Learning Provably Robust Motion Planners Using Funnel Libraries
This addresses the need for provably robust motion planning in robotics, offering incremental improvements by integrating existing tools for enhanced reliability.
The paper tackles the problem of learning motion planners with probabilistic guarantees of success in new environments under disturbances, by combining generalization theory and robust control to optimize PAC-Bayes bounds for composing robust motion primitives, achieving strong guarantees in simulated autonomous vehicle and drone navigation scenarios.
This paper presents an approach for learning motion planners that are accompanied with probabilistic guarantees of success on new environments that hold uniformly for any disturbance to the robot's dynamics within an admissible set. We achieve this by bringing together tools from generalization theory and robust control. First, we curate a library of motion primitives where the robustness of each primitive is characterized by an over-approximation of the forward reachable set, i.e., a "funnel". Then, we optimize probably approximately correct (PAC)-Bayes generalization bounds for training our planner to compose these primitives such that the entire funnels respect the problem specification. We demonstrate the ability of our approach to provide strong guarantees on two simulated examples: (i) navigation of an autonomous vehicle under external disturbances on a five-lane highway with multiple vehicles, and (ii) navigation of a drone across an obstacle field in the presence of wind disturbances.