Color-based Vision Tracking for an Astronaut EVA Assist Vehicle

Scenarios for a manned mission to Mars call for astronaut extravehicular activity teams to be accompanied by semi-autonomous rovers. These rovers must be able to safely follow the astronauts over a variety of terrain with minimal assistance. We propose a color-based visual tracking system as a high-speed robust sensory approach for astronaut following and present a quantitative analysis of its accuracy over a spectrum of astronaut motion profiles.

To evaluate performance, we provide astronaut motion profiles to a simulator and record actual and perceived target location. We characterize tracking accuracy as a function of rover-astronaut separation distance and astronaut speed and heading. Future plans include closed-loop control of the rover based on visual feedback, rover motion simulation, and target re-acquisition algorithms.