NASA Reference Publication 1207 August 1988 Derivation and Definition of a Linear Aircraft Model Eugene L. Duke, Robert F. Antoniewicz, and Keith D, Krambeer NASANASA Reference Publication 1207 1988 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 NASA National Aeronautics and Space Administration Scientific and Technical Information DivisionCONTENTS SUMMARY 1 INTRODUCTION 1 SYMBOLS 2 Vectors... eee 2 4 Matrices... 0... 4 Subscripts 5 Superscript... . - 1 NONLINEAR SYSTEM EQUATIONS 5 1.1 Definition of Reference Systems 6 1.2 Nonlinear State Equations : i 1.2.1 Rotational acceleration. . - n 1.2.2. Translational acceleration. . “4 12.3 Attitude rates. ©... . 15 1.24 Earth-relative velocity... . . 16 1.3 Nonlinear Observation Equations . . er oe 7 1.3.1. Accelerations. 5 gadao 18 1.3.2. Air data parameters... 60.020 eer 20 133. Flightpath-related parameters. 20 1.3.4 Energy-related parameters. a 1.3.5 Force parameters. ...... a1 1.3.6 Body axis rates and accelerations. . . . : au Instruments displaced from the vehicle center of gravity. . . pe 2 Miscellaneous observation parameters. .. 2 2 LINEAR SYSTEM EQUATIONS 22 2.1 Linearization of the State Equation . 23 2.2 Linearization of the Observation Equation 4 23 Definition of Matrices in Lineatized System Equations . G 26 2.4 Elements of the Linearized System Matrices... . . 7 3 CONCLUDING REMARKS 30 APPENDIXES 31 A—AERODYNAMIC FORCES AND MOMENTS 31 B—DERIVATION OF THE WIND AXIS TRANSLATIONAL PARAMETERS V, 4, AND § 35 Ba Preliminary Definitions 5 35 B.2 Derivation of V Equation... .. « 35 B.3. Derivation of é Equation ..... . 36 B.A Derivation of # Equation ...... a 37C—GENERALIZED DERIVATIVES ca c2 Generalized Derivatives of the Time Derivatives of State Variables . . Generalized Derivatives of the Observation Variables ............ D—EVALUATION OF DERIVATIVES Da D2 Da Preliminary Evaluation : D.1.1 Rolling moment derivatives. . D.1.2 Pitching moment derivatives. D.1.3 Yawing moment derivatives Dil Drag force derivatives. . . DLS Sideforce derivatives. D.L.6 Lift force derivatives. Evaluation of the Derivatives of the Time Derivatives of the State Variables D.2.1 Roll acceleration di D.2.2 Pitch acceleration derivatives... . . D.2.3 Yaw acceleration derivatives. . D.2.4 Decoupled roll acceleration derivatives. . . . D.2.5 Decoupled pitch acceleration derivatives. 1.2.6 Decoupled yaw acceleration derivatives. - D.2.7 Total vehicle acceleration derivatives. D.2.8 Angle-of-attack rate derivatives. D.29 Angle-of-sideslip rate derivatives. D.2.10 Roll attitude rate derivatives. . . D.2.11 Pitch attitude rate derivatives. . D.2.12 Heading rate deriv D.2.13 Altitude rate derivatives. D.2.14 North acceleration derivatives. D.2.15 East acceleration derivatives. . . oe Evaluation of the Derivatives of the Observation Variables . D.3.1 Longitudinal kinematic acceleration derivatives. 1.3.2 Lateral kinematic acceleration derivatives. D.3.3 Z-body axis kinematic acceleration derivatives. D.3.4_ x body axis accelerometer output derivatives. . D.3.5. y body axis accelerometer output derivatives. D.3.6. z body axis accelerometer output derivatives, D.3.7 Normal accelerometer output derivatives. . D.3.8. Derivatives of x body axis accelerometer output not at the vehicle center of gravity. . D.3.9 Derivatives of y body axis accelerometer output not at vehicle center of gravity. . oe D.3.10 Derivatives of 2 body axis accelerometer output not at vehicle center of gravity, D.B.1 Derivatives of normal accelerometer output not at velicle center of gravity... ban D.3.12 Load factor derivatives. D.3.13 Speed of sound derivatives, D.3.14 Mach number derivatives. tives. 39 39 49 49 49 51 51 52 53 53 54 55 56 87 57 58 59 60 62 62 63 64 65 65 66 66 or os 69 69 70 a n 73 a 6 76 1D.3.15 Reynolds number derivatives... +++ D.3.16 Reynolds number per unit length derivatives D.3.17 Dynamic pressure derivatives. D.3.18 Impact pressure derivatives wees D.3.19 Mach meter calibration ratio derivatives... | D.3.20 Total temperature derivatives. D.3.21 Flightpath angle derivatives... . 3.22 Flightpath acceleration derivatives... D.3.23 Vertical acceleration derivatives... . . - D.3.24 Specific energy derivatives. 1.3.25 Specific power derivatives. D.3.26 Normal force derivatives. D.8.27 Axial force derivatives. D.3.28 x body axis rate derivatives. . D.3.29 y body axis rate derivatives. - D.3.30 2 body axis rate derivatives... . . . D.3.31 x body axis acceleration derivatives. - D.3.32 y body axis acceleration derivatives. D.3.38 2 body axis acceleration derivatives. D.3.34 Angle-of-attack sensor output derivatives... . . D.8.35 Angle-of-sideslip sensor output derivatives. D.3.36 Altimeter output derivatives. i 1.3.37 Altitude rate sensor output derivatives. D.3.38 Total angular momentum derivatives. 1.3.39 Stability axis roll rate derivatives. 1.3.40 Stability axis pitch rate derivatives D.3.41 Stability axis yaw rate derivatives REFERENCES 80 98 101SUMMARY ‘his report documents the derivation and definition of a 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 linear 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. These models are widely used, not only for computer applications but also for quick approximations and desk calculations. 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 a 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 earth. 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 linear state equation is derived from the nonlinear six-degree-of-freedom equations of motion. ‘The linear observation equation is derived from a collection of nonlinear equations representing state variables, time derivatives of state variables, control inputs, and flightpath, air 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; McRuer and others, 1973; Northrop Aircraft, 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 flight and setting the asymmetric aerodynamic and inertia terms to zero, one can easily 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 Hoffman (1972), and Gracey (1980) provides a detailed discussion of speed and altitude measnrements. However, neither of these references present linear models of these measurements. This report relies heavily on these two references and uses 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 2 FORTRAN program called LINEAR that derives a linear aircraft model by numerical differencing (Dieudonne, 1978). The program LINEAR produces a linear aircraft model (both state and observation matrices) that is equivalent to the 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- linear state and observation equations (app. C), and the individual derivatives of the state and observation equations (app. D). The details of the principal results of this report are presented in appendix DSYMBOLS A a4 eK total aerodynamic axial force, Ib speed of sound, ft/sec normal accelerometer output, g output of normal accelerometer not at vehicle center of gravity, g output of accelerometer aligned with vehicle body 2 axis, g output of accelerometer aligned with body x axis, not at vehicle center of gravity, g Kinematic acceleration in vehicle body # axis, ¢ output of accelerometer aligned with vehicle body y axis, ¢ output of accelerometer aligned with body y axis, not at vehicle center of gravity, Kinematic acceleration in the vehicle body y axis, g output of accelerometer aligned with vehicle body 2 axis, ¢ output of accelerometer aligned with body z axis, not at vehicle center of gravity, g kinematic acceleration in vehicle body z axis, g reference span, ft generalized force or moment coefficient derivative of generalized force or moment coefficient with respect to arbitrary variable 2 reference aerodynamic chord, ft total aerodynamic drag, Ib L~ly L-L I-Ie 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 « body axis, slug-ft? product of inertia in 2-y body axis plane, slug-ft? product of inertia in z-z body axis plane, slug-ft? ‘moment of inertia about y body axis, shug-ft? product of inertia in y-z body axis plane, slug-ft? ‘moment of inertia about = body axis, slug-ft? Jele ~ [2, Teyle + Iyzles Teylys + Iyles Tels — 12, Telys + Teylz Fely ~ Izy total moment about + body axis, ft-lb; or, total aerodynamic lift, Ib unit length, ft total moment about y body axis, ft-Ib; or, Mach numberpopen eer sae vehicle mass, slugs total moment about z body axis, ft-lb; or, total aerodynamic normal force, Ib load factor specific power, ft/sec roll rate (about 2 body axis), rad/sec static or free-stream pressure, Ib/ft? stability axis roll rate, rad/sec total pressure, Ib/ft? pitch rate (about y body axis), rad/sec dynamic pressure, Ib/ft? impact pressure, Ib/ft? Mach meter calibration ratio stability axis pitch rate, rad/sec Reynolds number Reynolds number per unit length, ft“? yaw rate (about = body axis), rad/sec stability axis yaw rate, rad/sec surface area of wing, ft? total angular momentum; or, ambient or free-stream temperature, °R total temperature, °R time velocity along 2 body axis, ft/sec vehicle velocity, ft/sec velocity along y body axis, ft/sec velocity along z body axis, ft/sec total aerodynamic force along x body axis, Ib total gravitational force along z body axis, Ib total thrust force along « body axis, Ib vehicle position along 7 earth axis, ft total aerodynamic sideforce, Ib total aerodynamic force along y body a total gravitational force along y body axis, Ib total thrust force along y body axis, Ib vehicle position along y earth axis, ft total aerodynamic force along z body axis total gravitational force along z body axis, 1b total thrust force along z body axis, Ib vehicle position along 2 earth axis, ft 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 flightpath angle, rad ith control surface deflection pitch angle, rad coefficient of viscosity, Ib/ft-sec density of air, Ib/ft? arbitrary functionQEsk ce w Er mR ame ¢ co g aa Matrices a yr Lv Onxm laxm 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 control 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 generalized state equation, CX = Ax+ Bu control matrix of the state equation, x = A’x + Blu system matrix of the generalized state equation, CX = Ax+ Bu feedforward matrix of the generalized observation equation, feedforward matrix of the observation equation, y = H’x+ Fu derivative observation matrix of the generalized observation equation, y = Hx + G+ Fu observation matrix of the generalized observation equation, y = Hx +GX-+ Fu observation matrix of the observation equation, y = H’x-+ F’u 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, TX = f[x(t), X(t), u(t)] Xm matrix of 0 values an n x m matrix with values of 1 on the diagonal Hx+Gk+FuSubscripts a aerodynamic; or static or, free stream b body axis system D drag 5 gravitational h displacement of altitude instrument i displacement of altitude rate instrament 4 not at vehicle center of gravity a kinematic L itt t rolling moment m pitching moment 0 yawing moment n orthogonal P power plant induced 5 stability axis; or, specific T thrust t total y vehicle-carried vertical axis system w wind reference axis system 2 displacement in x body axis zy z-y body axis plane a 2-2 body axis plane Y sideforce v displacement in y body axis ve y-z body axis plane 2 displacement in the 2 body axis 0 at sea level, standard day conditions; or, nominal conditions Superscript 1 transpose 1 NONLINEAR SYSTEM EQUATIONS ‘The motion of an aircraft as a rigid body can be described by a set of six nonlinear simultaneous 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 Dieudonne (1978) as time-invariant system expressed as (0) = ffx(@), w(t)] (rt) where x(t) is the 12-dimensional time-varying state vector (¢ being time), (2) is the derivative of x(t) with respect to time, u(t) is the k-dimensional time-varying input or control vector, and f is a 12-dimensional nonlinear function expressing the six-degree-of-freedom rigid body equations. ‘Measurements of the vehicle state can be represented by the observation equation y(t) = alx(t), u(t) (1-2)where y(t) is an dimensional time-varying observation vector and g is an é-dimensional nonlinear func- tion expressing the relationship of the true vehicle state and control vectors to the observed parameters. ‘Typically, 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 Iliff, 1980, 1986). ‘The nonlinear system equations include X(t) terms in both the state and observation functions. In fact, in the most extended form 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 fEx(1), (0), u(e)] (13) y(t) = alx(t),X(¢), u(O)] (4) where Tis a constant 12x 12 angular velocity matrix. Txt) 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 topodetic reference systems. The stability axes are 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. Measurement 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 z,y, and z 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 (z and y) and altitude (4) are measured from the origin of this reference system. The vehicle-carried vertical axis system (ig. 2; Etkin, 1972) has its origin at the center of gravity of the vehicle. The 2y axis is directed north, the yy axis east, and the zy 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 #, 6, and 4, respectively) is described in terms of the orientation of the aircraft body axes with respect to the vehiele-carried vertical axes. The origin of the body axis system (fig. 3) is the vehicle center of gravity. The 2 axis is directed toward the nose of the aircraft, the y axis toward the right wing, and the 2 axis toward the bottom of the aircraft. The specific orientation of the actual body axes relative to the vehicle body is somewhat arbitrary. For symmetrical aircraft, the z and z axes are in the plane of symmetry; for asymmetrical aircraft, these axes are located 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 (1, M, and V about the 2, y, and 2 axes, respectively) are shown in figure 3. 6Ye (East) Worth) xy x (North) 24 (Down) (East) (Worth) x y (east) (own) — (Down) ne Figure 1, Topodetic azis system. Figure 2. Relationship between topode- tic and vehicle-carried vertical avis sys- tems. fed Se % ' Figure 3. Body avis system.‘The relationship between the vehicle-carried vertical and body axes is shown in figure 4, The Euler angles (, 8, and 4) define the orientation of the body axes with respect to the vehicle-cartied vertical axes, The rotations required to transform the vehicle-carried vertical axes to the body axes are shown in figure 3. The heading angle ¥ is a rotation about the = vehicle-carried vertical axis into a new axis system (designated (21, 11, #1) in fig. 5); the pitch attitude @ is a rotation about the yy axis into the (22, y2. 72) axes system; the roll attitude ¢ is a rotation about the yp axis into the body axes, Figure 4. Relationship between vehicle-carried vertical and body azis systems. These rotations are described by cos sind 0 Ly = | sinw cose 0 (5) Go cos@ 0 sin@ Ip = o 10 (1-6) =sin@ 0 cosé 10 0 Ls = | 0 cose -sind (7) 0 sing cose and the total rotation is described by cos 8 cosy cosdsinyy — — sind singsinO cos singsin@siny singcos® Ipvy=Lylele=| -cosdsind — +cos pcos (18) cosdsinBcosts cosdsindsinys sinocosd tsingsiny — sin gcosyray ey. sy ay m1) (a) Rotation through ys about 2 avis. (b) Rotation through 8 about yy azis. Yer (c) Rotation through ¢ about 2 avis. Figure 5. Rotation of azes through Euler angles.Because Zpy is a unitary matrix, the transformation from the body axes to the vehicle-carried vertical axes is LEy. By ‘The relationships between the body, wind, and stability axes are shown in figure 6. All three axis systems have their origin at the center of gravity of the aircraft. The 2 axis in the wind reference system (zw) is aligned with the velocity vector of the aircraft. The angle of sideslip and angle of attack a define the orientation of the wind axes with respect to the body axes. (‘The stability axes are shown in figure 6 also, This reference system is displaced from the wind axis system by a rotation 8 and from the body axis system by a rotation —a.) Figure 6. Relationship of body, stability, and wind azes. Also shown in figure 6 are the components of the velocity vector V in the body axes (u, v, and w) and the definition of positive rotations for a and 8. It should be noted that is a positive rotation in a left-handed coordinate system, whereas the positive sense of all other rotations used in aircraft analysis are positive in a right-handed coordinate system. ‘The definitions of the body axis velocities (fig. 6) are = Vcosacosf (1-9) =Vsing (1-10) w = Vsinacosf (a) The total velocity V, angle of attack a, and angle of sideslip 9 can be expressed in terms of these body axis velocities as V = |V]= (uP + 0? + w?)? (1-12) @ = tan? : (1-13) Basin tS (1-14) 101.2 Nonlinear State Equations For the aircraft problem, the state vector x is 12x 1 vector composed of four 3x 1 subvectors representing the vehicle rotational velocity, the vehicle translational velocity, the vehicle attitude, and the vehicle location: xe bf xf xd? (18) where x= [pao]? (1-16) x2 =[V a gl? (1-17) x9 = [¢ 6 yt (1-18) x4 = [hz yl” (19) with x1, x2, %3, and x, being the rotational velocity, translational velocity, attitude, and position subvectors, respectively. ‘The vehicle rotational and translational velocity are defined within the aircraft-fixed axis systems. In the formulation of the state used in this report, the vehicle rotations are body axis rates, whereas the vehicle velocity terms are stability axis parameters. The vehicle attitude and location parameters are earth relative. ‘The vector function f, relating the state vector its time derivative, and the control vector to the time derivative of the state vector with respect to time, is a 12-dimensional vector function composed of four 3-dimensional vector subfunctions: Efx(t), (0), w(t) = (PA PAT (1-20) where fi, fy, fs, and fj are the vector functions that relate the x(t), X(t), and u(t) vectors to the rotational acceleration, translational acceleration, attitude rate, and earth-relative velocity subvectors of X(t). In the, following sections, each of these subfunctions will be developed separately. The details of the derivation of these subfunctions can be found in any of the standard references on aircraft dynamics (Etkin, 1972; McRuer and others, 1973; Thelander, 1965). 1.2.1 Rotational acceleration—The subfunction f; of f from which the rotational acceleration terms in the X vector are derived is based on the moment equation d M=5H (1-21) where M is the total moment on the vehicle and H is the total angular momentum of the vehicle. ‘This expression can be expanded to M fum+9 x (19) (1-22) where 6/ét is the time derivative operator in a moving reference frame (such as the vehicle body axis system) and the substitution H=19 (1-23) has been used to replace the total angular momentum term with the product of the inertia tensor J and the rotational velocity vector @. (The inertia tensor is assumed to be constant with time.) The definition of the terms in equation (1-22) follow: EL] [Ltir M=|5M|=|M4 Mr (1-24) uN N+ Nr uwith L, M, and NV being the aerodynamic total moments about the 2, y, and z body axes, respectively, and Lr, Mr, and Ny the sums of all power-plant-induced moments; Te ~Tey lez t=|-by Wy -le (2-25) wles hye Te where Je, ly, and J; are the moments of inertia about the x, y, and z body axes, respectively, and Izy, en, and Iy; are the products of inertia in the z-y, 2-2, and y-z body axis planes, respectively; and Q=x,=[p qr" (1-26) where p, q, and r are the rotational rates about the 2, y, and z body axes, respectively. Because it is assumed that the inertia tensor is a constant with respect to time, equation (1-22) can be rewritten as fa =I-(M-2 x 19) (1-27) This is the vector subfunction for the rotational acceleration. Designating this subfunction as fj, the following definition applies: filx(),8(0), w(t] = MIM — 9 x (79)] (128) where Ao = flats) SC), u(t) (1-29) ot . 7 Raab aa (1-80) Since the inverse of the inertia tensor I~1 is given by 1 [tbo eoailt EE] on where det I = Lelyls — Lelgs ~ TeToy — Tyl2, - UysTeeley (1-32) h=Ul.- 2%, (1-33) Tn = Inyle+ Iyeles (1-34) Ty = Leghye + Iles (1-35) i= Ll.- 22, (1-36) Is = Lelyz + leyles (137) Ig = Tely ~ 1, (1-38) 2the expression for the rotational accelerations can be expanded as a set of scalar equations: aks detT pe [ELL + 2M Ip + ONIs — p*(Leed — Leys) + pqlesh = Iysl2 — Dela) ~ Pr(leyls + Dye ~ Tysls) + P(Iyel = Teyls) — gr( Doh ~ Inylz + Tesla) ~PUyeh ~ Ieeh)] (1-39) FIDL la + BALL + BN Is ~ Ueala ~ Ieyls) + Dales ~ Iya ~ Dale) = Pr(Teyla + Dyla ~ Iysls) + P(Iyels ~ Ieyls) ~ 9r( Dela ~ Ieyls + Tees) ~ P(Iyala ~ Teels)) (2-49) sg [EL Io + EMI + EN Ie — p'LezTs ~ Ieylo) + Pallerls ~ Iyelo ~ Delo) ae Dyls ~ Iysle) + 0°(Iyel — Inylo) ~ 7( Dela ~ Toys + leale) = (pels ~ Teal) (1-41) where (1-42) (1-43) (1-44) Equation (1-3) defines the generalized nonlinear state equations as TX) = f[x(2),X(2), w(0)) ‘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): M= fua)+0 x 10 Rearranging terms and assuming that the inertia tensor is constant with respect to time, the equation can be written as rh9 =M-2x 10 (1-45) ‘The rows of this vector equation are now scaled using the following scaling matrix: 0. 0 =| 0 1, 0 (1-46) o 0 Wh This matrix, when premultiplying equation (1-27), merely divides the first row by the roll inertia Jz, the second row by the pitch inertia Z,, and the third row by the yaw inertia [.. Using the definition Jess (1-47) 13the resulting equati Jha = I'M ~J'(Q x 19) (1-48) and J can be written as 10 —Iey/Ie ~ Tes/Te J —Teyfly 1.0 —Iys/ ly (1-49) Tell =Iy[l 10 Equation (1-48) can be expanded and expressed as y 10 = Iey/Te ~Iee/Te ] [ 3 vy) = Fey/Ty 10 aTyally | | a Y Tes/Ty —Iys/le 1.0 é BL Te tPlzy/ Ts + Paes] Te + 10ly/ Le + (Q? ~ 4) byel Te ~ gra | Te = | EM/1, ~ rple/ ly + raley/ly ~ Palys/ly + (0? ~ p?)les/Iy + prle/I, | (1-50) BN /Ie + gpls/Iz~ ares) Ie + Prlys/Iz + (v* ~ q? May/ Te — Paly/ Te where p’, 4’, and # are the decoupled rotational accelerations of the vehicle. Using the definition of J in equation (1-49), the matrix transformation T can be defined as T (1-51) which would be an identity matrix except for the presence of the inertia terms in the upper left-hand corner. Thus, the vector subfunctions for the generalized state equation defining vehicle translational acceleration, vehicle attitude rates, and earth-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. 1.2.2 Translational acceleration.—Derivation of the translational acceleration vector subfunction fy is based on the force equation (1-52) where F is the total force acting on the vehicle and mis the vehicle mass. TThis expression can be expanded to F=m(Svsaxv) (1.53) with the assumption of constant mass with respect to time and the following definitions of F and V: F (2x EY Ezlt (1-54) where BX, BY, and £2 are the sums of the aerodynamic, thrust, and gravitational forces in the x, y, and = body axes, respectively, and | Veluv ult (1-58) uRearranging the terms of equation (1-52) gives an expression for the translational acceleration bv . ir- QxV (1-56) This equation expresses body axis accelerations in terms of body axis forces, angular rates, and velocities. 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= Veosacos 8 v=Vsing w = Vsinacos 8 and equations (1-12) to (1-14) ~ " IV] = (0? +o? + Ww)? > t 8 ale ‘The wind axis translational acceleration terms (derived in app. B) are summarized as: (V 6d}? = Hfx(e), (4), u(e)] (57) where —Deos 6 +Y sin 8 + Xrcosacosf + Yrsing + Zrsin acos 9 — mg(cosa.cos 8 sin 8 — sin B sin 4cos 8 — sin a cos f cos dos 8)] (1-58) 1 4 = Taegegl E+ Zr cosa ~ Xr sin a + mg(cos aos 6.cos6 + sin asin 6)] +9~tanG(pcosa + rsina) (1-59) B= A lDsin8 + ¥ cos — Xpeosasin 8 + ¥p cos ~ Zrsinasin 3 + mg(cosasin fsin 8 + cos sin dcos 6 — sin asin 3 cos cos 8)] + psina —rcosa (1-60) with D being total aerodynamic drag; ¥ total aerodynamic sideforce; and Xr, Yr, and Zr total thrust force along the x, y, and = 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 100 =sing R=} 0 cosd singeosd (1-61) 0 -sing cosécos@where R is derived by Maine and Iliff (1986) from the total angular velocity of the aircraft expressed in terms of the derivatives with respect to time of the Euler angles (4, 8, i): P é ie 0: | fo 1 0 0 ] [cos 0 ~sing] [0 q 0] + ]0 cose sing} |6/+]0 cos@ sing}| 0 1 0 0 r 0 0 ~sing cos} [0 O-sing cosd| [sind 0 cosd | |b 1 0 ~sina ] fg 0 cos singcose | | 6 (1-62) 0 ~sing cos deosa | | ¥ ‘This transformation from earth-fixed to body axes can be expressed by the equation a =k (#) (1-63) where E is an attitude vector whose components are the Euler angles: E=(6y]" (1-64) Premultiplying both sides of equation (1-63) by R~! and rearranging terms yields the equation for the attitude rates, a GE=R0 (1-68) which can be expanded into the scalar equations O = p+qsingtand + rcosptand (1-66) 6 = qeosé—rsing (1-67) b= qvindsecd + rcospsecd (1-68) 1.2.4 Earth-relative velocity—The matrix Zmy that transforms earth axis system vectors into the body axis system is defined by equation (1-8) as cosy —siny 0) [ cos@ 0 sin@][1 0 0 Ipy = | sin’ cose 0 1 0 0 cos@ sing 0 0 1} [=sine 0 cosa} [0 sing cose cos 8 cos cos sin y ~sing = | singsindcosy—cosdsiny singsin@siny + cos dcos sin dcosd cosdsindcosp+singsiny cosdsinOsin y—singcosy cosdcosd The specific relationship between earth-relative velocities and body axis velocities is expressed by V= Ley (3x) (1-69) where R is the earth axis system vector defining the location of the vehicle: ey 2] (1-70) 16with z= —h. ‘The equation for the earth-relative velocity can be formulated as a qR= LghV an) 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 6: h = V(cosacos 8 sin — sin fsin ¢cos@ — sina cos cos 4 cos 6) (1-72) E = V[cosercos B cos @ cos w + sin A(sin @sin 8 cos y — cos dsin v) + sin a cos (cos sin # cos + sin gsin #)) (1-73) 41 = V [cos «cos 8 cos Asin y + sin A(cos bcos + sin gsin Bin ¥) + sin a-cos 8(cos $ sin @ sin p — sin $cos #)) (1-74) 1.3 Nonlinear Observation Equations No standard set 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 gix(t), %(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 bxtat at yt (1-75) where x and % are the state vector and time derivative of the state vector described previously, u is the control vector, and y’ is defined by = LV YS VS VR VS VS YF VT? (1-76) where Vi = [ae Oye Gee Ge dy Bs On Gai Opi ai Ami 7)” (1-77) yh= la M Re Re! G qe ge/Ps Pa T TJ" (1-78) ys = fy fpa iy (0-79) v= (Es PIP (1-80) ys= [LDN alt (1-81) Yea luv wid alt (1-82) feu: Bi hy hl? (1-83) YeelP ye nl (1-84) cuwith the elements of yi being terms related to the vehicle body axis acceleration, the elements of y', being air data terms, the elements of y4 being flightpath-related terms, the elements of y', being terms related to vehicle energy, y; being a vehicle force vector, the elements of y’, being body axis translational rates and the time derivatives of those terms, y+ being a vector of variables representing measurements from instruments not located at the vehicle center of gravity, and the elements of y, being 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 function g relating the state vector, the time derivative of the state vector, and the control vector to the observation vector is an é-dimensional function composed of four subfunctions: lx(t), 5), u(t)] = [x XP uF gi] (1-85) where x, %, and uw are identity functions on the state vector, time derivative of the state vector, and control vector, respectively, and g' is compored of voctor subfunctions dofining 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 elements 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, Clancy (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, 1.8.1 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 contzol analysis and design problem. These accelerations and accelerometer outputs are measured in units of gand are derived directly from the body axis forces defined in ection 1.2.2. The body axis acceleration vector a can be expressed as fyeSvsaxy 1.86) = av aes Qu It is important to note here that the i, 8, and t 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 etost-product terms, Thus, expanding equation (1-86) yields tex) [itqw—re as| ay | =| eru—pw (1-87) ax | | w+p-qu where az, ay, and a. are the kinematic accelerations in the vehicle body 2, y, and = axes, respectively. Using i (1/m)(Xp + Xq + Xy) trv qu & | =| Ufm)(rr+ Yat Yq) + pw — rw we] Osment 24%) + qu po (stated as eq. (B-1) in app. B), equation (1-87) can be rewritten as ek (1/m)(Xr + Xa + Xy) aye | =} (1/m)(¥r + Ya + Xe) (1-89) a (/m\(Zr+ 24%) (1-88) 18where Xz, Ya, and Z, are total aerodynamic forces and Xg, Yz, and Z, are total gravitational forces along the x, y, and z body axes, respectively. This can be expanded in terms of the gravitational and aerodynamic forces to give (in units of g) [33] 1 Xz — Dosa Lsina ~ gmsin8 Yr +¥ +gmsingcos6 Zy — Dsina ~ Leosa + gmcoscos@ = 1.90) gom (1-90) where go is the acceleration due to gravity at sea level ‘The outputs of body axis accelerometers at the vehicle center of gravity are simply the body axis accelerations due to the thrust and aerodynamic forces. The accelerometer output equations can be written directly from equation (1-90) as ‘gom Yr+¥ Zy —Dsina-Leosa (1-91) where az, ay, and a, are the outputs of accelerometers at the vehicle conter of gravity and aligned with the vehicle body 2, y, and 2 axes, respectively. Because the normal acceleration dy is defined by ty, = a; (1-92) ‘an expression for this variable can be extracted from equation (1-91): - 2p + 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) by as QVtaxv and the definition of inertial velocity 6 Va prtaxe (1-94) ‘The results from Gainer and Hoffman (1972) are reproduced here without rederivation: as, ae ~ [(q? + 1?)2e ~ (Pa Fe — (pr + G)2e]/90 ays | = | ay + [(pq + Fay ~ (p? + 1 )yy — (gr - B)24]/90 (1-95) a2, az + [(pr— d)zs + (ar + Bly: ~ (G+ P*)22]/90 where az, dy,i, and a,, are outputs at accelerometers aligned with the 2, y, and z body axes but not located at the vehicle center of gravity; the subscripts 2, y, and z refer to the z, y, and z body axes, respectively: and the symbols 2, y, and 2 refer to the 2, y, and 2 body axis locations of the sensors relative to the vehicle center of gravity. Because the normal acceleration is the negative of the z body axis accelerometer, the output of a normal accelerometer not at the vehicle center of gravity but aligned with the z body axis, anj, is given by nq = On [(9r — G2 + (gr + Bue — (@? + P*)eal/90 (1-96) ‘The 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 - ng (1-97) 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 pa, total pressure , ambient or free-stream temperature 7’, and total temperature T,. The selected reference and scaling parameters are Mach number M, dynamic pressure g, speed of sound a, Reynolds number Re, Reynolds number per unit length Re’, and the Mach meter calibration ratio ge/p,. The derivation of these quantities is treated extensively by Gracey (1980). ‘The nonlinear equations defining these quantities are 12 | (1-98) (1-99) (1-100) (1-101) (1-102) _ (2.0 + 0.2447)95 — 1.0]p. (M < 1.0) fe = Nites rates 1.0}p. (Mf > 1.0) (1-103) ge _ { (1.04 0.2M2)>> — 1.0 (Mf < 1.0) fa Pa \ 1.2M?[5.76M!?/(5.6M? —0.8)]?5— 1.0 (M > 1.0) (1-104) T, = T(1.0+0,2M?) (1-105) where p is the density of the air, is the coefficient of viscosity, and the subscript 0 refers to sea level, standard day conditions. Free-stream pressure, free-stream temperature, and the coefficient of viscosity are properties of the atmosphere and are assumed to be functions of altitude alone. 1.3.3 Flightpath-related parameters—Included in the observation variables are what might best be termed flightpath-related parameters for lack of better nomenclature. ‘These terms include fiightpath angle 7, flightpath acceleration fpa, and vertical acceleration fi. The variables are defined by the following equations: (1-106) Vv fpa= E 1-107) te Gl (2-107) h = a,x sin — a,xsin dcos 8 ~ a,y cos $cos 0 (1-108) 201.8.4 Energy-related paramete: tion variables considered in this report: —Two energy-related parameters are included with the observa specific energy E,, and specific power P., defined as (1-109) i (1-110) 9 1.8.5 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 N, and total aerodynamic axial force A, defined as, a5CL Qu) D=§5Cp (1-112) N = Leosa+Dsina (a3) A= —Lsina+ Deosa (114) where § is the surface area of the wing, Ci, 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 z 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, é, and to, respectively. ‘The definitions of the body axis rates are given in equations (1-9) to (1-11) as u=Veosacoss Vain 8 w = Vsinacos a 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 Xr- =Deosa+ Lsina o= 222m o=8 + bsing | wysinf— qV sinercosf (vis) YrtomsingeosO+¥ 5 yin acas f— rV cosacosf (1116) Dsina w = Zr tgmeosdcos + aV cosa cos 8 ~ pV sin 8 (117) 1.3.7 Instruments displaced from the vehicle center of gravity—The need to include measure- ‘ments from instruments displaced froin the vehicle center of gravity arises from the fact that not all aircraft ainstrumentation is located at the vehicle center of gravity. The most important of these quantities are un- doubtedly the accelerometer outputs treated in section 1. In this section four additional parameters are presented: angle of attack (ai), angle of sideslip (4), altitude (h.;), and altitude rate (h,) measurements from instruments displaced from center of gravity by some z, y, and z body axis distances. The subscripts a, 8, h, and h refer to the displacements of the angle-of-attack, angle-of-sideslip, altitude, and altitude rate instruments from the vehicle center of gravity. The equations used to compute these quantities are ay = a4 Se Pie (1418) Tp Pre 1-11 Ba = B+ (2-19) hig = h-+ ay sin 0 — ypsin dcos0 — x4 cos dcos 0 (1-120) h + 0(zj, cos 0 + y; sin dsin 8 + z, cos psin 6) ~ (yj, cos pcos — x, sindcos8) (1-121) 1.3.8 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 ps, stability axis pitch rate q., and stability axis yaw rate r., The equations used to define these quantities are T = 5(lep? ~ 2layog ~ 2lespr + Tyg? ~ 2a + Ter?) (1-122) Pe = peosa + rsina (1-123) G=4 (L124) 1 = —psina + rcosa (1-125) 2 LINEAR SYSTEM EQUATIONS ‘The standard state equation for a linear differential system has the form. X(t) = A'x(t) + Blut) (211) where, for a time-invariant system, A’ is a constant n x n matrix and BY is a constant n x k matrix. The standard output equation has the form y(t) = H'x(t) + F'u(t) (2-2) where H' is a constant £ x mn matrix and F” is a constant £x k 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 CX(t) = Ax(t) + Bu(?) (2:3) y(t) = Hx(t) + GX(t) + Fu(t) (2-4) where C and A are constant n x n matrices, B is a constant nx k matrix, H and G are constant £x n matrices, and F is a constant €xk matrix. The nonlinear system equations developed in section 1 (eqs. (1-1) to (1-4)) 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 uo(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 To(t) = £[%o(t), xo(t), wo(t)) Assuming that the system is operated at close to nominal conditions with u(t), x(t), and X(t) deviating only slightly from uo(t), xo(t), and Xo(¢), the following expressions can be written: u(t) = uo(t) + Su(s) (25) x(t) = xo(t) + 6x(t) (2-6) H(t) = Ko(t) + H(t) (2-7) where u(t), 6x(¢), and 6%(t) 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 %o(2), xo(4), uo(t), and assuming T constant with respect to x(t) yields op aie a ot, ot. ot Tivol) + 6%(4)] = flxo(t), Rot), w(t] + 5 Ox + Be Bi + where f/x, 9£/0%, and Of/0u 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 series expansion are defined by the following relationships: suth(t) (28) of _ af ee (2-9) Ox © Ox (0, X0,tto) at (210) = Ixo.a.t) of or | (211) Cee I(x0.%o,Ulo) the (i, j)th elements of which are defined as of of; (a) 7 8 on) ‘Of Of (3),,° 3 a af) _ ak (i), = ba; c respectively, where f; is the ith simultaneous equation of the nonlinear state differential function in equa- tion (1-3), x; the jth element of the state vector, %; the jth element of the time derivative of the state 23vector, ij the jth element of the control vector, and all derivatives are evaluated at the nominal condition (%o(t), Xo(#), wo(t)). Subtracting equation (1-3) from (2-8), rearranging terms and neglecting the higher order terms yields a linearized state equation, ot [r I 6X(t) = Sox + where the arguments of the matrix functions have been dropped to simplify the notation and where it is understood that the matrices are to be evaluated along the nominal trajectory. Su(t) (2-15) ou Letting at cer-= (2-16) of A-= (217) of | Bas (2-18) equation (2-15) can be written as CC 6%(t) = A 6x(t) + B u(t) (219) which is precisely the formulation of the generalized state equation desired. Premultiplying both sides of equation (2-19) by C-? results in the standard form of the linearized state differential equation, 53(0) = CTA Ex(t) + C71B Bult) (2-20) Letting A=CtA (2-21) Bl=C"B (2-22) equation (2-20) can be written in the more usual notation 5X(t) = A’ 6x(t) + BY 5u(t) (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), ¥(®) = elx(t), (4), u()] Performing a Taylor series expansion about the nominal trajectory (xo(2), Xo(t), wo(t)) yields ag 28 5s 9 Fe okt FE out hi) (2-24) volt) +4y(t) = ebeat) Xo(0), wo] + 98 bx + where ag a os 2 28 (2-25) . |(Xo,%a,tto), 24(2-26) xoxo) oe" Bs (80.9) 7 the (i, )th elements of which are defined by (3), See (2-28) (®), = (229) respectively, where gi is the ith simultaneous equation of the nonlinear observation equation (1-4). Again, all derivatives are evaluated at the nominal condition (xo(t), Xo(t), uo(t))- Subtracting equation (1-4) from equation (2-24), rearranging terms, and neglecting higher order terms results in a linear observation equation, 9 28 55 5 88 5, fy(t) = FE ox + SE ox (231) where the arguments of the matrix functions i been dropped to simplify notation. Letting og a= (2-32) = 2 c=# (2-33) og a (2-34) equation (2-31) can be rewritten as dy(t) = H 6x(t) + G é&(t) + Fbu(t) (2:35) which is the generalized linear observation equation desired. ‘The standard form of the observation equation can be derived by substituting for 6X from equation (2-23) into equation (2-33). This substitution results in by(t) = H dx(t) + GLA’ 6x(t) + BY éu(t)] + F éu(e) (2-36) which can be weitten as by(t) = [I + GA']dx(t) + [F + GB") bu(t) (237) By letting H! = H4GAa (2-38) Fos F+GB (2-39) ‘equation (2-37) becomes 25Sy(t) =H’ 6x(1) + F su(t) 2.3 Definition of Matrices in Linearized System Equations (2-40) ‘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 deriva nominal trajectory. Using the nonlinear state equation (1-3), Tx(t) = f[x(2), X(t), u(t)] the terms in the generalized form of the linearized state equation (2-19), C 6S(t) = A 6x(t) + B but) can be defined as of o=-T-— of A= ox of Bo oe The terms in the standard form of the linearized state equation (2-20), 63(1) = A’ 6x(t) + BY bu(t) can be defined as al) os wer ay du In a similar manner, the nonlinear observation equation (1-4), y(t) = alx(2), X(t), u(t)] can be used to define the terms of the generalized linearized observation equation (2-35), by(t) = H6x(t) + G X(t) + F bu(t) 26 ve of state, and control vectors. All derivatives are understood to be evaluated along the (2-41) (2-42) (2-43) (2-44) (2-45)(2-46) (2-47) (2-48) ‘The terms in the standard form of the linearized observation equation (2-40), by(t) = H’ 6x(t) + FY Sut) can be defined as fy"! at of) of -il oe (2-50) 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), ot Ox the element occupying the ith row and jth column of A, (A);,, can be represented as of Ox; (Adis (2:51) 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, H, G, and F matrices are defined in appendix D based on the generalized derivatives derived in appendix C. ing the state vector x defined in (1-7) as x=[pqrVasoovhay|” the elements of the A matrix can be expressed as A(p)/Ap A(p")/2q --- A(p)/Oy (a!) Op (4')/0q --- Hg) fy : : : (2:52) O(%)/Op A(X) /9q +++ A(X) /Oy Ay) H(G)/0q --- A(y)/Oy 2Substituting for these partial derivatives using the terms in appendix D gives (1/Te)[(45b?/2V0)Ce, + OL2/Op (A/Le){(qSbe/2V0)Ce, + OL1 [Oy + LesPo Ieyto + 12290) +2Iyz9 + To(Iy ~ Ie)] = | (C/E )GSbE/2V0)Cmy + OMr/Op_ — (1/Ty)(GSE*/2V0)Cm, + OMr/Iq (2-53) ~2LesPo — Tyedo + Polls ~ Ie] “Heyro~ IpsPo] The elements of the B, C, H, G, and F matrices can 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)), 8lx(0), (0), u(0)] = fx? 87 uP gi] and the definitions of the matrices for the generalized linear observation equations (2-46) to (2-48) , = 88 4 by _ os On =o Sain ‘These matrices may be expressed using a partitioning based on the vector subfunctions of g as (254) (2:58) (250) 28which become Lax? Orax12 H=|---. Oex12 Lexi upon evaluating the partial derivatives of the identity functions x, X, and u ‘The C matrix may be viewed as a partitioned matrix as where, from equation (1-48), and Cus lex|ly 1.0 Teal ly 1.0 ~Isy/Iz —le2/Te aTes[Te —Iye/Te 1.0 apy (av -a(H)/04 -a(p")/08 a(g)/aV -8(q)/84 -8(9°)/98 Ar) IV —A(r’)/I& -A(r)/AB 0 —(GS2/2Voly)Cms —(GSb2/2Voly)Cmy 0 ~(GSbE/2Vol=)Cng ~(G50"/2V0l-)Cng : —(GSbE/2Volz)Cz, —(GSb*/2Volz)Cr, (2-57) (2.58) (2-59) (260) (261) (2-02) 29Cn =0(4)/0V 1.0-2(a)/0a — -914)/08 -aayjov — -0(8)/04 1.0 a(8)/08 [: 10 (GS¢/2Vom)(cos Bo Cr, — sin Bo Cy.) (GSb/2Vom)(c0s fo Crs) re AV) /ae -avyyob 0 1.04 (98¢/2V9m cos Ao)Crs, (9Sb/2V2m cos Bo)Cr, (2-63) 0 (GSe/2VGm)(sin Bo Co, + c08 fo Cy,) 1.0 ~ (GSb/2V9m)(sin Bo Cp; + cos Bo Cy,) ‘The inverse of the C matrix, C~4, ean be expressed as a partitioned matrix in terms of the matrix subpar- titions of the C matrix as exe (2.64) | loxe ‘The elements of the A’, BY, H', and F’ matrices can be determined using the C~} matrix defined in equation (2-64), the A, B, H, G, and F matrices, and the definitions for 4’, B’, H', 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 aircraft 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 either the reference (nominal) trajectory about which the model is linearized or the symmetry of the vehicle mass and aerodynamic properties. Ames Research Center Dryden Flight Research Facility National Aeronautics and Space Administration Edwards, California, January 8, 1987 30APPENDIX A—AERODYNAMIC FORCES AND MOMENTS ‘The aerodynamic forces and moments acting on an aircraft are the result of multiple factors whose signif- 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 = O(a, 8,V,h,P, 4,1); 8, 61,---,6n) (A-1) where F is an arbitrary force or moment, @ is an arbitrary function, and the 6; are the n control surface deflections. These forces and moments are related to the nondimensional force and moment coefficients by the equations for the forces, D = qSCp (4-2) (A°3) (Ad) and the moments, gSbCe (4-5) GST (a6) N = GSC (a7) where b is reference span and ¢ is reference aerodynamic chord While the nondimensional aerodynamic force and moment coefficients are themselves nonlinear func 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 = Sto + Cine + Cig + Cigh + Cty $C + CyB + Chad + Chef + Cad + Ciy8 (As) mi Co = Co, + Co,@+ Cog + Opyh-+ Coy¥ $Y Cdgb + Cop + Ord + Cr,F + Cr,4+ C048 (49) a Cy = Cy + Creat C¥g8 + Cy ht Cry +30 Cva6i + Cy8+ Cv,d + Cy,F + Ov, + Cv, (A-10) Ce = Cay + Cig + CegB + Ch + CoV 4 36a, 6+ Ci + Cd + CaP + Cex + Coy8 (At) 31Cm = Cing + Crna + Cag + Cmyh + Coy V +0 Cing, 61+ CrngB + Cred + Ome? + Oma + Cg (A-12) ny + Cag + Cg + Oph + Cay V + Cu 6 + Cnph + Oud + Cnt + Ong + Eng (4-13) ct where Cy, is the value of the coefficient along the nominal trajectory and the notation Ce, is defined as 22 Oe with Cy being an arbitrary force or moment coefficient and 2 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 definitions of these nondimensional stability and control derivatives are given in terms of the coefficient Ce. The nondimensional stability derivatives are defined as Ce (aay Cu (a9) Cop = oe (A-16) Cy = xen (Aa7) Rag (ais) Ce oan (A19) Cte (4-20) cs (azn The two other stability derivatives are not nondimensional and are defined as oe (A-22) - (A-23) ‘The control derivatives are defined as Cu, a (A-24) 32The rotational terms in equations (A-8) to (A-13) are nondimensional versions of the corresponding, vari able with z (4-25) = (A-26) x (427) im (4-28) B z (429) Because the Cg, terms are included, the force and moment coefficients are total force and 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 3 ‘The derivation of the wind axis translational acceleration parameters is 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. Bl Preliminary Definitions Equation (1-56), fy-+p-axv a om can be expanded, using equations (1-54), (1-55), and (1-26), to §] [ipa ghee (1/m)(Yr + Ya + Yg) + pw ru (B-1) (Q/m)(Zr + Za + Zp) + qu po ‘The body axis aerodynamic forces can be rewritten in terms of the stability axis forces lift L, drag D, and sideforce Y: —Doosa + Lsina (B2) i (B:3) =Dsina - Leosa (B-4) The gravitational forces can be resolved into body axis components such that X_ = —mgsin® (B-5) Yq = mgsindcosé (B6) Zq = mg.cos ¢coe 0 (B7) ‘These equations will be used in the derivations of the V, 4, and # equations. Thus, the total forces in the body axes can be defined and expanded as EX = Xp-Deosa+ Lsina -gmsind (B-8) EY = ¥r+Y +gmsingcosd (B-9) EZ = Zp - Dsina - Leosa + gmcosdcosd (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), Va(vt+ ott uty?the equation for VV becomes * a d =4 wa cteuye Va GV = F(t +o tut) (ean) which after expanding the derivative and cancelling terms, becomes v= Pui + v4 we) (B12) By substituting the definitions for u, v, and w from equations (1-9) to (1-11) and cancelling terms, equa- tion (B-12) yields V = itcosacos 8 + bsin§ + wsinacos (B-13) ‘The definitions fori, #, and w in equation (B-1) are now used with equation (B-19) to give v ses x, Xr + Xq) + cos.acos A(rv — qu) + 8 + Yr +) +i Alpw — ru) + REP, + Ze + Zq) + sin acos (qu ~ po) (B14) Expanding (B-14) in terms of equations (B-2) through (B-7) and cancelling yields ¥ = 21 Dees +¥ sin 8 + Xrcos acon s+ Yrsin f+ Zr sinacos/t — mg(cos.a.cos f sin 8 — sin sin $cos@ — sin a cos f cos 6 cos 6)] + rvcos.acos J ~ qweosacos 8 + pwsin # — rusing + qusin aeos 8 ~ pusin cos B Equation (B-15) can be simplified by recognizing that the terms involving the vehicle rotational rates are identically zero, which becomes obvious after substituting for u,v, and w in these terms. Thus, the final equation becomes = Dcos B+ sin + Xp cos.acosf + Yr sin 8 + Zrsinacos 8 — mg(cosa_cos 8 sin 8 ~ sin f sin cos ~ sin ar cos f cos cos 8)] (B-16) B.3_ Derivation of & Equation ‘The equation for d can be derived from the definition of a in equation (1-13), ‘Taking the derivative of a with respect to time, daw qe (B17) then expanding and cancelling terms, the equation becomes oo a (uri — tw) (B-18) aw 36Substituting the definitions of u and w from equations (1-9) and (1-11) into equation (B-18) gives V cos 8 (B19) Using equation (B-1) to substitute for i and w and equations (B-8) to (B-10) to define the forces, equation (B-19) becomes, after rearranging terms, jl = VmeosB L + Zp cosa ~ Xp sina + mg(cosacos$cos 6 + sin asin )) 1 + Paeapttcoser= preota = rvsina + qusina) (B20) which after substituting for u,v, and w from equations (1-9) to (1-11) and combining terms gives A 1 A ‘ 4 = Tega glnh + Arcos — Xrsina + mg(cosa cos cos 4 + sin asin 6) +4 tan A (peosa + rsina) (B-21) B.4 Derivation of § Equation ‘The equation for (is derived from the definition of @ as given in equation (1-14), (B-22) which becomes, after expanding the derivative, substituting for V, and cancelling, oo : , A = F[-iteos asin B+ cos f — sisinasin 6] (B-23) Using equation (B-1) to substitute for i, 8, and t and equations (B-8) to (B-10) to define the forces, , 2 cosasin § (—Deosa + Lsina + Xp — mgsin#) + cos.8 (Y + Yr + mgsin dcos8) mV = sinasin 8 (~Dsina ~ Leosa + Zr + mgcos$cos8)] 1 + g7l- cosa sin 8 (rv — qw) + cos f (pw — ru) — sin asin 8 (qu — pr)] (B-24) Substituting into equation (B-24) for u, v, and w and rearranging terms yields the final equation 7 : 8 = plDsin 8 + ¥ coef — Xz cov asin + ¥r-cosf — Zrsin asin gs + mg(cos a sin B sin # + cos Asin 4 c080 — sin asin f cos cos 8)| + psina—reosa (B25)APPENDIX C—GENERALIZED DERIVATIVES ‘The equations defining the time derivatives of the state variables (derived in sections 1.2.1 to 1.2.4) 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 quantities. OL oM ON Oly OMy ONy {ngetn Fete Geek an Hey 1 [D(a ~ Ty) ~ {Tal ~ Iola ~ Dos) + 0g + Dyn fe) 4 [Cans ~ Iya ~ Dols) 42 ly ~ Dy) ~ {Daly ~ ela + el) = [PUleyli + Dyla ~ Iyels) + (Di - Teylo + Toss) + 2r(Iyzhi — Ieela)] s (C-1) ON |, Obr ,, OMr Nr OG) _ 1 [7 8b , 8M, ON |, Ober OE = wile ge th ae +Is 9 th 3E +h 8 ts oe — (2pUezTs - Teyls) ~ a{Teel2 ~ Iyela — Dels) + Tey + Dyla ~ Tyels)] 2 + (Dela ~ Tale ~ Dale) + 2Cyl Iyls)~ rDala~ Ile el) 38 ~ (P(Teyl2 + Dyla— Ils) + (Dela ~ Inyla + Iesls) + 2rUlyela ~ lesls) =} (c:2) 1 (0b, OM.) ON | alr, aMr, , ONt wr {szete Gp 18 Be th Ge tle Ge tle Ge ae og + [PUlseTs ~ IyeTs — Dee) + 2q(Iyels — Ieyle) — "(Del ~ Teyls + Ize le)] a ~ 2pUeels ~ Ieyle) — q{zsI ~ Iyels - Dzlo) + rUzyla + Dyls ~ Iy2To)] ~ Dll + Dyle~ lol) + Dols ~ lel + lol) + 2(lyts~ Ios) $2} (C3) ‘The quantities 11, Ir, Is, I4, Iss Ie, Dey Dyy 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 vehicle (p’, q’, and 7), which are used to determine the generalized derivatives of the decoupled quantities: ag) _ [2% lr oe - i Lae te (they ~ QTes) et (Ples + rly + 2gles ~ 11s) Or = (Ploy - a, lee + 12) =| (iy ~ ay + Pree + ah) FE a a op (C-4) 39omy 4 Germ (le + ahve + PPles ~ rhe) % + (rley ~ Ply) # ar - ~ (Pla ~ aay ~ lee ~ Phe) 5 (C5) Ca a oe *E (= + me + (ale + lye + 2Pley ~ Gly) xe (Pe = hes —2alny = Phy) HE ar ~ (ales ~ ple) F] (cs) 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 9) 1S. cong 22 con PAE 4 sing © 4 sin cos p 220 4 sina SE _ con Ge + eos cos SFE + sin Fe + sinxcoss FE + sin + [+ Xr sinacos + Zr cosa cos 8 + mg(sin Bsin a cos 3 a + os 8 cosdcosa.on 6) 92 + [Dsin 6 + ¥ cos 8 — Xp sin B cosa + Ypcos 8 — ZpsinasinB + mg(sin cos asin 8 + cos Bsin $.cos 8 ~ cos 4 cos $sin asin )] z ~ ma(~condcos sin + coed sin bn cos 8) 92 ~ ma(cos8 cona.cos + sn in in + sin Bcosinccosp) 2 (C7) aa) ot (2 22 nq 22) a “BE = maVveoad (ag +4 Ger si aE = tan Beosa 2? 4 94 _ tanpsina & a6 * 0¢ % av ~(ptaegl E+ Sr emea— Xesina rng(os6 cos cosa + sin in a)]} FF i {aiagl-Zrsina ~ Xx cosa ~ mg(condcongsina~sinBcora}] +tang (psina~ reona) } SB +{ que Pag CE + Zreosa — Xpsina + :mg(c086 cos cosa + sinBsin.a)| = —Lo(pcosa+ rsina)} 2 ~ (qezgerntamocene) 2 [pS cintoondcora costs] % (8 402) | 1 9D 4 566 & — cosasins 2Xt OF — in asin g 22 Dw 5 fang OP + ons Fe 0 Fe +0008 Fe 0] ap or sina SP — cosa Fe = =yalD sin b+ ¥ cov ~ Xr cos asin 8 + Yr cox ~ Zr sin asin 8 + mg(sind os asin 9+ c05 sin dcos ~ cos eos sin asin) ' { ptr sinarsin- Ze cosa sin 8 + mol sin sina sin 9 = 0080 cos 6 cos asin 8)] + peosa+ sina} 92 ze + ATID cos ~¥ sin 8 ~ Xz cosacos 8 — ¥rsinf ~ Zrsinacos + mg(sind cosa cos ~ cos8 sin sn ~ cost eon dxinacosp)] Se + (cos @ cos ¢c0s 8 + cos@sin psin asin 8) 2 + £(cos6 cos asin # ~ sin @sin bcos + sin Bcosdsin asin 8) 2 (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: (4) ae a FE + sin otane Se + omotano 3 fe + (acosgtand —rsingtand) 5 +(qsing sec? @ + rcosd sec? 4) 2 (C10) 8 Or 6 ir ren (C-11) = singsecd Fh + cosdsece Or + (gos #sec0 ~ rsin gsec8) 2% 7 + (qsin seca d + reondaectan6) oF (C12)Equations (1-72) to (1-74) define the earth-relative velocities of the vehicle. These equations are used to determine the generalized derivatives of these quantities: 2 = [cos cos a sin 8 ~ sin B sin 6086 — cos f sin a cos ¢cos 6] % nos: da V (cos sin asind + cos cos. cos cos) $F . . 7 a6 ~ V(Gin eos asind + con Asin bcos —sinfsin cos $cos8) FP ~ V (cin B c08 60080 — cos Asin asin cos 0) % +V (cos cos cos 8 + sin Asin g sind + cos sin a cos bsin 8) 2 (C13) “2 = [05 8 cos a cos 8 cos ¥ + sin B (sin dsin 8cos y — cos dsin ¥) . in uy OY + cos Bsin a (cos sin 4 cosy + sin dsin #)] ME = V{cos Aisin a cos 8 cas 4p — cos 8 eos & (e08 din 8 cos ¥ + sin sin W)] e — V[sin 8 cos a cos 8 cos ~ cos Bsin sin 8 cos — cos psiny + sin 8 sin a (cos sin @ cos p + sin sin p)] # + V[sin 6 (cos dsin @ cos» + sin g sin) — cos Asin a (sin dsin 8 cos # — cos psin ¥)] ie ~ V[eos 6 cos a sin # cosy — sin B sin dc0s8 cos ¥ ~ cos 8 sin a.cos cos 6 cos ¥] ie = V{[cos 8 cosa cov sin p+ sin f (sind sin Osin ¥ + cos pcos) + cos sina (cos sin BsinY ~ sin feos y)} SE (Ca) @ = [c08 cos a cos 8sin y+ sin f (cos dcos ¥ + sin dsinOsin y) + 08 sin @ (cos sin @ sin th — sin cos w)) Pa — V[cos 6 sin a cos 6 sin ¥ — cos 8 cos @ (cos psin @sin W ~ sin cos p)] 2 — V[sin 8 cos a cos @ sin y — cos 8 (cos cos p + sin dsin sin y) + sin B sin a (cos sin 9 sin p — sin pcos s)] ee 42— V[sin 6 (sin pcos — cos psin @ sin }) + cos 8 sin a (sin dsin O sin v + cos cos ¥)) = ~ V(co0s 8 cosersin Bsin sin Bing cosBsin — cos sinarcosdcos0 sin ¥) FP + V[c0s 8 cos a cos 0 cos ~ sin B (cos din W — sin dsin O cos #) + cos Asin a (cos sin 8 cos + sin dsin ¥)] % (C-15) 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: 1_ axe aD da 0) og. Mee) = [Ge -me B sina SE IL . (Dsina + Leora) 2% x ~ cose =] (c-16) Xena) . 1 [O¥r , OY 28 om sin 7 = ~l# FF + amcor cos S8 — gms sino 3) (cat) 1 [dz aD OL 7 da seal BE — wna $2 — cove SE - (Deoe a — Een) 5 ~am costing 32 — gmsind cos 5 zl (c-8) Vector equation (1-91) defines the output of body axis accelerometers at the vehicle center of gravity and is used to determine the generalized derivatives of the individual body axis accelerometers: Baz) _ 1 [OX aD a : da Haz) gon Pog oo Ge tne SE + (Dsina + Leona) 22] (c-19) 1 (0% | oY : aan (E+ 5e) 1 [0% _ ing OD aL ba : e walt ~ sina FP ~ cosa $2 - (Deosa ~ Lsina) $2] (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 fee = (-% 27 «sine 2 + cosa F 2 z+ (Deosa~ Lsina) =] (c-22) ‘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 43quantities: Oa, 1 op oq or OG oF) GE [love Hr20) GE + Cove ~ Dae) 30+ (ne — rae) Hse BE, ) (c-28 Say _ 1 ~qz,) 2% 124) 24 — (a2, - ry) 2 42, % 2, &) o- TE ~ py [20m ~ 20) BF — ny +959) 52 ry ~ ray) a, 2, 2] (02H da, 1 op og or Op a +. BE > 5 [20% — 720) BE aes ~ 10) BE — (oes tan) FE — ae + 2, BH] (025 Equation (1-96) defines the output of a normal accelerometer aligned with the z body axis but not located at vehicle center of gravity, aq. This equation is used to determine the generalized derivative of anq: Alani) _ Dan ag * 8g a ry ap % 3 [pes rae) FE (n= 10) 92 — (00s +002) 40 a 20) In equations (C-20) to (C-28), 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 equations, as before, the subscripts x, y, and 2 refer to the 2, y, and 2 body axes, respectively, and the symbols z, y, and z refer to x, y, and z body axis locations of the sensors relative to the vehicle center of gravity. U ing equation (1-97), the generalized derivative of the load factor can be defined as an) _ 1 aL LoL cx 3 ~ mg BE C2 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: Se” aomaeiesiay? 9 a) 2180 ie (c-20) Oa oct eae ee cow) ee at Ee oe way Paw Gey (oxy(aoro2ueys— 10 Be + 1AM(1.04 0.20)%p, OYE (rs 1.0) Hee) [120 (<étoa) -16] ) +m (2 au (sage) [eBay] 4} (ar 1.0) 14M(1.0-4 02M) SE (M < 1.0) 5.76M?_)?* a (_s.16M2_\'* {ear (OS M5) +3.0M" (aie Ma) (C-34) [si 9 wee om. 3 Fe roar Se (C35) Om) © 8E = (1.0+0.2M") 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 lightpath-related parameters are presented in equations (1-106) to (1-108). These definitions are used to derive the generalized partial derivatives of the fightpath-related parameters: Be ce Pee ean oe . 5 x (€-37) ” = [-ay4.c05 #6080 -+ a, sin e086] * + [ig 0080 + aya.sin psin 9 + 054 cosine +n St —singcose os - “se (C38) ‘The partial derivatives of altitude rate f 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). 45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.) _ V OV ah 7% ced vo va eared 7 ‘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 (eq. (1-114)) aze resented in terms of the generalized derivatives of the lift and drag forces: aN) _ ok Doe 8a “ag = C08 Ge + Hine GP - (sina ~ Deosa) Fe (c-41) A) __ ab op ina) 2% “ae 7 8G BE teosa Ge - (Leos + Daina) Fe (C-42) The body axis rates are defined in equations (1-9) to (1-11). ‘The time derivatives of these terms are defined in equations (1-115) to (1-117). These equations are used to derive the generalized derivatives of the body axis rates and accelerations: au) vg. ba 26 Be = tacos Ge — Vsin acos FF — V eosasins Fe (C-43) 2) = ing ™ 4 Vc0s 9 28 BE = Hind Ge + V cosh Fe (C-44) . ov. Oa yg OB sin acos 8 Ge + V cos acess 5p —Vsinasin s Fe (C-45) 3 (OXt 65 2D 4 ing DE) yg 20 vein g OF =( Be eG ina Fe) ~ Vsin acoso 2+ V ing + (rsin 8 — qsin cos 8) wet (Dsina + Leosa) ~ gV cos.a.cos a] 22 + (7V cos + 4V covasin 9) 3B soon oe (c-46) 8) 1 (OVE YY svn ene g ar em a (He oe) Visine cons 5E - V cona cos + (Psinacosp — roscoe) FF + (pV conacosp + rVsincxcos9) 22 7 08 ab a8 ~ GV sinasinl rV cosasin ) FF + goons coe 9 Se - asin sin b (c-47) 461 (d%_ 5, OD aL: bp og (GR sin Fe ~ cosa FE) — Vein E+ V con xeon Fe + (qcos cos — pain) Fe — -k (Deosa ~ Lsina) + gV sin acos a z : 08 ing 2% ~ (QV cosrsin + pV cosa) 32 — gost sing 38 — gsinBeoso 7 (C48) ‘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: ts 8 , ta 99_(92a- Pa) BV , Ba Vv ety a 5E -( A i ) OE CC) 29 Op , 29 Or _ (rzp—pz9\ OV , OB 7 -BR+ER- Ce) _ (C50) (+1 sn 008 $080 + 24.in poss) 2f 4 +(24c086+ ynsindsind + 24 cos sind) 3 (ci) [fan sin #cox0 4 cox cot 8) + Oy cox sind — x, sin din 6)] L ; ; ‘ ino] 2 + [asso manda x, c0s$c0s8) + dy, cos sind ~ x, sin dsind)] 57 = (yp cosdbeos0 ~ x, sin cord) 22 + (2, cov 8+ ysin sind + 2, condsind) ah +9 (cs) ‘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), (G-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: © . Lap — Legh ~ ent) 2+ (yn — Hey byt) 324 (Lar = Tesp = Iya) 2 x (C88) cn = Fe sina Fe (psina ~ reosa) (C-54) oe (C-55) Bs cosa Se + (- —pcosa— reine) 5 (C-56) a7APPENDIX D—EVALUATION OF DERIVATIVES ‘The generalized partial derivatives presented in equations (C-1) to (C-56) contain partial derivatives of the state variables, thrust forces, and total aerodynamic forces and moments with respect to the dummy variable & In this appendix, these partial derivatives are defined with respect to specific state, time derivatives of state, and control variables. The derivatives of atmospheric parameters are also discussed. D.1_ Preliminary Evaluation First, the partial derivatives of the state variables with respect to the state, time derivatives of state, and control variables are considered. All partial derivatives of the state variables with respect to the state variables are either equal to zero or unity. Thus, dp _0q_ Or _ OV _da_ 98 06 08 _O8 _Oh_ Oz _ dy _ dp 3g or OV Ga 08 36 30 O0 dh G2 dy™ and all other derivatives of state variables with respect to state variables are equal to zero. The partial derivatives of the state variables with respect to the time derivatives of the state variables (& and 8, particular) are equal to zero. ‘This is also true of the partial derivatives of the state variables with respect to the control variables. 1 (D-1) Segond, the partial derivatives of the aerodynamic forces and moments with respect to the state, time derivatives of state, and control variables are evaluated. Using the definitions of the force and moment coefficients presented in appendix A, the partial derivatives can be explicitly evaluated in terms of the stability and control derivatives. D.1.1 Rolling moment derivatives. z =o, (D2) oe oy, 3) oe, (v4) x = SbpV Cr+ gSbCuy, (p-5) ou = 950Ce (D-6) se = 50C ty (0-7) oe o 5s0v°C, ze + 9S8Co, (D8) oe (v9) A = ee, (D-10) 2 = 9S8Ce, (D-11) 49D.1.2. Pitching moment derivatives. — oM cy GSTCng, D.1.3 Yawing moment derivatives — ON _ gst? dp Cop an Oq an L an ne wr = SbpVCn + FSbCr, Be = 950C og, ye ew FSbC ay -_— |. Fh = ZV" Cn 5p + SbCoy, aN _ 958s “oy One an qs? ‘ap = vm ON 5 = aStC,,, Osi (D112) (D-13) (D-14) (D-18) (D-16) (D-17) (D-18) (D-19) (D-20) (D-21) (D-22) (D-23) (D-24) (D-28) (D-26) (D-27) (D-28) (p29) (0-30) (D-31)D.1.4 Drag force derivatives — So = SeVCp + 95D, = = a8Cp, GF = 180d, OD _ Lope, Oe = 55V%Cb 5p + B8CD, D.1.5 Sideforce derivatives — Sy = SeVCy + 9SCy, aSCy, asc, loyacy 2 4 2 5V7Oy 55 + a5Cy, (D-32) (D-33) (D-34) (D-35) (D-36) (D-37) (D-38) (D-39) (D-40) (p-41) (D-42) (D-43) (D-44) (0-45) (D-46) (D-47) (D-48) (D-49) (D-50) (p51) 51D.1.6 Lift force derivatives. — aL a (052) 2 (D-53) oe (0-54) GE = SpVO + 08C.y (0-55) & 80. (0-56) z (057) 2 (D-58) # (059) iG (0-60) ema (v6) Next, the partial derivatives of the powerplant-induced forces and moments with respect to the state, time derivative of state, and control variables are considered. The partial derivatives of the powerplant- induced forces and moments are assumed to be zero except for moments taken with respect to the body axis rates (p, q, r), moments and forces taken with respect to the velocity and velocity orientation terms (V, a, 8), and forces taken with respect to the control variables. These terms, assumed to be nonzero, are taken as primitives and not evaluated further. Thus, using Fy to represent a powerplant-induced force (Xp, Yr, and Zz) and M, to represent a powerplant-induced moment (Lt, Mr, and Nr), OF, _ OF» OF, _ OF _ Of) _ OF, rial Ory Oole Cel a Ovi Oh ° 2) 2Mp _ OMp _ My _ Mp _ Mp _ OMp os 86 ~ 00 ~ 8) ~ Oh ~ de and OF, Oly OF) OF, OM, OM, OMy Mp OMy |, ,2My WW Da" BB" B," Op” Oy” Br” BV” Ba 9B are taken as primitives and not evaluated further. ‘The final set of partial derivatives to be discussed are the derivatives of atmospheric parameters with respect to the state, time derivative of state, and control variables. In this report, all atmospheric parameters are assumed to be functions of altitude only. Thus, except for OF Op Yn 4 WP Oh? Bh’ aR Oh” all derivatives of ambient temperature, density, viscosity, and ambient pressure are assumed to be equal to zero. The nonzero quantities listed previously are dependent on an atmospheric model. Clancy (1975), Dommasch and others (1967), Etkin (1972), and Gracey (1980) present discussions of atmospheric models. In this report, the quantities will be taken as primitives and not evaluated further. 52D.2 Evaluation of the Derivatives of the Time Derivatives of the State Variables ‘The generalized derivatives of the time derivatives of the state variables are defined in appendix C, equa: tions (C-1) to (C-15). In this section, these generalized derivatives are evaluated in terms of the stability and control derivatives, primative terms, and the state, time derivative of state, and control variables. In this section, the notation 8(4;)/82% is used to represent the more correct notation 0 f:/02; that is employed in the discussion at the beginning of section 3, This notation is used because there is no convenient no- tation available to express these quantities clearly—particularly not the usual notation employed in fight mechanics texts such as Etkin (1972) and McRuer and others (1973). The notation that defines quantities such as Ly = 8(p)/dp and M, = 2()/2q is misleading in this context because the definitions of those terms (such as Zp, My) are based on assumptions of symmetric mass distributions, symmetric aerodynamics, and straight and level flight, and additionally do not include derivatives with respect to atmospheric quantities. D.2.1 Roll acceleration derivatives— a) _ 1 [Ra 10Ct, + 180m, + 1abCn,) + SE 2 Op ~ detl2Vo ~ 2po( Festa ~ Tay) + go(Fosl ~ a = Dela) ~rolayls + Dye ~Iyee)] (0-64) 3) SE, Oly | OMr , ONT We = ar ial (Cty + 180mg + 10Ong) Gar + Gee + GE + po(Dech ~Iyela ~ Dele) + 20lych ~ lyf) ~ Poel = Ieyla + Iesa)] (D-85) 3) Sb 7 OMr , ONr De Tact arg llttCee + 18m + abCn,) + SF 4 OE 4 OE ~ pol lal — Dyla~ Ips) ~ (Deby = Tyla + Ionla)~ 2rllyels ~ Ieela)] (0-88) a 1 BD = AE [sv(ovice + Ce) + 18e(pViCm + Any) “are Oly OMr a + LSU Vn + Amy) + TE + Fy EE 1, ONE (p67) oly 8. ONy’ Far [ICe, + oCne + 8G) +h FE + bp OE 1, | (0-68) 1 Oby oMr Fer [BSUANCY + Helmy + BC) +e Eo tp OA 4 1g OT] (p09) 0 (D-70) (D-71) (D-72) s ne aeet[ be (Freee $2 4 ace.) + 18 (FveCm FE 4 Cra) +10 (frac, 2+ aC) (0-73)Hg 8 14) ae 0 (D-75) @ . weenie, + LitCm, + IsbCna) (D-76) aa man + LC my + Is0Cn,) (D-77) a # Banca, + ns, + 0Cns,) (D-78) D.2.2 Pitch acceleration derivatives— aoa a Penbes, + L4iCmy + IabCp,) + oe +h Tee Is a ~ pe Foals = Ieyls) + deal Hal Bal) is + Iyets)] (0-79) wo Soldat, + 12mg + IghOg) + ot oa Tat, Se +h a + polls ~ Igela ~ Dals) + 260(Iyela ~ tain - cater ~Tegle+ Iesls)] (D-80) @ Punic, + LeiC im, + IebCo,) + I a +1 Mt oe ~ ollagla ~ Dye ~ Ils) ~ Go(Dela~ Ieyla+ lol) ~ 2ollyels~ oes)] (0-81 FD = A [ies6(ovice + aCe,)-+ 18H p¥6Cm + 9Cmy) Obr , , Mr.) ONT) + TsS0(0ViCn + 1Cmy) + FF te Se av ts ay] (D-82) war [astnicr, LGC ng + Ts8Cmg) + Ty EE 4 1, OME 1g el (D.83) Mos X[ascsbCe, + Heng + IsbCng) + la ots +h oe Is a (p-84) so (D-85) @ bal (D-86) @ 0 (087) D = 5 [no(Svece 2 + acu) + 1ae($¥8Cm # + aC) + 13(5v3e, 2 + aC)] (p88) cog! (D-89)(D-90) Teper labCes + eC my + 16bCng) (D-91) a5 Y= weer bce, + 1eiCmy + IsbCng) (D-82) @) © UnbCty, + LatCmg, + Isr MO ‘te, + LaeCmy, + Isl (D-93) ae) 1 abr ON Paste, + Toe my + To0On,) +15 GO + oe = = +h Op Ayls~ Delo) ~ rons + Dy ~ Iyele)] (D-94) Se, 7 Olt OM ON; BV Ce + ToeCmg + TebCng) + Ta Ge + Is at +h at 1 Pol Izzt3 ~ syle) + gol lzels + po(leels ~ Iyels ~ Delo) + ae ae ~ rolDela~ hey + lxto)] (0-95) b a I labCt, + IseCrm, + Leb.) + fa HE + 1, OME 4 1, ONE = a lag + Dyls ~ Iyele) ~ 90(Dels — Ieyls + ek) = 2ro(Iysls — tats)] (D-96) 1 Dir = Farr [SHOVE + Hey) + IsSElVOCm + mv) - oly oMr ONy + IoSU(pVCn + ICny) + Ia GE + Is OE 4 Ie a | (D-97) FaT[ASUHICe, + 6m, + lb) + Io EE +g HE 4 15 AN) (0-98) te Olt |, Mr |, ONr FaeT [PSIG HaeCg Lb ag) + fa EE a 1, OE 1g SA (0-99) 0 (D-100) 0 (D-101) =0 (D-102) s 2, a(t ap al ( V3C SP oP + aCn) + he(2V8Cm 5h + Cun) ap Lya, +10(Svacn 32 ee an)] (D-103) (D-104) (D-105) 55(IsbCey + IséCmg + LobCng) (D-106) 2Vodet I b - Ba det Fey + Is@Cmy + LobCny) (D-107) aS . FE plhates, + IeoCmg, + TebCng,) (0-108) D.2.4 Decoupled roll acceleration derivatives — ae "i ae p+ tyro Jest] (D-109) ae = -[feo. + oe + Tespo + 2Dyato + Polly — 1] (p-10) a al tee. + ot — Teypo + golly — Ie) = Ayer) (D-111) a) = 7 [seeovace +9Cy)+ (13) E (9580. 7 e) (p-113) («sic + FH) (0-114) (pais) (D-116) (D-117) (D-118) (p19) (D-120) (p21) (D-122) (D-123)D.2.5 Decoupled pitch acceleration derivatives. oa) 7 [Sagem + GE 2esPo — Isto + role — (D124) 7 i (Gz y+ EE Eyre = a0) (0-125) = fica 4 ME tt = ta # 2h] (er = £ [Sorin + 90m + FA] (om) = (tseon,+ 22) co 7 F, (#5%0ng + ae) (D-129) =0 (D-130) sf =0 (D-131) (D-132) (D-133) (D-134) (D-135) (D-136) (D-137) (D-138) D.2.6 Decoupled yaw acceleration derivatives — © - as 7 [Bea ONT oteyp0 + ale -1)+ har] (D-139) 7 ae pe? te fy) ~ Bayo ~ Tear] (D-140) (Bo +o + hur) (ou) [58 o¥0C + Cay) + (D142) 87fe 08 Bo Cp, + sin fo Cy,) BE (-cosfo Co, + info Cv.) J [~ 8 08/0 (o¥eC> + a0py) + 8sin a oe +4Cy,) Or + cos. cos o AF 4. sn ag cos bp ZF oe + sin fy OT) a 1S £08 Bo Op, + 48 sn By Cy, + 08 ao 608 fo aes Of + sin ag c08 fo F* + sin Go Mt _ x, sin a9 cos + Zr conc + malsinBo sin a e268 + 280 co concn os )] (D-143) (D-144) (D-145) (D-146) (D-147) (D-148) (D-149) (D-150) (p-151) (D-152) (D-153) (D-154) (D-158) (D-156) (D-157) (D-158)one 1G5(— cos fo Co, + sin So Cr + sin fo Cy, + cos So Cy) 3B ax: a2 ay + £08 an cos Boga" + sin eo 608 Aer + sin Bo = Xr sin Ap co8 ag ~ Ze sin ag sin Ay + Yr cos Bo -+ mag(sin 8 cos ai sin By + C08 Mp sin do C08 bp — c05 69 os é sin ao sin fo) (D-159) av) a Fa a = (0869 cos $y sin By — €08 fo tin dy sin a9 008 fo) (D-160) = 4(~€086o cos x 608 Bo ~ sin Bo sin do sin Bo ~ sin Bo c08 do sin a0 605 Bo) (D-161) (D-162) cos bo (S13 Co 9 + ac ) + sind (Frdey $F + acv,)] (D-163) 0 (5¥8CD FE + ACD, 0 v9 + acy.) - =0 (D-164) W) <6 (0-168) OV) oo FF cos io Co, + sin Bo Cr) (D-166) el ; = FERC cosh Co, + sin Cy,) (p.167) = $5 (— cos 80 Cx +sin fo Cy,,) oY, az, + £ (cosancos Et 35, * in By > 38; + sin a9 cos By ce) (D-168) D.2.8 Angle-o-attack rate derivatives— asb ~ HVE coset ~ tM Bocosao (0-169) ase at ROH +19 (.170) isb ~ aba ‘Ly — tan ig sin ag (D-171) 1 azn aXy ~arent {sc (VopCL + Ay) ~ c08 49 SEF + sin ay SE + LI-GSCL + Zpcosag- Xpsinag ¥

2 + 45Co,) +8 : FSVECLEE + a5Ci,) (-297) 0 (D-298) 0 (1-299) $52 (_ cosa Cp, + sin a9 Cig) (0-300) 2Vogom meat e080 Cp; + sin a0 Cig) (0-301) Fam [P(- c08 20 Co, + sin 29 Cha) + oa) (D-302) D.3.2 Lateral kinematic acceleration derivatives — Alayn) _ _ a0 2Vogam** - . neon (D-304) . nee. (D-305) ew) m (seve +98Cy, + ) (0-306) = (asex. + a) (D307) = (ascv, + a) (D-308) tua) Fp conf c08 bo (D-309) aan sin go (D-310) (0-311) 6735V8Cy SE 2% + 8x) (D-312) (D313) (D-314) € oy, (D-315) Cv, (p-316) Xp (cn) oa D.3.3. Z-body axis kinematic acceleration derivatives. — Aazx) _ __g5b = =H fein a9 Co, + c0800Ci,) (D318) = ~ Fagor tin-an Co, + 6080 Ci) (D-319) ~ fh ana Gp, + €08.0 Ch.) (D-320) Hox) _ “ee [sina (p¥6C> + 4Coy) + $08.0 (0VECL. + Cty) ~ a (0-321) wee 5 Ebvinoe (Co, ~ Ct) + 008 0 (Ci, +Cd)] + = oe (D-322) “ae! = - Keon 9 Cog + £0800 Cig) + vas (0-323) Meet) =~ Z costosin gs (D-324) ew) = ~ Lindo conte (0-325) (D-326) SVCD % + 35Co,) + 00800 (svac. Ha + aci,)| (D-327) (D-328) (D-329) Fogel *i0 0 Co, + 60809 Cig) (0-330) Tha (sinaa Cp, + €08 a0 Cr,) (D-331) = [rsine. Cy, + £0840 Cs.) ~ a (0-332) 68D.3.4_ x body axis accelerometer output derivatives. Ge) ae oo £08 a0 Cp, + sin 20 Cty) (D-333) Heat cote Co, + sin ao CL) (D-334) Hiaen(- core Cp, + sina Ct) (D-335) a [+S cot a0 (eVeCn + dCoy) + Seino (p¥aCL + Cty) + or) (D-336) Fae {A5t- cosa (Co, ~ C1) +n 20 (Cia + Co))+ FF} (0-397) = ed 0 Cy + sin a0 Chg) + ae (D-338) (D-339) (D-340) (D-341) a on saa ~ cor a0 (Fs vdc> Fe a svgo, 2 4 wC.)| (D-342) Ps 48Cp,) +sinao ( (D-343) (D-344) 752_, Cb, + sin a0 Ci) (D-345) Fogem( e080 Co, + sin 90 Cry as Figg (60840 Gp, + in 20 C4) (D-346) 1 [as(— cosa Cp,, + sin ao Cr,,) + 2XF (D-347) tom B,| D.3.5 y body axis accelerometer output derivatives — Hoy) - 456 cy, (D-348) ‘o (D-349) Ye (D-350) (D-351) 1 _, Yr aim (soVore + a80v, + FP) 69D.3.6 (8+ Te) mm (25C 5 +5 38 Be) te) z body axis accelerometer output derivatives — iS! °° (sin ag Co, + €08 a9 Ct.) gom Wogem (*i8 0 Cd, + 608.00 C,) if (sing Cp, + e080 Ci.) om Oz [sino (oViCo + apy) + 8 e080 (0VEC, + 4Chy) = 22) Fan {Alin a0 (Co, ~ C1) + e840 (Che + Co)) = Bah O2y oom 1 Jom [5(sin 20 Cg + 608.49 Cry) ~ (0-352) (D-353) (D-354) (0.355) (0-356) (D-357 (0-358) (D-359) (D-360) (:361) (D-362) (D-363) (D-364) (D-365) (D-366) (D-367) (D-368) (D-369) (p-370) (D-371) (v-372)D.3.7 ase ae (sin ag Cp, + 0840 CLs) ~ gam 1 os) = 7 [gs(sinaa Co, + e080 Cty) ~ (sin ae Co, + c08 a0 Cr) Normal accelerometer output derivatives — (sin. a9 Co, + 60809 C1) in 00 Cog + 608 09 Cia) 5 _(5p, [pose (gmat) a] Sp ~ Mean (gyi) nou {Sz aztaa — 4$P0M(1.0-+ 0.2M2)5 (M $1.0) 2 5.76M?_\'* 9.216 +9000 (FARMS) eee] O42 10 mG (Ms 1.0) (Mf > 1.0) (D-598) (D-556) (D-557) (D-558) (D559) (D-560) (D-561) (0-562) (D-563) (D-564) (D-565) (D-566) (D-567) (D-568) (D-569)D.3.19 Mach meter calibration ratio derivatives — =0 0 0 Ldy(1.0+0.2M7)25 : (fs 1.0) Mae) = | faa (gape) ~1S¥o (1.04 0.2m) Be (M 31.0) Agefvs) _ } _¥0| 7gq2_)** ae =| ~abaw (cdi) +o0ur (SiMe) aha] BE ore 1) Aaelr) _ 4 Pepeeteecte aera) _ (0-570) (571) (0-572) (D-573) (p-574) (p-575) (-576) (ps7) (0-578) (D-579) (-580) (D581) (D582) (D-583) (D-584) 8122 'D.3.20 Total temperature derivatives. op poTo(1-4p0/ p0To)/?, hg V3 — faye |} or oh (0-585) (D-586) (D-587) (D-588) (D-589) (D-590) (p-591) (D-592) (D-593) (D-594) (D-595) (D-596) (D-597) (D-598) (D-599) (D-600) (D601) (0-602) (0-603) (D-604)D.3.22 Flightpath acceleration derivatives — alfea) _ 4 (0-605) (D-606) (D-607) (0-608) (0-609) (D-610) (D611) (p-612) (D613) (D-614) (D615) (D-616) (D-617) (D-618) (D619) (0-620) (D-621) (D-622) (D-623) (D-624) (D-625)84 D.02 “86 oS alfa) (D-627) ap 2a) (D.628) D.5.28 Vertical acceleration derivatives — ath) _ _asb aa Se = itaem tinal cosao Co, + sin ay Cry) — sin do 08 % Cy, + £08 0 €058y (sin a Cp, + 08 a0 Ci,)] (0-629) a cos, Fagaril in Mo (— e080 Co, + sin a9 CL,) ~ sin do os % Cy, + 08 40 00880 (sin ao Cp, + 608 a9 CL,)] (D-630) Fae lain (cosas Op, + sin a0 C,) — sin do cos & Cy, + £08 #9 cos Mp (sin ao Cp, + cos ao Ct,)] (0-631) 1 axe FP = gan (tint [-8 cos 20 (0¥%C + #Cy) - 0 (HVC + 9Cuv) + FE] ~ in fo 08 ( SpViCy + aC ry + 5) + c08 4 0885 sin an (pVECD + aC ov) + Seosao (PVC. + #Ciy)— 924] } (0-632) gam {sito [95 cay (Cp, ~ Cy) + #8 sn 29 (Coa + Od) + oy Ov ~ sin guomay (asc, + 2) : azn + 084 00866 [25 sin a9 (Coy ~ Cx.) + 48 e080 (Cia + Cp) — Fy} (w-sss) i x, ay, uo 2 8 (-18co800 Cog + 98 sin a0 Cry + or) ~ sin cord (ascy, + BF) + eosdocosBo(48sina0 Gp, +98 cosan Ci, + 92t)] (0-034) = ~typ £08 6 C088 + a sin bp c08 Bp (0-638) ah) 2D 05, con + ay sin sin by day €08 4y sin Op (D636) oh) | 637) aero (D-637): ie i) 1 Lsvacp 2? 4 4s sin ag (5SV8CL +950. oh Fan { sin l0[-c0sa0 (}svacn 32 +45Cp,) + sinao (55vaC. $F I a Ls ~ sin da cos to (58VaCv SP + acy.) ae ++ c08 dy condos ag (SVCD SE + 45Cp,) i oe D038 + eovae (svi. 36-4 a5C..)] } ao ot) _y (D-639) on a) _ (0-640) oy OS ee aC ey ec. 1% ~ Wegom ; : +08 Go €08 8p (sin ao Co, + c08 a0 CL, )} (0-641) 2) 954 int (~ cos.ao Co, + sin ao Ci,)~ sn by cos Cy, 8 ago’ * ’ : 4 cosgacosdp (sin ao Co, + 60800 Cr (04) aXy fee OYy Fe = af sine [a8(—cosan Go, + snow 0.) + SA] — sin dont (48%, + 9) 06; gor + aay (D-643) + 0840 0 lo [25(sin 00 Co, + e080 Ci,,) - 34] } (D-643) D.8.24 Specific energy deri (644) (D-645) (D-646) (0-647) (0-648) (D-649) (-650) (D651) (D-652) (D-653) 8586 D.3.25 Specific power derivatives.— (D-654) (D-655) (D-656) (0-657) (D-658) (D659) (D-660) (D-661) (D-662) (D-663) (D664) (D-665) (0-666) (0-667) (D-668) (D-669) (D-670) (p67) (D-672) (D-673) (D.074)as Vo $32 coag Ci, +sine0Co,) 58 cosay ., + sino CD, £08.40 Cty + sin a0 Cp,) B30 cosa CL, + sin BY (008 19 Cl + Sin G0 Cp, ) = Sleoscas (pVoCt. + Gry) + sin a9 (pVCD + @Cdy)] §5(c08.40 Cg + sin a9 Cp, ~ sin ap Ci, + c08 a9 Cp) 1S(co8.a9 Cp + sina Cg) 5 [soran (308 36 + aCuy) + sina (Svc 2 + aCo,)] c08 49 Cy + sin a9 Co.) QV a5 (e080 Ci, + sin 9 Cp,) 45(c0800 Ct, + sin a0 C,,) b Fpl 8820 Cy + £08.00 Co,) Se, 2Va ~ sin ag Ci, + £08.49 Co,) Sb Fig sina CL, + 00809 Cp,) (0-675) (p.676) (p67) (D678) (D679) (D-680) (D-681) (D-682) (D-683) (D-684) (0-685) (D-686) (0-687) (D-688) (0-089) (D-690) (D-691) (D-692) (D-693) 87Sor = Sl- sin aa (pVeCL + FCiy) + c08 a0 (pVeCd + Coy )] = G5(~ sino CL, + 608 a9 Cp, — €08 a9 Cy ~ sin ag Cp) PAD) = 95(—sinag Cry + 608.09 C4) =0 =0 =0 Luan, Oe yap 0, on = 8 [-sinao (Fvdcr 3? + Ci) +0080 (jvéco ant aC.) =0 =0 gst, +, Fpl sit 20 Cog + 60880 Co) = $e sina Gr, + eos00 C0,) = 45(~sinag Ct, + 60840 C,,) €08 49 608 Ho Yo sin a9 608 Yo c05 a9 sin lo (D-694) (D-695) (D-696) (D-697) (D-698) (p-699) (D-700) (D-701) (D-702) (D-703) (D-704) (-708) (D-706) (D-707) (D-708) (D-709) (p-710) (p-711) (D-712) (D-713) (D-714)uO} oq uO} or aw) wv (0) da Hv) =0 =0 = sin By =0 = Vocos fo (p-715) (p-716) (p-717) (p-718) (D-719) (D-720) (p-721) (0-722) (D-723) (p-724) (p-725) (D-726) (D-727) (D-728) (D-729) (D-730) (p-731) (D-732) (D-733) (D-734) (0-735) 8990 D.3.30 D.3.31 2 body axis rate derivatives — Far 7 Yorosa0c08 Vo sin a sin ip x body axis acceleration derivatives. — aa) ai = cos a9 Cp, + sin ay C,) Se 72 (— cos a0 Cy + sin a9 CL,) ~ Vosin a9 8 fy (—cosa9 Co, + sina Ct.) + Vasin fo Bom 1 a {SI- co8a0 (oVeCD + 4p) + si. (pVECL. + GCLy)]+ onl +rosin Bo ~ go sin a9 c08 Bo (D-736) (D737) (D-738) (0-739) (D-740) (D-741) (D-742) (D-743) (D-744) (p-745) (D-746) (p-747) (p-748) (D-749) (D-750) (D-751) (p-752) (D-753) (D-754)D.3.32 ox. 08.0 Op, + sin ao Ch, + sin ag Co + cos a9 Ci) + F*| = qiV¥o.608 9 £08 By ax, =i [fst-covn Cog + sin 20 Cry) + Ga + roVoc0s Aa + qoVa 08 ao sin Ap ° =9 £08 Lge 88+ 40.) +sinan (342, 28+ C)] = eoseg (Svc 28+ 400) non (388, 22+ aC © gin ° Se = FS (- cose Cp, + sina Cra) 5b Wal £2240 Op5 + 8in a0 Cis) ae ri OX 25(— cos a0 Coy, + sin 0 Cia.) + Fe y body axis acceleration derivatives — OO) Ste 2Vom Se 2Vom"¥* Cy, + Vosin a9 608 8p oy, - Vocos a0 os 60 [StoVecy + a04y) + sa + posin a9 08 fy ~ 708 a 608 By © (asc, +22) + moesageni+ etasiao & (ase, + FE) ~ avosin aosin By ~ ro¥o cos 0 sin Ao = 908 €9co5 $0 =gsindosin do (D-755) (D-156) (D-787) (p-788) (0-759) (-760) (p-761) (0-762) (0-763) (v.16) (D-765) (0-766) (v-167) (0-168) (0-769) (0-770) (D-71) (D-772) (0-773) (D-774) m11 » = 5 (ev tin) a oe D.3.33 2 body axis acceleration derivatives — aSb 2Vam' Se fom! 2Vom wv + qo c08 a9 cos 3p — po sin Bo 1 ; : aan — sin ao Cp, — ¢08 a0 CL, — C08 a9 Cp + sinay Ch) + F] = GoVosin a9 ¢08 Bo iG i [25(~sinay Cb, — cosa CL) + a) = qoVo cos 09 sin By ~ poVo 608 Bo 9608 60 sin do —g sin 8 cos bo =0 S 2Vom 2 [-ssin a0 (pVeCb + GCoy) ~ S cos a0 (pVoCi. + FCuy) + (sin a9 Cp, ~ 608 09 Cty) — Vosin Bo (—sin ag Cp, ~ c0s a9 C14) + Vo cos a cos Bo (sin a0 Cp, ~ 608 a0 Ct,) 1) sna, (1v20, 2 + 0 2 6 0 sin ag ( VaCp ant Cos) ~ conan (Fee. wt 1.) al sin a9 Cog ~ c08.20 Ci.) (~sin ap Cp, ~ e089 C,,) (D-775) (D-781) (D-782) (D-783) (D-784) (D-785) (D-786) (p-787) (D-788) (D-789) (D-790) (D-791) (D-792) (D-793) (D-794)D.3.34 Angle-of-attack sensor output derivatives — D.3.35 (8) 06; 7 os 02; }75(— sin a9 Cp,, ~ 608 a0 Cy.) + ti Angle-of-sideslip sensor (0-795) (0-796) (0-797) (D-798) (D-799) (0-800) (0-801) (0-802) (0-803) (-804) (0-805) (D-806) (-807) (D-808) (0-809) (D-810) (D811) (D-812) (D-813) 934 Bs) _ _r0z8- pore roe — Pozo Vo Uh c08 do cos Mp + 24 sin gg c0s Bo = 21,008 89 + yp sin do sin 8 + 24 c0s dy sin By (D814) (D815) (D-816) (D-817) (D-818) (D-819) (D-820) (D821) (D-822) (D-823) (D-824) (D-825) (D-826) (p.827) (D-828) (D-829) (D-830) (D-831) (D-832) (D-833) (0-834)D.3.37 Altitude rate sensor output derivatives— (hs) op =0 (yj, sin do cos Oo + 2, COS do C08 80) + (yp, cos do sin Bo — 2; sin go sin Bo) (zy, sin 8 — yp, sin dy €08 A — 24,03 do C08 Mo) + dy; cos do sin 6 — 2, sin gy sin Ao) (D-835) (D-836) (D-837) (D-838) (D-839) (D-840) (D-841) (D-842) (D-843) (D844) (D-845) (D-846) (D-847) (D-848) (D-849) (D-850) (D-851) (D-852) (D-853) (D-854) 95,D.3.39 Stability axis roll rate derivatives — Se) = —pysinag + ro cosay (D-874) (D-875) (D-876) (D-877) (D878) (D-879) (D-880) (D-881) (D-882) (D-883) (D-884) (D-885) (D-886) (D-887) (D-888) (D889) (D-890) (D-891) (0-892) (D-893) (D-894) 9798 D.3.41 ee (4) a0 ~° 34.) _ ce Stability axis yaw rate derivatives — Py 608 a9 — To sin ao (D-895) (D806) (D-897) (D-898) (0-899) (D-900) (D-901) (D-902) (D-903) (D-904) (D-905) (0-906) (D-907) (D-908) (D-909) (D-910) (D-911) (p-912) (D-913) (D-914) (D915)(D-916) (D-917) (D-918)REFERENCES Clancy, L.J.: Aerodynamics. John Wiley & Sons, New York, 1975. Diewdonne, James E.: Description of « Computer Program and Numerical Technique for Developing Linear Perturbation Models From Nonlinear Systems Simulations. NASA TM-78710, 1978. Dommasch, Daniel 0.; Sherby, Sydney $.; and Connolly, Thomas F.: Airplane Aerodynamics. Pitman Publishing Company, Marshfield, Massachusetts, 1967. Duke, Eugene L.; Patterson, Brian P.; and Antoniewicz, Robert F.: User's Manual for LINEAR, A FORTRAN Program to Derive Linear Aircraft Models. NASA TP-2768, 1987. Edwards, John E: A FORTRAN Program for the Analysis of Linear Continuous and Sampled-Data Systems. NASA ‘TM X-56038, 1976. Etkin, Bernard: Dynamics of Atmospheric Flight. John Wiley & Sons, New York, 1972. Gainer, Thomas G.; and Hoffman, Sherwood: Summary of Transformation Equations and Equations of Motion Used in Free-Flight and Wind-Tunnel Data Reduction and Analysis. NASA SP-3070, 1972. Gracey, William: Measurement of Aircraft Speed and Altitude. NASA RP-1046, 1980. Kwakernaak, Huibert; and Sivan, Raphael: Linear Optimal Control Systéms. John Wiley & Sons, New York, 1973 Maine, Richard E.; and lif, Kenneth W.: User’s Manual for MMLES, a General FORTRAN Program for Mazimum Likelihood Parameter Estimation. NASA TP-1563, 1980. Maine, Richard E.; and Iliff, Kenneth W.: Application of Parameter Estimation to Aircraft Stability and Control—The Output-Error Approach. NASA RP-1168, 1986. ‘McRuer, Duane; Ashkenas, Irving; and Graham, Dunstan: Aireraft Dynamics and Automatic Control. Princeton University Press, Princeton, New Jersey, 1973. Northrop Aircraft, Ine.: Dynamics of the Airframe. Bureau of Aeronautics, Navy Department, Report AE- 61-4 II, Sept. 1952. Thelander, J.A.: Aircraft Motion Analysis. Air Force Flight Dynamics Laboratory, FDL~TDR-64-70, 1965. 101Report Documentation Page Eee 17 Report Re Government Reseason Wa Recents Catog Na. NASA RP-1207 “7 Tie and Subiie Ropar One Derivation and Definition of a Linear Aircraft Model August 1988 © Foriorming Oraarinnton Code 7 aor Eugene L. Duke, Robert F, Antoniewicr, and Keith D. Krambeer 3 Pororming Organization Roper Wo, #1391 6, Wok al Wo. ‘3. Perforing Orpariaton Nate and Address NASA Ames Research Center Dryden Flight Research Facility P.O. Box 273, Edwards, CA 93523-5000 RTOP 5065-66-11 TH Contest or Gant No, TH Type of Repor and Period Covered TE. Sponsoring Ranney Nave and Adarose National Aeronautics and Space Administration ‘Washington, DC 20546 Reference Publication TH Sponeotng Agency Code TE, Soppameniary Nowe Te Aras ‘This report documents the derivation and definition ofa linear aircraft model for a rigid aireraft of constant ‘mass flying over a flat, nonrotating earth. The derivation makes no assumptions of reference trajectory or vehicle symmetry. ‘The linear system equations are derived and evaluated along a general trajectory and include both aircraft dynamics and observation variables, 17. Key Words (Suggested by Autboriali Aircraft models Flight controls Flight dynamice 16 Danian Satenont Unclassified — Unlimited Linear models Subject category 08 1 Seca Gas Tota vapor BH. Seculy Casa. Tota pan) [a Wo: of epee] 2. Price Unclassified Unclassified 108 A0s NASA FORM 1625 OCT & “For sale by the National Technical Information Service, Springfield, VA 22161-2171. INASALangley, 1968