Motion-Visual Phase-Error Detection in a Flight Simulator

Abstract

An experiment was conducted on the University of Toronto Institute for Aerospace Studies Flight Research Simulator to determine the minimum phase lead of pitch motion cues relative to pitch visual cues that can be consistently detected by a human observer. The effects of pitch frequency and amplitude, motion gain, and visual scene complexity on the detection threshold of the phase error between the visual and motion cues was determined. The mean detection threshold of the phase error averaged across all subjects and conditions was 57 deg. Pitch amplitude significantly affected the detection threshold of the phase error. Higher amplitudes led to lower detection thresholds. Motion gain also had a significant effect on the detection threshold of the phase error when the frequency was 0.2 Hz or the visual complexity was low. Higher motion gains led to lower detection thresholds. The frequency had a significant effect on the detection threshold of the phase error when the motion gain was 0.5 or the visual complexity was low. Higher frequencies led to lower detection thresholds. The direction of the frequency effect suggests that subjects perform more like motion-visual phase detectors than motion-visual time delay detectors. The results of the experiment were used to analyze pitch high-pass washout filters. The analysis suggests that the break frequency for a second-order washout filter should be lower than 0.13 rad/s and the break frequency for a first-order filter should be lower than 0.2 rad/s to keep the motion-visual phase error below the measured human perception limit.