This paper introduces a new method of providing motion cues to a moving base six-degree-of-freedom flight simulator utilizing nonlinear filters. Coordinated adaptive filters, used to coordinate translational and rotational motion, are derived based on the method of continuous steepest descent, and the basic concept of the digital controllers used for the uncoordinated heave and yaw cues is also presented. The coordinated adaptive washout method is illustrated by an application in a six-degree-of-freedom fixed-base environment.

Coordinated Adaptive Washout for Motion Simulators

Theeffectsofe exibility on thee ightdynamicsoflargeaircraft havebeen shown to bequitesignie cant, especially as the frequencies of the elastic modes become lower and approach those of the rigid-body modes. The handling characteristics of such vehicles are altered signie cantly from those of a rigid vehicle, and the design of the e ightcontrol system may become drastically more complex. Consequently, the need to model accurately the dynamics of such vehicles, and to develop valid simulations, is becoming particularly acute. The theoretical development of a generic e exible-aircraft model is reviewed. This modeling technique allows for the e exible degrees of freedom to be added to an existing simulation model of the vehicle’ s rigid-body dynamics. The data necessary for modeling a specie c vehicle include aerodynamic stability derivatives, aerodynamic ine uence coefe cients, elastic mode shapes, modal frequencies and damping, and generalized masses. Two case studies are presented, both involving the developmentofmotion-basedsimulationsinNASALangleyResearchCenter’ ssimulationfacility.Thee rstexample investigatesanaircraftsimilaringeometrytotheB-1,whereasthesecondinvestigatesalargehighspeedcommercial transport. The dynamic responses (time and frequency domain ) for these vehicles are presented, along with the evaluations of the effects of dynamic aeroelasticity on their handling characteristics. The vehicle responses are critical to evaluating hardware and software requirements for simulation e delity, for example, visual and motion cues. Finally, a brief assessment of the effects of limitations of the simulation facility is presented. Limitations considered include digital time delay, motion hardware bandwidth, and motion washout logic.

Modeling and Simulation of Flexible Flight Vehicles

: The need for good aircraft handling qualities has been apparent since the days of the Wright Flyer. In the past decade, there has been a perception that this need has lessened as advanced concepts have evolved, in parallel with acquisition reform. The former has led those who are unfamiliar with the field of handling qualities to conclude that quantitative requirements are not necessary, as the latter has resulted in the elimination of the military specifications for handling qualities. This paper reviews the evolution of handling qualities and their specifications. It presents some ongoing challenges in the field to illustrate that handling qualities are still a critical issue for future aircraft.

Evolution, Revolution, and Challenges of Handling Qualities

The authors conducted further research with cueing algorithms for control of flight simulator motion systems. A variation of the so-called optimal algorithm was formulated using simulated aircraft angular velocity input as a basis. Models of the human vestibular sensation system, i.e. the semicircular canals and otoliths, are incorporated within the algorithm. Comparisons of angular velocity cueing responses showed a significant improvement over a formulation using angular acceleration input. Results also compared favorably with the coordinated adaptive washout algorithm, yielding similar results for angular velocity cues while eliminating false cues and reducing the tilt rate for longitudinal cues. These results were confirmed in piloted tests on the current motion system at NASA-Langley, the Visual Motion Simulator (VMS). Proposed future developments by the authors in cueing algorithms are revealed. The new motion system, the Cockpit Motion Facility (CMF), where the final evaluation of the cueing algorithms will be conducted, is also described.

/pdf/developments-in-human-centered-cueing-algorithms-for-control-4b2ussp7dw.pdf

Developments in Human Centered Cueing Algorithms for Control of Flight Simulator Motion Systems

Delayed visual feedback has long been known to effect the control input human beings make when they are an integral part of a system. The loss of time between system input and output forces human controllers to change how they maintain a desired system state. These changes have been documented, as have various schemes to compensate for a delay. These developments span work on human tracking and the control of movement, vehicle simulation (especially aircraft simulation for flight training), robotics for manufacturing and space exploration, and current work on virtual reality displays. The bibliography contains references to papers on human control of closed-loop systems with delayed feedback and on software compensation schemes for delays. Human input to systems where visual feedback about the results of the input is delayed has long been a topic of interest as the delay is an analog for the many ways time can be lost in the real world. These losses may be the time needed for communications to travel vast distances or they may come from the time expended on the calculations needed to define the response of a system and to display it graphically to a human operator. For experimentalists, delay of visual feedback has served as a reliable variable over the past four decades for, in the limit, lagging a system's output can always be expected to effect human control performance. Manual control, and changes to it caused by the presence of a feedback delay, can be thought of as determined partly by the dynamics of the controlled system, partly by the size of the delay (or the human controller's ability to compensate for it), and partly by the requirements of the task being performed. That is, the reasons delays cause problems can be related in part to relatively stable properties of human beings, and in part to characteristics of the system they are controlling and the task at hand. A sizable literature about the control of systems with delays has accumulated, and much of it is relevant to developments where human beings will operate either in artificial, computergenerated worlds or remotely through sensors. This work started in the late 1950s and early 1960s as studies of human tracking performance where individuals controlled single-axis systems with uncomplicated dynamic responses. The purpose was to investigate the acceptability of the firstand second-order lags of mechanical systems. At about that time also, the relatively long transmission times associated with remote space and planetary exploration stimulated interest in how well human controllers could accommodate delays of up to several seconds. The 1970s brought computer image generation to flight simulation, and while the delays involved were only on the order of 100 to 200 msec, they were long enough to produce demonstrable changes to the control behavior of pilots. The wide field-of-view displays of flight sinrulators made apparent the behavior of elements in the visual scene and many of the shortcuts taken in the shnulation's mathematical nrodel. Worry about the fidelity of visual or motion cues (and thus ultimately about the acceptability of the training devices themselves) stimulated the development of software compensations for display delays. Indeed, much of the work measuring the effects of delayed visual feedback has been performed in the context of flight simulation because of the need for an update rate of at least 30 Hz for the visual scene. Current effort is to form networks of distributed simulators and the bandwidth of the communication between thenr forces delay to be a problem still. While interest in simulation fidelity continues to the present, work on delay compensation changed focus to vehicle control when significant computational delays were introduced by the development of digital "fly-by-wire" aircraft. In addition, plans for a national space station included a remote manipulation capability so that astronauts could use a robotic teleoperation system to repair distant objects. As in simulated flight, a teleoperator may be using an artificial visual display which may contain significant computation or network delays. Current interest in graphical "environnrents" and the technology to allow people to interact with them (using head-mounted displays, position tracking devices, directional sound, and gesture gloves and suits) faces many of the same problems encountered during the widespread development of simulation for flight training. Addit ional time or computational power is required to increase the resolution or complexity or number of images a display system can provide. This equivalence of time and computational resources is the trade-off that has maintained interest in the problems caused by delays. The following bibliography lists studies that document human controllers' responses to display delays. In addition, a number of compensation schemes have been developed that reduce the effects of delays, and these are included also.

Manual control with delays: a bibliography

The inertial response characteristics of a synergistic, six-degree-of-freedom motion base are presented in terms of amplitude ratio and phase lag as functions of frequency data for the frequency range of interest (0 to 2 Hz) in real time, digital, flight simulators. The notch filters which smooth the digital-drive signals to continuous drive signals are presented, and appropriate compensation, based on the inertial response data, is suggested. The existence of an inverse transformation that converts actuator extensions into inertial positions makes it possible to gather the response data in the inertial axis system.

/pdf/compensation-based-on-linearized-analysis-for-a-six-degree-2ecc2p44ww.pdf

J. L. Copeland

Papers

Compensation based on linearized analysis for a six degree of freedom motion simulator