Multimodal Pilot Behavior in Multi-Axis Tracking Tasks with Time-Varying Motion Cueing Gains

Abstract

In a large number of motion-base simulators, adaptive motion filters are utilized to maximize the use of the available motion envelope of the motion system. However, not much is known about how the time-varying characteristics of such adaptive filters affect pilots when performing manual aircraft control. This paper presents the results of a study investigating the effects of time-varying motion filter gains on pilot control behavior and performance. An experiment was performed in a motion-base simulator where participants performed a simultaneous roll and pitch tracking task, while the roll and/or pitch motion filter gains changed over time. Results indicate that performance increases over time with increasing motion gains. This increase is a result of a time-varying adaptation of pilots' equalization dynamics, characterized by increased visual and motion response gains and decreased visual lead time constants. Opposite trends are found for decreasing motion filter gains. Even though the trends in both controlled axes are found to be largely the same, effects are less significant in roll. In addition, results indicate minor cross-coupling effects between pitch and roll, where a cueing variation in one axis affects the behavior adopted in the other axis.

Effects of Motion Cues on the Training of Multi-Axis Manual Control Skills

“…This result has also been observed in other multi-axis control studies. 18,24,25 Pilot control activity for roll and pitch are shown in Figures 7a and 7b was not expected based on previous training studies, for which control activity increased when training with motion. 12 The RMS of the roll control input during training was higher for HO and control activity at the end of training was more similar to that after transfer.…”

“…Even though the roll and pitch controlled dynamics didn't have any cross coupling, cross coupling in manual control behavior and performance was most likely still present, as found in previous studies with similar control tasks. 18,24,25 Analyzing the extend of this cross coupling, for example, using the method described in Ref. 18, might give additional insights, and might help explain some of the results found in this study.…”

“…Previous experiments have shown that using this approach in experimental settings that are less restricted, for example, by using a multiple-axis control task instead of a single-axis control task, may result in difficulties in finding global optimum pilot model parameter estimates using MLE. 25 A transfer-of-training experiment, where the goal is to analyze participants' changing control strategy between single experimental runs, exacerbates the situation. For this reason, we still averaged five consecutive runs in the MLE procedure, a common practice in non-training experiments, increasing the chances of finding global optimum parameter sets.…”

Manual Control Cybernetics: State-of-the-Art and Current Trends

“…Such methods are indispensable for studying what actually varies in HCs and which "function approximators" can best describe the adapting HC parameters. Once known which time-variations in the HC need to be modeled, promising approaches for the second step of fitting intrinsically time-varying manual control models to measured HC data are time-domain modeling [99], [146] or LPV modelbased methods [135], [153], [154]. The main challenge for time-varying HC identification lies in developing methods that are sufficiently sensitive and that can reliably pick out quick and short-duration temporal variations in HC behavior from inherently noisy data.…”

“…In reality, however, it is the adaptive nature of the HC, and how she or he is able to respond to sudden changes in the environment, that is of interest, yet still largely unknown [2], [110], [139]. Relevant real-world scenarios where HCs are forced to adapt their control behavior are, for example, time-varying changes in the controlled system dynamics (e.g., failure) [98], [99], [140]- [144], instantaneously modified task constraints (e.g., decreased road width) [135], loss-of-control [145], automaticto-manual control transitions, and control with time-varying information feedback (e.g., adaptive simulator motion feedback) [146]. Such time-varying HC adaptations are inherently highly variable, nonlinear, short-duration, and strongly taskdependent, making them immensely more complex to grasp than LTI HC behavior.…”