-—lUM ES iT lor : Chapter 7 Aircraft Equations of Motion 74. Introduction e process inv olved in carrying out any engineering analysis can be divided into two ahases. problem formulation and problem solution. The formulation phase requires tanding the physics of an engineering problem and converting that physics into a ent set of mathematical equations. The solution phase involves applying the rules ‘ematics and/or numerical methods to obtain a solution to the system of equations developed in the formulation phase. Both phases are equally important. A of equations is of little direct use in engineering analysis unless it can be y solved. Likewise, a mathematical solution is of no use unless the formulation ~ which it came provides a reasonable description for the physics of the original ng problem. wa n Chapters 3 through 6 of this text, the formulation and solution for many flight tal principles of Newtonian mechanics problems were obtained starting from fundament In each case we started with the physics of a particular engineering problem all appropriate simplifications as the governing mathematical equations were loped. This approach minimizes mathematical complexity and emphasizes ding the physics of each engineering proble1 simple problems, like those associated wil more complex problems, such as the analysis of tums and spins. However, as the has probably observed, there is a great deal of repetition involved in formulating but related problems in this manner. For most students, this repetition and the n from the simple to more complex are helpful in building an understanding of 1m. This approach also allows us to ith steady level flight, and progress gression understanding of the physics associated with fh that can be used for the formulation of starting with a very general formulation and obtaining the problems by direct mathematical reduction. For example, hanics with analysis of the steady coordinated turn, ‘could have been obtained from the more general ulation by simply setting the climb angle and bank . angle to zero. The ntage of this method is reduced formulation time once gineering problems, that is, mulation for more restricted we begun our study of flight mec! mulation for steady level flight r the more general fo ons beta developed. There is a less ‘obvious disadvantage. No formulation een completely general, Human limitations being what they are, we ‘endeney to forget the restrictions that apply to a “general” formulation, S dant engineers have used a so-called “ formulation” for a proble >not apply. Nevertheless, if the C “sing this method of form ulation © U 'n this chapter we ¢ a °f motion, Since the 'n flight mechani it in16 Chapter 7 Aircraft Equations of Motion cod ly throughout Chapters 3 through 6. This chapter se more or less independent of the presentation in Chapters 3 through 6. Thus. » is already familiar with the principles of stan « chapter to begin a study of aircraft dynamics, start with t motion for a physical system can be formulated must be chosen. Furthermore, this choice of c ofound effect on the complexity of the ensuing r . hus, before developing the more general rigid-body flight, it is worth spending some time considering +s of the coordinate systems that could be used. Here we a 25 of some common coordinate systems and settle > systems to be used in our development of the aircraft gins with the consideration of two-dimensional r sirfoils and long cylinders. When formulating any two- it is conventional to use an x-y coordinate Jow over an airfoil, it was quite convenient to have the x-axis in the general direction of flow with the origin at the T vas quite naturally chosen normal to the x-axis in the upward ngle of attack to be positive with the k in this direction increases the lift k was established, an coordinates, to choose the an reasing the angle of atta sn convention for angle of attac Ince the g 4 the pitching rate were naturally defined to be positive im reasing angle of attack rodynamics is extended to include three-dimensional flow the original x-y coordinate system and choosing @ requires the z-axis to be pointing in the spanwise hown in Fig, 7.1.1. Here, this coordinate system is ” rdinates. When using aerodynamic coordinates _it is ty << Figure 7.1.1. The aerodynamic coordinate system.l(c lC 7A. Introduction 717 _ anal to choose the origin to be at the leading edge of the object of study, in this von oe of an airplane, with the x-axis lying along a somewhat arbitrary fuselage > line. _ ovo primary objections to continuing with aerodynamic coordinates as the ginate system to be used in our study of aircraft dynamics. First, because the ic coordinate system is fixed to the aircraft, the position and orientation of the wy cannot be conveniently described in terms of aerodynamic coordinates. Second. convention for_the moment and angular velocity components used with jc coordinates does not follow the conventional right-hand rule. scribe the position and orientation of an aircraft relative to the Earth, we gly use a coordinate system fixed to the Earth, Position relative to the Earth is ribed in terms of latitude, longitude, and altitude above mean sea level because this coordinate system is not a Cartesian system, it introduces essary complexities to the dynamic formulation. Since the radius of the Earth is ree compared to the distance traveled by an airplane in the relatively short time jeriods that characterize aircraft motion, the formulation can be greatly simplified prusing 2 C artesian coordinate system fixed to the Earth. For our purposes here, gach 2 Ci artesian system can be considered to be an inertial coordinate system. Of course, this so-called “flat-Earth” approximatio vvction of an aircraft over long distances, While an Earth-fixed coordinate system allows us to conveniently describe aircraft and orientation, the components of the inertia tensor become time dependent in xed coordinates. The complexity associated with carrying a time-dependent nsor in the equations of motion is almost prohibitive. The mathematical cription of the inertia tensor can be greatly simplified by formulating the angular m equation in terms of a coordinate system that is fixed to the aircraft with the cigin located at the aircraft's center of gravity. rodynamic forces and moments acting on an aircraft depend not on the of the aircraft relative to the ground, but rather, on the velocity relative to the ding air. Thus, these forces and moments are most conveniently described in tems of a coordinate system fixed to the atmosphere. The atmosphere can have motion relative to the Earth, ie., wind. In our development of the aircraft equations of motion we shall consider the atmosphere in the immediate vicinity of the aircraft ‘ be moving at a constant velocity relative to the Earth-fixed coordinate system. ‘hus, within this approximation, an atmosphere-fixed coordinate system.can also be _ an inertial coordinate system,, Considered to be. in summary, position is not appropriate for tracking the Systems will all be dsirable to define the Moment and angular vé Earth-fixed and body-fi te shown in Fig, 7.1 “a718 Chapter 7 Aircraft Equations of Motion J “ Figure 7.1.2. © body-fixed coordinate system a si entions for roll, pitch, and yaw jow the right-hand rule relative to the x)- dd z,-axes, respectively. In Chapters 1 ro 6.2 ce aft c was denoted as e aerod: coordinates i that in fixed coordinates / is simp! b For the Earth- normal to the loca xed coordinate system sh 7.12 plane is and the yy-axis anded Cartesian s designated (x,, y;, =) and is often student may find a downward-pointing: tional, when the aircraft heading or azimuth angles described in terms of a conventional right-handed angular rotation abo is heading angles conform to conventional compass headings, that is, a heading angle: measused clockwise from north with east being 90 degrees, south being 180 west being 270 degrees, and north being either 0 or 360 degrees. ravitational vector with the pointing east. The stem z-axis then points dow In the present text, this coordinate system referred to simpl: as fixed coordinates. While axis to be somewhat unconyC—O 7A. Introduction 719 umosphere-fixed coordinate system is denoted (x,, y,, z,) and defined such N prec axes are always parallel to those of the Earth-fixed coordinate system ver, the atmosphere-fixed system moves at a constant velocity relative to the ‘xed system. The atmosphere-fixed and Earth-fixed systems are defined to be saa at time equal to zero. The atmosphere-fixed coordinate system will be used emporarily, in the development of the aircraft equations of motion. The final » ons will be expressed in terms of only Earth-fixed and body-fixed coordinates. apne vector Ve Will denote the velocity of the body-fixed coordinate frame relative arth-fixed coordinate frame. The magnitude of this velocity vector is commonly j the ground speed, which is simply the speed of the aircraft relative to the ground. " elocity of the body-fixed coordinate frame relative to the atmosphere-fixed ¢ frame is denoted here simply as V, and its magnitude is commonly termed need. Ground speed and airspeed are related through the simple vector equation V, = V+V, (7.1.1) V,, is the velocity of the atmosphere relative to the Earth, which is commonly wind. ne aerodynamic forces and moments acting on an aircraft could be expressed as of the translational and rotational velocity components relative to the g air. However, in the study of aerodynamic forces and moments, it is more t to express the translational dependence in terms of total airspeed, angle of nd sideslip angle. Of course, the velocity vector relative to the surrounding air ivalently described either in terms of three velocity components or in terms of tude and any two independent aerodynamic angles. The total airspeed is readily ed in terms of the body-fixed components according to the relation Ve Ve ty. (7.1.2) 4s standard convention, angle of attack has always traditionally been defined according ‘o the mathematical relation Unfortunately, there are two d “© sometimes used in the d In Chapter 5, the side fore Were estimated from the |720 Chapter 7 Aircraft Equations of Motion Bo = (7.1.4) Here the subscript @ is used with this definition of sideslip angle because the definition is ‘Most convenient for the analytical estimation of aerodynamic forces and moments. For the experimental determination of aerodynamic forces and moments, another Sefinition of sideslip angle is used almost exclusively. This definition comes about as a result of the way in which data are gathered in a wind tunnel. Typically, the rotationa| mechanism used to adjust the angle of attack is mounted on a turntable, which is used to set the sideslip angle. When this gimbal mechanism is used to orient a model in a wind tunnel. one of two gimbal angles 7s exactly the angle of attack as defined by Eq. (7.1.3), However, the other gimbal angle is not that defined by Eq. (7.1.4). The angle set by the turntable of such a mechanism is related to the body-fixed velocity components by Be = sin” ys C3 where the subscript e is used here to indicate the definition that is most convenient for experimental determination of aerodynamic forces and moments. The definitions given by Egs. (7.1.4) and (7.1.5) are identical only when the angle of attack is zero. Of course, these two definitions are essentially equivalent for small angles of attack, but as the angle of attack becomes large, the difference becomes significant. Clearly, both the aerodynamic angle defined by Eq. (7.1.4) and that defined by Eg. (7.1.5) are independent of the angle of attack defined by Eq. (7.1.3). Thus, either of these two sideslip definitions could be combined with the angle of attack and the total airspeed to uniquely describe the velocity vector relative to the air. However, the 4 aerodynamic angle defined by Eq. (7.1.4) has the same geometric relationship to a vertical surface as the angle of attack has to a horizontal surface. This cannot be said of — the aerodynamic angle defined by Eq. (7.1.5). Most textbooks define the sideslip only in terms of Eq. (7.1.5), but for analytical predictions this angle is often used as if were defined according to Eq. (7.1.4). The sideslip angle defined by Eq. (7.1.5) depends on the z,-velocity compone! while the angle of attack as defined from Eq. (7.1.3) does not depend on the yp-veloci component. This asymmetry in the two definitions is not at all attractive from a phy point of view, especially when considering large angles of attack. For example, an aircraft that is symmetric about both the x,-y) and the x)-z, planes, such as a roel with four identical tail fins. Let the velocity of this aircraft be such that the x5-, Yo | z-components of velocity are all equal. From Eqs. (7.1.3), (7.1.4), and (7.1.5) we | a=45°, fi,=45°, and f.=35.26°. If the sideslip angle were used to obtain the sit in the same way that the angle of attack is used to obtain the normal aerodynamic angles defined by Eqs, (7.1.3) and (7.1.5) would result in nic loading produced by this completely symmetric flight / not consistent with reality.OO i —E 7.4. Introduction 727 reason commonly given for defining the sideslip angle according to Eq. (7.1.5) 1 this definition gives a sidestip angle that is completely independent of the * ration of the x-axis within the x, 2s plane of symmetry. Clearly, the side force not depend on our choice for the direction of the xj-axis within this plane wever, the sweep angle that any vertical surface makes with the z;-axis does depend S orientation. Thus, the sideslip angle should also vary with the x-axis acon, such that the combined effects of sweep and sideslip make the airflow to a vertical surface independent of the x-axis orientation within the x)-zs plane The definition of sideslip angle that is given by Eq. (7.1.4) satisfies this nent, while that given by Eq. (7.1.5) does not. Thus, the definition in Eq. (7.1.5) rly as convenient for analytical prediction as that in Eq. (7.1.4). 5 her of these two definitions for sidestip angle may be used in a formulation of ¢ ngid-body equations of motion for aircraft dynamics. The important thing to remember is to be consistent. Whenever a relation for some particular aerodynamic Ticient has been obtained using one definition of sideslip angle, it should be applied at same definition. It is easy to violate this requirement unintentionally if some sdynamic coefficients are estimated analytically from airfoil section properties, while s are taken from experimental data. During most of the flight of a conventional airplane the aerodynamic angles remain ly small and the difference between the two definitions for sideslip angle is significant. However, for flight conditions that may involve large aerodynamic angles. care should be taken to avoid the common pitfall of using one definition for sideslip angle in a relation that was developed using the other. The transformation equations given by Eqs. (7.1.2) through (7.1.4) can readily be d to yield an inverse transformation, which can be used to compute the three xed velocity components from known values of V, a and #,. We start by solving 4 (7.1.2) for the x,-velocity component. This gives ve = 7 va (7.1.6) Solving Eq. (7.1.3) for the zs-velocity component yields Ve, = Vy tana = V,, S06 and solving Eq, (7.1.4) for!a «a 722_ Chapter? Aircraft Equations of Motion Solving 6g. (7.1.9) for the x-component of velocity and using the result in Eqs. (7.1.7) ‘and (7.1.8), we obtain the inverse transformation equations | ; ; 1 cosa cos fy Vg Oe = (7.1.10) Vi+ tana +tan?Z, ql-sin ‘a sin’ By " tan 8, cos@ sin By poe — =) ny yil+ tanta +tan* Bo yl-sin°a@ sin” By tan@ Vv sina cos Ba eee yp_ ee 7. yistan2a+tan2Z, —_1-sin2a sin2B, a ‘The relations on the far right of Eqs. (7.1.10) through (7.1.12) are preferred for large-angle computations where either of the aerodynamic angles may pass through 90 degrees. If Eq. (7.1.5) is used for the sideslip angle, the inverse transformation is V., = Vcosacos Be (7.1.13) Vy, = Vsin Be (7.1.14) Va, = Vsinacos Be (7.1.15) ce Notice the symmetry that exists between Eqs. (7.1.11) and (7.1.12). There is no such symmetry between Eqs. (7.1.14) and (7.1.15). ’ EXAMPLE 7.1.1. A rocket is flying at an airspeed of 1,000 f/sec. The angle of attack is 30 degrees and the sideslip angle, ,, is 20 degrees. Determine the axial, sideslip, and normal velocity components. Solution. From Egs. (7.1.10), (7.1.11), and (7.1.12), the velocity components are ea v a 1,000 f/sec e q{i+tan2a +tan2Z, . {1+ (0.577350)? + (0.363970)? ae 5 = 825.96 fl/sec Vv, Vtan Ba 1,000 ft/sec(0.363970) \ a = ne 1+ tan?a +tan? Byoe 7.1. Introduction 723 EXAMPLE 7.1.2. A rocket is flying at an airspeed of 1,000 ft/sec. The angle of attack is 30 degrees and the sideslip angle, 8, is 20 degrees. Determine the axial, sideslip, and normal velocity components. Solution. From Eqs. (7.1.13), (7.1.14), and (7.1.15), the three body-fixed velocity components are - Vn = Veosacos B. = 1,000 fi/sec(0.866025)(0.939693) = 813.80 f’sec | Vy, = Vsin Be = 1,000 f/sec(0.342020) = 342.02 fsec V2, = Vsinacos B. = 1,000 fi/sec(0.500000)(0.939693) = 469.85 f’sec In certain situations it may be necessary to estimate some aerodynamic coefficients using airfoil section properties and the definition given by Eq. (7.1.4), while others may ken from experimental results that were obtained using the definition in Eq. (7.1.5). In such cases, it is convenient to be able to transform an aerodynamic derivative with respect to sideslip from one definition of sideslip angle to the other. The mathematical relation between f, and f can be readily established by using Eqs. (7.1.13) and (7.1.14) in Eg. (7.1.4). This results in onyp =e Cae, cosa Now consider some aerodynamic coefficient, C, that is somehow known as a function of V. a, and £.. Suppose that we also know the partial derivative of C with respect to & at constant Vand @. The partial derivative of C with respect to #, at constant V and @ can then be expressed as 6C_ _ OC ope Of. Of. OPa (AAT) shown that After applying the tri eliminate tan fe,————— ———————— —— 724 Chapter 7 Aircraft Equations of Motion a cosa a a ae AA Ba cos* py +cos*asin® By OBe % Yn a similar manner, it can be shown that 6c cosa ac eeaesect sar 1.1, fe sin? 8, +cos*acos?f, Ba 20) EXAMPLE 7.1.3. An airplane model is mounted in a wind tunnel with the angle of attack set at 12.0 degrees and the sideslip angle, /,, set to 8.0 degrees, By varying £, slightly to both sides from this orientation, the change in the yawing moment coefficient with respect to f. is estimated to be 0.250. Find the angle, f,, and estimate the change in the yawing moment coefficient with Tespect to f,, Solution. From Eq. (7.1.16), if tan 2, y(ossansa ie = tan anf) _ , iH) = 01 = 82° Br ” ae e 0.978148) 02a Using this in Eq. (7.1.19) yields - cosa oC. Ba cos”, +cos*asin?B, OB. ss: 0.978148 = __.250 = 0.245 (0.989835) + (0.978148)? (0.142220) EXAMPLE 7.1.4. An airplane model is mounted in a wind tunnel with the angle of attack set at 40 degrees and the sideslip angle, Pr, Set to 30 degrees. By varying f slightly to both sides from this orientation, the change in the yawing moment coefficient with respect to f, is estimated to be 0.250. Find the angle,’ f,, and estimate the change in the yawing moment coefficient with respect to fi. Solution. From Eq. (7.1.16), -a( tan Be -1( 0.577350 0 fon, \ 0.577250 ) .. 0.645860 mie Beem ien (#4) jh Sree 0.645850 eg Using this in Eq. (7.1.19) yields a aa Pa cos? a + cos*asin® p, Pe 0.766044 = Parner pner sa 0.250 = (0.798588)’ + (0.766044)? (0.601878) {I ID li nS In7A. Introduction 725 From the results of Example 7.1.3 it can be seen that at angles of attack below hore there is no significant difference between our two definitions for sideslip angle a ior the relatively high angles used in Example 7.1.4, the difference between C te = Cu 's only about 10 percent. This should not be taken as a general license for care - 7 with regard to sideslip angle definition, There are important applications where a ¢ attack can approach or even exceed 90 degrees. For such cases, the difference be : the definitions in Eqs. (7.1.4) and (7.1.5) is quite large. For instance, when the angle anack i 80 degrees, a value for , of 30 degrees corresponds to a value of J, ao Spproximately 73 degrees. Nevertheless, in the vast majority of all flight, the angle o' attack remains small. For analysis of such flight the subscript on f will be dropped and no distinction will be made. 2. Newton's Second Law for Rigid-Body Dynamics The foundation for the formulation of most problems in flight mechanics is Newton’s second law (often stated as F=ma). Newton's second law applied to all points in a rigid body requires that the summation of forces must equal the time rate of change of wanslati aentum, and the summation of moments must equal the time rate of change of angular momentum. ‘Application of this principle is not always as simple as it sounds For the aoe formulation, we will be using a body-fixed coordinate system and allowing both translational and angular acceleration. At this point you may wish to review the chapter on accelerating coordinate systems in your introductory dynamics textbook. Newton’s second law for rigid-body motion can be expressed in the following vector equations, which are written in terms of a body-fixed coordinate system (see, for example, Hibbeler 1998 or Beer and Johnston 1997), Fs+W = 4 (mv) + ox(mv) 72.1) Ms = A (itlo)+ ox((Io) where Fs is the net surface force vector, W is the wei ight vector, My is the net mo vector about the body-fixed origin (i.e., the CG), mis the mass, [1] is the inerti Jay Lay, MM = |-J,, Tim ly In ~Ta, Im - Ip, Fan = Sffloi +23 yam " Joy = Sfflap + 22am h i ae.726 Chapter 7 Aircraft Equations of Motion Ia, = [ffi yp am (7.2.6) ion, = ly, © Jf fsoream (7.2.7) Tye, = Ie, = [flyozedm (7.2.8) Ty = Tey = [ffeozedn (7.2.9) and Vand @ are, respectively, the trai tional and rotational velocities of the body- fixed coordinate frame relative to some inertial reference frame. For convenience in describing the aerodynamic forces and moments, the atmosphere-fixed coordinate system is selected as our inertial reference frame. Mass is a scalar, so the second term on the right-hand side of Eq. (7.2.1) is written ix, iy, th, Oy Ve, —O2V yy @x(mV) = mloy ©, @,| = maz us — OV, (7.2.10) Vn Wey OAV) OMY, Equation (7.2.10) can be used in Eq. (7.2.1) to yield the body-fixed Cartesian components of the translational momentum equation, ( (Vz, (es +Wy, —m@y,V2, +mozVy, 4 mV, | = Fs, + Wy — me Vg +mOr,V ey (7.2.11) | |v, Fs, $W ey — MOV yy +MOy Vay The total surface force, Fs, is simply the vector sum of all pressure and shear forces acting on the surface of the control volume enclosing the aircraft. Aircraft mass typically changes with time as a result of fuel burn. Furthermore, fuel weight can be a very significant fraction of total aircraft weight, and the mass of an aircraft can change very rapidly with time. For example, a rocket may burn off a large fr: of its total weight in fuel within just a few seconds. During midair refueling, the mass of both airplanes can change quite rapidly. Also, bombers, crop dusters, and fire- fighting aircraft can drop a large fraction of their total mass in a very short time. When mass exits an aircraft, it carries momentum with it, Any mass that leaves an aircraft (or enters, in the case of an aircraft being refueled) could have velocity different from that of the aircraft, With a jet or rocket, the fuel mass leaves at a velocity that is significantly higher than the velocity of the aircraft. However, the difference between the ‘exit velocity and the aircraft velocity is commonly included in the thrust, force. Care ‘must be taken to see that this momentum flux is not counted twice when formulating ‘Newton’s second law. Thus, including mass transfer, Eq. (7.2.11) is written as7.2. Newton’s Second Law for Rigid-Body Dynamics 727 _ {Pa}. [%a} s Vex Fon, + Wa, —MOn Vz, +MO2V y, | = Lg, + Wp, —mOaVn, +mo,,V,> C212) - rg y. Figo, + We, —MOz,V yp, + MOnVa, =i $+i¥,, etme where it is the mass flow rate leaving the aircraft and Vz is the velocity of the mass leaving of entering the aircraft, as measured relative to the inertial coordinate system. The mass flow rate, m,, is negative for an aircraft being refueled. x In addition to Newton's second law, we must also assure that mass is conserved. an se, = 0 2.13) a Using Eq. (7.2.13) im Eq. (7.2.12), the body-fixed Cartesian components of the transla Donal momentum equation can be written F,, +W,, —mo,,V-, {F,, +W,, 214) F, +W., where F is a pseudo aerodynamic force including thrust. F = Fs —#(V,—V). and the ot Indicates @ time derivative. It is important to recognize that Eq. (7.2.14) is not based om the assumption that aircraft mass does not change with time. This equation is valid regardless of how rapidly the aircraft mass is changing. provided that the pseudo thrust force accounts for anv difference between the exit velocity and the velocity of the aircraft. As discussed in the preceding section, the Cartesian body-fixed coordinate system, which is commonly used in aircraft dynamics, has the x,-axis pointing forward from the ceater of mass along some convenient body-fixed axis in the aircraft plane of symmetry. ‘The y;-axis is normal to the plane of symmetry pointing out the right-hand side of the ~ aireraft, and the z,-axis points downward in the aircraft plane of symmetry. Here the designation (x;, ys, Zs) always refers to this body-fixed coordinate system. Since the coordinate system was chosen so that the airplane is symmetric in ys, we have Ty = Ip = I = I = 0 215) re yi Baas Mia 0 dy 0 a0. dn ‘The product of inertia terms in Eq. (7.2.15) could be carried However, virtually all aircraft possess this symmetry, which si_————— aa T2E_—_ Chapter 7 Aircraft Equations of Motion : .. Sn | Tay ny — ey Oey HeMo=/ 0 My 0 [ont = hn it, © ta ae V2, Ory —L1ay Ory (We) = ox fA, 4, P fe.0r, -0,,¥,, (lazy — Lyn yy Ory — Lez Oe, Dy, =le, @, $2, Hy —Oy Hey = {ey — Len, }e,@e, + Lae, (0% ~@2 ) W,, Hy [@x Hy, - Oy, Hy, (CLyyy — Lexy Ox, Oy, + Lez, Oy, Oz, (7.2.17) Using Bg. (7.2.17), the body-fixed Cartesian components of Eq. (7.2.2) are Tm 0 -Ie,][Ox Mx, + (yyy — Lez, Oy, Oxy + Lay Ory Oy, © Im 0 Ha@y f] = {Moy + Ulan ~ Len On Ory + Ian, (@2, 0? ) Pa a © +. jlo, M32, + Lax — 19 On yy — Lez, Oy, Oy . (7.2.18) As aircraft mass changes with time, so does the moment of inertia. However, when mass ts expelled from an aircraft, it can carry angular momentum with it. Thus, including the effects of mass transfer and applying conservation of mass in a manner similar to that used for translational momentum, Eq. (7.2.18) can be written as _. © —J,, (4, My, + (Ly, — Lexy Oy zy + Lez, Ox, Oy, ‘ O Ip 0 Hd = |My, + (lax, Len On Or +L, (02, —@2) ; Pe I, |\¢., Ma, + lay —Iyy, Oasys la OI (7.2.19) where M is 2 pseudo aerodynamic moment including any and all effects of thrust. This pseudo moment accounts for all the effects of mass leaving or entering the aircraft. Equation (7.2.19) does not require constant moment of ine The explicit notation (Vs,, Vy, Vz,) and (x, @y,, @z,), Which is commonly used in classical dynamics for the components of the translational and angular velocity vectors in body-fixed coordinates, is not normally used in the field of aircraft flight dynamics. The ‘notation that is used almost exclusively by the aircraft community is u Vu, the axial velocity Pp Oy, the rolling rate vy = 4~V,,7 = jthesideslip velocity, 447 =)" = the pitching rate as the normal velocity r Ox, the yawing rate ‘has always had the potential for causing some confusion with new students. ‘notation is well established in the literature and it should be memorized. to confuse the rolling rate with the pressure and remember that p and r does not stand for roll,7.2. Newton's Second Law for Rigid-Body Dynamics 729 As a memory aid, notice that p, g, and rare in alphabetical order, as are the rotation axes ® YA and 4. If we think of the rotation triad “roll, pitch, and yaw,” in that order, the symbols used for the angular velocity components and the axes of rotation form alphabene triads in the same order. There is another potential confusion associated with using lowercase m for mass, as is commonly done in classical dynamics. This is because in the field of aerodynamics, jowercase m is normally used to signify the aerodynamic pitching moment. To avoid this confusion, the x-, »-, and =y-components of the aerodynamic moment vector (excluding she effects of thrust) are denoted /, m, and n, whereas the mass is designated as W/g, with W denoting the weight and g signifying the acceleration of gravity. The flight mechanics community sometimes uses the symbols L, M, and N to Gesignate the x:-. yj-, and z,-components of the total moment vector, including any moment produced by the thrust vector, which in some cases may not act through the wer of gravity. This has been extremely confusing to new students, since L is also © signify lift and N is commonly used to denote the normal force. In at least one "tbook. the author has seen the symbol / used twice in the same equation, once to epresent lift and once to represent the rolling moment. In the present text, we shall not cary Tadition quite that far. Here we shall continue to use Mr,, My, and Mz, to designate the body-fixed components of the total moment vector. In the remainder of our development of the aircraft equations of motion we shall use the following notation: “= Xs-component of aircraft translational velocity relative to surrounding air. D = ys-component of aircraft translational velocity relative to surrounding air. 2-component of aircraft translational velocity relative to surrounding air. p = Xs-component of aircraft rotational velocity (rolling rate). ¢g = ¥e-component of aircraft rotational velocity (pitching rate). r = zcomponent of aircraft rotational velocity (yawing rate). F,, = xeomponent of total pseudo aerodynamic force vector, including thrust. F,, = Ye-component of total pseudo aerodynamic force vector, including thrust. F,, = zcomponent of total pseudo aerodynamic force vector, including thrust. x» = X»-component of total pseudo aerodynamic moment vector, including thrust. »» = Yercomponent of total pseudo aerodynamic moment vector, including thrust. ‘=, = Zp-component of total pseudo aerodynamic moment vector, including thrust. = xcomponent of aerodynamic force, excluding thrust (axial force). = y»-component of aerodynamic force, excluding thrust (side force), = zcomponent of aerodynamic force, excluding thrust (normal force), = x)-component of aerodynamic moment, excluding thrust (rolling moment), = = 8 W yyrcomponent of aerodynamic moment, excluding thrust (pitching moment), z,-component of aerodynamic moment, excluding thrust (yawing moment).i ( ””~”:C FRO Chapter > Anrcrah Eauations of Motion , ‘SteGents Show momorire this notation before proceeding with their study of this (Bhd Phe PoRiowing chapters, ‘Substiteting the aircraft notation just defined into Eqs. (7.2.14) and (7.2.19), we have & Sesto of ix fireorder differential equations for the six components of aircraft ‘welocity relative to the surrounding atmosphere. These are the velocity components mssnciated with the six Gogrees of freedom for the aircraft treated as a rigid body. This is (Gur Statement of Newton's second law. ( Fy + Wy, +(r0-qu)W/g = {Fy + Wy, +(pw-rw)Wig [Fs +W., +(qu—po)Wig My, + (yyy — Tz, qr + Licey PQ @ tm 8 HG} = |My + (le, — Ie) prt tn (0? - p”) eee Bir) | M+ —l,)pq—lear £4 ‘where ©. ©. and W., are body-fixed components of aircraft weight. It is important ° © keep i mumd that the forces and moments listed in Eq. (7.2.20) represent the total ~ ” ‘@utemettnes of forces and moments, including those due to thrust and gravity. Thrust can commribete s both the net force and the net moment if the thrust vector does not act ‘through the comer of gravity. The gravitational force can never contribute to the net moment. emce by definition it always acts through the center of gravity. Whee «= aircraft is operating in steady flight, with some equilibrium airspeed, V., some equilibeum angle of attack, a, no sideslip velocity, and no angular velocity, the umemateor of forces and moments will vanish and the terms on the right-hand side of ke 20) will be zero. For example, in steady level flight with the thrust vector ahigned wsth the darectuon of flight. the lift is equal to the weight, the thrust is equal to the and the summation of moments must be zero. If any of the parameters on the right Gevume from the equilibrium values, the aircraft will accelerate and the translational and coustions! velocities will change. The terms on the right may be thought of as dasutboances 10 steady equilibrium flight/ These disturbances are usually divided into two categones. The disturbances that a the motion only in the xs-z» plane are called ae wong tad: Penrmnces,/ u-Vpcoda,), w-V,8ioha,), 4, Fn +Wiy, Fay +Wa» My (7.2.20) disturvances are called the lateral disturbances, Pete Fy Wye Mayr Mer7.2. Newton’s Second Law for Rigid-Body Dynamics 734 and Bq. (7.2.20) becomes [We © 0 0 o 0 fw Fy + We, - quwig 0 Wig 0 yo 0 |\e 0 S eee 0 Oo | _ |Fatatm@Wle\ oy Seer «(0 - la, ||? ® See Oo Ut, © |(¢ My 0 0 ~ L . » es From Eq. (7.2.22), we see that the longitudinal disturbances affect only the longitudinal velocity components, u, w, and q. The longitudinal disturbances produce no change ie the lateral velocity components, 2, p, andr. This results from the assumed symmetry 01 the aircraft about the x,-z, plane. Eliminating the trivial results, # = p = * = 0, which are easily obtained from the second, fourth, and sixth rows of Eq. (7.2.22), the longitudinal equations of motion for the aircraft reduce to Wig 0 0 |fu Fy + Wx, — quis 0 Wig 0 \\w} =} F,, +We +quW/g (12.23) PO 7, \(¢ My, If, on the other hand, all of the longitudinal disturbances are zero, u-V,cos(a,) = w-V,sin(a,) = q = F+W, = F,+W, = My = 0 (278 and Eq. (7.2.20) becomes 6 momeno 6C0lCttO lo " 0 0 Wg 0 0 0 |lw meee 0 Je, 0 —In,\|\p se 8 Jy, 0 I\I4 See0 mls, 0 In, |\? roWlg Fy +Wy, +(p¥o sina.) ~rV, cos(a.)\W/g gs ~ poWig My, ay = Laxy pr + Lea, (vr? = p?) M, & oe . —732 Chapter 7 Aircraft Equations of Motion ‘Thus we see that while the longitudinal disturbances affect only the longitudinal velocity components, the lateral disturbances have an effect on both the lateral and the Jongitudinal velocity components. It should be noted that longitudinal motion (axial translation, normal translation, and pitching rotation) is simple planar motion confined to the x2, plane. On the other hand, lateral motion (sideslip translation, rolling rotation, and yawing rotation) is not a simple planar motion at all. Roll is rotation in the y,-z, plane and yaw is rotation in the x;-y; plane. The longitudinal velocity components are directly affected by the lateral velocity components, in part because of the Coriolis accelerations. For example, sideslip combined with a yawing rate results in a Coriolis acceleration in the axial direction. Similarly, sideslip combined with a rolling rate produces a normal Coriolis acceleration. Rolling and yawing rates also produce a pitching acceleration that results from the asymmetry of the aircraft in the x,-z, plane. The latter coupling term is usually referred to as the inertial cross coupling. EXAMPLE 7.2.1. Consider an object moving through space with no forces of any kind acting on the body. The body is initially moving relative to an inertial coordinate system with a velocity of Vo and is rotating about an axis normal to the velocity vector with an angular velocity of @. We wish to use Newton’s second law to determine the translational and rotational velocity of this object as a function of time. This problem is, of course, trivial to solve using an inertial coordinate system. However, to gain some familiarity and confidence in working with body-fixed coordinates, we wish to solve this simple problem using only body-fixed coordinates. Solution. Because the rotation vector is normal to the translation vector, this is planar motion. Choosing the body-fixed coordinate system so that at time ¢= 0 the x,-axis coincides with the direction of motion and the y,-axis coincides with the axis of rotation, the motion is planar longitudinal motion. Newton’s second law is given by Eq. (7.2.23) with all forces and moments set to zero, Wig 0 0 |i —quW/g 0 Wig 0 \\w} =4 quWlg (7.2.26) in MO By I 0 The initial conditions for this first-order system of differential equations is u Yo pl gecli0 (7.2.27) W120 CX) ‘The third equation in Eq. (7.2.26) is not coupled to the other two. This equation is _ easily integrated subject to the third initial condition in Eq. (7.2.27) to give7.2. Newton's Second Law for Rigid-Body Dynamics 733 Using this result in the remaining two equations from Eq. (7.2.26), we have [1 O}fw 0 a {r\ = {oh lo iflwf*|-oo 0 laf lo This is a homogeneous system of first-order linear differential equations. At this point students may wish to review the chapter on systems of linear differential equations from their introductory course on ordinary differential equations. Rewriting this system of equations in the usual differential operator notation, we have [ D oo)ful _ {oh (7.2.28) [-o DJJ lo where D the differential operator. Differentiating the first equation and multiply- ing the second equation by @ gives 2 0-1 -a% aD} 9 Subtracting the second equation from the first, we have “Sle e -o§ = aD |v 0 In a similar manner the system can be further reduced to yield (D? +03) 0 {nf id ‘2 0 (D? +09) |\w 0 We have now reduced a system of two coupled first-order equations to a two uncoupled second-order equations. Each of these is easily solved to give {it ss ie sin(@ot) + C2 cos(@ot) w C3 sin(@ot) + Cy cos(@ot) Using this result in Eq. (7.2.28), we have { Cio COs( Mot) ~ Czy Sin(@ot) + Cy sin(@ot) + Cy@o —C\@q sin(@ot) — Cz@ COs(@ot) + Cy COS(Wol) —734 Chapter 7 Aircraft Equations of Motion which requires that Q Cy « (7.2.34 fed {3} y Using Eq. (7.2.30) in Eq. (7.2.29) gives u) _ needa (7.231) w C2 sin(@ot) — Cy cos(@ot), Applying the initial conditions, u(0)| _ [% ; wf [0 : we obtain Gi fio {e} = {ef (7.2.32) Using Eq. (7.2.32) in Eq. (7.2.31) and including the result obtained for the angular velocity gives the complete solution, u Vo cos(@ot) ws = 4Vosin(wot) q a This problem is, of course, trivial to solve relative to an inertial c system. For an inertial coordinate system, Eq. (7.2.26) reduces to the co uncoupled system, Wig 0 0 |[%, 0 0 Wig 0) | aa 0 0 Ty, |\\ay, 0 which can be directly integrated subject to the initial conditions to give7.2. Newton's Second Law for Rigid-Body Dynamics 735 Thus, to a fixed observer in the inertial coordinate system, the velocity of the object appears to be constant with time. On the other hand, to an observer in the body-fixed coordinate system, the velocity appears to be changing sinusoidal with time. These are simply two equivalent ways to describe exactly the same motion. The advantage of using the body-fixed coordinate system is not at all obvious from Example 7.2.1. This very simple problem is much easier to solve using the inertial coordinate system. We do see, however, that Newton’s second law can be stated and solved either in terms of inertial coordinates or in terms of body-fixed coordinates. The advantage of using a body-fixed coordinate system will become more apparent in the following sections. 7.3. Position and Orientation: The Euler Angle Formulation “in the preceding section, Newton's second law was written in terms of a coordinate system fixed to the aircraft. For complex geometry and/or complex motion, this coordinate system can simplify the equations of motion by making the inertia tensor independent of time. Furthermore, onboard instruments report flight parameters only as viewed from a body-fixed coordinate system. Thus, the body-fixed coordinate system extends considerable utility to the study of aircraft dynamics. Unfortunately, this moving coordinate system cannot easily be used for describing the position and orientation of the airplane. The aircraft position and orientation are most conveniently described in terms of a coordinate system fixed to the Earth. The position of an aircraft relative to the Earth is normally described in terms of latitude, longitude, and altitude. However, over short time periods, the distance traveled is small compared to the radius of the Earth and we can describe position in terms of a Cartesian system. As previously discussed, the Cartesian system commonly used has the x-axis pointing north, the y-axis pointing east, and the z-axis pointing down. This Earth- fixed coordinate system has been designated (xy, yy, zy). The position of the aircraft is specified by the location of the origin of the body-fixed coordinate frame relative to the Earth-fixed coordinate frame. The orientation of an aircraft relative to the Earth can be described in terms of what are called Euler angles. The orientation of the body-fixed coordinate frame, ( relative to the Earth-fixed coordinate frame, (xy, y;, zy), is described in terms consecutive rotations through three Euler angles, in the specific order as fi (a) Rotate the (xy, yy, zy) coordinate system about the Zy-axis the coordinate system (x), y), 21), as shown in Fig, 7.3. 1a. (b) Rotate the (x), yj, 21) coordinate system about the y\-axis the coordinate system (x2, 2, 22), as shown in Fig. 7.3.1b, (c) Rotate the (x2, y2, 22) coordinate system about the x» the coordinate system (x,, yp, 2), as shown in Fig, 7.3.% 5 \ (©) View projected from view (b) with fold line perpendicular to true view of x-axis © (b) View projected from view (a) with fold line perpendicular to true view of y,-axis. Figure 7.3.1. True views of the three Euler angles shown following the standard conventions of engineering graphics and descriptive geometry. An isometric view of these three rotations is shown in Fig. 7.3.2. The order of rotation is very important. In general, if these three rotations were to take place in a different order, the final orientation would be different The three Euler angles, g, 0, and y are, respectively, called the bank angle, the elevation angle, and the azimuth angle or heading. These three angles are sometimes incorrectly referred to as the roll, pitch, and yaw. There is a subtle but very important difference between the Euler angles and roll, pitch, and yaw, The azimuth angle is a rotation about the z,-axis and yaw is a rotation about the z)-axis. Similarly, the elevation angle is 2 rotation about the y)-axis and pitch is a rotation about the y»-axis. Roll, pitch, and yaw are orthogonal. The Euler angles are not, The Buler angles can be related to roll, pitch, and yaw, but this is not a one~ relationship. For example, it is possible to change the bank angle of an aircraft using any roll input whatsoever. To demonstrate this, consider an airplane in a attitude with all three Euler angles at zero, as shown in Fig, 7.3.3. Now let the move through a pitch angle of 90 degrees, followed by a yaw angle of 90 pitch angle of ~90 degrees. The airplane now has a bank angle of re was no roll input throughout the process. Try it witha model.7.3. Position and Orientation: The Euler Angle Formulation 737 West South Pitch up, 90° =0°,0=90", y=0" Zz y Yaw right, 90" Pitch down =90",0=0", y= 90 =90",0= Figure 7.3.3. Relationship between Euler atTSB Chapter? Amerah Equations of Motion } angles describes a unique orientation of the aircraft (ie, ne iicanmae a an three Baler angles). A specific set of Euler ‘angles specifies an instantaneous orientation that does not depend in any way on how the ‘mrcrat Came to he so oriented. While a given set of Euler angles specifies a unique ‘oniemtation for the aircraft. orientation is a periodic function of the Euler angles. ‘Thus, ‘Sty Grentation of an aircraft could be described by more than one set of Euler angles, ‘Gleatty, adding 350 degrees to any one of the Euler angles will result in a different but ‘cgurvalent set of Euler angles. If we are to define the Euler angles so that any given ‘Qmentation has e unique set of Euler angles, then we must restrict each Euler angle to fall within some range including no more than 360 degrees. However, even this is not ‘sufficient. This is demonstrated in Fig. 7.3.4. In this figure we start with an airplane pommed Gue north with both the wing span and the fuselage reference line parallel to the ounc. Normally. we would specify this orientation as having all three Euler angles qual t© zero. From this starting point, shown in Fig. 7.3.4a, we let the airplane move Shrough = patch angie of 180 degrees to the orientation shown in Fig. 7.3.4b. Clearly, ‘ts omemtanon could be described as having a bank angle of zero, an elevation angle of 80 Geprees. and azimuth angle of zero. The airplane could be placed in this same % From (a) pitch 180° = 0°, 0 = 180°, y=0° (b) : From (a) yaw 180° = 0°,0=0", y= 180° «© Figure 7.3.4. Two ‘equivalent Euler7.3, Position and Orientation: The Euler Angle Formulation 739 orientation by starting with all Euler angles at zero and rotating the airplane through a yaw angle of 180 degrees to the orientation shown in Fig. 7.3.4¢. From this point the airplane is rolled through 180 degrees to the orientation shown in Fig. 7.3.44. This final onentation could be described as having a bank angle of 180 degrees, an elevation angle of zero, and an azimuth angle of 180 degrees, Since the orientation shown in Fig. 7 — is identical to that shown in Fig. 7.3.4d, the Euler angle set #= 0°, @= 180°, and y= 0 is obviously equivalent to the set #= 180°, @= 0°, and y= 180°. To avoid this duality and make the Euler angle definitions unique, the Euler angles are by standard convention usually restricted to the range —180° < $< 180°, -90° < @< 90°, and 0° < y< 360°, To obtain the transformations between coordinate systems (xs. Vs. 24) aNd (xy. ¥/» Z))s we now review some principles from calculus and analytic geometry. In terms of the three Euler angle rotations shown in Fig. 7.3.1, we wish to develop two transformation matrices, one to transform from body-fixed to Earth-fixed coordinates and another to transform from Earth-fixed to body-fixed coordinates. We first consider each of the three rotations shown in Fig. 7.3.1 as a separate transformation and then combine the three transformations into one transformation matrix. ; c Consider first an arbitrary vector v having components (V's, Yyys Ys) in coordinate system (%), ¥), 2;) and having components (v;, ¥y,. Vz) in the (xj, 7 ) coordinate system. From the geometry in Fig. 7.3.1a, we sce that the x-component of the vector v in coordinate system I can be broken up into two components in coordinate system f. Thus we have Vals, = Vx COS(W)ix, + Ym SiN(Y)iy, (73.1) Similarly, the y-component transforms as Yyiy, = —Vy Sin(Y)ix, + Vy COS(Y)iy, (73.2) and the z-component transforms as teeny Combining Eqs. (7.3.1) through (7.3.3), we obtain the transformation equation Vay cos(y) —sin(y) 0) {v,, vy, 7 =| sin(y) cos(y) 0 Vy, Vey 0 0 1} va In a similar manner, the inverse transformation can be written Yay cosy) sin(y) O]){vs, vy, ¢ =| —sin(y) cos(y) 0 Vyy Vay 0 1] [vey740 Chapter 7 Aircraft Equations of Motion | Likewise, from Fig. 7.3.10, (%. YM) can be expressed in terms Of (Ve, ¥y, Ve). i > : va) [ cos(@) 0 sin()] {vss eet 0 1 yy (7.3.6) Ys =sin(@) 0 cos(9)} (v=, | ‘and the inverse transformation is found to be | Ve cos(@) 0 -sin(@)] {Ys ae 8 4 DO he (73.7) sin(@) 0 cos() |{v., From Fig. 7.3.1c we obtain 4 Toray Yo live >} =| 0 cos(d) —sin(d) |) ¥», (7.3.8) vn} [0 sin(g) cos(4) | [vz } and Vay 0 0 V[ve vy, 7 =|0 cos(g) sin(g) }4 vy, (7.3.9) V4 [0 —sin(d) cos() } (vz, Combining Eqs. (7.3.4), (7.3.6), and (7.3.8), we have |v [oosty) -sin(y) Of cos(@) 0 sin(@)]1 0 0 Vrs %, ¢ = | simiy) cosy) 0] 0 1 0 JO cos(g) -sin(g),|}y, %2,) 0 0 1 | -sin(@) 0 cos(#)}0 sin(g) cos(p) | vz. (7.3.10) and combining Eqs. (7.3.5), (7.3.7), and (7.3.9) gives the inverse transformation, =|0 cos(g) sin(p) 6 3 0 ~sin(y) cos(v) 0}) ry, 0 —sin(g) cos(g)| sin(@) 0 cos(A) 0 0 1 Ym Vang 1 0 0 Ycos(@) 0 —sin(A)| cos(y) sin(y) 0}| vx, Ys, ‘The individual transformation matrices in Eq. (7 those in Eq, (7. be multiplied together to obtain a single t each7.3, Position and Orientation: The Euler Angle Formulation 741 Vay CoCy SpSoCy -CySy CySeCy + SpSp |[Vnn Vy, ft =|CoSy SpSpSy +CyCy CySeSy —SpCy |} (7.3.12) ‘ | -Se SpCo C4Co Yen and from Eq. (7.3.11) we obtain the inverse transformation equation, Vr, CoCy Cae -So |[vx, vy} =| 5pS0Cy—CpSy SpSoSy+CeCy SoCo 11%, (7.3.13) : CpSoCy +SpSy CSoSy -SpCy CoCo | (vz, Equation (7.3.12) can be used to transform the components of any vector from body- fixed to Earth-fixed coordinates, and Eq. (7.3.13) can be used to transform vector components from Earth-fixed to body-fixed coordinates. It should be noted that the inverse of the transformation matrix in Eq. (7.3.12) is its transpose. ‘ In body-fixed coordinates, the velocity of the aircraft relative to the surrounding air has been designated (u, v, w), and in Earth-fixed coordinates, the velocity of the aircraft relative to the Earth is the time rate of change of the position vector (x, ¥y,2/). The ground speed is related to the airspeed and wind through Eq. (7.1.1). Thus, combining Eqs. (7.1.1) and (7.3.12), we can write xy CoCy SpSeCy —CySy CySoCy + SpSy | u Vx, Dep =|CoSy SySeSy+CyCy CpSoSy —SgCy |} Vp + Vay, ¢ (7.3.14) zy -So S¢Co CyCo wl Vrey where Vex, Vy,, and Vz, are the components of the constant wind vector in Earth-fixed coordinates. Integration of this equation will yield the airplane's position as a function of time, relative to the Earth-fixed coordinate system. Since we are using Newton’s second law in body-fixed coordinates, we must express the gravitational force in body-fixed coordinates as well. In Earth-fixed coordinates, the representation of the gravitational force vector is (0, 0, W). Thus, from Eq. (7.3.13), the gravitational force vector in body-fixed coordinates is represented as Ws, —sin(0) Wy, + = W4sin()cos(@) (7.3.15) We, cos(¢) cos(A) In a similar manner, the time derivatives of the Euler angles, ( g, 0, y/), are related to the body-fixed components of the angular velocity’ vector, ( p, g, 7). However, we must remember that each of the Euler angles is defined relative to a different coordinate system. The bank angle, ¢, is defined relative to the coordinate system (x2, y2, 22), the elevation angle, 9, is defined relative to (x), y;, 2)), and the azimuth angle, y, is defined relative to (xy, yy, zy). Thus, using Eqs. (7.3.5), (7.3.7), and (7.3.9), we can write es742 Chapter? Aircraft Equations of Motion inverting Eg. (7.3.17), we get the Euler rates in terms of the roll, pitch, and yaw rates, |p) P 0 0 (| 1 0 O]Ce 0 -So | Jes =]0 G S,Hos+]o cy SfO 1 0 | ir} [0 -s c lol 0-5, Cy] Se 0 Co i 0 Of 0 -Sof CG Sy 0 +]0 G S]o 1 0 |-s, CG 0 [0 -S, Cy] Se 0 Co] 09 0 1 ‘or after combining terms, D i 0 —So l[¢ : | JIp =|0 Cy SoCo 10 (73.17) r} [0 -Sp C4Co |v : | ; ] {d) fl SpSa/Co CgSo/Co|{p {Or =|0 & -Se |g (7.3.18) | Ww) [0 So/Co Cy/Co | Ir | | Here, we see that the inverse of the matrix in Eq. (7.3.17) is not its transpose. Integration of this equation will yield the airplane's orientation as a function of time in | terms of the Euler angles ¢, 8, and y. ; Combining Eqs. (7.3.14) and (7.3.18), we have a system of six first-order differential equations that relate the time rate of change of the position and orientation of the aircraft in Earth-fixed coordinates to the translational and rotational velocity components, relative to the surrounding air, in body-fixed coordinates, {zr Caly SpSoCy-CoSy CpSoCy +SgSy ](u) (Vr, ; 4 =|CoSy SpSoSy+CyCy CySoSy -SpCy |} 0 ++) Voy, ' zy -So ; SgCo C4Co w) Vue, 73.19) ¢ 1 S4Se/Co CsSo/Co|(p brm40 IC; -Se |iq v 0 Sy/Co Cy/Co |Ir These six equations are called the kinematic transformation equations, in Euler angles. This transformation contains a singularity. This singularity can be the last two terms in both the fourth and sixth equations in Eq. (7.3.19). four terms become singular when the Euler angle, 0, is 90° and cos@7.3. Position and Orientation: The Euler Angle Formulation 743 these points the transformation breaks down, and integration of the Euler angles becomes indeterminate. This singularity in the integration of the Euler angles is commonly known as gimbal lock. Due to its existence, virtually no commercial or widely used six-degree- oFfreedom (6-DOF) computer code employs the transformation given by Eq. (7.3.19) a determination of the Euler angles. Rather, other methods are generally employed, whic! avoids these singularities through the use of either a quaternion transformation or a direction cosine transformation. However, the Euler angle transformation is widely used in the linearized equations of motion) (Clearly, under normal operation, most airplanes stay within the range of motion that can be described by Eq. (7.3.19). Except for vertical launch vehicles, fighter aircraft, and stunt planes that might pass through these se the Euler angle singularities at +90 degrees are not a problem. ‘The main disadvantage of using the quaternion formulation for aircraft attitude is the increased complexity of the physical interpretation. Thus, the student should have a good understanding of the Euler angles discussed in this section before attempting to understand the quaternion formula- tion to be presented later in this text. : For pure longitudinal motion, the bank angle is typically zero and the azimuth angle must remain constant. For simplicity, we could align the x,-axis of the Earth-fixed coordinate system with the x;-z5 plane of the body-fixed coordinate system and make the assumption of no wind. Thus, we can write v=p=r=¢=y=0 (7.3.20) For this pure longitudinal motion, from Eq. (7.3.15), the gravitational force described in body-fixed coordinates reduces to W,, —sin(@) Wyb =W} 0 (7.3.21) We, cos(0) and Eq. (7.3.19) becomes iy Co (0 Sp ONO Olam wy 0? 71:0 40 0eNON:| (a) iy -So0C 00 0 |lw = 7.3.22) @ 0 0 0 1 0 S/Cy\\o ‘ o 0 0 0 0 clmanO q W 0 0.000 IC, jlo Here again we see that longitudinal motion has no tendency to introduce lateral Eliminating the trivial results, Wy, =» = = y = 0, longitudinal motion requit Beas oe —sin(@) nea 3 wf aaan mite 4 _ int) oad 74a chapte ao) ot) (6h in) SL 7 i and elevation angle must normal veloc’ ty must be Zero. | For _ Sse ’ ear Thus, we can write oY prium flight path ae ., 7.3.19) become Thus, fo re lateral 7 a (7.3.26) W wi sin() W {cos(@), ss, 00 0 [Yocos(a)] [cy -CoSv S05v ; << 6 Toa Bras, p00 0 sin(@) | Re 0 $ | | aio (0 (Onl 0.20) pa | é ( || 6 o 0 0 0 Cy ~So\! 0 saa y 0 0 0. os) Gl al rad that while the longitudinal disturbances affect only the longitudinal ral disturbances have an effect on both the lateral and the longitudinal UEXAMPLE 7.3.1. Determine the position and orientation of the object in Exampl as a function of time. Starting with the body-fixed solution to Example 7.2.1, describe the position in terms of Earth-fixed coordinates, and the orientation in terms of Euler angles. Solution. From the solution to Example 7.2.1, {u Vy cos(aot) 7 {Pecos \,| Vo sin(wot) 4) % | Since this is long tudinal motion, we can apply this result to Eq. (7.3.24):___“_=<——™™™ Motion ations of 744 Chapter? aircraft Equ: cos(@) sin(A) O};u sin(@) cos) 0})w 0 o Il¢ elocity, and elevation the axial velocity, normal i yea Must be poe Mug For pure lateral motion, component of the angular 2 ones 2et0, For ! Temain constant =n a 2 oa and consider the special casi Thee en 4 simplicity, we shall neglec : le is zero. 5 toan pee flight path where the elevation ang! Fr Atiy, ‘aN Write 3 V..cos(@) = w-Vosin(ao) = 4 =9=0 0.3.25) u-T, costa, 19) become ‘Thus, for this pure lateral motion, Eqs. (7.3.15) and (7.3.19) (We, . 1, = Wi sin(d) (7.3.26) (We, cos(g) c CoSy SSy 0 0 0 Tf, cos(a,) | Cy -Sly 0 0 0 en % 0 0 0 |r, sina) a? ‘ os (3.27) 0 ool P Bl io 0 0G -s, 0 lw} L 0 0 Oh, OSG, r ( Here again we find that while the longitudinal disturbances affect only the longitudina) motion, the lateral disturbances have an effect on both the lateral and the longitudinal motion. ~PXAMPLE 73,1, Determine the position an Example 7.2.1 as a functi Solution. From the solution to Example 7.2.1, u Yo cos(wot) w Yo sin(aot) 2. a Since this is longitudinal motion, we can apply this result to Eq. (7.3. OO #7.3. Position and Orientation: The Euler Angle Formulation 745 ay cos(A) sin(@) 0] {Vo cos(aot) Zp} = |-sin(@) cos(@) 0}; Vo sin(at) (7.3.28) 6 0 uC a % The initial conditions are xy 0 7.3.29 zy 0 ¢ ) 0 The third equation in Eq. (7.3.28) is easily integrated subject to the third initial condition in Eq. (7.3.29) to yield 0= at Using this result in the first two equations in Eq. (7.3.28), we have ky cos(@ot) _sin(wot) | {Vo Seo 27) — |—sin(@ot) cos(@ot) | | Yo sin(@ot) Performing the indicated multiplication gives ep) Vo cos? (cont) +Vp sin? (wot) a et zy ~V~ cos(aot)sin(@ot) + Vo sin(@ot) cos(@ot), 0 Integrating this result subject to Eq. (7.3.29), the complete solution is xy Vot zy =40 0 @ot WEAMPLE 7.3.2, The rocket in Example 7.1.1 has a bank angle of 40 an elevation angle of 20 degrees, and an azimuth angle of 70 degrees. no wind, what is its velocity in Earth-fixed coordinates? Solution. From Example 7.1.1,746 Chapter 7 Aircraft Equations of Motion From Bg. (7.3.19), CySoCy + SoSy CySoSy —SeCy |4 v CoCo Coy SpS0C - COS¥ CoSy SeS0Sy +OOCv nS, S6Co Ce = cos40° = 0.766044, Sg = sin40° = 0.642788, .= gee Ce = c0s20° = 0.939693, Se = sin20° = 0.342020, Cy = cos 70° = 0.342020, Sy = sin70° = 0.939693 Substituting, we have A 0.321394 - 0.644654 0.693633 ] {825.96 fi/sec v,, =| 0.883022 0.468591 0.026356 |, 300.63 ft/sec Ve, 0.342020 0.604023 0.719846 || 476.87 fi/sec and the matrix multiplication gives Vey 402.43 fi/sec Vy, + = 4882.78 f/sec V,, 242.37 fi/sec y As a cross-check on our math we can compute the magnitude of this vector. It should be the same as the magnitude in body-fixed coordinates, 1,000 f/sec. V2 +V, +V2 = 402.43" + 882.78" +242.37° f/sec = 1,000.00 fi/see EXAMPLE 7.3.3. The flight of a stable projectile, such as an arrow or a st bomb, can be approximated as pure longitudinal motion. If the projectile is enough, the angle of attack remains small and we can approximate the longi aerodynamic forces and moment as 4O -1 pu?s,-(ko +ka?) Z = -1pwS,kna m = ~+ puSrl, (kya + kal, q/u), <—- 7.3. Position and Orientation: The Euler Angle Formulation 747 where X is the x-component of the aerodynamic force, Z is the z)-component of the aerodynamic force, m is the aerodynamic pitching moment, ko, ki, kas ka, and ke are dimensionless constants, S, is a reference area, /, is a reference length, and a is the angle of attack, which can be approximated as re) (7.3.33) u determine the Using these approximations and assuming no wind, we wish to Formulate the position and orientation of the projectile as a function of time. problem in a manner suitable for solution by fourth-order Runge-Kutta. Assume that at time equal zero, the position and orientation are (vo, Zo, 4), the magnitude of the velocity is Mo, while the angle of attack and angular velocity are both zero. Solution. Fourth-order Runge-Kutta requires a system of first-order differential equations and initial conditions in the form am LOI Inv) 21,23," yi(O) = Y; Using Eq. (7.3.33) in Eqs. (7.3.30) through (7.3.32), we can write EX Kou? +k w> Zh= -3PS, knuw m kal, uw +kal; qu The total force is the aerodynamic force plus the gravitational force. Since this is longitudinal motion, we can apply this result along with Eq. (7.3.23) to Eq. (7.2.23), Wie 0 0 |fu 0 We 0 lw 0 0 ty, \l4 ou? + kw —Wsin(@)| (-quWig = -5PS, knuw +4 Woos(@) }+) quWig kyl, uw kal; qu 0 0 Solving for the time derivatives, we have tt —Kou?—K,w? - gsin(0)-qw Wr = ~K,uwtgcos(A)+qu 4q ~Kyuw-KyquoO OO ——— FEB chapter? Arerah Equations of Motion where Re 1 pS,kox/W is LpSke/W {xyh = { dpSo9/W Ix] | $05: &y/tom xe} [$05:12ka/lyn Combining Bg. (7.3.24) with Eq. (7.3.34), we obtain a complete system of Sc SErential equations ina form suitable for solution by fourth-order ye Kuta, [-Kou?-K,w?-gsind-qw | -Kzuw+gcosd+qu —K3uw-Kyqu ucosO+wsind (7.3.35) weosd—usind q The appropriate initial conditions are u(0) VY w| | 0 q0)| _ Jo is elie 27(0) Zo a0) % Starting with the initial values given in Eq. (7.3.36), Eq. (7.3.35) can be integrat forward in time using fourth-order Runge-Kutta. Computation time can be decreased with no loss of fourth-order Adams-Moulton method. The details of the fourth-order Rung and the fourth-order Adams-Moulton numerical methods can be found in undergraduate text on numerical methods (see, for example, Hoffman 1992 details of obtaining this solution are left as an exercise for the student, After the first four steps b accuracy by switching to EXAMPLE 7.3.4, An arrow is shot from a bow with a velocity of 210 fU/s an elevation angle of 5.0 degrees, Using the formulation developed 7.3.3, we wish to obtain the Position, orientation, and airspeed of tl yards downrange. For this arrow the aerodynamic coefficients following Eq. (7.3.34) in Example 7.3.3 are Ky = 0.00061 tt", = 7.8. Position and Orientation: The Euler Angle Formulation 749 Ky = 0.00059 8, Ky = 0.0016 A, and Ky = 0.0064 ft!, Also plot the elevation of the arrow in inches above the starting point, its airspeed in feet per second, its clevation angle in degrees, and its angle of attack in degrees as a function of horizontal distance downrange in yards, from zero to 100 yards. Solution. The details of obtaining a Runge-Kutta solution are straightforward and left as an exercise for the student. However, the formulation given by Eq. (7.3.35) is written with time as the independent variable and we wish to obtain a solution for a specified downrange distance of 100 yards. We have no way of knowing, prior. the length of time required for the arrow to travel 100 yards. Since xy is monotonically increasing function of time, rather than using the formulation with time as the independent variable, which requires stepping forward in time, it would be convenient to have x, as the independent variable, so that we could step forward in range. To facilitate this change of variables, we recognize that aF _ dF [oy dxy dt/ dt where F could be any one of the dependent variables. The change in xy with respect to time is given by the fourth equation in Eq. (7.3.35). Thus we can write a = 45 [ycoso+ sind) (73.37) Using this result in Eq. (7.3.35) gives wu’ Kou? —K,w? -gsind-qw w' —K>uw+gcosO+qu This 1 —Kyuw-Kyqu aya errs) ucos0+wsin zy weos0—usind . ao q where the prime indicates differentiation with respect to x;. With this change of variables, the fourth equation becomes trivial, and if we do not care about time, we can reduce the order of the system from six to five. However, since computation tme is of no concern with this very simple system, we could as well compute time as a dependent variable. Using Eq. (7.3.37) for = t, we have dt _ dt [acy 1 diy dtl dt ~ ucosO+wsind Replacing the fourth equation with this result, we obtain the complete first-order differential equations with x,as the independentrf —— | FSO Chapter 7 Airerat Equations of Motion = Kou? Ki w? —gsing— 9) 2uwW+gZcosO+ gy -K,uw-Kaqu 1 weos8—usind q ~ ( Avera Equations of Motion mAX © = 2CV,AR-kAx This is the Tinearized second-order differential equation that Boverns @isturbance motion of the trailer relative to its equilibrium Position beh 3” Moving tractor. hind ‘Smallisturbance theory can also be applied to systems of differenti ‘The rigié-body 6-DOF equations of motion, written in terms of Euler angles, 12 coupled nontinear differential equations in 12 unknowns. Using small-d u Sheory, these equations can be linearized. The small-disturbance approximation that the motion of the aircraft be confined to small deviations about some , ‘equilibrium Might condition. Under normal operating conditions, the force; ‘Scting on an aircraft change slowly and do not deviate far from equilibrium, Thus ai Gisturbance theory can be applied to aircraft motion for many commonly eneo tm the application of small-disturbance theory to aircraft dynamics, we Write 1 tions of motion as its value at the equilibrium reference state pj, W= Wo + dw q= Go+ Aq r= lo+4r Ay = Xo + Axy Ve = Yor Avy Zp = 2o+Azy @= 6+Ad 0=0,+A0 Y= Wor Ay F,, = Fn. +AF,;, Fy, = Fyo + AFy Fz, = Fao + MF, W,, = W,,. + AW,, Wy, = Wyo + AWy, We, = Wyo + Wey M,, = Mi.+4M,, My, = Myo +4My, M2, = Mio + 4My, 6. = 6a, + 46, Se = Seo + Abe Or = 67 + Ab, ‘The first 12 variables in Eq. (7.5.1) are commonly known as the state variables, they define the instantaneous state of the aircraft in terms of velocity, position, an orientation. Since there are only 12 scalar components in Eqs. (7.4.4) through (7.4.1), th remaining 12 variables must somehow be known as functions of the state variables an pilot contro] inputs. Here we consider the reference condition to be steady flight, at : airspeed. V,, with no sideslip, no bank angle, and no angular velocity. alignment of the axes is arbitrary, we shall align the x,-axis with the aircraft’s direction of flight and place the x,-axis so that the equilibrium azimuth Furthermore, since the summation of forces is zero in steady flight, we have Uo =Vor Xo = Vox, + Vy CO8A)t, Yo =Vuyyty Zo = (Vuey —Vo Fo = Woyos Fey = Wey, Vo = Wo = Po = Yo = Yo = bo = Wo = Fro =7.5, Linearized Equations of Motion 757 By substituting the results from Eq. (7.5.2) for the equilibrium values in Eq. (7.5.1), the system variables are expanded in terms of the disturbances from this equilibrium flight condition. This gives u= Vo + du v= dv w= Aw p= 4 q= a Xp = Pra, + Vo cos, t+ Axe yy = Vuy,tt+ Aye zp = Voz, —Vo sino) + Azy o= 49 O= 0,+40 y= Ay (7.5.3) Fa = —Wryo + AF x, Fy = Aly Fy, = —Wayo + 4F x, W,, = Way + 4Ws, Wy, = Wy, We, = Weyo + AW, M,, = 4M, My, = 4My, M., = 4Mz, 62 = Sao + Aba 5, = 510 + 46, Se = Seo + Ae Newton’s second law of motion can be written in terms of the disturbance variables by using Eq. (7.5.3) in Eq. (7.2.20). Since the disturbances are assumed small, we can neglect the product of disturbances, and Ne wwton’s second law can be approximated as ae. 2 9 0 0 Ya AF, + AWx, 0 Wig 0 Ot 40-8 OF AG AF,, + 4Wy, — ArVoW/g. 0 0 We 0 0 0 |\Aw| _ | AF:, +AW., + 4qVoWe (ee re 0. In, 0. —Ix,|| 45 4M,, END 0) (0 Ty, 200 |g AMy, menor 20 Tor 0s sla 4M., The aerodynamic forces and moments are functions of the translational and_rotational velocities, the translational acceleration, an id the deflection of the control surfaces. Thus, for small-disturbance theory, the aerodynamic force is approximated as OFr, OF, OF xy AF», ou dv Ow OFy, OFy, OFy, Ciel =\i9a, «acer cwe AF z, OF, OF, OF, du ov w | OF, OF, (Ole fs OF, OF y OF y On 0 lw OF, OF, OF 5 Ou = OF;, OF, OF x, Au op oq oF \(4y on OF, OF, OF yy 4 q or ae pO x OF, OFOEE EEE —=—_ — TSS Chapter > Aircraft Equations of Motion ‘and For the serodynamic moment we use the approximation am, @Mxy OMy OM, — sp oy ae er | |om, eM, amy, lo éMy, » y 2 Shela aay aoe ao 6M, ||4~) | aM. OM. @M., ||& ow Gp Gq oa 7: a 2M, OMy My @ Sai a aur, we | | 2M BM OM m1) sob4|oe om I Ag ow 65. 66. 06, OM,, |(40} | @M:, @Mz, OM, |(45, ow 05, 06. 66, ‘Where all derivatives are evaluated at the equilibrium flight condition. The body-fixed aerodynamic force and moment components, which are expresseq inl Eas. (7.5.5) and (7.5.6), are shown as independent of the Euler angles. The aerodynamig force and moment components in body-fixed coordinates do not depend on the aii Onentation relative to the Earth, but only on the motion of the aircraft relative to the Surrounding air. This is why atmosphere-fixed coordinates were chosen as the inertial ‘Coordinate system in the formulation of Newton’s second law. In Egs. (7.5.5) and (7.5 6) we have assumed that the force and moment components depend on the translational acceleration components. The forces and moments depend on the translational accelerations because the vorticity generated by the wing requires a finite time to travel from the wing to the tail. As the angle of attack changes, the lift and vorticity generated by the wing also change. This change in Vorticity alters the downwash at the tail and hence alters the forces and moments produced by the tail. However, since the vorticity travels from the wing to the tail at the speed of the aircraft, it takes a finite time for the altered downwash to reach the tail. Since sideslip has lit effect on lift or vorticity, the sideslip acceleration terms are not significant. In additi several of the remaining terms can be eliminated on the basis of symmetry. Thus, for linearized equations of motion we shall use OF y, e @. OM,, _0M,, _ i aa ao z Several more of the derivatives in Eqs. (7.5.5) and (7.5.6) can be eliminated basis of symmetry. For example, at the equilibrium flight condition, the change ir force with respect to the sideslip velocity must be zero, because the aircraft is symmetric about the x,-z, plane and the equilibrium sideslip velocity is zero.ali ea 7.5, Linearized Equations of Motion 759 mind that the derivatives in Eqs. (7.5.5) and (7.5.6) are all evaluated at the equilibrium flight condition. Clearly, the axial force is not totally independent of sideslip, However, from the assumed symmetry, a positive sideslip should have the same effect on the axial force as a negative sideslip. Thus, the axial force must be a symmetric function of sideslip, and the change in axial force with respect to sideslip velocity must be zero, when the sideslip velocity is zero. By similar reasoning we can eliminate all of the following derivatives evaluated at the reference flight condition: ae oe er) a ag op Oa OTA (7.5.8) 2M, _@My _@My Ms _ 4 a (pC OF, OF _ Fy OF yy _ Fy _ 06, 05. 86, 06, @My _OMy _@My, 2M _4 05. «86, + +85, + O5e The _body-fixed c mn vitational force vector change with aircraft Qtientation. Since the body-fixed gravitational components only depend on the Euler angles, from Eq. (7.3.15) we haye Wey —sin(9) Wy, + = Wsin(g)cos() (7.5.9) W., cos(¢) cos() which at the equilibrium flight condition gives OWs, OWs, OWr a, pes pe Ea Ag 0. ~WeeaGai 0) a = |Mn Wy Wy, a Ant = |\"a5° a0" “ay \|29} = |e 0 Ol aay | MW, OW,, OW, OW,, \\AY 0 -Wsind 0|\4y} 0g 00 oy By definition, the gravitational force does not produce a moment about Using Eqs. (7.5.7) and (7.5.8) in Eqs. (7.5.5) and (7.5.6) and small-disturbance force approximations are given by