Pilot-induced Oscillations Research Articles

Investigation of limb-side stick dynamic interaction with roll controll

A fixed-base simulation was performed to identify and quantify interactions between the pilot's hand/arm neuromuscular subsystem and such features of modern fighter aircraft roll-rate command control system mechanizations as force-sensing side-stick-type manipulator, vehicle effective roll time constant, and flight control system effective time delay. The results provide insight to high-frequency pilot/vehicle oscillation (roll ratchet), low-frequency pilot-induced oscillation, and roll-to-right control input problems previously observed in experimental and production fly-by-wire control systems. The simulation configurations encompass and/or duplicate several actual flight situations, reproduce control problems observed, and show that the highfrequency nuisance mode known as ''roll ratchet can derive primarily from the pilot's neuromuscular subsystem. The simulations show that force-sensing side-stick manipulator force/displacement/command gradients, command prefilters, and flight control system time delays need to be carefully adjusted to minimize neuromuscular mode amplitude peaking (roll-ratchet tendency) without restricting roll control bandwidth (with resulting sluggish or low-frequency pilot-induced oscillation prone control). The results further demonstrate that roll-ratchet tendency, difficult to detect in fixed-base simulations, is readily apparent from application of frequency-response spectral analysis techniques.

Space Shuttle longitudinal landing flying qualities

The Space Shuttle program took on the challenge of providing a manual landing capability for an operational vehicle returning from orbit. Some complex challenges were encountered in developing the longitudinal flying qualities required to land the Orbiter manually in an operational environment. Approach and landing test flights indicated a tendency of pilot-induced oscillation near landing. Changes in the operational procedures reduced the difficulty of the landing task, and an adaptive stick filter was incorporated to reduce the severity of any pilot-induced oscillatory motions. Fixed-base, moving-base, and in-flight simulations were used for the evaluations. Overall, the Orbiter control system and operational procedures produced a good capability to perform routinely precise landings with a large, unpowered vehicle that has a low lift-to-drag ratio.

Prediction and occurrence of pilot-induced oscillations in a flight test aircraft

Flight testing of a high-performance aircraft was conducted several years ago and again during the past two years. During the first flight tests, a pilot-induced oscillation was experienced in the pitch axis during the time between the landing flare and touchdown. During the more recent flight tests two pilot-induced oscillations were experienced in the pitch axis during up-and-away flight. One of these occurred while the pilot attempted to track a cockpit display. Math models of the aircraft were used with three handling qualities criteria to determine the predicted handling qualities of the aircraft. These predictions were compared with flight test results. The handling qualities criteria used were the equivalent systems, bandwidth, and R. Smith criteria. The handling qualities predicted by each of these criteria agreed well with flight test results. The R. Smith criteria were especially interesting because they predicted the likelihood of pilot-induced oscillation explicitly and correctly and also predicted the frequency of each pilot-induced oscillation correctly. The predictions and flight test results also highlighted the impact that pilot displays can have on handling qualities. g n/a.

Suppression of biodynamic disturbances and pilot-induced oscillations by adaptive filtering

A method for reducing the effects of biodynamic interferences due to whole-body vibrations is described. Manual control tasks are particularly addressed. The method, based on adaptive filtering, enables a substantial attenuation of additive stick involuntary control commands due to pilot body vibrations (stick feedthrough). By means of computer simulations the feasibility of the proposed filtering method is demonstrated for two examples 1) adaptive filtering of biodynamic interferences due to stick feedthrough in manual vehicular control, 2) suppression of pilot-induced oscillations (PIO) due to stick feedthrough. This form of PIO is particularly anticipated for flexible aircraft having a high-gain, stiff stick. It is demonstrated that adaptive filtering is capable of suppressing the biodynamic interferences without substantially affecting the control characteristics of the aircraft/pilot system.

Influence of roll command augmentation systems on flying qualities of fighter aircraft

Roll command augmentation systems (CAS), in which commanded rate is directly proportional to stick input, provide high authority and precise control over the entire flight regime. However, operational experience with high-gain CAS has revealed problems with their use. Oversensitivity to small control inputs, roll ratcheting, and pilot-induced oscillation (PIO) are commonly encountered in the early stages of development. Recent flight test and operational experiences with high-authority CAS of fighter aircraft are reviewed. Possible sources of the problems encountered are suggested, and guidelines are proposed for improving the flying qualities of aircraft equipped with CAS. y Fas L'?

Analysis of Aircraft Attitude Control Systems Prone to Pilot-Induced Oscillations

An optimal control model of the human pilot is applied to the study of aircraft attitude control systems. Attention is focused upon documented examples where an aircraft's linear pitch attitude dynamics have led to the incidence of pilot-induced oscillations. In particular, the effects of control system time delays, or higherorder vehicle dynamics that can be represented by time delays, are examined. A simple criterion for determining an aircraft's susceptibility to pilot-induced oscillations is offered. The criterion is based upon the open-loop pilot/vehicle crossover frequency as predicted by the optimal control model. The modeling procedure offers insight into the manner in which higher-order vehicle dynamics effect pilot equalization requirements and closed-loop pilot/vehicle performance.

An application of invariance principle to pilot model for NT-33 aircraft with variable coefficients and delays†

A method is presented for analysing pilot-induced oscillations (PIO) for the NT-33 closed-loop pilot model when retardations and coefficients are not constant. The variation of retardations and coefficients results from the effect of wind shear and the neuromuscular dynamics of the pilot reported in available data. Non-linearities in the model are also considered. The method is based on the use of a new description of such systems in terms of convolution equations. Spectral factorization is applied to the entire functions of exponential order. The result is a criterion for the PIO-system with variable coefficients and variable delays. The criterion assumes continuity and boundedness of the coefficients and delays. A Lyapunov functional is constructed which gives a criterion on the roots of a certain ‘ quasi-polynomial ’, i.e. a polynomial in a variable and the exponential of that variable. The largest domain of attraction is obtained from the ‘ invariance principle’ of LaSalle (1975).

An Adaptive Stick-Gain to Reduce Pilot-Induced Oscillation Tendencies

An Adaptive Stick-Gain to Reduce Pilot-Induced Oscillation Tendencies

Influence of Bobweights on Pilot-Induced Oscillations

Historically, a major contributor to longitudinal pilot-induced oscillations (PIO) in service airplanes has been the use of bobweights, which cause the control system to couple dynamically with the airframe's short-period mode. The purpose of this article is to review and discuss the influence of bobweights on flying qualities, and to show how the unwanted effects can be minimized by careful design. A method of analysis is presented, followed by application of the method to several actual PIO situations. Finally, the various options available to the airplane designer are discussed.

Higher-order control system dynamics and longitudinal handling qualities

An experimental investigation of the effects of higher-order control system dynamics on the longitudinal handling qualities of a fighter airplane are discussed. This research was undertaken using the USAF/CAL variable stability T-33 airplane as an in-flight simulator. Different higher-order responses were simulated by altering the elevator feel system, elevator actuator, and airplane short-period characteristics. Essentially the same configurations were evaluated by two pilots using a revised pilot rating scale. One pilot also rated the configurations for their pilot-induced oscillations (PIO) tendencies. Comments and ratings were related to a response delay parameter. Many of the higher-order control systems investigated produced pronounced PIO tendencies in flight, and some were considered unflyable with certain higher-order characteristics. A comparison of fixed-base and in-flight evaluations indicated that configurations with significant PIO tendencies were rated poorer in flight, and configurations with little or no PIO tendencies were rated better in flight.

A regression analysis of pilot-induced oscillation ratings

Pilot-induced oscillation (PIG) has been studied by fitting a regression surface to flight test data published by the Air Force Flight Dynamics Laboratory. The independent variables, characteristics of the pitch rate step-response, include time to first peak, effective time delay, slope after first peak, and stick force per g. These were selected mainly from theoretical considerations and partially from their fit in the regression analysis. These variables were correlated to the PIO numerical ratings, and least-square regression surfaces were calculated. Various models were considered, including logarithmic, linear, and second-order models. The results of this study are a set of parameters that can be measured on the time history of the pitch rate step response, and an ordering of these parameters according to their relative importance (correlation) to PIO ratings. On the basis of this analysis, a complete simulation study is planned to investigate further the connection of these parameters with pilot ratings.

正誤表

正誤表

Some consideration on pilot-induced oscillations

Pilot induced oscillation of a stable second order system is investigated as an instability that involves the closed loop consisting of the human pilot, the control system and the controlled element.Transfer functions of the human pilot and the control system are assumed. Nondimensionalized closed loop gains on the instability boundary are used as the extent of the instability.Fixed base simulator tests are conducted and the experimental results are compared with the results of theoretical analysis.These results seem to offer good insight into the phenomenon of the pilot induced oscillation.

Experimental investigation of pilot dynamics in a pilot-induced oscillation situation.

Results are presented for an experimental fixed and moving-base flight simulator investigation of a generalized aircraft longitudinal pilot induced oscillation (PIO) situation. Data are given relative to four handling-quality areas 1) pilot dynamic performance when tracking sinusoidal inputs following the occurrence of PIO, 2) the influence of motion cues on such performance, 3) the effects of varying stick force on pilot dynamic behavior in the PIO situation, and 4) the effect of varying the vehicle short-period transfer function numerator term 1/T02* Increases in this term to values above the normal level associated with the simulated airframe yielded experimental PIO's. The intentional increases were accomplished at a high input rate in an effort to preclude significant initial pilot gain adaptation. Approximately five times the increase in 1/T0 2 which produced the moving-base PIO was needed to produce instances of fixed-base PIO. With only external visual cues available during an oscillation, the pilot did not appear to operate in a synchronous (pure gain) manner. The availability of full-scale motion cues with visual cues causes the pilot to appear more nearly synchronous in the visual loop; however, the same data more consistently show the pilot operating with lag dvnamics on the load factor cues.

Erratum: "Sample Estimates of B-66B Low-Level, Clear-Air Gust Field Parameters"

References 1 A'Harrah, R. C., Low-altitude, high-speed handling and riding qualities, J. Aircraft 1, 32-40 (1964). I. L., Jex, H. R., and McRuer, D. T., Pilotinduced oscillations: their cause and analysis, Northrop Corp., Norair Div., Kept. NOR 64-143 (June 20, 1964); also Systems Technology, Inc., Rept. TR-239-2 (June 20, 1964). 3 I. L., Further comment on 'Low-Altitude, highspeed handling and riding qualities,' J. Aircraft 1, 377 (1964). 4 Soliday, S. M. and Schohan, B., A simulator investigation of pilot performance during extended period of low-altitude, highspeed flight, NASA CR-63 (June 1964). 5 A'Harrah, R. C., Reply by author to I. L. Ashkenas, J. Aircraft 1, 223-224 (1964). 6 I. L., Comment on 'Low-altitude, high-speed handling and riding qualities/ J. Aircraft 1, 222-223 (1964). 7 Richardson, J. D. and A'Harrah, R. C., The application of flight simulators to the development of the A-5A Vigilante, J. Aircraft 1, 171-177 (1964).

Erratum: "Some Opportunities for Progress in Aircraft Performance"

References 1 A'Harrah, R. C., Low-altitude, high-speed handling and riding qualities, J. Aircraft 1, 32-40 (1964). I. L., Jex, H. R., and McRuer, D. T., Pilotinduced oscillations: their cause and analysis, Northrop Corp., Norair Div., Kept. NOR 64-143 (June 20, 1964); also Systems Technology, Inc., Rept. TR-239-2 (June 20, 1964). 3 I. L., Further comment on 'Low-Altitude, highspeed handling and riding qualities,' J. Aircraft 1, 377 (1964). 4 Soliday, S. M. and Schohan, B., A simulator investigation of pilot performance during extended period of low-altitude, highspeed flight, NASA CR-63 (June 1964). 5 A'Harrah, R. C., Reply by author to I. L. Ashkenas, J. Aircraft 1, 223-224 (1964). 6 I. L., Comment on 'Low-altitude, high-speed handling and riding qualities/ J. Aircraft 1, 222-223 (1964). 7 Richardson, J. D. and A'Harrah, R. C., The application of flight simulators to the development of the A-5A Vigilante, J. Aircraft 1, 171-177 (1964).

An analytical and flight-test approach to the reduction of pilot-induced oscillation susceptibility

FEW = effective bobweight force, Ib FE = force error signal, Ib Fs = pilot's applied force, Ib fn = undamped natural frequency, cps g = gravitational acceleration constant, ft/sec GI = inner loop transfer function, gr/lb Gp = pilot's transfer function, Ib/gr j = (-1)1/2 Kp = pilot's gain, Ib/deg \BW = distance from center of gravity to the mass center of the bobweight effectiveness, ft 1 panei = distance from center of gravity to the front cockpit, ft Nz = normal load factor, g s = Laplace operator d/dt, X TN = neuromuscular lag of pilot's transfer function, sec T8 = hydraulic servo valve time constant, sec dH = horizontal stabilizer displacement, deg d$ = longitudinal stick displacement, in. f = damping ratio w = circular frequency, rad/sec wn = undamped natural circular frequency, rad/sec WD = natural damped circular frequency, rad/sec T = pilot's reaction time constant, sec do = pitch angle output of loop, deg 0/ = pitch angle input of loop, deg

The Human Pilot and the High-Speed Airplane

A s AIRCRAFT ATTAIN higher and higher speeds, the •*• ** ability of the human pilot to maintain control becomes marginal in some cases. Recognizing tha t pilot-induced oscillations of large magnitude have been experienced in high-speed low-altitude flight, it is necessary to predict accurately whether the pilot will be able to achieve stable flight with a particular aircraft. Consequently, analytical expressions, which interpret the dynamic characteristics of a human pilot, are needed for stability analysis. Investigations of pilot response which provide data on such physiological factors as reaction t ime and muscular lag have been supported by both the United States Air Force and Navy Bureau of Aeronautics. Under specific test conditions experimental da ta establishing human transfer functions have been obtained. Some of these investigations have taken place at the Franklin Inst i tute, Goodyear Aircraft, and Langley Field, and their findings are included in reference 2. The purpose of the present s tudy is to investigate human dynamics as it affects the airplane in pitch, and to present a means of predicting the pilot transfer function during pitching flight. In detailing the longitudinal stability of a human pilot-aircraft combination, the type of control the pilot a t tempts to apply has been rationalized, and the physiological factors of reaction time and muscular lag have been included.