Risk-Sensitive Learning and Decision Making for Autonomous Space Robots

This program will develop and demonstrate a risk-sensitive framework for learning and decision-making that allows intelligent physical systems (IPS) to operate in unknown and uncertain environments without continuous control or supervision. By combining a Bayesian approach to learning capable of rapid adaptation online, with a risk-sensitive decision-making approach that is capable of tunable conservatism based on online uncertainty, the framework will allow online learning and safe operation of new autonomous systems in space. The developed framework will be validated on representative scenarios, specifically (1) autonomous traversability assessment for rover navigation, and (2) robust grasping and manipulation of non-cooperative, free-floating objects. The research will be conducted at the Autonomous Systems Laboratory at Stanford and the Robotic AI and Learning Laboratory at UC Berkeley. Both labs combine expertise in online learning and scalable approaches to decision-making in unstructured environments.

The framework will allow online learning and safe operation of new autonomous systems in space, via a combined Bayesian approach to learning capable of rapid adaptation online and a risk-sensitive decision-making approach that is capable of tunable conservatism based on online uncertainty.

This is early/mid-stage (TRL 4) — the most realistic path in is NASA SBIR/STTR, which funds small businesses and research institutions to develop technology aligned with NASA's needs (equity-free, phased funding). Check whether a current SBIR/STTR solicitation topic overlaps with this project's technology area, or contact the project directly (above) to ask.

None of these are guaranteed paths for this specific project — TechPort itself doesn't have an "apply" button. Reaching out to the contact(s) above with a specific question is usually the fastest way to find out what's actually open.