Autonomous Science during Kilometer-Scale Robotic Survey

In general today’s planetary exploration robots do not travel beyond the previous day’s imagery. However, advances in autonomous mobility will soon permit single-command site surveys of multiple kilometers. Robots will allocate time and energy resources to characterize large areas from scientific observations. They must navigate and collect data autonomously since scientists cannot see the explored terrain in advance. Advances in pattern recognition, learning and planning can improve these site surveys through onboard science data understanding. We report on field experiments near Amboy Crater, California that demonstrate fundamental capabilities for autonomous surficial mapping of geologic phenomena with a visible near-infrareed spectrometer. We develop an approach to “science on the fly” that dynamically adapts the robot’s exploration decisions based on collected science data. We demonstrate feature detection and visual servoing to permit autonomous data collection from dozens of targets without human intervention. The rover can interpret collected measurements by learning onboard spatial models of science phenomena that guide it toward informative areas. It discovers spatial structure (similarities from one locale to the next) and cross-sensor structure (correlations between different sensing scales). The rover reinterprets remote sensing data in light of the surface materials it discovers on the site and leverages the result to improve exploration efficiency. The Amboy experiments demonstrate that onboard data understanding can facilitate remote site survey on kilometer scales.