Planetary Rover Absolute Localization by Combining Visual Odometry with Orbital Image Measurements

Visual Odometry (VO) has established itself as an important localization technology for planetary exploration rovers, being capable of yielding location estimates with small error over medium-sized trajectories. However, due to VO’s incremental mode of operation, the estimation error accumulates over time, resulting in considerable drift for long trajectories. This paper proposes a global localization method that counters VO drift by matching boulders extracted from overhead and ground images of a planet and using them periodically to re-localize the rover and refine VO estimates. The performance of the method is evaluated with the aid of overhead imagery of different resolutions. Experimental results demonstrate that a very terse representation, consisting of approximate boulder locations only, suffices for bringing significant accuracy improvements to VO over long traverses.