American Institute of Aeronautics and Astronautics

1

Evaluation of the Flying Qualities of a Half-Scale Unmanned Airplane via Flight Simulation

Pedro J. González R.1 Universidad Nacional Experimental Politécnica de la Fuerza Armada, Caracas, 1060, Venezuela

and

Pedro J. Boschetti,2 Elsa M. Cárdenas,3 and Andrea Amerio4 Universidad Simón Bolívar, Naiguatá, Estado Vargas, 1160, Venezuela

The purpose of the present work is to evaluate the flying qualities of a half-scale radio controlled airplane based on the Unmanned Airplane for Ecological Conservation via flight simulation. The stability coefficients of the half-scale airplane were obtained by vortex lattice method using the code Tornado. The radio controlled aircraft simulator CRRCSim was used to perform the virtual flights. Aerodynamic, geometry, thrust and mass characteristics of the airplane were introduced to the simulator by a code written in C++. Virtual test flights were carried out and the flying qualities were obtained. It could be concluded that the half-scale airplane has excellent flying qualities, rating the aircraft in level 1. The simulator is recommended as a training tool for test pilots of the real model.

Nomenclature CLq, CMq = variation of lift, and pitching moment coefficients with pitch rate CLα, CDα, CMα = lift, drag, and pitching moment slopes CLδe, CMδe = variation of lift, and pitching moment coefficients with elevator deflection Cℓp, Cnp, CYp = variation of rolling, yawing and side force coefficients with roll rate Cℓr, Cnr, CYr = variation of rolling, yawing and side force coefficients with yaw rate Cℓβ, Cnβ, CYβ = variation of rolling, yawing and side force coefficients with sideslip angle Cℓδr, Cnδr, CYδr = variation of rolling, yawing and side force coefficients with rudder angle Cℓδa, Cnδa, CYδa= variation of rolling, yawing and side force coefficients with aileron angle c = medium chord gx, gy, gz = components of gravity acceleration vector Ix, Iy, Iz = rolling, pitching and yawing moments of inertia Ixy, Iyz, Izx = products of inertia moments about x-y, y-z and z-x axes L, M, N = rolling, pitching and yawing moments acting on the aircraft m = mass p, q, r = roll, pitch and yaw rate u, v, w = longitudinal, lateral and vertical components of velocity X, Y, Z = components of resultant aerodynamic forces acting on the airplane

I. Introduction ANDLING or flying qualities are those characteristics of an aircraft that govern the ease and precision with which a pilot is capable of executing the tasks required in support of an aircraft role.1 Flying qualities are

related to the dynamic and control characteristics of an airplane.2 The flight simulation could be used to assess the flying qualities of a new aircraft design. These allow to recreate adverse atmospheric condition, and some problems

1 Undergraduate Student, Department of Aeronautical Engineer, Av. La Estancia, Chuao. 2 Assistant Professor, Department of Industrial Technology, Camurí Grande, Senior Member AIAA. 3 Assistant Professor, Department of Industrial Technology, Camurí Grande, Senior Member AIAA. 4.Aggregate Professor, Department of Industrial Technology, Camurí Grande, Senior Member AIAA.

H

48th AIAA Aerospace Sciences Meeting Including the New Horizons Forum and Aerospace Exposition4 - 7 January 2010, Orlando, Florida

AIAA 2010-298

Copyright © 2010 by The Authors. Published by the American Institute of Aeronautics and Astronautics, Inc., with permission.

American Institute of Aeronautics and Astronautics

2

associated with handling as stall, loss of control and spin. Pilot opinion on aircraft response characteristics is then used to evaluate the flying qualities.

The Unmanned Airplane for Ecological Conservation (ANCE) is an unmanned aerial vehicle (UAV), designed to perform reconnaissance missions over the extraction towers and petroleum pipelines to look for possible leakages which could damage the ecosystem of the lake of Maracaibo.3 The ANCE is a twin-boom light airplane, driven by a pusher-propeller joined to two-stroke engine, with a straight rectangular wing of 5.187 m of span, and 3.13 m2 of surface. The maximum take-off and the empty mass are 182.055 kg and 123.28 kg, respectively. Figure 1 shows a virtual geometry of the ANCE. Early wind tunnel test4 and computational fluid dynamic analyses based on potential flow techniques5 had been carried out to date and some modifications have been made to the original design.4-6 Static and dynamic stability analyses have been in process.7,8

A half scale model prototype of the ANCE has been made to perform flight tests and evaluate the flying qualities of the aircraft. Remote control scale airplanes have been used to estimate experimentally the aerodynamic characteristics and flying qualities of unmanned aerial vehicles9,10 and light aircrafts.11 The half-ANCE is a remote control aircraft and requires an in-ground pilot for its operation. The half-ANCE is an airplane built with the basic techniques of radio controlled planes; it is made of wood, foam, and some aluminum joints, with a two-stroke engine of 2.60 kW, and the control surfaces connected to servo-mechanisms activated by a radio control. The main dimensions of the half-ANCE are 2.324 m of longitude; 2.596 m of wingspan, a wing surface of 0.78 m2, maximum take-off mass of 14 kg, and an empty mass of 8 kg.12 Figure 2 shows a picture of this aircraft.

The objective of the research presented here is to evaluate the flying qualities of the half-ANCE via flight simulation. The Cooper and Harper scale is used to describe the flying qualities of this airplane, after simulated flights are performed.

II. Stability Coefficient Calculation The stability coefficients or the non-dimensional stability derivatives represent the partial derivates of the force

and moment coefficients acting over the airplane with respect to the non-dimensional form of the variable indicated by the subscript.13 The aerodynamic forces and moments over the half-ANCE were computed herein using the vortex lattice method code Tornado, version T131b.14

Tornado is an open source code written in Matlab. This code models any number of three-dimensional lifting surfaces and calculates three-dimensional forces and moments via vortex lattice method.14 This method represents a lifting surface covered with quadrilateral panels, on which a horseshoe vortex is superimposed.15 In Tornado, the horseshoe-vortex arrangement is replaced with a vortex-sling arrangement. This one works in the same way, but the legs of the shoe are flexible and consist of seven elements instead of three with the same strength.14 This method solves velocity into each singularity using the law of Biot-Savart. A set of linear algebraic equations for the horseshoe-vortex strengths is obtained when all control points on the wing are summed, satisfying the boundary condition of no flow through the wing. The wing circulation and the pressure differential between the upper and lower surfaces are connected to the vortex strengths. Finally, the forces are obtained by integration of the pressure differentials.15,16 The vortex lattice method have been used in conceptual and preliminary design to compute aerodynamic forces and moments, and pressure distribution.17

Figure 3 shows the vortex-lattice geometry of the half-ANCE, which has 1,100 control points. The fuselage and booms were idealized with cruciform shapes; this body simulation is highly computationally efficient for load distribution.17 The nose and main landing gears are not included in the lattice model since their contribution in potential flow is assumed negligible. A similar vortex lattice representation was previously used to model the full-

Figure 1. The configuration of the ANCE.

Figure 2. The half-ANCE under construction.

American Institute of Aeronautics and Astronautics

3

scale ANCE without wing twist, and the lift and drag coefficients were in good agreement with the experimental data present in Ref. 6. No compressibility effects were assumed in the simulations.

A. Longitudinal and Lateral-Directional Aerodynamics The angle of attack, the elevator deflection angle and the pitch rate were used as variables to study the

longitudinal aerodynamics of the airplane. The lift coefficient at zero angle of attack results equal to 0.374, and the lift slope is 5.4603 rad-1. For five elevator angles, the pitching moment coefficient curves were computed. The pitching moment slope is –2.7330 rad-1, indicating that the airplane is longitudinally statically stable. For a baseline center of gravity at 0.25c, the neutral point is located at 0.751c.

The lateral-directional aerodynamic coefficients were calculated as a function of the sideslip angle, rudder and aileron deflection, and yaw and roll rate, all of them at zero angle of attack. The static roll stability has a negative slope, -0.0057 rad-1, and the directional stability has a positive slope, 0.1719 rad-1. These indicate that the airplane is laterally and directionally statically stable. Table 1 presents the stability coefficients computed for the half-ANCE by Tornado.

III. Flight Simulation of Half ANCE

A. Design of the Virtual Model of the Half-ANCE The simulated flights of the half-ANCE were performed in CRRCSim version 0.9.8.18 CRRCSim is a flight

simulator open source program based on LaRCSim, the NASA Langley Research Center Simulator.18,19 CRRCSim allows the simulation of diverse radio-controlled airplane models in diverse atmospheric conditions. For a given radio-controlled airplane; geometry, mass, inertia moments, stability coefficients and thrust condition of the power plant are required.18

The mathematical formulation used in CRRCSim solves the equations of motion of a rigid-body aircraft in atmospheric flight. Equations (1) and (2) show the force and moment system of equations, respectively,20 in their compact form by assuming a symmetrical distribution of mass with respect to the fore-and-aft plane of symmetry; this means Iyz=Ixy=0. They are not solvable in this form except by numerical techniques, in which integration methods are used and then the terms X, Y, Z, L, M and N can be obtained.19

( )( )( )⎪

⎩

⎪⎨

⎧

⋅+=⋅−⋅+

⋅+=⋅−⋅+⋅+=⋅−⋅+

z

y

x

gmZuqvpwm

gmYwpurvmgmXvrwqum

&

&

&

(1)

( ) ( )( ) ( )( ) ( )⎪

⎪⎩

⎪⎪⎨

⎧

=−−+−

=−−+−

=−−+−

NpqIIqrpIrI

MprIIprIqI

LqrIIpqrIpI

yxzxz

xzzxy

zyzxx

&&

&

&&

22 (2)

Table 1. Stability coefficients of the half-ANCE given at 1/rad.

CDα 0.2153 CYδa 0.0052 CLα 5.4603 Cℓβ -0.0057 CLq 20.4159 Cℓp -0.8761 CLδe 0.6761 Cℓr 0.1249 CMα -2.7330 Cℓδr -0.0057 CMq -51.5771 Cℓδa -0.2005 CMδe -2.8533 Cnβ 0.1719 CYβ -0.5959 Cnp -0.0038 CYp -0.1251 Cnr -0.7506 CYr 1.3762 Cnδr 0.1490 CYδr -0.2693 Cnδa 0.0057

Figure 3. Vortex lattice model used in Tornado.

American Institute of Aeronautics and Astronautics

4

For a specific radio-controlled airplane model, the subroutines describe the aerodynamic and propulsion system. Once combined with the vehicle specific routines, it provides a desktop-based near-real-time simulation of the vehicle. CRRCSim models the six rigid-body degrees of freedom, assumes the vehicle x-z symmetry, uses quaternions to determine the angular orientation (although equivalent Euler angles are also calculated) to avoid the singularity at ±90 degrees pitch angle, incorporates modified atmosphere, turbulence, and steady winds into the simulation, and does not model rotating machinery effects.18,19

The execution of the simulation depends of the model file of the half-ANCE; this is a code with “.xml” extension, exported from a text file written in Wordpad in language C++, a format compatible with the simulator. The model file of the half-ANCE is composed by four sections. The first section explains the type of airplane, the designer, a brief description of it. This part is made principally for authors’ rights. Figure 4 presents this section of the code.

Then, the “air file” or flight file is written. This section includes the geometric characteristics of the airplane (chord, span, wing surface, etc) in English or International System, the non-dimensional stability derivatives, and the cruise flight speed. The stability coefficients present in Table 1 are introduced in a pre-established order inside the program, in radians. The cruise speed was previously estimated by Morales12 and is 16.321 m/s. Figure 5 presents the flight code of the half-ANCE.

Figure 6 shows the section code which describes the mass, inertia moments, thrust conditions and the engine sound. The inertia moments were obtained from remote controlled aircraft similar to the half-ANCE.11 The thrust conditions were estimated experimentally and analytically by Morales, 12 via torque test.

The last section of the code refers to the half-ANCE geometry. This was created in Blender,21 a three-dimensional graphic design program of open source, capable of creating virtual solids with a high level of

precision. The geometry file was saved with the extension “.ac”. Figure 7 shows the geometric model of the half-ANCE drawn by Blender.

The limits of the geometry are defined by hard points. The hard points are the contact points of the aircraft inside the simulator. The wing and empennage tips, the nose of the airplane, the fuselage and the landing gear are the hard points described as hard points. Figure 8 shows this section of the code. The ability to turn the nose wheel joint with the rudder control to the ground control were introduced in this section.

After the code and the geometry of the half-ANCE are completed, the model is ready to be loaded in the simulator. An E-sky EK2-0905A, with 4 channels and USB port control was used to be in command of de virtual airplane. This is similar to a real control of radio controlled planes. Figure 9 shows the virtual model during flight simulation in CRRCSim.

B. Evaluation of the Flying Qualities of the half-ANCE To estimate the flying qualities of the half-ANCE, the “Briefing Guide and Rating Information for Handling

Qualities Experiments” was adapted to evaluate a radio controlled aircraft. This is an inquiry that has the purpose of obtaining the pilot opinion about the flying qualities of a given airplane under different tasks, missions and

Figure 4. Introduction section of the code for thehalf-ANCE.

Figure 5. Flight code for the half-ANCE.

Figure 6. Mass file, engine description and soundversion for the half-ANCE.

American Institute of Aeronautics and Astronautics

5

atmospheric conditions.1 This guide is used to evaluate the pilot confidence factor. The evaluation was developed in five stages

The first stage is about the adaptation of the pilot to the simulator. The pilot has the opportunity to operate the airplane in the track-way and is free to drive any maneuver during the flight. This stage lasts seven minutes. Then, the power plant is turned off, and the pilot must perform an emergency landing. The half-ANCE glide flight is evaluated in this stage.

The second, third and fourth stages have the objective of evaluating the capabilities of the half-ANCE during a specific mission. In this case, the pilot has to perform a 360 deg flight around the track-way, starting with the take-off, a steady flight at constant altitude, and finishing with the landing. The wind velocity is different for each stage. In the second stage is zero, in the third stage 1.524 m/s against take-off direction, and in the fourth stage 9.144 m/s with a direction of 5 deg to the track-way.

The fifth and last stage consists in basic reconnaissance mission over a fixed target in the simulator environment. The pilot must complete a take-off, and a steady flight at constant altitude. Then, the airplane must fly over the target in an “8” pattern at constant altitude. Finally, the radio controlled aircraft has to land. Figure 10 shows the mission graphically. During this mission, the wind velocity was 1.524 m/s with direction of 180 deg to the track-way.

To carry out the evaluation, two pilots were selected based on their experience in the flight of radio controlled aircrafts. Pilot 1 has 35 years of experience and also is a flying instructor for radio controlled planes, Pilot 2 is the Venezuelan national trainer class champion of radio controlled planes and has plenty of experience flying airplanes of similar weight of the half-ANCE.

IV. Results and Discussion After each task was completed, the pilots were inquired into the behavior of the virtual model of the half-ANCE.

For the first experience, the pilots do not report problems. For the second stage the pilots report a satisfactory longitudinal stability and control. At lateral maneuver, the

behavior of the airplane was suitable for the task. At directional maneuvers, small compensation was required to

Figure 8. Graphics and hard points of the half-ANCE.

Figure 7. Half-ANCE geometry created in Blender.

American Institute of Aeronautics and Astronautics

6

keep the turn, although this one was easily achieved by the pilots. Directional trim was highly satisfactory in sideslip angle.

During the second and third stages, the pilots report that the approach and landing speed is higher than that one in radio controlled airplanes of the same size. This observation does not cause any problem during the test. For the severe environmental conditions of the fourth stage, the pilots were able to control the aircraft without an intolerable workload.

For the fifth stage previously mentioned, the pilots completed the mission demonstrating the ability of the airplane to fulfill a recognition mission over a fixed target without pilot workload.

The pilots rate the radio controlled airplane in the Cooper and Harper scale after the evaluation ends. The pilot described that the airplane is controllable during all the maneuvers; it does not cause any intolerable pilot workload and is satisfactory without improvement. This rates the half-ANCE as Level 1.

The pilot confidence factor in the simulator was evaluated by means of the airplane reactions to disturbances during the simulation caused by the environmental conditions and/or aircraft controls. An excellent correlation was found between the surface control movements and the airplane initial reactions. The changes in the performance of the aircraft with the variation of thrust, atmospheric conditions and aircraft orientation were appreciable. If the airplane flies in a yaw angle, a roll rate is induced. The stall is perceptible in the simulator at a given angle of attack. CRRCSim does not include the ground effect and the inadequate behavior of the model during the landing and take-off phases is achieved. CRRCSim has a pilot confidence factor of class A, because it offers a relatively high degree of confidence, although it neglects the ground effect.

V. Conclusion This paper presents a series of simulated flight tests to determine the flying qualities of a radio controlled

airplane. An aerodynamic and stability analysis was done to compute the stability coefficients of the aircraft, and it concludes that the airplane is statically stable.

From the simulated flight tests performed in CRRCSim by the pilots, it could be concluded that the airplane is easily maneuverable and has excellent flying qualities, for level 1 in the Cooper and Harper scale.

The simulator presents a relatively high degree of confidence, but the ground effect is neglected by the flight simulation program, and the pilot confidence factor is class A. CRRCSim installed in a personal computer, with E-sky EK2-0905A control are recommended to be used as ground station in the training of pilots.

Acknowledgments The authors thank the financial support of the Decanato de Investigación y Desarrollo, Universidad Simón

Bolívar, and the Grupo IDEA, both in Caracas, Venezuela.

References 1Cooper, G. E., and Harper, R. P., “The Use of Pilot Rating in the Evaluation of Aircraft Handling Qualities,” NASA TN

D-5153, 1969.

Figure 9. Flight simulation of the half-ANCE in CRRCsim.

Figure 10. Fifth stage flying trajectory.

American Institute of Aeronautics and Astronautics

7

2Nelson, R. C., “Longitudinal Motion,” Flight Stability and Automatic Control, 2nd ed., McGraw Hill, Boston, MA, 1998, pp. 165–167.

3Boschetti, P., and Cárdenas, E., “Diseño de un Avión No Tripulado de Conservación Ecológica,” Engineer Thesis, Department of Aeronautical Engineer, Universidad Nacional Experimental Politécnica de la Fuerza Armada, Maracay, Venezuela, 2003.

4Boschetti, P. J., Cárdenas, E. M., and Amerio, A., “Drag Clean-Up Process of Unmanned Airplane for Ecological Conservation,” Aerotecnica Missile e Spazio, Vol. 85, No. 2, 2006, pp. 53–62.

5Boschetti, P. J., Cárdenas, E. M. and Amerio, A. “Aerodynamic Analysis of the Unmanned Aerial Vehicle for Ecological Conservation,” AIAA Paper 2009-1480, Jan. 2009.

6Boschetti, P. J., Cárdenas, E. M., and Amerio, A., “Increasing the Lift–Drag Ratio of an Unmanned Aerial Vehicle Using Local Twist,” Journal of Aircraft, Vol. 45, No. 1, 2008, pp. 10–15. doi: 10.2514/1.33353.

7Cárdenas, E. M., “Evaluación de la Estabilidad del Avión No Tripulado de Conservación Ecológica (ANCE X-3),” M.S. Thesis, Coordination of Mechanical and Civil Engineer, Universidad Simón Bolívar, Caracas, Venezuela, 2008.

8Cárdenas, E., Boschetti, P., and Amerio, A., “Stability and Flying Qualities of an Unmanned Airplane using the Vortex Lattice Method,” Journal of Aircraft, Vol. 46, No. 4, 2009, pp. 1461–1465. doi: 10.2514/1.44306.

9Howard, R. M., Tanner, J. C., and Lyons, D. F., “Flight Test of a Half-Scale Unmanned Air Vehicle,” Journal of Aircraft, Vol. 28, No. 12, 1991, pp. 843–848.

10Howard, R. M., Bray, R. M., and Lyons, D. F., “Flying-Qualities Analysis of an Unmanned Air Vehicle,” Journal of Aircraft, Vol. 33, No. 2, 1996, pp. 331–336.

11Yip, L. P., Ross, H. M., and Robelen, D. B., “Model Flight Test of a Spin-Resistant Trainer Configuration,” Journal of Aircraft, Vol. 29, No. 5, 1992, pp. 799–805.

12Morales, M. C., “Diseño del Plan de Pruebas de Vuelo para un Avión Radio Controlado: Caso Prototipo Medio ANCE,” Engineer Thesis, Department of Aeronautical Engineer, Universidad Nacional Experimental Politécnica de la Fuerza Armada, Maracay, Venezuela, 2008.

13Etkin, B., Dynamics of Flight, John Wiley & Sons, Inc., Toronto, Canada, 1959. 14Melin, T., “A Vortex Lattice MATLAB Implementation for Linear Aerodynamic Wing Applications,” M.S. Thesis,

Department of Aeronautical and Vehicle Engineer, Kungliga, Tekniska, Högskolan (KTH), Stockholm, Sweden, 2000. 15Moran, J., “Wings of finite span,” An Introduction to Theoretical and Computational Aerodynamics, Dover, New York,

2003, pp. 130–135. 16Bertin, J. J., and Smith, M. L., “Incompressible Flow about Wings of Finite Span,” Aerodynamics for Engineers, 3rd ed.,

Prentice–Hall, Upper Saddle River, New Jersey, 1998, pp. 287–288. 17Miranda, L. R., Elliott, R. D., and Baker, W. M., “A Generalized Vortex Lattice Method for Subsonic and Supersonic Flow

Applications,” NASA CR-2865, Dec. 1977. 18CRRCsimWiki, CRRCSim [online database], URL: http://crrcsim.berlios.de/wiki/ [cited 20 January 2009] 19Jackson E. B., “Manual for a Workstation-based Generic Flight Simulation Program (LaRCsim) Version 1.4,” NASA TM-

110164, April, 1995. 20Rolfe, J., and Staples K., Flight Simulation, Cambridge Aerospace Series, Cambridge, United Kingdom, 2004. 21Blender, Blender Handbook [online database], URL: http://www.blender.org [cited 15 February 2009]