measurement) equa- tion, The system equations have been broadly formulated to accommodate a wide variety of applications. The lincar state equation is derived from the nonlinear six-degroe-of-freedom equations of motion. The Jinear equations representing state variables, Tinear observation equation is derived from a collection of non time derivatives of state variables, control inputs, and fiightpath, ait data, and other parameters. The linear model is developed about a nominal trajectory that is general Whereas it is common to assume symmetric aerodynamics and mass distribution, or a straight and level trajectory, or both (Clancy, 1975; Dommasch and others, 1967; Etkin, 1972; MeRuer and others, 1 Northrop ‘Aireraft, 1952; Thelander, 1965), these assumptions limit the generality of the linear model. The principal contribution of this report is a solution of the general problem of deriving a linear model of a rigid aircraft without making these simplifying assumptions. By defining the initial conditions (of the nominal trajectory) for straight and level fight and setting the asymmetric aerodynamic and inertia terms to zero, y obtain the more traditional linear models from the linear model derived in this report. Another significant contribution of this report is the derivation and definition of a linear observation (measurement) model. The observation model is often entirely neglected in standard texts, A thorough treatment of common aircraft measurements is presented by Gainer and Toffiman (1972), and Gracey (1980) provides a detailed discussion of speed and altitude measurements, However, neither of these references present linear models of these measurements. ‘This report relies heavily on these two references and nses their results as one of the bases for the nonlinear measurement equations from which the linear measurement model is derived. Also included in this report is a large number of other measurements or variables for observation that have been found to be useful in vehicle analysis and control law design, Duke and others (1987) describe a FORTRAN program called LINEAR that derives a linear aircraft model by numerical differencing (Dieudonne, 1978). The program LINEAR produces a linear aircraft model (botl state and observation matrices) that is equivalent (0 Ue linear models defined in this report. ‘This report is divided into two main sections that define the reference systems and nonlinear state and observation equations (section 1) and derive a linear model presented in the appendixes (section 2). ‘The appendixes contain a definition of the linear aerodynamic model used in this report (app. A), a derivation of the wind axis translational acceleration parameters (app. B), generalized linear derivatives of the non- Tinear state and observation equations (app. C), and the individual derivatives of te state and observation equations (app. D). The details of the pri ipal results of this report are presented in appendix D. SYMBOLS a 45,5 ax yx total aerodynamic axial force, Ib speed of sound, ft/sec normal aceclerometer output, g output of normal accelerometer not at vehicle conter of gravity, g output of accelerometer aligned with vehicle body # axis, g output of accelerometer aligned with body 2 axis, not at vehicle center of gray kinematic acceleration in vehicle body 2 axis, g output of accelerometer aligned with vehicle body y axis, g output of accelerometer aligned with body y axis, not at vehicle center of gravity, & Kinematic acceleration in the vehicle body y axis, ¢ output of accelerometer aligned with vehicle body 2 axis, g output of accelerometer aligned with body = axis, not at vehicle conter of gravity, g Kinematic acceleration in vehicle body z axis, ¢ reference span, ft generalized force or moment coefficient derivative of generalized force or moment coefficient with respect to arbitrary variable reference acrodynamic chord, ft total aerodynamic drag, 1b L-l In-le -k specific energy, ft arbitrary force or moment, fightpath acceleration, g acceleration due to gravity, ft/sec? acceleration due to gravity at sea level, ft/sec’ altitude, ft altitude measurement not at vehicle center of gravity, ft inertia tensor moment of inertia about x body axis, shug-ft? product of inertia in 2-y body axis plane, slug-ft? product of inertia in xz body axis plane, slug-ft? moment of inertia about y body axis, slug-ft? product of inertia in y-z body axis plane, sing-ft moment of inertia about z body axis, slug-ft? Tele — 2 Teyle + Iysles Teylys + Tylee Tels ~ 12, Telys + beylee Tely ~ 12, total moment about 2 body axis, ft-lb; or, total aerodynamic lift, Ib unit length, ft total moment about y body axis, ft-lb; or, Mach number vehicle mass, slugs total moment about = body axis, ft-Ib; or, total aerodynamic normal force, Ib Toad factor specific power, ft/sec roll rate (about « body axis), rad/see static or free-stream pressure, Ib/ft? stability axis roll rate, rad/see total pressure, Ib/ft? pitch rate (about y body axis), rad/see dynamic pressure, Ib/ft? impact pressure, Ib/ft? Mach meter calibration ratio stability axis pitch rate, rad/see Reynolds number Reynolds number per unit length, ft-" yaw rate (about = body axis), rad/see stability axis yaw rate, rad/see surface area of wing, f1? total angular momentum; or, ambient or free-stream temperature, *R total temperature, °R time clocity along x body axis, ft/sec vehicle velocity, ft/sec velocity along y body axis, ft/sec velocity along = body axis, ft/sec total aerodynamic force along 7 body axis, Ib total gravitational force along 2 body axis, Ib total tlirust force along 2 body axis, Ib vehicle position along + earth axis, ft total aerodynamic sideforce, Ib total aerodynamic force along y body axis, 1b total gravitational force along y body axis, 1b total thrust force along y body axis, Ib vehicle position along y earth axis, ft total acrodynamic force along z body axis, Ib total gravitational force along z body axis, 1b total thrust force along z body axis, Ib vehicle position along = earth axis, f angle of attack, rad angle-of-attack measurement not at vehicle center of gravity, rad angle of sideslip, rad angle-of-sideslip measurement not at vehicle center of gravity, rad fightpath angle, rad ith control surface deflection pitel angle, rad coefficient of -osity, Ib/ft-see density of air, 1b/t? arbitrary function W p Lev ? Onxm Tnxem bank angle, rad heading angle, rad body axis acceleration vector attitude vector of Euler angles total force vector state vector function observation vector function total angular momentum vector sum of higher order terms in Taylor series total moment vector position vector in earth axis system input or controt vector vehicle velocity vector state vector observation vector perturbation of control vector perturbation of state vector perturbation of time derivative of state vector rotational velocity vector state matrix of the generalized state equation, Cx = Ax + Bu state matrix of the state equation, X= A'x-+ B’u control matrix of the gencralized state equation, CX = Ax + Bu control inatrix of the state equation, & = A'x + B'u system matrix of the generalized state equation, Cx = Ax + Bu feedforward matrix of the generalized observation equation, y = IIx +GX-+ Fu feedforward matrix of the observation equation, y = I'x + Iu derivative observation matrix of the generalized observation equation, y = Hx + G&+ Fu observation matrix of the generalized observation equation, y = IIx+OX+ Fu observation matrix of the observation equation, y = H'x +" intertia tensor scaling matrix for inertia tensor transformation matrix from earth to body axes transformation matrix from earth to body axes angular velocity matrix in the generalized state equation, nx m matrix of 0 values an nx m matrix with values of 1 on the diagonal TR = E[x(1},X(1}, u(2)] Subscripts a aerodynamic; oF static or, free stream body axis system D drag g gravitational h displacement of altitude instrument i displacement of altitude rate instrament vi not at vehicle center of gravity ak Kinematic L hit c rolling moment, m pitching moment a yawing moment n orthogonal P power plant induced s stability axis; or, specific r thrust t total v vehicle-carried vertical axis system w wind reference axis system z displacement. in x body axis zy -y body axis plane 22 z-2 body axis plane Y sideforce y displacement in y body axis ve yz body axis plane z displacement. in the body axis, 0 at sea level, standard day conditions; or, nominal conditions Superscript T transpose 1 NONLINEAR SYSTEM EQUATIONS ‘The motion of an aircraft as a rigid body can be described by a set of six nonlinear simultancous second- order differential equations. These equations, representing the translational and rotational motion of the vehicle, can be formulated in the notation of Kwakernaak and Sivan (1972) and Diewdonne (1978) as a ne-invariant system expressed as H(t) = f[x(t), u(t) ce) where x(2) is the 12-dimensional time-varying state vector (t being time), X(t) is the derivative of x(t) with respect to time, u(t) is the k-dimensional time-varying input or control vector, and £ is a 1 nonlinear function expressing the six-degree-of-freedom rigid body equations. -dimensional Measurements of the vehicle state can be represented by the observation equation ¥(0) = alx(O, w(9] (ry where y(t) is an &-dimensional time-varying observation vector and g is an &dimensional nonlinear func tion expressing the relationship of the true vehicle state and control vectors to the observed parameters ically, the function g characterizes the dynamics and location of the sensors. For the aircraft analysis and design problem, both the nonlinear and linear system equations are formu- lated more broadly than just described (Edwards, 1976; Maine and Tlif, 1980, 1986). The nonlinear systein equations include %(t) terms in both the state and observation functions. In fact, in the most extended fori the state equation is expressed in terms of transformed variables (discussed in section 1.2.1). These generalized equations form the basis of the analysis in this report. The generalized system equations are TH) = {lx(t), (0), w(t) (13) ¥@ = wlx(4),*(0), (9) (a) where T is a constant 12x 12 angular velocity matrix. 1.1 Definition of Reference Systems While numerous reference systems are used in aerospace applications, this report is limited to four reference systems: the body, the wind, the vehicle-carried vertical, and the topodetie reference systems, The stability axes aro also defined even though this reference system is used only to define the stability axis rotational rates (section 1.3.8). Within this report the translational equations are referenced to the wind axes, and the rotational equations are referenced to the body axes. Meastrement equations are primarily referenced to the body axes when the use of a reference system is needed. ‘The use of this mixed axis system definition in both the nonlinear and linear models is related to the measurability and meaningfulness of quantities. Because the aerodynamic forces act in the wind axes, this reference system is used for the translational equations For instance, angle of attack, velocity, and angle of sideslip are either directly measurable or closely related to directly measurable quantities, while the body axis velocities (u,v, and w in the x,y, and = directions, respectively) are not. The body axis rotational rates are measured by sensors fixed in the body axes; wind axis rates can be derived only from these quantities through axis transformations, ‘The first reference system to be described is the topodetic reference system, also called the earth-fixed reference frame (Etkin, 1972), the earth axes (Thelander, 1965), and the Eulerian axes (Northrop Aircraft, 1952), The topodetic reference frame is considered fixed in space (and hence, inertial) with the orientation of the axes as shown in figure 1; the z axis is directed north, the y axis east, and the z axis down. The vehicle position (x and y) and altitude (i) are measured from the origin of this reference system The vehicle-carried vertical axis system (fig. 2; Etkin, 1972) has its origin at the center of gravity of the vehicle, The 2, axis is directed north, the ye axis east, and the =, axis down. This axis system is obtained by a translation of the topodetic axis system to the vehicle center of gravity. The attitude of the aircraft (heading, pitch, and bank angles ¥, 9, and 4, respectively) is described in terms of the orientation of the aircraft body axes with respect to the vehicle-carried vertical axes. ‘The origin of the body axis system (fig. 3) is the vehicle center of gravity. The x axis is directed toward the nose of the aircraft, the y axis toward the right wing, and the = axis toward the bottom of the aircraft ‘The specific orientation of the actual body axes relative to the vehicle body is somewhat arbitrary. symmetrical aircraft, the x and z axes are in the plane of symmetry; for asymmetrical aircraft, these axes are Toeated in a plane approximating what would be the plane of symmetry. The positive direction for the body axis rates (roll, pitch, and yaw rates, p, q, and r, respectively), the body axis velocities (u,v, and w), and the body axis moments (, M, and N about the x, y, and z axes, respectively) are shown in figure 3, 6 yy (East) (North) xy — (North) 2 (Down) (East) (North) x —__—--+ yeas (Down) m0 (Downy mt Figure 1. Topodetic azis system. Figure 2. Relationship between topode- tic and vehicle-carried vertical azis sys- tems. Hy . 1 Figure 3. Body azis system. The relationship between the vehicle-carried vertical and body ax is shown in figure 4. The Euler angles (25, 8, and 9) define the orientation of the body axes with respect to the vehicle-carried vertical axes. The rotations required to transform the vehicle-carried vertical axes to the body ay fignre 5. The heading angle y is a rotation about the = vehicle-carried vertical axis (designated (4. ys, +1) in fig. 5); the pitch attitude @ is a rotation about the yy asi Figure 4. Relationship between vehicle-carried vertical and body azis systems. These rotations are deseribed by cos ty = | sing 0 Ls " body axes, respectively, and Loy Lees and Ty, are the products of inertia in the 2-y, 2-z, and y-z body axis planes, respectively; and Q=x. = [pq ot (1-26) where p, q, and rare the rotational rates about the z, y, and z body axes, respectively. Because it is assumed that (he inertia tensor is a constant with respect to time, equation (1-22) ean be rewritten as Oe ea Ale M2 x 19) (127) ‘This is the vector subfunction for the rotational acceleration. Designating this subfunction as f,, the following definition applies: filx(t), (1), u()] = I'M ~ 9 x (79) (1-28) where 6 : Fe = Glx(),¥(E), w()] (1-29) So = ta (1-30) Since the inverse of the inertia tensor It js given by a [é Z | (1-31) See te oe la where det l = Tplyle ~ Tel, — [elby - 112, - Uyslsaley (1-32) hehh-h. (1-33) = Teylet Iysles (1-34) Teylys + Iyles (2-35) Tals — 12, (1-36) Ts = Felye + Teylee (37) =hly-2, (1-38) the expression for the rotational accelerations can be expanded as a set of sealar equations: Be qe7en EM Ip + SN fy ~ p%(Leela leyls) + pa(Tooli = Tyels - Del) ~ pr(Teyh + Dyla ~ Tyals) + (Tye — Teyls) ~ ar Dal ~ Heyla + Tesla) = (iyzhi — Teal) (1-39) a= eth + EMI, + ON Is — peel ~ Leyls) + pa Teste ~ Uysta ~ Dels) — preys + Dyla — Iyels) + ¢UIyels ~ Leyls) ~ ar(Dela ~ Ieyla + les) = PUyela — Leal] (1-49) WL + EM Ig + ON So — p*(LeaTs — Foyle) + a Tesla ~ Iysts — Delo) = pr(Zeyls + Dyls — IyeTs) + 4°(Iyels ~ Leyte) — 9r(Dels ~ Teyls + Testo) =P Uyels — Leels)] (ain where (1-42) (1-43) (4) Equation (1-3) defines the generalized nontinear state equations as TH) = fix), H(0), WC] ‘This equation, although more complicated than the nonlinear equations defined by equation (1-1), allows for ‘a more tractable formulation of the state equation by using the matrix T to provide a means of addressing the rotational accelerations in a decoupled axis system. ‘The derivation of the rotational acceleration terms is based on the moment equation (1-22) é M= 519) +9 x 19 Rearranging terms and assuming that the inertia tensor is constant with respect to time, the equation can be written as é 6 TE =M-Ox 19 (1-45) ‘The rows of this vector equation are now scaled using the following scaling matrix: Wh 0 0 0 Wl 0 (1-46) 0 0 Wh ‘This matrix, when premultiplying equation (1-27), mercly divides the first row by the roll inertia J,, the second row by the pitch inertia J, and the third row by the yaw inertia 1,. Using the definition Jad at) 13 the resulting equation is sia = J'M —J'(O x 1) (1-48) and J can be written as 10 —Aayfle —Ses/ Te Js | -ley/ly 10 = Tye/ly (1-49) Teefle -Ipfl, 1.0 Equation (1-48) can be expanded and expressed as [*| 10 .~Iny/Te he [4] ¢ Ley[ly 1.0 = Tyelly | | 4 —FoefTz = Tye/Te 10 ¢ where j/, qf, and are the decoupled rotational accelerations of the vehicle u BL [Te — rPley{Te + Palee/Te + 1 ly/Ia + (P= 1 Mlye/ Te = Gre [Te = | EM/ly — role ly + raley/ Ty ~ palys/ ly + 07 ~ PY Foe] ly + Pre /Ty (1-50) EN/I, + qple [Te ~ Isa] Le + prlys/ Lz + (P? — 4? )hey/ te — Paly/ Te Using the definition of J in equation (1-49), the matrix transformation 7 can be defined as J {0353 | ES | exe ra | Boe tna | Toxo 5x6 which would be an identity matrix except for the presence of th Thus, the vector subfunctions for the generalized state equation defining vehicle translational acceleration, vehicle attitude rates, and carth-relative velocities are the same as those defined for the standard nonlinear state equations in sections 1.2.2, 1.2.3, and 1.2.4, respectively. inertia terms in the upper left-hand corner. 1.2.2 Translational acceleration.—Derivation of the translational acceleration vector subfunction fy is based on the force equation a F=—(mV) 5: env) (1-52) where F is the total force acting on the vehicle and mis the vehicle mass. This expression can be expanded to F=m(Sv+axv) (1-53) with the assumption of constant mass with respect to time and the following definitions of F and V: F=[2X fy Ez)" (1-54) where EX, EY, and £2 aro the sums of the acrodynamic, thrust, and gravitational forces in the 2, y, and = body axes, respectively, and . Vafuv wl" (1-55) Rearranging the terms of equation (1-52) gives an expression for the translational acceleration: by HY im F-2xV (1-56) and velocities. This equation expresses body axis accelerations in terms of body axis forces, angular rates However, the desired form of this relation requires the translational accelerations in the wind axis system; that is, in terms of the magnitude of the total vehicle velocity V, angle of attack a, and angle of sideslip 8, which are expressed by equations (1-9) to (1-11) u = Vcosacos 8 v=Vsing Vsin cos 9 and equations (1-12} to (1-14) ‘The wind axis translational acceleration terms (derived in app. B) are summarized as: [Va SIT = fhfx(t), X(t), u(t)] (1-57) where V = Af -Deosp+¥ sind + “pcos a.cos 9 + ¥rsin 8 + Zrsinacos 8 = mg(cos.a cos 8 sin @ — sin 8 sing cos6 ~ sin arcos cos 4 cos 6)] (1-58) i . Vineosal “2+ Zz cosa ~ Xz sina + mg(cos.a cos cos 4 + sina sin 8)] +4 ~tan§(peosa + rsina) (1-59) B= FDsing + ¥ cos ~ Nrcosasind + Vrcosp ~ Zrsinasing + mg(cosaasin sin 8 + cos S sin $cos9 — sina sin Scos pcos 4)] + psina — reosa (1-60) with D being total aerodynamic drag; ¥' total acrodynamic sideforce; and Nr, Yr, and Zr total thrust force along the 2, y, and z body axes, respectively. 1.2.3. Attitude rates.—The matrix R that transforms angular velocities in the earth-fixed axis system into body axis angular velocities is defined by 1 0 ~sing R=|0 cos singcose (1-61) 0 =sind cosdcos@ where R is derived by Maine and Tliff (1986) from the total angular velocity of the aireraft expressed in er angles (¢, 8, i): P 0] [1 © © ]feos@o~sine] fo 4 + 0 cos an 6)+]0 cosd sing} | 0 1 0 0 r o-sind cos] Lo} [0 -sind cose} [sine o cosa } [db sind ] [db =|o oa oS 6 (1-62) terms of the derivatives with respect to time of the Ei 0 —sin 9 cos 6cos 8 This transformation from earth-fixed to body axes can be expressed by the equation a= n(4n) (1-63) where E is an attitude vector whose components are the Euler angles: E=[¢@ yl" (1-64) Premultiplying both sides of equation (1-63) by R-? and rearranging terms yields the equation for the attitude rates, dn= ra (1-65) which can be expanded into the scalar equations O = ptasindtand + rcostane (1-06) 4 = qeos$—rsing (1-67) B= qsingsecO +r cos psecd (1-68) 1.2.4 Earth-relative velocity.—The matrix Lpy that transforms earth axis system vectors into the body axis system is defined by equation (1-8) as cosy —sin O][ cosd 0 sina] [1 0 0 sin’ cose 0 o 10 0 cosg —sing 1 Lay = 0 0 =sind 0 cosa || 0 sing cos 05 6 cos cos Osin ~sind = | sindsinOcosy—cosdsin y sin sin dsin y+ cosdeos ys sin dcosd cos dsin 8 costs + sin dsin cosdsin sin y—singdcosy cosdcosd Tho specific relationship between earth-relative velocities and body axis velocities is expressed by V = Lav (=) (1-69) where R is the earth axis system vector defining the location of the vehicle: =e y J" (1-70) 6 with z= -h. ‘The equation for the earth-relative velocity can be formulated as (1-71) in which these velocities are expressed in terms of body axis velocities. Using equation (1-72) and the definitions of the body axis velocities in equations (1-12) to (1-14) allows the earth-relative velocities to be expressed in terms of V, a, and 8: V'(cos.a.cos § sin 9 — sin sin cos 8 — sin a-cos Geos ¢ cos 4) (1-72) i V [cos 0.cos 8 cos 8 cos + sin A(sin psin @cos y — cos@sin ¥) + sin.a.cos B(cos¢ sin @ cos qh + sin dsin ¥)] (1-73) V {cos a cos 8 cos 8 sin y + sin A(cos pcos w + sin dsin Bin Y) + sin « cos (cos gin Bin y — sin cos ¥)] (74) 1.3 Nonlinear Observation Equations No standard sot of observation variables exists for the aircraft analysis and control design problem, However, for any guidance and control problem, the main observation variables generally will be a subset of the state variables. Other common observation variables are the vehicle body axis translational accelerations and air data parameters. Thus, the dimension of g[x(t), X(t), u(t)] is not fixed and varies from application to application. The set of observation variables described in this section was selected to address a wide range of problems. The basic composition of the observation vector y as used in this report is given by yo [tatut yt? (1-75) where x and £ are the state vector and time derivative of the state vector described previously, w is the control vector, and y" is defined by yl sly yi vB vi ¥F yt yF yST (1-76) where Y= [rk Gye Bek Me My A: Oy Mey ys Oz4 Ine)” (1-77) yh = [a M Re Re’ 7 qe d/Pa Pa mT Td? (1-78) ys = [7 fpa i? (1-79) yi = 1B PJ” (1-80) yh=(L DN alr (st) Yo=leow ae wl” (1-82) fas Ba hs hal (1-83) vea(T na nl (180) 7 with the elements of yj being terms related to the vehicle body axis acceleration, the elements of y4 being ir data terms, the elements of y4 being fightpath-related terms, the elements of y4 being terms related to vehicle energy, yf, being a vehicle force vector, the elements of yf being body axis translational rates and the time derivatives of those terms, y5 being a vector of variables representing measurements from instruments not located at the vehicle center of gravity, and the elements of yf boing a collection of miscellaneous terms, Obviously, this grouping of terms is somewhat arbitrary and is done primarily to ease the definition of these terms in the following sections of this report. This grouping of observation variables parallels that used by Duke and others (1987). ‘The vector funetion g relating the state vector, the time derivative of the state vector, and the control veetor to the observation vector is an €-dimensional function composed of four subfunctions: slx(2),8(0), u(d)] = [x7 &™ uP g] (1-85) where x, %, and u are identity functions on the state vector, time derivative of the state vector, and control vector, respectively, and g’ is composed of vector subfunctions defining the y’ vector. The state vector, time derivative of state vector, and control vector components of the observation vector are not discussed in detail in this section of the report. The equations for the clements of the time derivative of the state vector were developed in section 1.1. The observation equations for the state and control variables are simply identities. ‘The equations for the remaining observation variables are obtained from a variety of sources. In addition to the previously cited sources, Claney (1975), Dommasch and others (1967), Gainer and Hoffman (1972), and Gracey (1980) provide the background and derivation of the observation equations used in this report 1.3.1 Accelerations.—The vehicle body axis accelerations and accelerometer outputs constitute the set of observation variables that, after the state variables themselves, are most important in the aircraft control analysis and design problem. These accelerations and accelerometer outputs are measured in units of g and are derived directly from the body axis forces defined in section 1.2.2. The body axis acceleration vector a can be expressed as tos GY = AV tOxV (1-86) It is important to note here that the i, #, and w body axis velocity rates, derived in appendix B and defined by equation (B-1), are not the body axis accelerations. The body axis accelerations contain not only the body axis velocity rates but also the rotational velocity and translational velocity cross-product terms, Thus, expanding equation (1-86) yields ark ttqw-re ayy | =| 64 ru—pw (87) ay w+ po-qu where dz,e, dye, and dae are the kinematic accelerations in the vehicle body x, y, and 2 axes, respectively. Using (fm) (Vr + Yat Vg) + pre — ru (1-88) (1fmn)( Zr + Za + Zq) + qu po app. B), equation (1-87) can be rewritten as ‘ai (lfm)(Xr + Xa +X) an |; (Lfm)(Xr + Xa +X) + rv - qw Boe (stated as eq. (B-1) (l/im)(Sr + Yat Yq) (1-89) (1/m)(Zr + Za + 25) 18 where Xq, Ya, and Z, are total aerodynamic forces and Xz, Y,, and Z, are total gravitational forces along. the 2, y, and 3 body axes, respectively. This can be expanded in terms of the gravitational and aerodynamic forces to give (in units of g) Ok Spa ei where go is the acceleration due to gravity at sea level Yr +¥ + gmsingcosd (1-90) 97 | Z_— Dsina ~ Lcosa + gmcos¢ cos @ 1’ Xp — Deosa + Lsina~ gmsiné ‘The outputs of body axis accelerometers at the vehicle center of gravity are simply the body axis accelerations duc to the thrust and aerodynamic forces. The accelerometer output equations can be written directly from equation (1-90) as ay 1 Xz - Deosa + Lsina a, |= $Y (1-91) a} "| 2, -Dsina~Leosa where ag, ay, and a, are the outputs of accelerometers at the vehicle center of gravity and aligned with the vehicle body 2, y, and 2 axes, respectively. Because the normal acecleration ay is defined by y= — ay (1-92) an expression for this variable can be extracted from equation (1-91) a, = (Zr + Dsina + Lcosa)/gom (1-93) ‘The equations defining the output of accelerometers aligned with the vehicle body axes but displaced from the vehicle center of gravity are derived by Gainer and Hoffman (1972) using the definition of inertial acceleration given in equation (1-86) é a v FV tax and the definition of inertial velocity 5 Va prtaxr (1-94) The results from Gainer and Hoffman (1972) are reproduced here without rederivation: [=] where az. @y,i, and a,, are outputs al accelerometers aligned with the z, y, and 2 body axes but not located at the vehicle center of gravity; the subscripts x, y, and 2 refer to the x, y, and 2 body axes, respectively; and the symbols z, y, and 2 refer to the 2, y, and z body axis locations of Ue sensors relative to the vehicle center of gravity. Because the normal acceleration is the negative of the = body axis accelerometer, the output of a normal accelerometer not at the vehicle center of gravily but aligned with the z body a is given by ay + [(pg + F)zy ~ (0? + 7? )uy — Car ~ B)2u]/90 (1-95) ae ~ [(9? +1 )te — (Pa Fue ~ (Pr + G) 241/90 a, + [(pr — Ge + (gr + Bue — (77 +P? e190 Gg = dn = (pr ~ Gee + (ar + Aue ~ (4? + P*)=2)/90 (1-96) ‘Tho final quantity included in the general category of accelerations is load factor n. This quantity is defined without inclusion of the z body axis force component as L mg (97) 19 1.3.2 Air data parameters.—The air data parameters having the greatest application to aircraft dynamics and control problems are the sensed parameters and the reference and scaling parameters, Chosen for inclusion as the sensed parameters are impact pressure ge, static or free-stream pressure p, total pressure Px, ambient or free-stream temperature T, and total temperature Ty. The selected reference and scaling parameters are Mach number Af, dynamic pressure q, speed of sound a, Reynolds number Re, Reynolds number per unit length Re, and the Mach meter calibration ratio g-/Pa- The derivation of these quantities is treated extensively by Gracey (1980) The nonlinear equations defining these quantities are mo)! a b + fer] (1-98) M= e (1-99) pve fare OMS, La Res Fe (1-100) Rel = (1-101) 7 ra toy a= sev (1-102) ((1.0 + 0.24f7)55 — 1.0)p, (Mf < 1.0) 15103 =) (1.215.766? /(5.6M? — 0.8)?5—1.0}y. (Af > 1.0) (1-103) _ [040.2075 ~ 10 (Mt < 1.0) a po > (12M? [5.7602 /(5.60? — 0.8))25— 1.0 (M2 1.0) (08) 7, = T(1.0 + 0.2M?) (1-105) where p is the density of the air, jt is the coefficient of viscosity, and the subscript 0 refers to sea level, standard day conditions. Free-stream pressure, free-stream temperature, and the cocflicient of viscosity are properties of the atmosphere and are assumed to be functions of altitude alone. 1.3.3 Flightpath-related parameters—Inchided in the observation variables are what might best be termed flightpath-related parameters for lack of better nomenclature, These terms include flightpath angle 7, flightpath acceleration fpa, and vertical acceleration h. The variables are defined by the following equations: (1-106) ¥ fpa=% pa= 7 (1-107) ht = azysin# ~ ayy sin pcos @ ~ 2,4 cosdcosd (1-108) 20 Two energy-related paramoters are included with tle observa- 1.3.4 Energy-related parameters. tion variables considered in this report: specific energy Ey, and specific power P,, defined as (1-109) (1-110) 1.3.5 Force parameters —The set of observation variables being considered also includes four force parameters. These quantities are total aerodynamic lift L, total aerodynamic drag D, total aerodynamic normal force 1V, and total aerodynamic axial force A, defined as L= 4q8Cy (4am) D = @SCp (L112) N = Leosa+Dsina (1-113) A= -Lsina + Deosa (1114) where S is the surface area of the wing, Cy, coefficient of lift, and Cp coefficient of drag, 1.3.6 Body axis rates and accelerations—Because they are of interest in the control analysis and design problem, six body axis rates and accelerations are included as observation variables. These include the @ body axis rate u, the y body axis rate v, and the z body axis rate w. Also included are the time derivatives of these quantities, i, #, and to, respectively. ‘The definitions of the body axis rates are given in equations (1-9) to (1-11) as u = Veosacos§ v= Vsing w= Vsinacosf ‘The time derivatives of these terms can be defined using equation (B-1) and equations (B-8), (B-9), (B-10), and (1-56) as a = Atmamsnd— Desa b ISIE 4 nV sin 8 — QV sina cos (1-115) Yr +.gmsingcos@ + Y + DV sinaccos — rV cos.acos 9 (1-116) ™ 1 + gin coscos 0 ~ Dsina ~ we rd ameosdcosb— Deine — L064 4 Gy cosacossi~ pW sin (2-117) 1.3.7 Instruments displaced from the vehicle center of gravity The need to include measure- ments from instruments displaced from the vehicle center of gravity arises from the fact that not all aircraft 21 instrumentation is located at the vehiele center of gravity. The most important of these quantities are un- doubtedly the acccterometer outputs treated in section 1.3.1. In this section four additional parameters are presented: angle of attack (az), angle of sideslip (9), altitude (f,,), and altitude rate (h,:) measurements from instruments displaced from center of gravity by some z, y, and 2 body axis distances. The subscripts a, 9, li, and he refer to the displacements of the angle-of-attack, angle-of-sideslip, altitude, and altitude rate nstruments from the vehicle center of gravity. The equations used to compute these quantities are ay = a+ Pa Pe (1-118) 729 = pep By = a+ ee (119) hy = ht 2,sind — ysin cos — 2 cos cos @ (1-120) hy = h + (2; cos6 + y;, sind sin ® + 2, cos sind) — d(y, cos$cos@ — 2; singcos®) (1-121) : ch a i h a 8 Miscellaneous observation parameters—The final set of observation parameters considered in this report is a miscellaneous collection of parameters of interest in analysis and design problems. These parameters are total angular momentum T, stability axis roll rate p,, stability axis pitch rate q, and stability axis yaw rate rs. The equations used to define these quantities are T= ile = ey — QUeapr + Tyg? — By.ar + Ler) (1122) Py = peosatrsina (1-123) b=9 (1-124) ts —psina +reosa (1-125) 2 LINEAR SYSTEM EQUATIONS ‘The standard state equation for a linear differential system has the form X(t) = A'x() + B'u(s) (21) where, for a time-invariant system, A’ is a constant m xn matrix and B’ is a constant n x k matrix, The standard output equation has the form y(t) = W'x(t) + Fra(t) (22) where JI’ is a constant € xn matrix and FY is a constant €x & matrix. The generalized linear system equations used with an extended formulation compatible with the generalized nonlinear equations (1-3) and (1-4) can be characterized by CR{t) = Ax(t) + Bult) (23) y(t) = x(t) + GR) + F(t) (24) where C and A are constant n x n matrices, B is a constant n x k matrix, 17 and G are constant xn matrices, and F is a constant €xk matrix. The nonlinear system equations developed in section 1 (096. (1-1) to (1-1)) can be linearized about a trajectory, and a linear model can be formulated that is similar to either the standard or the generalized linear system equations 22 2.1 Linearization of the State Equation If up(t) is given input to a system described by the state differential equation (1-3), and if xo(t) is a known solution of the state differential equation, then approximations to the neighboring solutions can be found for small deviations in the initial state and in the input by using a linear state differential equation. The nonlinear state differential equation (1-3) can be linearized about a general trajectory, as by Kwakernaak and Sivan (1972) and Dieudonne (1978), s0 that xo(t) satisfies Tko(t) = FXo(t), xo(4), wolt)] Assuming that the system is operated at close to nominal conditions with w(t), x(z), and X(#) deviating only slightly from up(2), xo(t), and Xo(t), the following expressions can be written: u(t) = p(t) + 6u(0) (25) x(t) = xa(t) + 6x(t) (2-6) H(t) = Kolt) + 63(8) (27) where u(t), 6x(1), and 6&(1) are small perturbations to the control, state, and time derivative of the state vectors, respectively. Substituting equations (2-5) to (2-7) into the nonlinear state differential equation (1-3), expanding in a ‘Taylor series about Xo(t), Xo(t), Uo(!), and assuming T constant with respect to x(t) yields ; of, | of at Ahxolt) ol), WD] + 5 OH Ge HK F Sut W(H) (28) where 0f/Ox, 9£/A%, and 9£/du are defined in equations (2-9) to (2-11) and h(t) represents the sum of the higher order terms in the Taylor series, assumed to be small with respect to the perturbations. ‘The matrices used in the Taylor seties expansion are defined by the following relationships: Thkolt) + 6X(1)] ot _ at Ox = dx| (2-9) xo tous) af _ at ees (2-10) Ox ax| (Xo to to) of of fu = Dal . (2-11) xo.) the (i, j)th elements of which are defined as of ( ie (2-12) ( = (2.13) at (aa), oa respectively, where fj is the ith simultaneous equation of the nonlinear state differential fanction in equa- tion (1-3), x; the jth element of the state vector, Xj the jth element of the time derivative of the state 23, vector, i the jth clement of the control vector, and all derivatives are evaluated at the nominal condition (xolt), Xo(), wo(t))- Subtracting equation (1-3) from (2-8), rearranging terms and neglecting the higher order terms yields a linearized state equation, a [r ot) ax a = Fax $ sun (2-15) where the arguments of the matrix functions have teen dropped to simplify the notation and where it is understood that the matrices are to be evaluated along the nominal trajectory. Letting of c=T- (2-16) ar A= (2:17) at a= (218) equation (2-15) ean be written as C 6X(0) = A x(t) + B 6u(t) (219) which is precisely the formulation of the generalized state equation desired. Promultiplying both sides of equation (2-19) by C~? results in the standard form of the linearized state differential equ: 5(t) = C714 bx(t) + C71B 6u(t) (2-20) Letting A=C7tA (2-21) Bi=C7B (2-22) equation (2-20) can be written in the more usual notation EX(t) = A 6x(t) + BY Sut) (2-23) 2.2 Linearization of the Observation Equation ‘The technique used in section 2.1 to linearize the state equations can be applied to the nonlinear observation equation (1-4), y(t) = elx(), (4), 0()] Performing a Taylor series expansion about the nominal trajectory (xo(t), Xo(#), Wo(t)) yields volt) + 6y(t) = glxolt), Xo(t), wo(t)] + i Out h(t) (2-24) where og Fa (2:25) a (2-26) Ox the (i, j)th elements of which are defined by a om respecti all deriv: ly, where g; is the ith simultaneous equation of the nonlinear observation equation (1-4). Again, tatives ate evaluated at the nominal condition (xo(t), Xo(t), ol?) Subtracting equation (1-4) from equation (2-24), rearranging terms, and neglecting higher order terms results in a linear observation equation, a FB ine 4 8 554 88 5. fy(t) = Geox 5g OR 5 ou (2-31) where the arguments of the matrix functions have been dropped to simplify notation. Letting =i. H=e (2-32) G (2-33) (2:34) equation (2-31) ean be rewritten as Sy(t) = HT Ex(t) + G 8X(t) + F bul?) (2:35) which is the generalized linear observation equation desired. ‘The standard form of the observation equation can be derived by substituting for 6% from equation (2-23) into equation (2-33). This substitution results in Sy(t) = HT &x(t) + G[AT 5x(4) +B! 6u(t)] + F 6u(s) (2:36) which can be written as by(t) = [10 + GA'] &x(t) + [F + GB] Sut) (237) By letting Hl =1+GA' (2-38) P=F+cp (2-39) equation (2-37) becomes Sy(t) = HW 6x(t) + F bu(t) (2-40) 2.3 Definition of Matrices in Linearized System Equations The results of sections 2.1 and 2.2 can be used to define the matrices in the linearized system equations in terms of partial derivatives of the nonlinear state and observation functions taken with respect to the state, time derivative of state, and control vectors. All derivatives are understood to be evaluated along the nominal trajectory. : Using the nonlinear state equation (1-3), Tx(t) = flx(t),X(2), uC) = the terms in the generalized form of the linearized state equation (2-19), C 6X(1) = A Bx(t) + B bul) can be defined as of _ CeT-& (2-41) : of A= (2-42) or 5 B=a (2-43) ‘The terms in the standard form of the linearized state equation (2-20), 6(t) = A’ Bx(d) + BY bu(t) can be defined as at)! at [rs] ae oan » fp af] at . = fr ‘eal oe (2-45) Ina similar manner, the nonlinear observation equation (1-4), y(t) = Bix), 54), u()] can be used to define the terms of the generalized linearized observation equation (2-35), Sy (t) = H Ex(t) + G 6X(0) + F bul?) 26 (2-46) (247) (2-48) ‘The terms in the standard form of the linearized observation equation (2-40), by(t) =H! 6x(¢) + F’ éu(t) can be defined as , cp ae|-* af > W [r- | 5 (2-49) 1 = 26 4 26 fr of or ‘ +a lox ow i) 2.4 Elements of the Linearized System Matrices “The elements of the linearized system matrices derived in sections 2.1 and 2.2 are determined by applying the linearization method employed with the vector equations in those sections to the individual scalar equations constituting the vector equations that define the time derivatives of the state and observation variables. ‘Thus, for a matrix, such as the state matrix A defined by equation (2-42), ar Ox the clement occupying the ith row and jth column of A, (A)iy. can be represented as A (Ay where fj is the scalar function defining the time derivative of the ith state and x; is the jth state. The individual terms used in the A, B, C, Hf, G, and F matrices are defined in appendix D based on the generalized derivatives derived in appendix C. Using the state vector x defined in (1-7) as x=(parVasgovheul the elements of the A matrix can be expressed as {p/p OU/O4 --- O(HVfy Alay A(a’)/04 «++ Aa/Oy A(%)/Op (3/94 «+» OG%)/Oy AHO» OG)IO4 -- IG)/O4 Substis ting for these partial derivatives using the terms in appendix D gives (1/Te)(GS0?/2V0)Ce, + OL /Op. (L/Te)[(GSbe/2Vo)Ce, + OL7/Dy + TeePo —Teyro + Fee) 422,290 + ro(Ty ~ I2)] . As | C/MGSbE/2W)Cm, + OMr/Op— (1/Ty)(GSE7/2M0)Cmy + OM /Oq ” (2-53) —2esPo — Tye + rol I — Ie)) +Iryr0 — IyzPo] The elements of the B, C, H, G, and F matrices ean be determined in a similar fashion, although some care must be taken in determining the elements of the matrices for the observation equation and the C matrix. To determine the elements of the matrices for the observation equation, one must consider the definition of the nonlinear vector function g defining the observation variables (eq. (1-85)), abx(t),8(0), (] = [x 87 aT gi] and the definitions of the matrices for the generalized linear observation equations (2-46) to (2-48) , = 8 a = og : rz og - - au ~~ ‘These matrices may be expressed using a partitioning based on the vector subfunetions of g as (2-54) (2-55) (2-56) 28 which become (2-58) (2-59) upon evaluating the partial derivatives of the identity functions x, x, and u. The C matrix may be viewed as a partitioned matrix as Cu ica a, C= t (2-60) Ocxe | toxe where, from equation (1-48), 1.0 ~ley/Le ~Tee/ Te Cn =F = | Tes/ly 10 a Tes/Ty (2-61) wTes[ Tz ~Iye[Tz 10 and -A(pJOV —O(W)/da -A(p}/03 0 -(9Sbe/2Vole)Ce, —(GSU7/ Vale )Cey Cin = | -(q)/OV -a(¢)/04 -I(q)/93 | = | 0 -(GS7*/2V01y)Cm, -(GSbE/2V01}Cm, | (2-62) =A(r/)/aV —A(r")/Oa -A(r'\/93. 0 ~(GSbE/2VoI2)Cn, (GSP /2Wol-)Cng 29 Cx -0(a)/aV 1.0-A(4)/da — -a(a)/a8 (Aja — -9(8)/da 10-0(8)/a8. i 0 (GS¢/2Vom)(cos 8 Cr, — sino Cy,) (G5b/2Vem)(cos fo Cv,) eo -AV) fda = -avyas 0 10+ (gS¢/2V2m cos 6o)Cr, (@Sb/2V3m cos Ao)Cr, (2-63) 0 @ The inverse of the C matrix, C-?, can be expressed as a partitioned matrix in terms of the matrix subpar: @/2Vgm)(sin Bo Cp, + €0s fig Cy,) 1.0 ~ (@Sb/2VFm)(sin Bo Cp, + cos 8a Cy,) titions of the C’ matrix as Cz ee baa (2-64) exe lexe The elements of the A’, BY, H', and F" matrices can be determined using the C-! matrix defined in equation (2-61), the A, B, H, G, and F matrices, and the definitions for A’, B', Hf", and F” given in equations (2-21), (2-22), (2-38), and (2-39) 3 CONCLUDING REMARKS This report derives and defines a set of linearized system matrices for a rigid aireraft of constant mass, flying in a stationary atmosphere over a flat, nonrotating earth. Both generalized and standard linear system equations are derived from nonlinear six-degree-of-freedom equations of motion and a large collection of nonlinear observation (measurement) equations. ‘This derivation of a linear model is general and makes no assumptions on cither the reference (nominal) trajectory about which the model is linearized or the syimmetry of the vehicle mass and aerodynamic properties. Ames Research Center Dryden Flight Research Facility National Aeronauties and Space Administration Edveards, California, January 8, 1987 APP. INDIX A—-AERODYNAMIC FORCES AND MOMENTS ‘The aerodynamic forces and moments acting on an aircraft are the result of multiple factors whose signi icance varies with flight condition as well as from vehicle to vehicle. In general, these forces and moments are nonlinear functions primarily of Mach number, angle of attack, angle of sideslip, altitude, rotational rates, and control-surface deflections. For the purposes of this report, the aerodynamic forces and moments are assumed to be functions having the following form: F = 8(0,8, Vb, 7,957) 8,8, 8150++56n) (A-1) where F is an arbitrary force or moment, ® is an arbitrary function, and the 6; are the m control surface deflections. ‘These forces and moments are related to the nondimensional force and moment coefficients by the equations for the forces, (42) (a3) (As) and the moments, SC (5) Wm (A-6) N = GSbC, (A-7) where b is reference span and @ is reference aerodynamic chord. While the nondimensional aerodynamic force and moment coefficients are themselves nonlinear fune: tions of the vehicle states, time derivatives of the vehicle states, and the control surface deflections, these coefficients are commonly expressed in linear form in terms of partial derivatives of these coefficients with respect to the functional variables. These linear equations for the aerodynamic force and moment coeffi- cients are derived in the same way as the linearized system equations (section 2); therefore, this derivation will not be repeated here. These linear equations are CL = Cty + rg + Cig + Cig + Cy +L Cty bi + CyB + Chad + CLP + Cig + C1 g8 (as) a Cp = Coy + Cpo.a + Cf + Cpyh + CoV +0 Cd,4 + Cd, + C,4 + Co,F + Crgd + Co,8 (A9) a 2 " Cie + Cvs + Cy gh + Crh + Cry¥ + Cy b+ CyB + Ord + CrP t Cvsd + Cy,8 (A-10) a Ce Cog + Cry + Cry + Ch + CV +L Cub + Cyb + Cyd t Caf + Ce + Cod (Al) ct 31 Cin = Cm + Cma® + Crag 8 + Cmyh + CmyV + 1 Cing 5: + Cap + Cg + Cnet? + Cnadi + Omg) a: 5 = Ong + Ong + Ong + Cnyht + Coy : +30 Cuy 6: + Cag + Ong + Ona + Ong + Ong (An) i & where C,, is the value of the coefficient along the nominal trajectory and the notation Ce, is defined as (aM) with C being an arbitrary force or moment coefficient and x being an arbitrary state, time derivative of state, or control-related parameter that for the usual derivatives is nondimensional. However, the derivatives with respect to altitude and velocity are not taken with respect to a nondimensional quantity. ‘The def of these nondimensional stability and control derivatives are given in terms of the coefficient Cg. The nondimensional stability derivatives are defined as aCe (15) (A16) (417) (ais) * (a-19) (A-20) : (A21) = The two other stability derivatives are not nondimensional and arc defined as Coy = (A-22) Coe La (Az) - ~The control derivatives are defined as Cu, a (A-24) ‘The rotational terms in equations (A-8) to (A-13) are nondimensional versions of the corresponding vati- able with ._ bp _ 2. A-25) Be (A-28) Be (A260) :_ 48 baa (4.29) Because the Cg, terms are included, the force and moment coefficients are total force anc moment coefficients The state, time derivative of state, and control parameters on the right-hand side of equations (A-8) to (4-13) are differentials. 33 APPENDIX B—DERIVATION OF THE WIND AXIS TRANSLATIONAL PARAMETERS V, a, AND 8 ‘The derivation of the wind axis translational acceleration parameters js based primarily on the definitions in equations (1-9) to (1-14), the body axis translational acceleration equations (1-56), and the expression of the force terms defined in equation (1-53). In the following sections, each of the wind axis transla tional acceleration terms is derived separately after stating some preliminary definitions applicable to all calculations. B.1_ Preliminary Definitions Equation (1-56), oy-tp-oxv HY in can be expanded, using equations (1-54), (1-55), and (1-26), to ie) [Osm)( Xr + Xt Xp) + rv — aw &| =] C/m)(¥rt Ya + Yq) + pw ru (B-1) (Lmn)(Zp + Za + 2g) + qe po ‘The body axis aerodynamic forces can be rewritten in terms of the stability axis forces lift 1, drag D, and sideforce ¥: X, = —Deosa + Lsing (B2) Ye=¥ (B-3) %q = —Dsina— Leosa (BH) ‘The gravitational forces can be resolved into body axis components such that Xg = —mgsin 9 (B-5) Y_ = mgsin dcos8 (B46) 2g = mg cosecos0 (B7) ‘These equations will be used in the derivations of the V, @, and J equations. Thus, the total forces in the body axes ean he defined and expanded as EN = Xp — Deosat Lsina - gmsind (B-8) EY = ¥r+¥ ¢gmsindcos? (B-9) DZ = Ze — Dsina ~ L-cosa + gmcosdcosé (B-10) B.2 Derivation of V Equation Beginning with the definition of V in terms of u,v, and w in equation (1-12), Via (40? + wh)? 35 PRECEDING PAGE BLANK NOT FILMED the equation for 1” becomes i @ pa ty a Feta ty yy Bat Vey sqlite tury (Baty which after expanding the derivative and cancelling terms, becomes wit + vi + wi) (B12) By substituting the definitions for w, v, and w from equations (1-9) to (1-11) and cancelling terms, equa- tion (B-12) yields - V = ticosa cos + ésin § + thsin acosf (3-13) ‘The defini jons for i, 8, and th in equation (B-1) are now used with equation (B-13) to give SOS CSB x 4 Xp 4 Ny) + coseacos (re — qu) : In B y+ Vp + ¥5) + sin A(pw- ru) 7 : 4+ MALO y+ 20 4 Z—) + sin axcos Agu — pu) (Bt Ve i Expanding (B-14) in terms of equations (B-2) through (B-7) and eaneclling yields 1 v =[— Deos 8 +¥ sin 8 + Xr eosacos 8+ Vr sin B+ Zr sina cos? = mg( cos. cos isin 0 — sin Asin $ cos @ ~ sin a cos 8 cos $.cos6)] + rv cosa cos 9 — qweosacos 3 + pwsin 8 — rusin§ - + qusinacosf ~ pusin acos a (B15) Equation (B-15) can be simplified by recognizing that the terms involving the vehicle rotational rates are identically zero, which becomes obvious after substituting for u, v, and w in these terms. Thus, the final ‘equation becomes l= DeosA + ¥ sin + Xrcosacos + Yesing + Zpsina cos = mg(cos a cos f sin 6 — sin f sin 6cos@ ~ sin «cos f cos # cos 6)] (B-16) B.3 Derivation of d Equation The equation for é can be derived from the definition of a in equation (1-13), aw ie 7 (B-17) utd = iw) (B18) 36 Substituting the definitions of u and w from equations (1-9) and (1-11) into equation (B-18) gives wheosa~ isina (0-19) “ V cos Using cquation (B-1) to substitute for i and t and equations (B-8) to (B-10) to define the forces, equation (B-19) becomes, after rearranging terms, 1 Vin cos [-L+ Zr cosa — Xpsina + mg(cosacosdcosé + sin asin 8)) (queosa — pecosa —rusina + qusina) (B20) 1 VeosB which after substituting for tt, v, and w from equations (1-9) to (1-11) and combining terms gives 1 Vincos 8 4+ q- tan (pcosa + rsina) (B21) [-L-4 Zrcosa— Xrsina + mg(casa coscos@ + sin a sin 8} B.4_ Derivation of 3 Equation "Phe equation for A is derived from the definition of 9 as given in equation (I-14), =sin? Basin" ‘Taking the derivative of with respect to time yields p44? B= Fas Gant (B-22) which becomes, after expanding the derivative, substituting for V, and cancelling, B= Pl-weosasind + 00s — tosinasin f} (B-23) Using equation (B-1) to substitute for i, 8, and w and equations (B-8) to (B-10) to define the forces, z ipl cosesins (-Deosa + Lsina +X — mgsin#) + cos (¥ + Yr + mgsin gos) = sin asin f (~Dsina — Leosa + Zp + mg cos #cos)] cos asin J (rv ~ qw) + cos 6 (pw — ru) axsin 9 (gu ~ pe)] (B-24) Substituting into equation (B-24) for u, v, and w and rearranging terms yields the final equation 1 [Dsin 8 + ¥ cos P - Xqcosasin f+ Yreos 8 - Zpsinasing mv + mg(cosa sin f sin 6 + cos Hsin cos # — sin a sin B cos 4 cos 8)] + psine ~ reose (B-25) APPENDIX C—GENERALIZED DERIVATIVES ‘The equations defining the time derivatives of the state variables (derived in sections 1.2.1 to 1.24) and those defining the observation variables (presented in sections 1.3.1 to 1.3.8) are used to determine the generalized partial derivatives of the quantities with respect to a dummy variable €. The purpose of these generalized derivatives is primarily to facilitate the derivation of the terms in the linearized equations presented in section 2.4; however, these equations have also proved to be useful for computer programs and were used to verify the results obtained using LINEAR (see Duke and others, 1987), C.1 Generalized Derivatives of the Time Derivatives of State Variables Equations (1-39) to (1-41) define the rotational accelerations of the vehicle. These equations are used to determine the generalized derivatives of these quant . OMy th opth sth eth Se a) __ 1 {1 OL |, aM.) ON. , Okr aE ~ detT 1 5E ap ~ [20a = Ieyls) ~ a(Tosth ~ Iyela ~ Dela) + rUayhs + Dyla ~ lyelo) 5p a. + [Desh ~ Iyela~ Dela) + 24{Iyelt~Iayls) — (Del ~ Ioyla + Teel) 5g [pUleyls + DyTa — Iysls) + q(Dhh = Leyla + Feels) + 20 yeh = Tesla) at (C1) OF OM ON, Obn , p OMe, 5, Nr {ngeth Det Bs et ta Get le Get de Ge = [20 Feds = fyds)~ Ula ~ Fela ~ Dele) + Wlyla + Dole ~ fal] + [PUesT2 ~ Iyels — Dells) + 2g(Iyela ~ Heyl) — 7( Dela — Teyla + Teyls)] q = (Unyla + Dyls = Iyets) + (Dela — Teyls + Tes ls) + 2eLyela ~ Teel] 5 (c-2) A (1 Mg OBE Ie a EE nM 4 1 OME DE = de U8 oe th Ge tle Bet oe tle ae the Ge = [2p( Foss = Teylo) ~ Feels ~ yes — Delo) + r(Lepls + DyTs ~ Fyele)] 2 4 (DUTsals ~ Tyas ~ Delo) + 2allyela ~ ley le) — r(Dolo ~ Ieyls + lesle)] # = [p(Tayls + Dyls — Iysle) + (Dela — Teyly + Lesle) + 2r(Iyels ~ Lezls)) a} (c-) ‘The quantities I, Izy Is, Lay Is, Io, Dey Dys Dz, and det I are defined in equations (1-32) to (1-38) and (1-42) to (1-44). Equation (1-50) defines the decoupled rotational accelerations of the vehi used to determine the generalized derivatives of the decoupled quantities: (ds and #), which are Ov) _ L [aL , abr op aq Of = Te fe + pe 7 (lav — alee) 5¢ + (Plex + Thy + Dales — rhe) Fe = (Play = a, +2rIes +41) 5 (c-4) 30 PRECRDING PAGE BLANK NOT FILMED 1 [dat | OMy Op 4 Tae “ag ~ (let aye + Biles — 1H) 56 + (rly ~ Pe) 36 ~ (Phe ~ they ~ les ~ ph) FE (cs) a + (ple ~ Pes ~ 24ley ~ Ply) % (C6) Equations (1-58) to (1-60) define the translational accelerations of the vehicle. These equations are used to determine the generalized derivatives of these quantities: < Fr sing OVE : ox + cos ecos 8 7G rm a’) oD oe af ~ eos 2 a +[-Xrsina cosh + Zr cosecoss + vin sin a.cos * +sing 2 re tsinacorp + cos cos pcos cos p)) 22 e + [Dsin 8 + ¥ cos 8 — Xz sin B cosa + Yr cos — Zrsinasin A + mg(sin cos a sin B + cos sin $ cos ~ cos # cos dsin asin )) 22 7 : 5 : og z ~ mal e080 on in + cosOsin dain acon a) 32 _rtenaans senna aatensanacnt (C7) 8 5 cose 22 sng XE) - 2 ar 5 (Fe tone FE — sin 2) tan A cosa 32 + OE ~ tan sina 5 a {sapteaghet Zr eosa ~ Xysina + mg(cosOcos dcosa + sindsin ay} oe + {iratagl-Arsine = Xr cosa — mg(cos8cos ¢sin a — sin 8 cos a)] + tan 8 (psina ~ reosa) }7E {apasgte Zycosa ~ Xtsina + mg(cos@ cos dcosa + sin Osin a)] — aarglrora +rsin oh - (pEqqcortsin dora) e- [irEgoingeon cosa — cos sina)] 2 (cs) 40 OB) 21 fg PP, ogg ® OXr Or _ snasin 2 6 fang 924 cop casas 2 + na ME ning sina 2? — coxa sina - cosa $e 1 nv? [D sin B + Y cos ~ Xx cosasin f+ ¥r cos 8 ~ Zrsinasin a ‘i A saan ay + mg(sin d cos asin 8 + cos 8sin #c0s/9 — cos 9cos sin asin B)] Fe + {aplxr sinasin 8 - Zp cosasin 8 + mg(— sin Osin asin 8 — cos 0 cos deosa sin) da + peosa + rsina} 7 . ip cos ~¥ sin 9 = Xr cos «cos 3 — Yq sin — Zp sin acoso + mg(sin 8 cosa cos — cos@ sin dsin 3 — cos # cos ¢ sin a ¢0s 8)]} ae g in 8) + Z(cosdcor dos + cosdsin dsin asin 8) 5 + Frlcos cos a sing — sin @sin pcos 8 + sin 8 cos sin asin 8) x (C-9) Equations (1-66) to (1-68) define the vehicle attitude rates. These equations are used to determine the generalized derivatives of these quantities: ae = 38 4 sin ptand $2 + cosgtane FE t (9eostand— sin dtan8) 4 $(qsin dsec? @ + rcos dsec? A) 2 (C-10) 20) 5428 cing 38 2D = cons FE sing Fe (asing +r 00s 6) Je (cy) % = singsecd x eos sece x + (qc0s dec 6 ~ rsin $sec 8) % 4 (qsindsecOtand + rcosgsced tang) 2 (C12) 0 Equations (1-72) to (1-T4) define the earth-relative velocities of the vehicle. These equations are used to determine the generalized derivatives of these quantities: a [cos 8 cosa sin 8 ~ sin sin ¢ cos 8 — cos sin acos $cos 6] ie . ek aa = V{(cosAsinavsind + cos Peos. cos dos) Fe — V(sin 8 cosa sin 9 + cos Asin $cos@ — sin sin «cos $cos 8) 8 2 ~ cos sin asin cond) °% ~ V(sin cos bcos! ~ cos sin asin cos 8) Fe + ¥(cos cosa cos0 + sn Bin gsind + ox sia acs sind) FP (C-13) [cos 8 cos a cos cos ¥ + sin G (sin dsin 8 cosy ~ cos sin #) + con sine (cos dsinBeos sin dsin vy] ~ V{c08 f sin a cos cosy — cos A cos. (cos dsin 8 cos y+ sin dsin p)] e = V[sin 8 c08 « c0s 8 cos ¢$ ~ cos Asin dsin Ocos ¥ — cos sin + sin 8 sina (cos sin 6 cos p+ sin dsin B)] 2 + V[sin 8 (cos psin 8 cos ys + sin sin y) — cos A sin a (sin } sin Ocos p ~ cos sin ¥)] Ha ~ V[cos f cosa sin # cos ¥ — sin B sin d cos 8 cos yp — cos sin a-cos cos cost] i ~ V[cos 8 cos a cos # sin y + sin @ (sin dsin Osin y + cos 4 cos ) + cos Bsina (cos dsin Bsin y— sin dcos ¥)] * (cu) HH) . ” 5th SFE. = [608 con cos Bsin + sin (cos cos + sin bsinOsin &) + cos sin a (cos sin # sin ys — eos] SE — V[cos J sin a.cos @ sin y — cos 8 cos a (cos dsin @sin ys — sin pcos W)] z — V[sin 8 cos. cos @ sin & — cos 8 (cos 4 cosy + sin sin Osin ¥) + sin B sin a (cos @sin 8 sin w — sin g.cos)] # a — V [sin B (sin $ cos y — cos din @ sin #) + cos 8 sin a (sin dsin Bsin ¥ + cos deos ¥)] i in B sin cos 6 sin y — cos fsin a cos pcos 4 sin }) z = V (cos 8 cosarsin Bsin 4 V [cos 8 cos « cos 0 cos y ~ sin f (cos@sin y — sin @sin 9 cos #) 4 cos sin a (cos sin Bcos Win sin #5 (C15) C.2 Generalized Derivatives of the Observation Variables ‘The vector equation (1-90) defining the body axis kinematic accelerations is used to determine the gener alized derivatives of the individual body axis accelerations: ax walk KT _ cosa 2B tsina SE + (Daina Leova) Fe — gmcosd fom 0G “1 [avn | av ++ e056 22 — gmsind sin a am (= gm cost cos 32 — ginsin sind Fe ag aE 1 [azr_ OP ‘ aol ae 9 aE cosa SE = (Deosar~ bsina) aa og ~amcos0sing $8 ~ gmsindcors ] (e138) Vector equation (1-91) defines the output of body axis accelerometers at the vehiele center of gravity and is used to determine the generalized derivatives of the individual body axis accelerometers: Olas) _ 1 [aXe oD wale = cosa sina SE 7 + (Dsina+ Leesa) Fl (C19) 1 (a¥r , OY am (36 * ae) cn 2: aL a sina FP coma BE ~ (Deora - sina) (C-21) Using equation (1-93), the generalized derivative of the output of a normal accelerometer at the vehicle center of gravity can be expressed as Bay 2 Se [- OE ~ gom gine 2 + cosa ag * 8" OE der ¢ + (Deosa ~ Lsina) Fe (C22) ‘The vector equation (1-95) defining the output of orthogonal accelerometers aligned with the body axes but displaced from the vehicle center of gravity is used to determine the generalized derivatives of these quantities: + [love 720) BE + Cove = 2080) 30 + (nee ~ Bre) OE 4 20 BE ae FE] (C29) a 21 Gon, ory) % 4g OP oe). Soya oon BE (a 2ray) Fes, HE a, Fe) (C2 Aaes) _ Boe _ 1 ory) toe 2%) eos NOE = Fen gg lemme FE ae — 109) Eee me) em oe Ba ns HY (025 Equation (1-96) defines the output of a normal accelerometer aligned with the : body axis but not located at vehicle center of gravity, ay,;. This equation is used to determine the generalized derivative of ayy: a op oF .) + (tee — ran) FE = (es + 0m) SE aa gett +5 (¢-26) ani) _ Aan | 1 [ 3 ap DE = DE 3g [Pe — P88) 9g In equations (C-20) to (C-23), the partial derivatives of the vehicle rotational rates with respect to the dummy variable € are defined by equations (C-1) to (C-3). The partial derivatives of the outputs of the body axis accelerometers at the vehicle center of gravity are defined by equations (C-16) to (C-19). In these jons, as before, the subscripts z, y, and = refer to the x, y, and = body axes, respectiv Using equation (1-97), the generalized derivative of the load factor can be defined as an) 1 ob Og mg O€ (C27) Equations (1-98) to (1-105) define the air data parameters of interest for this report. These equations are used to determine the generalized derivatives of the air data parameters: 0.7p0, or poDoliN(vofoota™ E (es) : ie - a FA (C-29) (€-30) (C-31) (€-82) (1.0 + 0.247)95 — 1.0) he 41 4M(1.0 + 0207p, YE (M < 1.0) s.7gM2_)** Ops 9 - [ oar (Sgr) — 1.0 Be (c-33) 2 Pe 2 a (SSG a) 9.216M__] oar lz ou? | } (te 1.0) 14M(Lo+ o.2ar7ys Se (uf < 1.0) vam (-eigee _)* oan (, zg) cad {2 (sd Ma) +800 (sees fem) 9.210M. uw lz eM “ital } aE GBS 10 ar om 2M?) <= + 04TM (C- (1.04 0.297) Fe Hoare Se (C35) In the preceding equations, the generalized derivative of Mach number appears several times. This term can be expanded using equation (C-29). ‘The definitions of the flightpath-related parameters are presented in equations (1-106) to (1-108), These definitions are used to derive the generalized partial derivatives of the flightpath-related parameters: A av , ak —t— [4% , 36) | V2 | (C-36) (€-37) eynos 080 + ogy sin dose] SE . . 08 + [a4 cos 0+ ayy sin dein + tex covsin de asx _ 5 Bay Det s+ sin Te — sin cond “GEE ~ cos heosd Fe (C38) ‘The partial derivatives of altitude rate h and velocity rate V that appear on the right-hand side of these equations are defined in equations (C-13) and (C-7), respectively. The partial derivatives of the body axis accelerations appearing in equation (C-38) are defined in equations (C-16) to (C-18). Using equations (1-109) and (1-110), the generalized derivatives of the energy-related parameters are defined. The partial derivatives of altitude rate and velocity rate appearing in equation (C-40) are defined in equations (C-13) and (C-10), respectively: OE) _ VV, ah (c20) oy oT (C40) ag The derivatives of the force parameters, lift (eq. (1-111) and drag (eq. (1-112)), are defined in see- tion D.1. The generalized derivatives of the normal force (eq. (1-113)) and the axial force (oq. (1-114)) are presented in terms of the generalized derivatives of the lift and drag forces: = cosa eg tna 22 = (sina ~ Deosa) 3 cnt OL aD . da = ~sina FE tease Ze — (Lease + Daina) 5 (ca) = The body axis rates are defined in eq defined in equations (1-115) to (1-117). the body axis rates and accelerations: tions (1-9) to (1-11). The time derivatives of these terms are ese equations are used to derive the generalized derivatives of = we emacosa Fe Vain coe 32 Veosasing 2 (€-43) He) og May 8 2 “ae = BN8 FE + V coo Fe (C-44) wi, é a incor Fe + Veosacos FF ~ Vsinasing 32 (C-45) 1(@Xr_ OD, ak) 20 ven g 5B cosa Fe tna Ze) Vin acon 5E-4 ¥ sina 3 + (rsing— asin cos) + plDaina + Leona) g¥ cos cos, a - y ing) 28 26 + OV cos8 + aV conasina) FE ~ gcor0 32 (c-46) da + (psinaccos 8 ~ rcos.acos 8) GE HOV cosa cos + r¥ sin aos) & ~ (DV sinasin f - rV cosasin 8) % + 9008 9c0s g % ~gsin@sing a (e-47) 1 (22 ng 2? cena 22) - vein 2 4 2 aE iQ sina a) Vsin8 5¢ + V cosa cos oe ov O + (qoscos 8 = pain) Fe ~ [cease ~ Lsina) + V sin acosd| a = (qV cos eesin f + pV cos 8) 3B - gcosdsing 2 — gain Boon 5p (c-48) ‘The outputs of various instruments displaced from the vehicle center of gravity are defined in equa- tions (1-118) to (1-121). ‘These equations define angle of attack, angle of sideslip, altitude, and altitude rate instrument outputs, The generalized derivatives of the quantities are based on these equations: Ales) _ ta OP , 2a 04 _ (42a = Pia WV, Ba : af Vv ut V a ( v ) 3 a silees (85) _ _ 78 OP , 29 Or _ (729 pee OV , 28 eae Bo (“) Fete (C50) a6 (-¥nc0s $ e080 + z4sin de0s@) 5 ah z (C51) +(zn.cos 0 + mm sin dsin 8 + 25.cos sind) at sin gsin8)] 3 = [yj sin dc0s 0 + x, cos cos) + A(y;, cos dsin 8 + [= d(zysind~ yy sin dos cos $0058) + H(yj, cos dsin 8 — 24 sin dsin#) a = (yj, 008 6 e050 — 24 sin dos 8) $ + (2jcosd + y, singsind + 25, cos isin 8) Fe oh 5 1% (€-52) ‘The generalized derivatives of bank angle rate, pitch attitude rate, and altitude rate with respect to the dummy variable € are defined in equations (C-10), (C-11), and (C-13), respectively. The final set of observation variables is defined in equations (1-122) to (1-125). These equations, defining total angular momentum and the stability axis rotational rates, are used to determine the generalized derivatives of these quantities: OP)

    “ance tho (D-170) ay = -xp eae — tan ig sin ey (D-171) FO = A {avin + 10g) ~ e080 Te sine SP + EIHASC. + Zecosap— Xrsinao 4+ mg( cos 9.08 Ga¢0s a0 + sindosinag)} (0.172) 59 az ax, ‘ “ @SC\,, — c08 a9 Ta +sin ag Fe + Zypsinay + Np cosao + mg(cos By cos do sin ag ~ sin Go c08 00) + tan fo (posin ay ~ rocos 2a)] aX ° OB = tan fo [-95CL + Zr cos.ag — Xy sin ay aa) “oa a: ay {asc., = 08 a9 Se + sin + mg(cosM% cos do ¢05.a9 + sin By sin ool} 1 ~ aargltos0%A0 + rosin) 4 05 Be 08 fy sin do cos a9 (sin Bp c0s dy c05 ay — cos Hy sin a9) 2 Toeos fy 1 0, Bo FvbCi $e + ain) ase ~ Vom cos BoM Sb ~ BV coset Ss [ 2s, os, D.2.9 Angle-of-sideslip rate derivatives — a semen Bo Cp, + c08 By Cy,) + sin ay ~~ IEE (sin Bo Co, + c08 2a Cy,) we io in Po Cp, + 608 fo Cv.) — £08 a 60 (D-173) (D-174) (0-175) (D176) (D-17) (D-178) (D179) (D-180) {D-181) (D-182) (D-183) (D-184) (D-185) (D-180) aa) [sino (o¥a-+ apy) + S05 Ba (WVaCy + Ce) av ax: 7 _ saosin iy 2 = £08 ay sin fo Gye + cose yr — sin ao sin Bo Giz = warglastcin Bo Cr + €08 fo Cy) — X-005 eosin Bp $+ ¥7 089 ~ Zrsinap sin A] 2) A [asteindo Co, + condo Cv ) = cosas Bo 2X + cosy BE da ~ mVo ph Pe eye da 8a —sin ap sin Sy a + Xysin ag sin Bo — Zr cos ay sin By = ma(sin sin a9 sin Bo + C08 By C05 $y COs agin | + po cosao + rosin a9 s = {asin a (Co, ~ eo + cos fo (Co + Cy,)I mo OY, " dz = cosagein fy - + 008 Bo OB ~ sin.ag sino Fe — Xr 08 c10 ¢08 By — ¥r sin fig — Zr sin a9 cos + mg(sin Bp cos 29 C05 fJo — C08 Hp Sin dy sin Hy ~ cos Ao COs Go sin ancos o)} 5. (c0s 8 cos by 05 fo + €08 By sin hy sin ao sin 90) ‘608 0p C08 49 sin fly — sin A sin dp cos Bo + sin pcos dy sin a sin fo) fa (3v8C> 2 + aCr,) + cont (54 1) 15° (sin a Cn, + 080 Cv) ‘es gm sin ly Cp, + £08 Bo Cv) 5(sin Ao Coy, + £08 By Cy,,) ~ €08 a9 sin Bo “GE + €08 0 a4, ~ snaysin a, 2] mal me (D187) (D-188) (D-189) (D-190) (p-191) (D-192) (D-193) (D-194) (D-195) {D-196) (D-197) (D-198) D.2.10 Roll attitude rate derivatives — ag) _ : Seat (D-199) 2) sin do tan dy (0-200) 9 i 38 — cong tanty (D201) = 28) _ 02) Fe =0 (D-202) ag) _ .! ai ao (D-203) a8) _ Z - eo (D-204) = 7 = qocos dy tan Oy ~ ro sin dy tan 6 (D-205)- FE) = sin do sec? 0 + ro cos do sec My (0-200) ~ 4) _ =P (D-207) (D-208) - (D-209) - (D-210) (D-211) (D-212) (D-213) D.2.11 Pitch attitude rate derivatives. 6) no (D-214) a6) ; y= 60800 (D-215) a =

    You might also like