Virtual Reality and Eyetracking during Docking Training

Abstract

Introduction: The ability to manually dock a spacecraft on a space station can be crucial for mission safety. The computer-based learning program “6df” is an abstract docking simulation that teaches and refreshes the needed skill of controlling six degrees of freedom. During the AGBRESA bed rest study additional technologies can be tested that might improve this learning process. Common two-dimensional (2D) presentation of the learning program “6df” for docking is compared to a stereoscopic three-dimensional (3D) presentation. Since the beginning of docking simulation at the end of the last century, it has been of interest to analyze the operator’s eye movements. Several eye tracking systems have been prepared for use in space but did not reach practicality. With proceeding commercialization, for example in the gaming industry, plug-and-play systems emerged that suggest easy and robust applicability of eye tracking. The study investigates if there is faster learning progress with 3D presentation compared to standard 2D presentation. Moreover, eye tracking is used to answer the question whether there is a relationship between docking quality and extent or timing of visual information processing regarding speed and distance of the spacecraft. Methods: Until now twelve subjects (33.33% female) participated in the ongoing study. Each of them completed 20 training sessions which lasted approximately 45 minutes and were conducted twice a week. The learning program is self-sufficient and adapts itself to the individual learning speed. Half of the participants were presented with an UNITY-based stereoscopic visualization of docking, whereas the other half used the standard version of the learning program “6df”. Learning progress was measured as the number of tasks needed to reach a target task in the middle of the learning range. In the 2D group an eye tracking device (Tobii C4) could be used to assess eye movements as well as pupil dilation. Results: Results showed no significantly faster learning progress while using 3D technology. Looking at the eye movements, there was a continuous decrease in pupil dilation while approaching the docking point. Various task types produced specific eye patterns and subjects differed in eye transition frequency. Furthermore there was a significant positive relationship between visual control of speed and distance to the docking point and the quality of the docking maneuver. Conclusions: Preliminary results suggest that the learning process does not benefit above average from stereoscopic presentation during early training stages. As the real maneuver has to be flown with a two dimensional view, one may favor common 2D training in comparison to the (yet) costlier 3D system. The analysis of eye movements during docking training proved itself to be feasible and an opportunity to gain additional insight into the learning process. There are first hints at the possible suitability of eye tracking for giving learners crucial feedback about productive information processing strategies. Further conclusions are expected by the end of the second AGBRESA campaign.