Gait Patterns in an Ambulatory Virtual Reality Exploration Extravehicular Activity Simulation

Exploration extravehicular activity (EVA) places extraordinary physical and cognitive demands on crewmembers. To characterize these demands, an ambulatory virtual reality (VR) simulation was developed. During the simulation, subjects traverse along a virtual lunar surface by walking on a passive treadmill. Each physical step is mapped to a virtual step using VR ankle trackers. During the traverse, the subjects communicate waypoints in the scene while managing resources and monitoring systems using information in a heads-up-display projected through the VR headset. The immersive, shirtsleeve environment serves as a testbed and research tool for future technology and operations concepts with the ability to adjust the cognitive and physical demands of the simulation. To objectively quantify cognitive workload during the simulation, performance is assessed on the subject�s ability to accurately and quickly complete the communication, resource management, and system monitoring tasks. Complementing this set of embedded performance metrics, the aim of this work is to investigate the effect of training and cognitive demand on gait patterns during the VR simulation. Similar to the task performance metrics, gait performance metrics were expected to change with familiarization of the system and availability of cognitive resources. Data was collected at NASA Johnson Space Center in the Human Physiology, Performance, Protection, and Operations (H-3PO) Laboratory with eight subjects completing a training simulation and two testing simulations, in high and low cognitive workload configurations. Spatiotemporal metrics were calculated from the data recorded by the VR ankle trackers. Walking on the passive treadmill in VR improved from the training session to the testing sessions with faster average walking speeds and less variable stride, stance, and swing times. This knowledge will help tune the training protocol for future testing. Further work is needed to assess the sensitivity of the gait metrics to the cognitive workload level.