Variable Stability Flight Simulation in Aerospace Engineering Education

Variable Stability Flight Simulation in Aerospace Engineering Education Dr Peter Gibbens, Mr Nathan Rickard The University of Sydney, Sydney, Australia pwg@aeroemech.usyd.edu.au nathan.rickard@aeromech.usyd.edu.au Abstract. The University of Sydney s School of Aerospace, Mechanical and Mechatronic Engineering has been developing its Variable Stability Flight Simulator since 1998. Although the simulator has been functional in this role since 2003, a major overhaul of the software systems has facilitated its first formal integration into the Aerospace Engineering curriculum in September 2006. The simulator is designed to represent the flight dynamics and handling qualities of a broad range of aircraft types in a generic cockpit with Three Degree-of-Freedom motion. Its purpose is to demonstrate to engineering students the effects of aerodynamic design parameters on the flight stability and handling qualities of an aircraft. These effects are observed via changes in the characteristics of the aircraft s modes of motion, hence allowing students to experience first-hand the theoretical analyses that they undertake during the course. The presence of motion in the simulation has been found to be critical to the realism of the variable stability experience. The importance of this simulation capability is that it allows students to directly connect the effects of aircraft configuration characteristics with the nature of the aircraft s dynamic stability and handling qualities. In aircraft design, this means that the handling qualities can be assessed earlier in the design stage, rather than later during a prototyping phase. This paper gives an overview of the simulation system and gives details of the nature of the modelling and how the variable stability capability is achieved. A description of the simulation experiments is given, and the effectiveness of the teaching exercises is appraised in terms of learning improvement. The improvements in graduate engineers knowledge in flight stability and handling qualities provided by this capability will be of great importance to employers in aerospace engineering research and development industries in Australia. 1. INTRODUCTION Aircraft dynamics, stability and handling qualities are of utmost importance in the aerospace industry. It is these characteristics by which an aircraft is judged by pilots. They also have an enormous impact on an aircraft s effectiveness in its role. It is therefore important that aerospace engineering graduates who venture into aircraft design, flight control or flight simulation industries have a well established and exhaustive knowledge of aircraft stability and handling qualities. Through teaching experience it was identified that engineers who lacked the real world application of flight stability and controllability theory struggled to grasp the concepts developed through the course. It is known, however, that personal experiences can make significant changes in a person s understanding. Therefore, to augment the lectures and notes provided to the engineering students in Flight Mechanics courses, a broad experience of simulated aircraft responses was hypothesised to be advantageous to their educational development. It was proposed that by experiencing the flying characteristics of a variety of aircraft types, a comparison of the different handling qualities would yield a greater understanding of underlying principles. Furthermore, it was proposed that to help students understand the importance of fundamental static and dynamic stability parameters, the effects that these parameters have should be demonstrated and preferably experienced first hand. Students gain some flying experience through a course of Ab Initio flight instruction lessons. However this course doesn t provide the experiences required since the course utilises one aircraft type and only focuses on teaching students the basics of flight. On the other hand, simulation is inherently flexible and controllable since governing parameters can easily be modified. This means that simulation can provide a large variety of repeatable experiences. Due to this capability, a simulator was considered a viable method of providing students with a broader educational experience in the engineering aspects of aircraft handling. At the University of Sydney, a Variable Stability Flight Simulator (VSFS) has been developed to help engineers and students to develop a broader understanding of flight stability and controllability characteristics. Most traditional simulators are built to represent a single generic aircraft type. A Variable Stability Simulator is substantially different in that it is not only able to simulate multiple aircraft but also vary individual aerodynamic stability characteristics of any one aircraft. It enables the operator to manipulate a variety of flight simulation parameters and implements these changes in real time. This allows for the rapid reconfiguration or manipulation of the aircraft being simulated and allows changes in aircraft responses to be evaluated quickly. SimTecT 2007 Refereed

2. VARIABLE STABILITY SIMULATION 2.1 Background The VSFS is based on a LINK BOEING 707 simulator that was acquired from QANTAS Airways, see Figure 1. The package consisted of a Boeing 707 cockpit, a 3 axis hydraulic control loading system, as well as a 3 degree of freedom motion base. The system has been re-engineered to run on a cluster of personal computers. This software upgrade reached a stage in 2006 where the simulator was able to facilitate an interactive coursework exercise. The interior of the simulator can be seen in Figure 3, which shows students engaging in a simulation exercise. The displays and controls are all visible, with the IOS operator on the left. Figure 3: Cockpit interior showing displays, controls and IOS Figure 1: University of Sydney s VSFS in motion This system has been operational since 2001 and has had variable stability capability since 2003 [1] for research and development purposes using software based on a Modula 2 software suite [2,3]. The basic system development and the educational foundations for the creation of the simulator are given in [4]. In 2005, a software upgrade was initiated. This aimed to have all core flight simulation, motion base and control loading algorithms running in real-time using MathWorks Simulink xpc/target environment. Outside world, instrumentation and system displays, along with Instructor/Operator Station (IOS) functionality is facilitated through the communication of data with a number of X-Plane [6] clients as shown in Figure 2. This system has been implemented by disabling the internal X-Plane flight model, and communicating all flight, navigation, configuration, weather and instrumentation variables between the flight model and X-Plane using UDP and dedicated plug-in software modules. Figure 2: VSFS computing system architecture 2.2 Cockpit, Motion and Control Loading System The simulator motion base provides pitch, roll and heave degrees of freedom from three vertically mounted hydraulic jacks with 36 stroke. Control is via a conventional yoke and rudder pedals. These are hydraulically loaded, providing realistic feedback. Control loading is essential during any simulation, but sessions can be performed with or without the motion base in operation. 2.3 Flight Modelling All real-time critical computations are hosted on a single PC using the MathWorks Simulink xpc/target environment. The flight model implements the nonlinear equations of aircraft motion and aerodynamic modelling via linear, polynomial or look-up table representations. The system implements system failures, weather conditions and wind and turbulence settings as determined by the IOS. It also models undercarriage and engine dynamics. 3. VARIABLE STABILITY CAPABILITY For educational purposes, the Variable Stability Module (VSM) represents the key capability of the simulation facility. In order to demonstrate the effects of aerodynamic, configurational or design parameters on flight stability and handling qualities, it is crucial that they be altered in real-time so that the changes in handling can be observed in quick succession. Manipulation of the simulation is done through the use of an interface designed using the Matlab GUI 1 framework. Figure 2 shows the VSM used in the experiments described in this paper. It shows two primary menus. The leftmost menu shows how 1 Graphical User Interface

aerodynamic derivatives can be manipulated through a change in slider position or through entering a value directly. The rightmost menu is the primary menu for operation of the simulation. These define the controllability or manoeuvrability of the aircraft, and have a great impact on pilot workload. For example, an aircraft that has low stability (slow responses) will be greatly affected by turbulence. This results in high pilot workload and limits the attention the pilot can put into other cockpit tasks. 4.1 Natural Modes of Aircraft Motion The primary modes of motion for aircraft of conventional configuration are; the Short-period Phugoid Mode (known simply as the Short Period Mode) 2, the Long-period Phugoid Mode (the Phugoid Mode) 3, the Roll Mode 4, the Dutch Roll Mode 5, and the Spiral Mode 6. Figure 2: Variable Stability Module interface Below is a summary of VSM capabilities. The ability to load and save a range of aircraft models, The ability to manipulate the geometry, inertia or aerodynamic derivatives, of a given aircraft. The aerodynamic characteristics can be modified either directly via aerodynamic derivatives or by movement of Centre of Gravity (CG) position so that individual or overall effects can be observed, The ability to load predefined flight scenarios and save scenarios in terms of aircraft position, attitude and speed information, The ability to fail aircraft components such as landing gear, engines and instruments, The ability to inject control input forms additional to the pilot control commands. This is useful for the repeatable demonstration of transient dynamic behaviour, The ability to set or modify wind speed, direction, and turbulence intensity, The ability to start and stop data recording. This feature allows students to take their data away for analysis, and utilisation within the simulation programs that they develop as part of their course. 4. EDUCATIONAL DEMONSTRATIONS The study of flight dynamic and aircraft handling qualities amounts to the assessment of an aircraft s static and dynamic stability. This is done through the analysis of its natural dynamic modes of motion. Static stability addresses the aircraft s tendency to remain at an equilibrium flight condition, or to return to that equilibrium when disturbed from it. Dynamic stability refers to the way an aircraft responds to deterministic or stochastic disturbances. Attributes of particular importance to pilot s are speed of response, damping, overshoot and settling time. Students learn the characteristics of these modes in the flight mechanics courses, and engage in analyses of their properties both mathematically and in simulation. By helping students experience these natural responses they should gain a greater appreciation of the natural responses and handling qualities of an aircraft. 4.2 Aims of Educational Exercises Specifically, the aims of the simulation based laboratory exercise are to improve the understanding of; what effect aerodynamic derivatives have on an aircraft s natural modal dynamics as well as its static and dynamic stability the relationship between handling qualities and modal stability what effect dynamic stability has on pilot workload, and what effect modal stability has on an aircraft s response to turbulence, and how that relates to pilot workload 4.3 Programme of Activities 4.3.1 Outline The majority of the exercise uses a Pilatus PC9 flight dynamic model. This model was chosen because it represents a capable aerobatic training aircraft with middle of the range handling qualities. Substantial differences in the speed of its responses can also be easily detected by students. It is a very good model to highlight the key concepts of natural frequency and damping factor of the primary modes of motion. 2 A rapid second order pitching motion with relatively high natural frequency and damping. It the primary mechanism for manoeuvring the aircraft in the vertical plane though angleof-attack changes 3 A slow roller-coaster like exchange between speed and altitude. It has low natural frequency and low damping 4 Rapid first-order roll subsidence. The primary mechanism for manoeuvring the aircraft in the horizontal plane 5 A coupled roll/yaw motion with relatively high frequency and low damping. Easily excited by sideslipping or rudder input. It has a great impact on lateral manoeuvring. 6 Slow exponentially convergent or divergent coupling in roll/yaw. It is the measure of static lateral-directional stability and has impact on how steady turns are flown.

For the basic PC9, students are given the opportunity to firstly spend time familiarising themselves with the handling of the standard aircraft with a nominal centreof-gravity position. This is then followed by a series of exercises designed to highlight the sensitivity of the Short Period Mode (SPM), which is the primary mechanism of longitudinal manoeuvrability, and the Dutch Roll Mode (DRM), which causes the most significant lateral-directional handling limitations. These effects are achieved by introducing changes in the primary longitudinal and lateral-directional aerodynamic stiffness and damping derivatives. A reduction in aerodynamic stiffness is analogous to a reduction of the static margin, the primary measure of static stability. This is an operational parameter associated with the loading of the aircraft and amounts to backward movement of the aircraft s centre-ofgravity. Reduced aerodynamic damping is more a matter of aerodynamic design but whilst it is somewhat linked to the stiffness, it is demonstrated separately to highlight its effect. The different aircraft configurations are then flown in turbulent weather. This helps students evaluate how the controllability and hence pilot workload is affected. The exercise is concluded with general flying of a Boeing 747 and a jet fighter model to demonstrate the extremes of an aircraft s dynamic response rate. 4.3.2 Simulation Scenarios The simulation exercise involved a sequence of 17 individual scenarios, performed in succession. These scenarios and their individual purposes are described in Table 1. After some familiarisation exercises, control impulse increments are added to the pilot input sequence to disturb the aircraft from its nominal path (being constant pitch orientation, altitude and heading). Students are then instructed to control the aircraft back to its nominal flight path. The purpose of this exercise is for the student to feel the rate, magnitude and number of oscillations of the free response (natural frequency and damping factor of the relevant mode of motion), and to associate these with the handling qualities that they imply. Each situation is followed by a flight with turbulence exciting the natural response. Students can gauge the workload involved in managing the flight path with the stability modified. They are also asked to make changes to various cockpit controls to divert their attention from the flight path. This demonstrates to the students how quickly the aircraft can diverge from the desired flight path if the stability is reduced. Due to time constraints, scenario 15, concerning the effect of airspeed, was not performed during the exercise in 2006. No Configuration Impulse Turbulence Purpose 1 Nominal (no motion) - Off To get a basic feel of the aircraft response in all axes 2 Nominal - Off To refine feel of aircraft response with motion feedback 3 Nominal 2 o Elevator Off To expose students to the natural SPM response 4 Nominal - On To observe the extent to which turbulence necessitates increased attention to pitch and altitude management 5 Reduced Pitch Stiffness 2 o Elevator Off To observe the reduction in pitch response speed and additional difficulty in control of the flight path (pitch) 6 Reduced Pitch - On To observe the increased pitch sensitivity to disturbances Stiffness 7 Reduced Pitch 2 o Elevator Off To observe the increase in pitch oscillation and additional difficulty in control of the flight path (pitch) 8 Reduced Pitch - On To observe the increased pitch sensitivity to disturbances 9 Nominal 3 o Rudder Off To observe natural DRM response 10 Nominal - On To observe the extent to which turbulence necessitates increased attention to sideslip and heading management 11 Reduced Yaw Stiffness 3 o Rudder Off To observe the reduction in yaw response speed and additional difficulty in control of the flight path (heading) 12 Reduced Yaw Stiffness - On To observe the increased yaw sensitivity to disturbances 13 Reduced Yaw 3 o Rudder Off To observe the increase in yaw oscillation and additional difficulty in control of the flight path (heading) 14 Reduced Yaw - On To observe the increased yaw sensitivity to disturbances 15 Nominal at higher speed - Off To observe the dependence of dynamic response speed on airspeed 16 Boeing 747-400 - Off To observe the dramatic reduction in pitch, roll and yaw response speed, and reduced manoeuvrability 17 Jet Fighter (no motion) - Off To observe the dramatic increase in pitch, roll and yaw response speed, and improved manoeuvrability Table 1: Flight simulation scenarios

4.4 The Importance of Motion The subject of simulator motion in training is a much contested issue. Some consider that imprecise replication of motion actually detracts from the quality of training. However, the main function of a motion base is specifically the replication of the transient modal dynamics. The importance of motion becomes evident very quickly at the start of the simulation session described in Table 1. At the start of each session the student is asked to manoeuvre the aircraft in all axes to get a feel for the basic aircraft stability and handling. Their reaction, particularly with students with little flying experience, is to throw the aircraft around aggressively. They are then asked to do the same immediately thereafter with motion switched on. The motion feedback very quickly limits the size and speed of their control inputs, which thereafter become more consistent with those of experienced pilots. This limiting induces some realism to the session. Because students experience and respond to the simulated events in a realistic manner, the motion feedback has been found to be critical in creating a realistic, educational experience. 5. ASSESSMENT OF THE EDUCATIONAL BENEFITS OF THE VARIABLE STABILITY SIMULATOR The evaluation of the educational benefits of this simulation exercise was undertaken via questionnaires taken before and after the exercise. The pre-simulation questionnaire focused on gathering information regarding the students knowledge of the technical aspects of aircraft handling qualities. The postsimulation questionnaire pursued the same topics but focused on students indicating how much their knowledge had improved. The questions were of written form, so some interpretation and judgement of the responses before and after was required to assess knowledge improvement. However, this question design proved positive in that detailed information was received rather than the limited information provided by multiple choice answers. 5.1 Survey of pre-simulation knowledge The following is a list of the first six questions extracted from the pre-simulation questionnaire. 1. How will an aircraft that is longitudinally marginally stable respond differently from a stable aircraft? 2. In the longitudinal sense, how will turbulence affect the controllability of a marginally stable aircraft as compared to a stable aircraft? 3. How does a smaller a) pitch stiffness and b) pitch damping affect the aircrafts controllability? 4. How will an aircraft that has marginal lateraldirectional stability respond differently to a stable aircraft? Refereed 5. How does a smaller a) yaw stiffness and b) yaw damping affect the aircrafts controllability? 6. Rate your understanding [of these concepts] (refer Figure 3) Rate your understanding of the following; Very Poor Average Good Excellent Poor Flight Mechanics [ ] [ ] [ ] [ ] [ ] Generally Flying an Aircraft [ ] [ ] [ ] [ ] [ ] Longitudinal Stability [ ] [ ] [ ] [ ] [ ] Lateral-directional [ ] [ ] [ ] [ ] [ ] Stability Turbulence Effects [ ] [ ] [ ] [ ] [ ] Speed affect on Control Responsiveness [ ] [ ] [ ] [ ] [ ] Figure 3: Self assessment response sheet 5.2 Survey of post-simulation knowledge The post simulation questionnaire addresses the same topics as questions 1-5 of the pre-simulation questionnaire. However students were asked to assess, in each case, how their understanding had changed. They were again asked to rate their understanding of the key concepts after the exercise via Figure 3. 5.3 Results of simulation learning effectiveness A wide variety of data was extracted from the simulation experiment. Information was gleaned from response levels within the questionnaire, the level of understanding expressed in answers, the change in self evaluated understanding levels, and the averaged results from the students. 5.3.1 Self Evaluated Levels of Understanding For analysis, values were assigned to the responses, from 1 (Very Poor) to 5 (Excellent) and collated, see Figure 4. The students self evaluated level of understanding showed a distinct increase in understanding across all subjects. On average across all 5 topics students evaluated an increase in their understanding of 1 rating point, which equates to roughly a 20% improvement. These results were then grouped into those who had indicated the same change in response levels. From these groupings, the initial response levels were averaged. The results are as shown in Table 2. From this analysis a few groups were identified. The students who felt that they had achieved the greatest benefit (+2 to +3 change or 40-60%) were, on average, those who felt they had a limited understanding before the exercise. Students who, on average, felt some to little change, (0 to +1 change), indicated a moderate level of understanding before the exercise. Finally a handful of students, who indicated a decrease in understanding, (-1 change) initially thought they understood the topics well. This would indicate that they were probably trying to re-evaluate their understanding and to accommodate [5] this new experience. Overall this analysis indicates that the SimTecT 2007

students with poorer understanding of the underlying principles gained the most from the experiment. In particular, it shows that the students were brought to about the same level of understanding after the event. This is the goal of the teaching initiative. simulator. Student s also commented on how good the experience was but would like more time within the simulator. Finally when asked if they thought the experience would improve their results, 91% of the students said yes. 5.00 5.00 4.00 3.00 3.5 3.8 3.8 2.9 2.9 2.9 2.7 2.6 2.5 3.7 3.9 3.7 4.00 3.00 3.5 3.6 3.6 3.7 3.6 3.6 2.9 2.9 3.0 2.8 2.8 2.6 2.00 1.00 0.6 1.0 0.9 1.1 1.4 1.0 2.00 1.00 0.6 0.8 0.7 1.0 0.6 0.7 0.00 Flight Mechanics in General Flying An Aircraft Longitudinal Stability Lateral- Directional Stability Turbulence Effects Pre-Simulation Post-Simulation Difference Average Across all Topics 0.00 Stiffness Stability Controllability Turbulence Averaged Result Pre-Simulation Post-Simulation Difference Figure 4: Self-assessed understanding level Rating Change Question a) b) c) d) e) Average 3-1.0 2.0 2.0 1.3 1.4 2 2.2 1.9 2.1 1.9 2.2 2.0 1 2.6 2.9 3.0 2.8 2.7 2.8 0 3.3 3.5 3.6 3.0 3.3 3.4-1 4.0 4.0 - - 4.5 4.2 Table 2: Average pre-simulation rating grouped by change in self assessed understanding levels 5.3.2 Assessor Evaluation of Responses To help cover the wide variety of questions and topics asked within the questionnaires, results were grouped under the particular topics that they addressed. The assessors assigned a mark from 1 to 5 according to correctness. Figure 5 shows the results of the analysis. It can be seen that across all topics an improvement of 12-21% was realised. On average, responses were 15% better in the post simulation questionnaire. 6. CONCLUSION Sydney University has developed a Variable Stability Flight Simulator for educational and research purposes. September 2006 saw the simulator integrated into the flight mechanics curriculum. The simulator was used to create an educational experience that would help student s understand aircraft static and dynamic stability, as well as controllability issues. The educational benefits were evaluated through the use of pre- and post-simulation questionnaires. From student responses the simulator was found to be a worthwhile experience. Through self assessment students indicated that their levels of understanding increased on average by roughly 20%. Through assessor analysis of pre- and post-simulation responses, it was evaluated that responses received after the exercise were 15% better than those received before. From both sets of analysis it is evident that students gained a greater understanding of aircraft stability and controllability characteristics through the use of the Figure 5: Assessor appraised knowledge level Through the use of a Variable Stability Flight Simulator, The University of Sydney is providing an experiential component to augment the traditional didactic educational process. Through these experiences, graduates will be well equipped when entering into the aircraft design, flight control or simulation industries. 7. ACKNOWLEDGEMENTS The integration of the VSFS into coursework in 2006 was achieved via the enormous efforts of students on the new software structure, and with their enthusiastic assistance in running simulator classes. They were; Dylan Reynolds, Eran Medagoda, Michael McWhinnie, Tammy Taylor, John Ducat and Scott Miekle. REFERENCES 1. Scamps, A. Gibbens, P (2003) Development of a Variable Stability Flight Simulator as a Research / Education Tool Paper AIAA-2003-5753 AIAA Modelling and Simulation Technologies Conference and Exhibit, Austin, Texas, 11-14 August, 2003. 2. Allerton, D.J. (1999) The Design of a Real-Time Engineering Flight Simulator for the Rapid Prototyping of Avionics Systems and Flight Control Systems Transactions of the Institute of Measurement and Control, vol. 21, no.2/3, ISN 0142-3312. 3. Allerton, D.J. (1998) A Distributed Approach to the Design of a Real-Time Engineering Flight Simulator Paper A98-31444, 21st ICAS Congress, Melbourne, Australia, 13-18 September, 1998. 4. Rickard, N. Gibbens, P. W. (2006) Variable Stability Simulator for the Experiential Education of Flight Dynamic Principles. SimTecT, Melbourne, Australia, June 2006 5. Piaget, J. (1983). "Piaget's theory". In P. Mussen (ed). Handbook of Child Psychology. 4th edition. Vol. 1. New York: Wiley. 6. X-Plane by Laminar Research: PC based flight simulation system. See www.x-plane.com