Evaluation of High-Angle-of-Attack Handling Qualities for the X-31A Using Standard Evaluation Maneuvers

NASA Technical Memorandum 104322

Evaluation of High-Angle-of-

Attack Handling Qualities for

the X-31A Using Standard

Evaluation Maneuvers

Patrick C. Stoliker and John T. Bosworth NASA Dryden Flight Research Center Edwards, California National Aeronautics and Space Administration Dryden Flight Research Center Edwards, California 93523-0273 PIO pilot-induced oscillation PST poststall RPC roll performance classification SSLA slow-speed line-abreast STEM standard evaluation maneuver INTRODUCTION Controlled flight at high angles of attack (AOAs) provides a modem fighter aircraft with the ability to turn rapidly, providing enhanced nose-pointing capability. The ability to accurately point the nose of the aircraft in a timely manner is the basis for handling qualities criteria and ratings.

With the exception of recent flight programs such as the F-16 Multi Axis Thrust Vectoring (MATV), 1 F-18 High Angle of Attack Research Vehicle (HARV), 2 and X-29A, 3 an opportunity for flight test evaluations at poststall (PST) angles of attack has not existed. The Handling Qualities Military Standard (MIL-STD-1797) 4 provides a summary of criteria for handling qualities that have been derived primarily for a more conventional flight regime. Simulation-based criteria 5,6 have been developed to specifically address flight in the PST regime. Additional criteria 7-9 have also been developed to address handling qualities of modem augmented aircraft. Using the X-31A linear simulation, analytic evaluations of the handling qualities at high AOAs were performed to predict the characteristics of this aircraft.I° Designed specifically for investigation of flight in the PST regime, the X-31A Enhanced Fight- er Maneuverability (EFM) program evaluated the benefits of thrust vectoring in a close-in combat (CIC) environment with emphasis on PST or flight at greater than 30 ° AOA. Following the com- pletion of the original X-31A CIC objectives, a high-AOA handling qualities flight test program was performed. Standard evaluation maneuvers 11 (STEMs) were used to assess longitudinal and lateral gross acquisition and fine tracking at high AOAs. Pilot ratings and comments were collected immediately following each maneuver. These data were analyzed and compared with existing han- dling qualities criteria.

The development and preparation for the high-AOA handling qualities flight testing, a summa- ry of the flight test data, a comparison of the results with existing handling qualities criteria, and a summary of lessons learned during the flight testing are covered in this paper.

AIRCRAFT DESCRIPTION The X-31A airplane (fig. 1) is a single-seat fighter configuration with an empty weight of approximately 12,000 Ibm that uses a single GE-F404-400 engine (General Electric, Lynn, Massachusetts). Fuel capacity is approximately 4000 Ibm. Two aircraft were built by Rockwell In- ternational (Downey, California) and Daimler-Benz Aerospace (Germany). The wing planform is EVALUATION OF HIGH-ANGLE-OF-ATTACK HANDLING QUALITIES FOR THE X-31A USING STANDARD EVALUATION MANEUVERS Patrick C. Stoliker and John T. Bosworth NASA Dryden Flight Research Center Edwards, CA ABSTRACT The X-31A aircraft gross-acquisition and fine-tracking handling qualities have been evaluated using standard evaluation maneuvers developed by Wright Laboratory, Wright-Patterson Air Force Base. The emphasis of the testing is in the angle-of-attack range between 30 ° and 70 ° . Lon- gitudinal gross-acquisition handling qualities results show borderline Level 1/Level 2 perfor- mance. Lateral gross-acquisition testing results in Level 1/Level 2 ratings below 45 ° angle of attack, degrading into Level 3 as angle of attack increases. The fine-tracking performance in both longitudinal and lateral axes also receives Level 1 ratings near 30 ° angle of attack, with the ratings tending towards Level 3 at angles of attack greater than 50 °. These ratings do not match the expec- tations from the extensive close-in combat testing where the X-31A aircraft demonstrated fair to good handling qualities maneuvering for high angles of attack. This paper presents the results of the high-angle-of-attack handling qualities flight testing of the X-31A aircraft. Discussion of the preparation for the maneuvers, the pilot ratings, and selected pilot comments are included. Evalu- ation of the results is made in conjunction with existing Neal-Smith, bandwidth, Smith-Geddes, and military specifications.

NOMENCLATURE AOA angle of attack, deg CHR Cooper-Harper rating close-in combat CIC EFM Enhanced Fighter Maneuverability HARV High Angle of Attack Research Vehicle HUD head-up display KIAS knots indicated air speed LOES lower-order equivalent systems MATV Multi Axis Thrust Vectoring MAX AB maximum afterburner

EC 94-42478-1

Figure 1.X-31A airplanein poststallflight.

a double delta with an inboard leading-edge sweep of 56.6 ° and an outboard sweep of 45 °. The wing area, span, and mean chord are 226.3 ft 2, 22.833 ft, and 12.35 fl respectively. Four trailing- edge flaps on the wing can be deflected symmetrically for pitch control and differentially for roll control. The leading-edge flap is scheduled to deflect as a function of AOA. The aircraft has an all- moving canard for pitch control and to meet the requirement for aerodynamic recovery from ex- treme AOAs. The vertical tail contains a rudder for directional control at AOAs less than 40 °. Pitch and yaw moments can be generated by the three thrust-vector vanes. The inlet lip is moveable and is deflected as a function of AOA. These control effectors were all integrated into a control system 1°,12 that provided the capability for good control throughout the AOA range.

In the longitudinal axis, the control system uses load factor command to a maximum 30 ° AOA.

In the PST regime, from 30 ° to 70 ° AOA, deflections of the control stick command a specific AOA. Three in. of aft stick commands 30 ° AOA; and full deflection, or 4.5 in., commands 70 ° AOA. This characteristic results in a stick sensitivity in AOA command of 33.3 deg/in of stick de- flection. The nominal stick force is 5 lbf/in. The rate of change of AOA command was limited to 25 deg/sec. The longitudinal control system also includes an AOA command limiter that was set by the pilot. The AOA limiter provided the capability for the pilot to set the limit for the AOA com- mand in 5 ° increments from 30 ° to 70 ° AOA.

For the lateral--directional axes, deflection of the control stick commands velocity-vector roll rate. The roll stick deflects 3 in. left and right. The maximum allowable roll rate is 240 deg/sec at a low AOA. In PST, the velocity-vector roll rate is between 30 and 50 deg/sec, scheduled as a func- tion of dynamic pressure and AOA. During envelope expansion, the pilots had difficulty using full- lateral stick when using full-aft pitch stick because of interference with their legs. To accommodate this, the lateral-stickdeflection-to-roll command gain waschanged linearly from 1 to 2 between

30° and70° AOA. This changeresultsin full-roll rate command beinggenerated with half-stick

deflectionat70° AOA. Therudderpedalscanbeusedto command sideslipatlow AOAs, andtheir

commandauthorityis reducedto0 ° at an AOA greater than 40 °. The basic operation of the aircraft is designed for "feet-on-the-floor" flying.

The primary source of information for the pilot was the head-up display (HUD) (fig. 2). This display contained a conventional pitch ladder and heading display. Altitude and altitude rate were displayed on the upper right, while airspeed and Mach number were shown on the upper left. On the left side of the display were two tapes that showed the AOA and load factor. These data were displayed digitally at the top of the tapes. The current AOA command limit was indicated by an ar- row next to the AOA tape. The HUD also contained a 2-mrad fixed pipper, depressed 2 ° from the waterline with an inner 20-mrad and outer 40-mrad reticle. Flight test instrumentation allowed in- flight recording of the HUD.

F Rate of Mach number 350 000 010 -1200 climbldescent Airspeed _ I I I I I _"_6_ _ Altitude Aircraft g _ +4.6 +3.8 Jl___l _ Sideslip angle Heeding Angl_o_ +5.0 attack +9 - +8- s -VV- s +7- Aircraft reference Angle-of- +6 - attack • /-- Pitch ladder limit---_ +5 -

4_q/-g

f "_ _ 40-mred outer reticle +2-- +3-- J ___ 20-mrad inne_reticle (3£110 _ +1 - / _> +0- / % _2_ / V, \ _ g [ _.,-- Bank angle 960376 Figure 2. Head-up display symbology.

AIRCRAFT SIMULATIONS Three simulations were used in the preparation for and analysis of the flight test maneuvers: a six-degrees-of-freedom, nonlinear simulation 13 that incorporated flight hardware and a fixed-base cockpit mock-up; a batch version of the six-degrees-of-freedom simulation; and linear simulations of the longitudinal and lateral--directional axes.

The cockpit for the piloted simulation incorporated the pilot displays and controls. A 5 ft by 6.5 ft fiat screen projection provided the pilot a limited view out of the cockpit. The field of view for this screenwasapproximately 30 deg laterally and 20 deg vertically. One feature of the simulation was the capability to project a target aircraft that could be used for practicing the maneuvers. The target aircraft trajectory could be "flown" and recorded to allow for training with a repeatable maneuver.

The batch version of the simulation was used primarily for the generation of linear state-space models. Using these plant descriptions from the batch simulation, the linear simulation was used to generate transfer functions for use in the handling qualities criteria. These transfer functions could be used directly in the criteria evaluation or in the calculation of lower-order equivalent systems (LOES) parameters. The aerodynamic models for the linear simulation were fourth order. The con- trol system included sensor models, filters, and high-order actuator models.

HANDLING QUALITIES EVALUATION During the X-31A flight testing, an informal handling qualities evaluation was conducted dur- ing the CIC testing and formal evaluation using STEMs. The CIC testing was performed to evalu- ate the effectiveness of PST maneuverability. 14 From a predetermined set of starting conditions, the X-31A airplane was flown against an adversary aircraft. Both pilots were free to maneuver as re- quired to try to establish a tracking situation. In addition to the test pilots assigned to the program, service pilots demonstrated the ability to become quickly familiar with the aircraft and to fly ag- gressively without any limitations on control stick inputs in the PST flight regime. In all of the CIC engagements, the pilots flew the aircraft aggressively to try to "win" the simulated combat.

CIC testing is used as a comparison with the formal handling qualities testing because of the demonstrated ability of the X-31A pilots to successfully accomplish gross acquisitions and per- form fine tracking in a high-gain environment at high AOAs. During CIC evaluation, the X-31A aircraft was generally able to outperform adversary aircraft by using PST maneuvers. Although no handling qualities ratings were made during these tasks, the general consensus was that the X-31A aircraft had good handling qualities (Level 1 or Level 2) in this flight regime, and no major han- dling qualities deficiencies were noted. Similar handling qualities were expected from the STEM evaluations. A disadvantage of using CIC to evaluate handling qualities is that the AOA varies con- siderably and the handling qualities characteristics cannot be sorted out as a function of AOA.

A method for providing consistent techniques for flight-test handling quality evaluation has been addressed by the definition of a set of STEMs. 11 These maneuvers can obtain evaluations at a constant AOA that can then be compared to analysis. During a limited flight test evaluation, the X-31A aircraft used four evaluation maneuvers: three STEMs, and a maneuver developed from CIC testing. The flight test maneuvers were derived from STEM 10 (High-AOA Longitudinal Gross Acquisition), STEM 3 (High-AOA Lateral Gross Acquisition) and STEM 2 (High-AOA Tracking). The chase airplane for the X-31A aircraft, an F-18 aircraft, was used as the target air- plane. Data were collected using a pilot rating sheet that was completed immediately following each maneuver, postflight interviews, a review of in-flight video recordings made through the HUD, and a comprehensive set of telemetered data. The techniques for performing these maneu- vers were developed using experience gained from the F-18 HARV program. In order to emulate the acquisition and tracking tasks that were performed during the CIC investigation using the

X-31A airplane,an additionalevaluationmaneuver was flown. This maneuverusedslow-speed

line-abreast (SSLA) initial conditionsandresultedin acquisitionandtrackingtasksat a varietyof

AOAs. The formal handlingqualitiestestingcovereda 5-monthperiod and usedfive different

pilots duringthe performance of 19flights.

Flight preparationinvolvedpracticein the simulatorto establishguidelinesfor maneuvering

the testandtargetaircraft.Theinitial startingpositions,targetmaneuver, andtiming weredefined so that the gross-acquisition or fine-trackingtasksoccurredat a specific AOA. To accurately achieveconsistent initial startingconditions,two operational groundradarswererequiredbecause the X-31A aircraftwasnot equippedwith a radar.During testing,the pilots couldachieveconsis- tentspacingwithout thegroundradars by comparingtherelativetargetsizewith theHUD reticle.

Pilot comments wererecordedon a questionnaire immediatelyfollowing eachmaneuver. The

completionof the questionnaire requireda pilot rating using the Cooper-Harper rating (CHR)

system 15(fig. 3) andanevaluation of theconfidence class(fig. 4). Theconfidence classratingwas usedto help assess the effectiveness of themaneuvers for ratinghandlingqualities.Changingthe Aircraft Demands on the pilot in selected Pilot I characteristics task or required operation* rating I Excellent Pilot compensation not a factor Adequacy for selected Highly desirable for desired performance task or required operation* Good Pilot compensation not a factor Negligible deficiencies for desired performance Fair - some mildly Minimal pilot compensation unpleasant deficiencies required for desired performance l Yes Minor but annoying Desired performance requires deficiencies moderate pilot compensation Is it Moderately objectionable Adequate performance requires satisfactory without warrant deficiencies considerable pilot compensation Deficiencies improvement Very objectionable but Adequate performance requires tolerable deficiencies extensive pilot compensation Adequate performance not Major deficiencies attainable with maximum tolerable Yes pilot compensation. Controllability not in question Is require Considerable pilot compensation with a tolerable pilot Deficiencies _-4 Improvement Major deficiencies is required to control workload?

Major deficiencies Intense pilot compensation is required to retain control Yes I I Is it Control will be lost during some Improvement I-_ Major deficiencies controllable?

mandatory portion of required operation

! ]

* Definition of required operation involves designation of flight phase and/or subphases with accompanying conditions.

960377 Figure 3. Cooper-Harper rating scale.

initial conditions or additional practices improved the confidence class ratings. Pilot comments were solicited regarding difficulty, predictability, aggressiveness effects, and control system ef- fects. Following these comments, a pilot-induced oscillation (PIO) rating (fig. 5) and a second CHR were recorded. For lateral gross acquisitions, a rating using the roll performance classifica- tion (RPC) 16 (fig. 6) was also solicited. The RPC was developed through simulation studies to ad- dress the open-loop nature of lateral gross acquisition. The RPC is intended to judge the initial rate and rate onset and is not based on the ability to arrest the roll rate. The pilot comments were tran- scribed using the HUD video recordings that were available for every flight.

Each maneuver was also evaluated using the telemetered data. Linear models were calculated for each maneuver based on the AOA, airspeed, altitude, and estimated fuel state. The linear models were used to generate the parameters and frequency responses required for the handling qualities criteria.

Classification Description A The pilot rating was assigned with a high degree of confidence.

B The pilot rating was assigned with only a moderate degree of confidence because of uncertainties introduced by moderate differences in environmental conditions, or in aircraft configuration or state, or in task, from what was desired.

C The pilot rating was assigned with minimum confidence because of important differences between the desired and actual environmental conditions, aircraft configuration or state, or task, requiring considerable pilot extrapolation.

960378 Figure 4. Classification of pilot confidence factor.

Numerical Description rating No tendency for pilot to induce undesirable motions.

Undesirable motions tend to occur when pilot initiates abrupt maneuvers or attempts tight control. These motions can be prevented or eliminated by pilot technique.

Undesirable motions easily Induced when pilot Initiates abrupt maneuvers or attempts tight control. These motions can be prevented or eliminated but only at sacrifice to task performance or through considerable pilot attention and effort.

Oscillations tend to develop when pilot initiates abrupt maneuvers or attempts tight control. Pilot must reduce gain or abandon task to recover.

Divergent oscillations tend to develop when pilot initiates abrupt maneuvers or attempts tight control. Pilot must open loop by releasing or freezing stick.

Disturbance or normal pilot control may cause divergent oscillation.

Pilot must open control loop by releasing or freezing stick.

960379 Figure 5. Pilot-induced oscillation rating scale.

Improvements In Roll performance for Numerical roll performance mission effectiveness Enhancing - None warranted tactically superior Satisfactory - May be warranted, but not required mission requirements met 3 Unsatisfactory - Required mission requirements not met Unacceptable - Mandatory 4 tactically useless 960380 Figure 6. Roll performance classification.

Longitudinal Gross Acquisition STEM 10 was used as the basis for the longitudinal gross-acquisition task. The X-31A airplane started 3000 ft in trail of the target aircraft. At the initiation of this maneuver, the target aircraft en- tered a steady turn to the conditions indicated in Table 1. After predetermined time delays, the X-31A pilot selected maximum afterburner (MAX AB), roiled the aircraft so that the target aircraft was in the pitch plane, and then aggressively pulled to capture the target in the pitch plane within the criteria (fig. 7). The pipper and reticles in the HUD provided a reference for evaluating the gross-acquisition and fine-tracking tasks. The goal of the tasks was not to drive the pipper to the target, but to acquire or track the target within the specified criteria in relation to the pipper. The timings were selected so that the gross acquisition would occur at the desired AOA of either 30 °, 45 °, or 60 °. The AOA limiter was not used during this testing. Table 1 shows the maneuver timing for each flight condition.

Lateral Gross Acquisition Lateral gross acquisitions were flown using STEM 3 as a baseline. For these maneuvers, the target aircraft established a steady turn at specified conditions, and the pilot of the X-31A aircraft maneuvered the aircraft to the target AOA (30 °, 45 °, or 60 °) at maximum afterburner. Depending on how rapidly the pilot applied aft stick, the aircraft could be at 1 g or an elevated load factor at the desired AOA. When the target aircraft was at a prespecifled angle away from the nose of the X-31A aircraft, the X-31A aircraft was maneuvered aggressively using only lateral stick to acquire the target in the roll plane within the criteria (fig. 8). Table 2 shows the initial conditions for these maneuvers. To assist the pilot in remaining at the targeted AOA and to try to constrain the maneu- ver to the lateral axis, the AOA command limiter was set to the desired value.

Table 1. Task descriptions for longitudinal gross acquisition.

Maneuver Maneuver description timing Flight condition T=0 Target begin maneuver: MAX AB, constant 30 ° AOA, Mach 0.45 20 ° AOA turn, maintain 200 KIAS.

X-31A advance throttle to MAX AB.

T + 4 sec T + 4 sec X-31A roll in plane with target, perform rapid pull to 30 ° AOA.

T=0 Target begin maneuver: MAX AB, constant 30°AOA, Mach 0.60 20 ° AOA turn, maintain 200 KIAS.

X-31A advance throttle to MAX AB.

T + 4 sec T + 5 sec X-31A roll in plane with target, perform rapid pull to 30 ° AOA.

T=0 45°AOA, Mach 0.50 Target begin maneuver: MAX AB, constant 25 ° AOA turn, maintain 170-180 KIAS.

X-31A advance throttle to MAX AB T + 5 sec T + 7 sec X-31A roll in plane with target, perform rapid pull to 45 ° AOA.

T=0 60°AOA, Mach 0.50 Target begin maneuver: MAX AB, constant 25 ° AOA turn, maintain 170-180 KIAS.

T + 5 sec X-31A advance throttle to MAX AB.

T + 8 sec X-31A roll in plane with target, perform rapid pull to 30 ° AOA.

Desired: Aggressively acquire target within 25* or 40** mrad longitudinally of pipper with no overshoot and within a desirable time to accomplish the task.

Aggressively acquire the target within 25* or 40** mrad longitudinally Adequate: of pipper with no more than 1 overshoot and within an adequate time to accomplish the task.

* Criterion for 30°and 60 ° AOAs ** Criterion for 45 ° AOA 960381 Figure 7. Performance criteria for longitudinal gross acquisition.

Aggressively acquire target within 25* or 40** mrad laterally of pipper with Desired: no overshoot and within a desirable time to accomplish the task.

Adequate: Aggressively acquire the target within 25* or 40** mrad laterally of pipper with no more than 1 overshoot and within an adequate time to accomplish the task.

* Criterion for 30 ° AOA ** Criterion for 450 and 600 AOAs 960382 Figure 8. Performance criteria for lateral gross acquisition.

Table 2. Task descriptions for lateral gross acquisition.

Angle of Attack Test Condition Test Description 30 ° 170 KIAS F-18 (target): Roll and pull to 170 KIAS/30 ° AOA, X-31A 1500 ft Echelon adjust power/attitude to maintain conditions.

and behind F- 18 X-31A: MAX AB, pull to 30 ° AOA. When target is AOA limit = 30 ° 30 ° off nose, acquire target laterally.

45 ° 170 KIAS F-18 (target): Roll and pull to 170 KIAS/30 ° AOA, X-31A 1500 ft Echelon adjust power/attitude to maintain conditions.

and behind F- 18 X-31A: MAX AB, pull to 45 ° AOA. When target is AOA limit = 45 ° 300-45 ° off nose, acquire target laterally.

60 ° 170 KIAS F-18 (target): Roll and pull to 170 KIAS/30 ° AOA, X-31A 1500 ft Echelon adjust power/attitude to maintain conditions.

and behind F- 18 AOA limit = 60 ° X-31A: MAX AB, pull to 60 ° AOA. When target is 300-45 ° off nose, acquire target laterally.

Fine-Tracking Evaluation The fine-tracking evaluation consisted of two phases. Phase 1 testing was performed at AOAs of 10% 15, ° and 20 ° to establish a reference point for comparison with other conventional AOA evaluations and testing in the PST regime. During phase 1, fine tracking was performed only in the longitudinal axis. Phase 2 testing, based on STEM 2, evaluated fine tracking at AOAs of 30 °, 45, ° and 60 ° for the longitudinal and lateral axes. The AOA command limiter was not used in fine- tracking evaluations.

Initial testing in Phase 2 concentrated on longitudinal fine-tracking evaluations while the ma- neuver setup was refined. Because only one axis was being evaluated at a time, the maneuver had to be set up with the target approximately in the reticle so that maneuvering could be performed only in the axis being evaluated. After an acceptable set of starting conditions was developed, the same setup was used for the longitudinal and lateral tracking tests at each AOA. The X-31A pilot

wouldpracticethemaneuver toensure thatthesetupwouldresultin the desired AOA and then per-

form the maneuver twice. First, a longitudinal fine-tracking task was performed and pilot ratings were given. Then a second maneuver was performed where lateral tracking and ratings would be done. Table 3 shows the maneuver sequence and figure 9 shows the criteria. To test the ability to make precise longitudinal changes in track point, the maneuver description called for the pilot to move the pipper from nose to tail. Similarly, the lateral tracking task required the movement of the pipper from wing tip to wing tip.

Table 3: Task descriptions for fine tracking.

Angle of Attack Test Condition Test Description 10 ° 0.80 Mach number F-18 (target): Roll and pull to 3 g, adjust power/ attitude to maintain conditions.

X-31A 1500 ft behind F-18 X-31A: Roll and pull to 10 ° AOA for longitudinal tracking.

(Repeat with target at 1.8 g and initial Mach number of 0.60.)

15 ° 0.75 Mach number F-18 (target): Roll and pull to 3.5 g, adjust power/ attitude to maintain conditions.

X-31A 1500 ft behind F-18 X-31A: Roll and pull to 15 ° AOA for longitudinal tracking.

(Repeat with target at 2.1 g and initial Mach number of 0.55.)

15 ° 0.70 Mach number F-18 (target): Roll and pull to 4.0 g, adjust power/ attitude to maintain conditions.

X-31A 1500 ft behind F-18 X-31A: Roll and pull to 20 ° AOA for longitudinal tracking.

(Repeat with target at 2.4 g and initial Mach number of 0.50.)

30 ° 180 KIAS F-18 (target): Roll and pull to 180 KIhS/25 ° AOA, X-31A 1500 ft behind F-18 adjust power/attitude to maintain conditions.

X-31A: MAX AB, at 20 ° angle off, roll and pull to 30 ° AOA for tracking.

45 ° 180 KIAS F-18 (target): Roll and pull to 160 KIAS/30 ° AOA, X-31A 1500 ft behind F-18 adjust power/attitude to maintain conditions.

X-31A: MAX AB, at 30 ° angle off, roll and pull to 45 ° AOA for tracking.

60 ° 180 KIAS F-18 (target): Roll and pull to 170 KIAS/30 ° AOA, X-31A 1500 ft behind F-18 adjust power/attitude to maintain conditions.

X-31A: MAX AB, at 45 ° angle off, roll and pull to 60 ° AOA for tracking.

Desired: Pipper within +/- 5 mrad band for 50 percent of task and within +/- 25 mrad for the remainder of the task; no objectionable PIO.

Adequate: Pipper within +/- 5 mrad band for 10 percent of task and within +/- 25 mrad for the remainder of the task; no objectionable PIG.

_0_3 Figure 9. Performance criteria for fine-tracking tasks.

Combined Maneuvers Pilots consistently commented on the difference between the types of maneuvers used in the handling qualities evaluations and the maneuvering performed during CIC. To address the per- ceived handling qualities differences between CIC and STEMs, a combined maneuver was evalu- ated during one flight. For this maneuver, the starting conditions were those of the SSLA setup from the CIC flight tests. The X-31A and F-18 aircraft started side by side at the same speed and altitudem215 knots indicated airspeed (KIAS) and 25,000 ft--separated by 1500 ft. For the han- dling qualities evaluation, the maneuvering began on the call of the X-31A pilot. The aircraft initially turned towards each other with the X-31A aircraft going over the target aircraft. Then the F-18 aircraft performed a single heading reversal and maintained a steady turn at 30 ° AOA and 170 KIAS. The X-31A aircraft maneuvered as required to acquire and track the target. Multiple acquisitions were achieved by lagging off of the target aircraft and then maneuvering aggressively to reacquire the target. Figure 10 shows the rating criteria.

Gross acquisition Desired: Aggressively acquire target within 25 mrad of pipper with no overshoot and within a desirable time to accomplish the task.

Aggressively acquire the target within 25 mrad of pipper Adequate: with no more than I overshoot end within an adequate time to accomplish the task.

Desired: Fine tracking Pipper within +1- 5 mrad band for 50 percent of task and within +/-25 mrad for the remainder of the task; no objectionable PIO.

Adequate: Pipper within +1-5 mrsd band for 10 percent of task and within +1- 25 mrad for the remainder of the task; no objectionable PIG.

960384 Figure 10. Performance criteria for the combined maneuvers.

HANDLING QUALITIES RESULTS Handling qualities testing was done during 19 flights over a 5-month period in 1994.

Five pilots participated in the testing, using both X-31A aircraft. When acquiring the pilot

commentsat the completionof eachmaneuver, a CHRwassolicitedbeforeandafterthe detailed

comments. Havingthe pilot repeattheCHRattheendof thequestionnaire alloweda reassessment

of theratingin light of themoredetailedcomments anddiscussion. Thesecond ratinggivenis used asthe reference for this report.Thefirst andsecond CHR weregenerallythe same.

LongitudinalGrossAcquisition

Longitudinalgross-acquisition taskswereflown on five flights by threepilots.The initial tim- ingsfor thesemaneuvers werebased onthepilotedsimulation.Because of the limited field of view provided by the projection television display in the simulator, transferring this simulation experi- ence to flight was difficult. A total of 49 gross-acquisition tasks were performed with 28 receiving pilot ratings. Twenty tasks were practices and one task was an unsuccessful gross acquisition. Elev- en of the practice maneuvers occurred on the first flight. Results from this first flight were used to refine the maneuver timing, and consequently, each of the other pilots typically required only one practice at each target AOA. The goal was to collect data at 30 °, 45 °, and 60 °, with the actual AOA for acquisition falling between 22 ° and 65 ° .

It became apparent after testing started that horizontal bands located in relationship to the pip- per as specified by the performance criteria (25 or 40 mrads) rather than a circular reticle would have provided the pilot with the appropriate reference for the task. Review of the HUD data and te- lemetry data showed that, during acquisition, if the target was entering the HUD field of view on either side of the reticle, a lateral input to bring the target within the reticle often occurred.

Figure 11 shows the CHRs plotted as a function of AOA. These data show a trend for CHRs in- creasing from "2" to "4" as AOA increased from 20 ° to 65 °. The one CHR of "5" was the result of a very large overshoot during capture. For these maneuvers, the pilots developed a technique to put in a nearly full-aft stick initial input and then leading the AOA capture with forward stick. As a compensation technique, Pilot B noted, "I'm starting to get a feeling for when I need to lead the pitch rate to get the capture task." Figure 12 shows this phenomenon where maximum pitch rate during the maneuver is plotted as a function of AOA. At the higher AOAs, the pilots would hold aft stick longer, allowing a larger buildup of pitch rate prior to the countering control movement.

A confidence class rating of "A" was given for all but one of the maneuvers, meaning that the pilots' ratings were assigned with a high degree of confidence. The PIO ratings for 18 of the 28 tasks were "1," indicating that the pilots observed no undesirable motions. The remaining tasks received a PIO rating of "2," indicating undesirable motions that did not compromise task perfor- mance. These data indicate that the X-31A aircraft would have Level 1 performance at less than 40 ° AOA. The trend would be for borderline Level 1/Level 2 at AOAs greater than 40 °. These rat- ings matched the expectations from the CIC testing.

In conjunction with these pilot ratings, a number of pilot comments add insight into the data.

During the testing where the target AOA was 30 °, Pilot A reported, "Thirty is the critical point. It' s better [for the evaluation] to be above 30; below 30 is too easy." For the PST AOAs, the pilots

consistentlynotedthatthestickforcesweretoo heavyandthatthe stickmotionwastoo large.For

the acquisitions at45oand60°,thepilotsnoteda lateraldisturbance thatcomplicatedthetask.This

disturbance wasnotedduringenvelope expansion andwasattributedto asymmetricforebodyvor-

tex coresthatchanged asa functionof AOA.

10-- Pilot O A 9-- [] B

Oc

8-- 7-- Level 3 6-- Cooper-Harper [] 5-- rating O 4-- Level2 _000 0

3-- Level 1 0 _ _' 0

2-- O0 __0

1--

t I t I I I I

0 10 20 30 40 50 60 70 Angle of attack, deg 960385 Figure 11. Cooper-Harper ratings as a function of angle of attack for longitudinal gross acquisition.

40 -- Pilot

O A O

0 o

[] 35-- [] B

O []

Oc

30-- [] []

O0

25-- O•

oO o

Maximum pitch rate, 20 deg/sec

O

15-- 10-- 5--

I I I I I I I

10 0 20 30 40 50 60 70 Angle of attack, deg 960386 Figure 12. Maximum pitch rate as a function of angle of attack for longitudinal gross acquisition.

Figure13showsanexample time historyfor grossacquisition at45 ° AOA. To show pitch-stick

movement, a comparison of AOA command with AOA response and pitch-rate response are shown. Nearly full-aft stick is used to initiate the maneuver, followed by a number of stick inputs on the order of one-half inch. These small stick displacements result in a rate-limited AOA com- mand. An inspection of the trailing-edge flaps and thrust-vector vanes also showed periods of rate limiting. None of the pilot comments indicated that rate limiting in either the command path or in the control surface response affected the handling qualities.

Lateral Gross Acquisition The lateral gross-acquisition task was performed by four pilots during five flights. Nineteen of the total 49 acquisitions received pilot ratings (fig. 14). The remaining 30 maneuvers were practic- es. Two of the pilot ratings are not included in the summary of data because the AOA varied from 60 ° to 35 ° during attempted gross acquisitions at 60 ° AOA. The large number of practices required for this task shows the increased difficulty over the longitudinal gross acquisitions. Unlike the lon- gitudinal acquisitions, where the task was primarily confined to one axis after the X-31A aircraft was banked into the correct plane, the lateral acquisitions required motion in multiple axes. First, the aircraft is performing velocity-vector rolls that result in a significant coning motion at high AOA. This motion is further complicated by the fact that the velocity vector settles during the ma- neuver. During extended maneuvers, the velocity vector is almost straight down, allowing the "he- licopter gun attack." It should be noted that Pilot E had two sorties on one day and required the same level of practice maneuvering in both flights. This pilot had also practiced similar maneuvers in a domed simulation, which increased familiarity with the task being performed.

During the initial flight practices, the acquisition was not occurring at the desired AOA with the target in the HUD field of view. Adjustments were made in the distance the X-31A aircraft was trailing the target aircraft, the lateral displacement from the target aircraft, and the offset angle after the target began maneuvering before the X-31A pilot initiated acquisition. Typical difficulties with the performance of these maneuvers were loss of sight of the target aircraft by the pilot under the nose of the X-31A aircraft, causing termination of the maneuver for safety concerns, and acquisi- tion of the target above or below the HUD field of view as a result of improper initial lateral offset.

Figure 14 shows a comparison of CHRs with AOA, revealing a degradation in handling qual- ities as AOA is increased. The cases near 30 ° AOA generally fall into the Level 1 category. At 45 °, the pilot ratings are consistent with Level 2 handling qualities. At 60 o, the trend is for Level 3 han- dling qualities. For this task, the majority of the maneuvers (11 of 17) were given a confidence class rating of "B," which shows only a moderate degree of confidence in the ratings. All of the data at AOAs greater than 50 ° were rated confidence class "B." Based on pilot comments, this rating can be attributed to the difficulties with adjusting the initial conditions to account for the multiple-axis maneuvers required of the X-31A aircraft. The general trend for increased CHRs with increasing AOA is present in the ratings regardless of the confidence class rating.

I

i A ............... .4 ............... L, ........... d ............. .L ............... L ..............

q_ I I I I t I I I I I 4 V i i Pitch-stick deflection, 2 ............... i ............... i .......

i i

in ! !

1 : ....... _ ....

,, ,, 0 _ ........ .L .............

i i i i I I I I I i i i i i t i E i -- ._ ........... u .............. J ......... J- ............... _- ..............

," ', , , ,. l 4o Angle-o,-e.eck / /i \ W '_ ,'_'\ ", ii command _ / / _Angle of 'I i \ %\ Angle of , , / / ; attack i I \ \ a t t a c k, 3 0 . . . . . . . . . . . . . . . 4 . . . . . . . . . . . . . . . r . . . . . _ . . . . . . . . . . . _ . . . . . _ . . . . . . _ i I i I I I I I i i i i 30 . . . . . . . . . . . . . . . I T . . . . . . . . . . . . . . . . . . . . . . . . . _ . . . . . . . . . . . . . . . T . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

p i t i i i i i i i ............... _ ............... r .................... T ............... _ ..............

I i i J Pitch rate, deg/sec i i i i ' ' -10 2 4 6 8 10 12 Time, sec 960387 Figure 13. Time history from longitudinal gross acquisition.

10 -- Pilot [] B 9-- AD 8-- eS

•

7-- Level 3 6-- Cooper-Harper mm rating 5 -- Level 2 4 -- Level 1 O • • m

I I I I I I I

0 10 20 30 40 50 60 70 Angle of attack, deg 960388 Figure 14. Cooper-Harper ratings as a function of angle of attack for lateral gross acquisition.

This task did not emulate the lateral acquisitions performed during CIC testing. During simu- lated combat, the X-31A aircraft typically maneuvered within the turn radius of the target aircraft.

The X-31A velocity vector was nearly straight down, resulting in a "helicopter gun attack." The CIC results did not indicate a tendency for Level 3 handling qualities at the higher AOAs. While this task did identify handling qualities deficiencies, it is not clear that the STEM task is represen- tative of the maneuvering pilots may be required to perform in the PST flight regime. For the STEM, the pilot had to aggressively initiate the maneuver with full-lateral stick; while in CIC test- ing, the pilot input was proportional to the change in nose-pointing angle required.

The PIO ratings tended to increase as a function of AOA. Three cases had a rating of "2"; un- desirable motions were present but did not affect task performance. An additional three cases had a rating of "3," indicating that undesirable motions did compromise task performance. One case was given a rating of "2-3," also falling into the category of undesirable motions. Two cases showed nondivergent oscillations and received a PIO rating of "4." Two cases did not receive a PIO rating, and six cases had a rating of "1."

Figure 15 shows stability-axis roll rate for each maneuver plotted as a function of AOA and shows that, for the PST range, that rate was relatively constant at approximately 40 deg/sec. The peak rate occurred for a maneuver at 25 ° AOA, and in general, the higher roll rates were the result of the pilot using roll stick before achieving the desired AOA while the aircraft was still pitching up. Nine maneuvers received an RPC rating of "2," or satisfactory. Eight cases received ratings of "2.5," which falls between the satisfactory and unsatisfactory levels. One maneuver received a RPC rating of "1," which equates to enhancing or tactically superior. This maneuver had the sec- ond highest stability-axis roll rate. The pilot commented, "I would say it' s just fine tactically. I got

_ound asfastasI wantedto." Althoughtheonsetratewasgood,thepilot wasunableto accurately

arrestthe roll rate,resultingin a CHRof "8" anda PIO ratingof "4." Addressingthe undesirable motions,the pilot stated,"Lots of them.Many overshoots; borderlinePIO at the end." The pilot

alsonotedthatthetaskwas"very difficult."

-- Pilot [] [] B __O C 6O A D • E I O A O 4O Maximum roll rate,

• 8

deg/sec

I I I I t I I

30 40 50 60 70 0 10 20 Angle of attack, deg 960389 Figure 15. Stability-axis roll rate as a function of angle of attack for lateral gross acquisition.

Figure 16 shows time histories. It can be seen that the pilot used full-stick displacement three times during the maneuver with a peak stability axis roll-rate command of 40 deg/sec. Although the stability-axis roll rate was high for this flight condition, the pilot had difficulty using it effectively for the aggressive gross-acquisition task. These data indicate that using the RPC to assess only the roll onset does not necessarily equate to good gross-acquisition performance.

During the performance of this task, the pilots regularly used full-roll stick displacement, regardless of the AOA. For the high AOAs, this displacement would be more than required to get maximum roll command because of the modification in the relationship between stick deflection and full-roll rate command discussed above. One reason for this excess displacement would be that no feedback to the pilot exists when full-roll command is generated. No pilot comments were di- rected towards any effects caused by the limiter that result when the pilot stick deflection is larger than actually required for full-roll command. Examination of the time histories do not indicate any particular effects from the control inputs. Analogous to the longitudinal task, rather than a circular reticle, vertical bands at the specified distance from the pipper would have provided the pilot with a more appropriate reference for the task.

i i i ................ i ................ ::/ _-, ....... ,_?-i ..............

1 r, f ', _i- _, ;AN.

................. _ .............. _:_ _,,:, , ----\;_:\-\ ............

Angle Angle-of'apc_I//: ! ' _'i _. \ of command_ ] _, i ', _. \ i ......... _ ........... -_-I---_ ..........

attack, 30 ................ ', ...... 7 _ Angle of ! _.

deg ......... i............ ._17 ..................................... "............

'_ ..... /J ."3. _'T _-__ S__ _- ....................................................... _. .................

I I I I t r I

' i

................ _ i ................ i:.................... i ................

Lateral stick displacement, in.

-2 i I I I T , i i i ' _J', I .................................................. .%-_, ............................

i R°ll'rate ' / \ _ i : command _/! \_ --1 ,, \/, \ _ IX,_ .:

........... -D' ----k_--t:._

Stability-

! ivlt U \'t; " -

axis ............. :, 3'_', ....... /:_. ..... v_ .........................

roll rate, -20 deg/sec

rt

-4O _------," .... ,_-'-# ..... 4.................................

I |/ ', i * i i I 0 5 10 15 20 25 Time, sec 960390 Figure 16. Time histories for lateral gross acquisition.

Fine-Tracking Evaluation Fine tracking was evaluated during eight flights by three pilots. In Phase 1 testing at low AOAs, 17 tracking tasks were performed and 9 maneuvers rated. Of the eight practices, six were required in the first flight. During Phase 2 testingathigh AOAs, 45 trackingtaskswereperformed.A total of 19practiceswererequired,and 16 longitudinaland 10lateral fine-trackingtaskswererated.

Duringthefirst flight, six practices wererequiredto getinitial conditionsthatallowedonescorable task.Forthenext 3 flights, efficiencyimproved,with 9 practices requiredto get 13maneuvers that

couldberated.AsthetargetAOA increased, two or threepractices wererequiredto achievethede-

siredaircraft positioningatthetargetAOA. Thelast 2 flights requiredonly 1practicefor 13 scor- ablemaneuvers.

One factorthat affectedthe pilot ratingswasthe amountof time spenttracking.The original

flight cardscalledfor 4 secof tracking.However,thepilots oftenspent20 secor moreperforming

the trackingtask,resultingin significantvariationsin flight condition(particularlyAOA). In one

casewherethe intendedAOA was30° but the trackingoccurred between30° and23°, the pilot

commented, "There weretwo distinctiveairplanes. WhenI wasatthe initial AOA around30°, it

wasquitea bit harderto trackthanwhenI settledin. My ratingwill be associated with the initial valuesof thetracking." Not all of the pilotswereasconcisein identifyingthe AOA rangefor their rating,andtheengineers hadto identify the AOA.

LongitudinalFineTracking

Initial difficulties with fine trackingresultedfrom the initial conditionsof the aircraft. The

spacingof 3000ft usedduringthe longitudinalgrossacquisitions wasreducedto 1500ft, but the

maneuvertiming usedfor grossacquisitionwasnot changed. This changeresultedin the X-31A

aircraft going a considerable distancedownrangewhile the targetwas maneuvering. When the

X-31A airplanewasmaneuvered, it wasoutsidetheturnof theF-18airplane.Suggestions from the

pilot in thecontrolroomtobasethemaneuver ontherelativeanglebetween the aircraftallowedthe

onescorable maneuver in thefirst flight. Duringthesubsequent flights,the starttime for theX-31A maneuverwasbasedon theoff-boresightangleandresultedin morerepeatable tasks.

Figure 17 showsCHRsplotted as a function of AOA and showsan increasein rating (or

decrease in handlingqualities)asAOAs increases. ForAOA lessthan30°, the ratingsareconsis-

tently "3" or less,indicatingLevel 1 handlingqualities.Between30° and50° AOA, the ratings

rangedbetween"3" and"7." The highest ratingsareatthehighestAOAs. This rangewould berat-

edLevel 2 with two Level 3 ratingsnear50° AOA.

All of theratingswerein confidence class"A" for AOAs lessthan30°. ForthePSTratings,ten

werein confidenceclass"A" andsix wererated"B." Theseratingsreflecta high degreeof confi-

dencefor mostof the ratings.All thepilots notedthatthetrackingtaskusedfor thehandlingqual-

ities evaluationwasdifferentfrom the type of trackingthatwasdoneduringthe CIC evaluations.

Onepilot summarized it by saying,"The trackingwe're trying to dohereis kind of dynamic-pitch

tracking andnot the kind of trackingwe typically did during the end game,which tendedto be

morein matchingyaw rates."PIO ratingsalsotendedto increase with AOA for this task.The rat-

ings rangedbetween"2" and"4," indicatingundesirable motionsandoscillationsthroughoutthe

PSTrange.

The initial instructionsfor the fine-trackingtaskscalledfor nose-to-tailtracking.Because of

the unique geometriesthat could result during the high-AOA maneuvering, the tracking tasks

requiredboth lateralandlongitudinalstickinputsto performthenose-to-tailtrackingbecause the

maneuver planeof theX-31A airplane wouldnot correspond with theplaneof symmetryof thetar-

get aircraft.This instructionwasmodified to statethattrackingwasnot necessarily from noseto tail, but shoulduseonly pitch stickinputsandusethe appropriate aircraft featuresasa reference.

Even with the modified instructions,the pilots would often usediagonalstick inputsduring the trackingtasks.

10 -- Pilot O A g--_ C Ao 8 _

O0

7 -- Level 3 6 -- Cooper-Harper

0 O0

5-- rating

00o

4 -- Level 2

Level 1 A

3--

O

2-- 1 --

I I I I I I

I

20 30 40 50 60 70 0 10 Angle of attack, deg 960391 Figure 17. Cooper-Harper ratings as a function of angle of attack for longitudinal fine tracking.

Longitudinal Fine-Tracking Handling Qualities Criteria The X-31A data were evaluated using the Neal-Smith, bandwidth, and Smith-Geddes criteria to assess the applicability at high AOAs. These criteria are all based on the pitch stick-to-pitch at- titude transfer function. An analytic study l° had shown that other criteria based on LOES were not applicable to high-AOA flight. Transfer functions were generated using the linear models based on the mass properties and flight conditions (Mach number, altitude, and AOA) associated with the pilot ratings. The transfer functions were used in the criterion assessment and correlated with the pilot ratings. With the exception of a few data points, the linear analysis results correlate with han- dling qualities ratings obtained in flight. The low-AOA data and the data with CHRs of "3" tend to fall into the Level 1 regions for all of the criteria. The data with the higher CHRs seem to fall in clusters, and for all the criteria, these clusters move away from the Level 1 regions.

Figure 18 shows X-31A data plotted using the Neal-Smith criterion. 7 The Neal-Smith criterion uses a simple compensator model to close the loop of the pitch stick-to-pitch attitude transfer func- tion. The magnitude of the resonant peak in the resulting closed-loop transfer function is compared

with thephase angleof thecompensation. The high-AOAdataindicatethatlessleadcompensation

canbeallowed.With two exceptions, datawith Level 1ratingsrequiredlessthan 15° of leadcom-

pensation. A clusterof datanear40° of leadcompensation with adjacentCHRsof "5" and"7" ex-

ists that may indicate the proximity of the Level 3 boundary.Figure 14 showsthe existing

boundaries as solid lines, andboundaries indicatedby the X-31A PST dataareshownas dashed

lines.Additional dataarerequiredto determineif theseboundaries arevalid.

-- Neal/Smith [bandwidth = 3 rad/sec, _ Criterion time delay = 0.3 sec] ----- X-31 data CHR \ [] 3 \ 4 Level 3 -- A _-'_'-._ A s v \\ _ e 6 Resonance peak, \\\\ _ • 7 dB Level 2 <_> \\\ -- Ax\ i [3o a_ Level 1

o I

40 60 80 - 20 0 20 Lead compensation, deg 960392 Figure 18. Comparison of X-31A data with the Neal-Smith criterion.

For the bandwidth criterion 9 (fig. 19), all of the data show an estimated equivalent time delay of approximately 0.04 sec. This criterion has correlated handling qualities with the estimated equivalent time delay and bandwidth frequency calculated from the pitch stick-to-pitch attitude transfer function. A reduction in bandwidth exists that is consistent with an increase in AOA and CHR. The X-31A data indicate that the Level 1 boundaries are reasonable. Several data points exist with a bandwidth of approximately 3 rad/sec that have CHRs of "5" and "7," indicating that it might be appropriate to move the Level 3 boundary to this bandwidth as shown by the dashed line.

When compared with the Smith-Geddes criterion s (fig. 20), the X-31A data indicate that the slopes of average CHR as a function of phase angle at the bandwidth frequency need to be steep- ened. The criterion calculates the bandwidth frequency based on the slope of the gain relationship from the pitch attitude-to-pitch stick transfer function. In general, the tolerance bands for the av- erage CHR would be valid for most of the data points with a pilot CHR of "3" or "4." An altemate relationship between average CHR and phase angle at the bandwidth frequency is presented as a dashed line.

.07 MIL-STD bandwidth requirements - Category A .06 -- Level 3 .05 m .04 -- Criterion Equivalent ------ X-31 data time delay, sec CHR .03 m Level 2 O 2 [] 3 .02 m <> 4 L_ s • 6 .01 -- • 7

I

I I \

5 10 Bandwidth, rad/sec 960393 Figure 19. Comparison of X-31A data with the bandwidth criterion.

10 E 9B 8 ...... _\ \ 7. "\, _ \, :\" __ CHR "\' _ \'\ \ , Average

o= ", \ ,\

Cooper-Harper rating "" ZXZX 5 3-- H [] uriterJon "----.........---._ Criterion bounds "_._.._ ....

O 2 .... X-31 data

, I I I I I ] I

-220 -200 -180 -160 -140 -120 -100 Phase angle at bandwidth frequency, deg 960394 Figure 20. Comparison of X-31A data with the Smith-Geddes criterion.

The one data point that is anomalous for all three criteria is the 30 ° AOA tracking case that received a CHR of "6." The confidence class rating was "A," indicating a high degree of confi- dence in the rating. In addition, the PIO rating of "3" indicated that undesirable motions affected the pilot's ability to performthe task.The pilot did attributesome of the difficulty to aggressive- ness, commenting, "The more aggressive you are, the more you oscillate." Another pilot perform- ing a similar maneuver gave a better CHR of "4," but also commented, "If you are aggressive, you get undesired motions." Other than pilot technique, one difference noted between the two tasks was that the task that received the degraded rating was performed at a higher airspeed. An analytic investigation of handling qualities 1° did show a degradation in predicted handling qualities during PST flight as airspeed increased with a constant AOA.

Lateral Fine-Tracking Figure 21 shows CHRs plotted as a function of AOA. As with the longitudinal tracking data, some scatter in the ratings exists near 30 ° AOA, but the trend is toward higher CHRs as AOA in- creases. The three data points at an AOA at or greater than 40 ° had confidence class ratings of "B."

The lower AOA data received a confidence class rating of"A." The PIO ratings are consistent with the other tasks in that an increase in undesirable motions as AOA increased existed, with oscilla- tions being reported at the highest AOAs.

10 m Pilot o A 9-- <> C A 8-- 7-- O Level 3 6-- O Cooper-Harper O 5-- rating Level 2 O O 1_ 4-- Level 1 O_ 3-- 2-- A 1--

I I I I I I I

0 10 20 30 40 50 60 70 Angle of attack, deg 960395 Figure 21. Cooper-Harper ratings as a function of angle of attack for lateral fine tracking.

A consistent pilot comment was, "The more aggressive you are, the harder the time you have tracking." As well as the impact of aggressiveness on the task performance, the pilots also com- mented that the task frequently required diagonal stick inputs as opposed to pure lateral stick mo- tions. Lateral tracking initially required wing tip-to-wing tip tracking. The tracking task was redefined to use only lateral stick inputs, but the pilots continued to use diagonal inputs.

LateralFine-TrackingHandlingQualitiesCriteria

UsingLOESderivedfrom thelinearmodels,dutchroll frequency, dutchroll damping,theroll-

modetime constant,andthe equivalent time delaywerecalculated andcomparedwith the criteria

from MIL-STD-1797(fig. 22 and23). ThesecriteriapredictLevel 1handlingqualitiesthroughout

theAOA range,which is not consistent with thehandlingqualitiesratings.Thesedataindicatethat

dutch roll frequencyanddampingandthe roll-modetime constantare not the factorsaffecting

high-AOA handlingqualities.

CHR 1.2 w 1.0 m I-]3 Equivalent .8 -- dutch roll Z_s © damping, .6 Level 1 Qs rad/sec .4 • 7 Level 2 •2 --

I

I I I I I I

-- Comparison of dutch roll mode with MIL-STD criteria Equivalent dutch roll

_

frequency, <} • Z_ rad/sec © -- Level 1

T i---I

t Level2 1..... T 1 50 60 20 30 40 0 10 Angle of attack, deg 960396 Figure 22. Comparison of X-31A data with MIL-STD-1797 dutch roll frequency and damping criteria.

The lateral fine-tracking ratings were also compared with the Smith-Geddes criterion 8 (fig. 24).

Although a limited amount of data exists, there appears to be general agreement with this criterion.

Some caution must be used when applying the results of linear analysis to the lateral- directional high-AOA tracking tasks. Several nonlinear effects are evident in the data. The flight condition changes rapidly during the task from a high-speed, high-AOA condition to a low-speed, reduced-AOA condition. The maximum roll-rate command is scheduled as a function of airspeed so that the pilot experiences a reduced command authority as the airspeed decreases. The rate limit for the stability-axis roll-rate command was reached several times during the fine-tracking tasks.

The high workload demand on the thrust-vectoring system resulted in rate limiting of the thrust- vector paddles. Also, at high AOAs, moving the pipper from wing tip to wing tip required a com- bined lateral and longitudinal stick input.

.12 -- CHR Level 2 O 2 .10 Level 1 [] Comparison of time delay B with MIL-STD criteria .08 <> Equivalent Zx m time delay, .06 sec .04 m .02

I 6

1.2 Level 2 1.0 Level 1 Comparison of roll-mode time .8 constant with MIL-STD criteria Equivalent roll-mode .6 time constant, sec .4 .2

o

I

I

0 10 20 30 40 50 60 Angle of attack, deg 960397 Figure 23. Comparison of X-31A data with MIL-STD- 1797 roll mode time constant and time delay criteria.

R0,D

<>

Average Cooper-Harper 6 Z_,

-o2 ',,\

rating 03 \ \\, A s ,\ .....

z_e6 • 7 " ".----.

Criterion _ "_" "-- Criterion bounds

1 I I I I I I I

-220 -200 -180 -160 -140 - 120 -100 Phase angle at bandwidth frequency, deg 960398 Figure 24. Comparison of X-31A data with Smith-Geddes criteria for the lateral axis.

Even a full six-degrees-of-freedom nonlinear simulation did not entirely reproduce the dynamics observed during some of the lateral fine-tracking tasks. Figure 25 shows some of the ex- cursions in yaw rate and sideslip angle that were not duplicated with the nonlinear simulation.

These excursions approximately correlate with target overshoots where the target wanders outside the 20-mrad reticle and may be related to asymmetric forebody vortex cores. To accurately predict handling qualities requires an analytic model that includes all of the dynamics, so the effect of these vortices should be included.

Flight test ------ Simulation i z_-_ _ ............... -4-.................... t............................... _......

Roll rate, deg/sec -5 i i i : , ,, i _1 i .

-10 I I i : t i/_ : A i i _ !Z_ __'_at_ ....... _' ............ __1. ............................................... ,_ ......

°k _ /._ ....._-- i J \ .-,,. !

Yaw rate, I',./ V \ --\ // \ t / / ; 1 : ,., deg/sec - .5 ............................... 4-_J- _12 ..................

1.0 .5 Sideslip angle, 0 deg --.5

"_' _ i \/ \ / i i

.............................. -i ............................. I................. i - 1.0 I I I 5 10 15 Time, sec 960399 Figure 25. Comparison of flight and simulation data for a lateral fine-tracking case.

Combined Maneuvers

The combinedmaneuverwas flown four times during oneflight by one pilot. Two of the

maneuvers wereusedfor practice.Comments andCHRratingsof "3" and"4" weregiven on the

othertwo maneuvers. No distinctionexistedin the ratingsfor lateralor longitudinaltasks,but fine

trackingandgrossacquisitionwereratedseparately. In summary, theratingsweregiven with high

confidenceandwereborderlineLevel 1/Level2 for both trackingandacquisition.No undesirable

motionswerepresentduringthe grossacquisition,andthemotionsdid not affect the taskduring

fine tracking.Following the flight, the pilot reported,"The SSLA setupwasanexcellentstarting

conditionto evaluatehandlingqualitiesin thePSTregime."

Figure26 shows time historydatafrom thesecond ratedmaneuver, which spanned 60 sec.The

AOA rangedbetween30° and70°. Pitch-sticksensitivitycanbeseen in the AOA command where

15° excursions in thecommand attheAOA command ratelimit of 25deg/sec canbeseen. Full-roll

stickwasusedearlyin themaneuver, andapproximately 66percentofthe stickdeflectionwasused

in alateracquisition.Peakvelocity-vector roll ratesof nearly40 deg/sec wereobserved. The fourth

traceshowsthetiming for thethreegrossacquisitions performedandtheperiodsof tracking.This

maneuver wasinitiatedat analtitudeof 25,000ft andwascompletedatan altitudeof 14,000ft.

As with theothertasks,thepilot commented thatthestickforceswere"too heavy"andthemo-

tions were "too large."Because this maneuverintentionallyuseddiagonalstick inputs,the pilot

wasableto commenton stickharmony,"The stickmovement is muchtoo high; andyou havethe

nonharmony betweenthepitch stick,which is so sensitive, andtheroll stick, which is not so sen-

sitive."

Because thepilot ratingscovermaneuvers thatspanalargeflight envelope andencompass two

axesof control,comparing themwith analyticresultsis difficult. Thepilot liked thismaneuver bet- ter andfelt it wasmorerepresentative of the typeof flying doneduringthe CIC investigation.The maneuveralsoresultedin the Level 1/Level2 ratingsthatwereexpected. Additional testingis re-

quired,but thistype of maneuver mayprovidea bettermeans of evaluating thePSThandlingqual-

ities, but like CIC, it is of limited value for analysisor designbecauseof the varying flight conditions.

LESSONSLEARNED FORHIGH-ANGLE-OF-ATTACK HANDLING QUALITIES TESTING

Whenflying a newtask,backupcardsshouldbepreparedfor anestablished taskin the event

the first taskis not working out. During the first PSTfine-trackingflight, it wasquickly apparent to the pilots in the airplaneandon the groundthat the testasdesigned would not resultin an ac- ceptablefine-trackingtask.Almost anentireflight wasusedto get onedatapoint.Testingof alter-

nateflight cardswould havecollectedadditionaldata,andgroundreview would haveadjustedthe

testsetupfor the acquisitionof PSTfine-trackingdata.

4 -- Stick If :: :: _J,-. i ...... , ..... ] deflection, 0 ---- "_ ..................... I.......... -_'_ .... ;_ .... ._ ,-- ....... _.- .... _-'-T .... ,-p .........

in.

I ",.,_. ,' "_._:_ ............ l

'[ ....... _: ..... 'i- ...... -'- ............... i ...... i ........ ]

: I \ 60 ......

,_,:: ...... -,_:- .................... :-__ I-', ......

0 t

Angle of attack, deg 40 ]-- --//_/- .........

.... ___ _:- i --_-_ , ............. _,_;:-2;.-;.'L_ ................................................

20 [._I ............... command I1! ................................ ,- &-__ "-___

.... ::_vi_________,.. _pv/ _'r -- '..[,l__,___'?___ ...._.i ,:_,a "

Stability- - 10 ....... I ............ _--_-H _.........

axis roll rate, deg - 20 ....... Iii_---9'---- _ ......................... .,,--- ' ............

- 3O ......... L .................... I_ --_-'R:lJ-ret:n d - 4O : T i i Gross : Gross Gross ................ acquisition .......................... acquisition ............. acquisition ......

........................ I

Indication of task iiiiiiiiiiiiiiiiiiiii .... ..... ...........

0 20 40 60 Time, sec 960400 Figure 26. Time history from a combined maneuver.

Some of the pilots thought a domed simulation would have helped them better prepare for the tasks. But it is interesting to note that during the lateral gross acquisitions, the one pilot who had performed the maneuvers in a domed simulation required the same amount of in-flight practice as the other pilots.

Careshouldbetakenin taskdefinition.For thefine-trackingtasks,thepilots wereaskedto do

separate longitudinal and lateral tracking tasks. Even with instructions that the inputs should be limited to pitch or roll inputs, the pilots continued to use diagonal stick motions to perform the more classical tracking tasks of nose-to-tail and wing tip-to-wing tip. The task definition should also include a reasonable time limit for the performance of the task. One of the reasons 4 sec was initially chosen was to try to minimize variation in flight condition during the performance of the task. This time limit was not enforced during the testing and resulted in a tracking task that lasted 20-30 sec with large AOA variations.

Modifications to the HUD could have provided the pilots with the proper cues for the tasks. For longitudinal gross acquisition, horizontal bars at 25 and 40 mrads would have provided the proper reference for the task that was being rated. Similarly, vertical bars could have been used for lateral gross acquisition in place of the circular reticles. This display might reduce the tendency of the pi- lot to try to place the pipper on the target.

CONCLUSIONS The Standard Evaluation Maneuvers (STEMs) provided repeatable tasks that could be com- pared with analytic linear and nonlinear simulation results. With suitable initial conditions and practice, gross acquisition and fine tracking could be performed at the desired angle of attack (AOA). Pilot comments indicated that these maneuvers were not consistent with the types of ma- neuvering performed during the close-in combat (CIC) evaluations. This testing identified prob- lems that may not be significant in actual tasks. Further testing is needed to resolve these differences.

The pilot-assigned ratings for gross acquisition and fine tracking for both the longitudinal and lateral axes were dependent on AOA. More undesirable motions and then oscillations existed as AOA increased.

The longitudinal gross-acquisition task was well-defined and provided an easily repeatable task. The pilot ratings and comments indicated a high degree of confidence. These ratings reflected the expectations from CIC testing with the aircraft having Level 1 or Level 2 handling qualities.

The lateral gross-acquisition task was one of the most difficult. The task required a significant amount of flight time to adjust the starting conditions to achieve the desired AOA with the target aircraft in the head-up display field of view for the X-31A airplane. The pilot proficiency for this task did not improve as significantly as it did for the other acquisition and tracking tasks. The pilot comments and ratings indicated a degradation in handling qualities as AOA increased, with Level 3 handling qualities at an AOA near 60 °. The pilot comments noted that this type of acqui- sition was not similar to the acquisitions performed during CIC testing. The degradation in han- dling qualities was not expected from the CIC testing where the general assessment would have been Level I/Level 2 handling qualities.

For the longitudinal fine-tracking task, consistent trends existed in regard to the Neal-Smith, bandwidth, and Smith-Geddes criteria. The maneuvers that received Level 1 ratings in flight were

ratedLevel 1 by the criteria.The Levels 2 and3 datafrom flight tendedto produceconsistent

resultswhencompared with thelinearmodels andindicatedpotentialmodificationsfor the criteria.

TheX-31A handlingqualitiesratingsshoweda degradation with AOA thatwasnot observed dur-

ing the CIC testing.

Lateral fine tracking showed a degradation to Level 3 handling qualities as AOA increased.

The X-31A program provided only a limited amount of data that could be compared with the ex- isting criteria. The data showed good general agreement with the Smith-Geddes criterion. These data did not provide sufficient information to offer modifications to the existing criteria that pre- dicted Level 1 or borderline Level 1/Level 2.

For both lateral and longitudinal fine tracking, the effect of the velocity-vector settling during the maneuver had a significant impact. Future use of this STEM may require modifications to allow a more stabilized starting condition for the fine-tracking tasks. During the X-31A testing, fully sep- arating the lateral and longitudinal tasks was not possible. The pilots generally used diagonal stick inputs regardless of the axes being evaluated.

For control stick harmony, the majority of the comments were noted during the fine-tracking tasks. In these cases, the pilots were using diagonal stick inputs to perform the wing tip-to-wing tip and nose-to-tail tracking. The one other task that elicited a comment on control stick harmony was the combined maneuver. The pilot commented on the disparity in motion for longitudinal and lateral stick displacements (large for roll and small for pitch). Although the control implementation resulted in a limiter for large roll-stick deflection, no particular comments were given by the pilots.

The limited testing with the combined maneuver was commented upon favorably by the pilots, but these ratings are not amenable to comparison with analytic results because of the rapidly vary- ing flight conditions. This type of maneuver may be useful for providing an overall evaluation of aircraft performance in the post-stall flight regime and should be considered as an additional STEM. Like the CIC results, these data are of limited value for analysis because of the varying flight conditions.

Dryden Flight Research Center National Aeronautics and Space Administration Edwards, California, June 5, 1996 REFERENCES 1Sweeney, Joseph E. and Gerzanics, Michael A., "F-16 MATV Envelope Expansion: Testing For Controllable High AOA Maneuvering," Society of Experimental Test Pilots Thirty-Eighth Symposium Proceedings, Sept. 1994, pp. 285-295.

2Wichman, Keith D., "High Alpha Handling Qualities Flight Research on the NASA F/A-18 High Alpha Research Vehicle," High-Angle-of-Attack Technology Conference, NASA Langley Research Center, Virginia, Sept. 17-19, 1996.

3Webster,Fredrick R. and Purifoy, Dana, X-29 High Angle-of-Attack Flying Qualities, AF-FTC-TR-91-15, Jul. 1991.

4U°S. Department of Defense, Flying Qualities of Piloted Vehicles, MIL-STD-1797, Mar. 1987.

5Krekeler, Gregory C., Jr., Wilson, David J., and Riley, David R., "High Angle of Attack Fly- ing Qualities Criteria," AIAA-90-0213, Aug. 1990.

6Wilson, David J., Riley, David R., and Citurs, Kevin D., Flying Qualities Criteria for 60 ° Angle of Attack, NASA CR-4535, vol. I, Dec. 1993.

7Neal, T. Peter and Smith, Rogers E., An In-flight Investigation to Develop Control System Design Criteria for Fighter Airplanes, AFFDL-TR-70-74, vol. I, Dec. 1970.

8Smith, Ralph H., "The Smith-Geddes Criteria," SAEAerospace, Control & Guidance Sympo- sium, Reno, Nevada, Mar. 1993.

9Hoh, Roger H., Mitchell, David G., and Hodgkinson, John, "Bandwidth --A Criterion for Highly Augmented Airplanes," AGARD CP-333, Jun. 1982, pp. 9-1-9-11.

l°Stoliker, P.C., Simulation Prediction of High-Angle-of-Attack Handling Qualities for the X-31A, NASA TM-4758, 1996.

llCord, Thomas J., Leggett, David B., Wilson, David J., Riley, David R., and Citurs, Kevin D., "Flying Qualities Evaluation Maneuvers," AGARD CP-548, Mar. 1994, pp. 18-1-18-8.

12Beh, H. and Hofinger, G., "X-31A Control Law Design," AGARD CP-548, Mar. 1994, pp. 13-1-13-9.

13Noflin, Ken A., Flight Simulation Software at NASA Dryden Flight Research Center, NASA TM-104315, Oct. 1995.

lnEubanks, D., GUtter, R., and Lee, B., "X-31 CIC Flight Test Results," Four Power Senior Nation Representative Full Envelope Agility Workshop, Eglin AFB, Florida, Mar. 1995.

15Cooper, George E. and Harper, Robert P., Jr., The Use of Pilot Rating in the Evaluation of Aircraft Handling Qualities, NASA TN-D5153, Apr. 1969.

16Foster, John V., Ross, Holly M., and Ashley, Patrick A., "Investigation of High-Alpha Lateral-Directional Control Power Requirements for High-Performance Aircraft," AIAA- 93-3647, Aug. 1993.

Form Approved REPORT DOCUMENTATION PAGE OMB No.0704-0188 Public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and mainta n ng the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this col- lection of information, including suggestions for reducing this burden, to Washington Headquarters Services, Directorate for Informat on Operations and Reports, 1215 Jefferson Davis Highway, Suite 1204, Arlington, VA 222024302, and to the Office of Management and Budget, Paperwork Reduction Project (0704o0188), Washington, DC 20503.

1. AGENCY USE ONLY (Leave blank) 2. REPORT DATE 3. REPORTTYPE AND DATES COVERED September 1996 Technical Memorandum 4.TITI.E AND SUBTITLE 5. FUNDING NUMBERS Evaluation of High-Angle-of-Attack Handling Qualities for the X-31A Using Standard Evaluation Maneuvers WU-505-68-30 6, AUTHOR(S) Patrick C. Stoliker and John T. Bosworth 8. PERFORMING ORGANIZATION 7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) REPORT NUMBER NASA Dryden Flight Research Center P.O. Box 273 H-2128 Edwards, California 93523-0273 10. SPONSORING/MONITORING 9. SPONSORING/MONOTORING AGENCY NAME(S) AND ADDRESS(ES) AGENCY REPORT NUMBER National Aeronautics and Space Administration NASA TM-104322 Washington, DC 20546-0001 11. SUPPLEMENTARY NOTES Presented at the High-Angle-of-Attack Technology Conference, NASA Langley Research Center, Hampton, Virginia, Sept. 17-19, 1996. Also presented at the AGARD Flight Vehicle Integration Panel Symposium on Advances in Flight Testing, Lisbon, Portugal, Sept. 23-26, 1996.

12a. DISTRIBUTION/AVAILABILITY STATEMENT 12b. DISTRIBUTION CODE Unclassified--Unlimited Subject Category 08 13. ABSTRACT (Maximum 200 words) The X-31A aircraft gross-acquisition and fine-tracking handling qualities have been evaluated using standard evaluation maneuvers developed by Wright Laboratory, Wright-Patterson Air Force Base.

The emphasis of the testing is in the angle-of-attack range between 30 ° and 70 ° . Longitudinal gross- acquisition handling qualities results show borderline Level 1/Level 2 performance. Lateral gross- acquisition testing results in Level 1/Level 2 ratings below 45 ° angle of attack, degrading into Level 3 as angle of attack increases. The fine-tracking performance in both longitudinal and lateral axes also receives Level 1 ratings near 30 ° angle of attack, with the ratings tending towards Level 3 at angles of attack greater than 50 ° . These ratings do not match the expectations from the extensive close-in com- bat testing where the X-31A aircraft demonstrated fair to good handling qualities maneuvering for high angles of attack. This paper presents the results of the high-angle-of-attack handling qualities flight testing of the X-31A aircraft. Discussion of the preparation for the maneuvers, the pilot ratings, and selected pilot comments are included. Evaluation of the results is made in conjunction with exist- ing Neal-Smith, bandwidth, Smith-Geddes, and military specifications.

14. SUBJECTTERMS 15. NUMBER OF PAGES Flight controls; Handling qualities; High angle of attack; Pilot-induced oscillation; ;16. PRICE CODE Standard evaluation maneuvers; X-31A aircraft AO3 20. LIMITATION OF ABSTRACT 17. SECURITY CLASSIFICATION 18. SECURITY CLASSIFICATION 19. SECURITY CLASSIFICATION OF REPORT OFTHIS PAGE OF ABSTRACT Unlimited Unclassified Unclassified Unclassified Standard Form 298 (Rev. 2-89) NSN 7540-01-280-5500 Available from the NASA Center for AeroSpace Information, 800 Elkridge Landing Road, Prescribed by ANSI Std. Z39-18 Linthicum Heights, MD 21090; (301)621-0390 298-102