Learning Provably Robust Motion Planners Using Funnel Libraries

Abstract

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.

References in corpus (15)

• Adam: A Method for Stochastic Optimization
• Solving Rubik's Cube with a Robot Hand
• Emergent Tool Use From Multi-Agent Autocurricula
• Certifying Some Distributional Robustness with Principled Adversarial Training
• A Study on Overfitting in Deep Reinforcement Learning
• Computing Nonvacuous Generalization Bounds for Deep (Stochastic) Neural Networks with Many More Parameters than Training Data
• Multi-Goal Reinforcement Learning: Challenging Robotics Environments and Request for Research
• Evolved Policy Gradients
• Improving Generalization in Meta Reinforcement Learning using Learned Objectives
• CausalWorld: A Robotic Manipulation Benchmark for Causal Structure and Transfer Learning
• Distributionally Robust Reinforcement Learning
• Distributional Robustness and Regularization in Reinforcement Learning
• PAC-Bayes with Backprop
• Generalization Guarantees for Imitation Learning
• Distributionally Robust Policy Learning via Adversarial Environment Generation