Gameplay Filters: Safe Robot Walking through Adversarial Imagination

arXiv:2405.00846v1 Announce Type: cross
Abstract: Ensuring the safe operation of legged robots in uncertain, novel environments is crucial to their widespread adoption. Despite recent advances in safety filters that can keep arbitrary task-driven policies from incurring safety failures, existing solutions for legged robot locomotion still rely on simplified dynamics and may fail when the robot is perturbed away from predefined stable gaits. This paper presents a general approach that leverages offline game-theoretic reinforcement learning to synthesize a highly robust safety …

abstract adoption advances adversarial arxiv cs.lg cs.ro dynamics environments filters imagination legged robot legged robots novel policies robot robots safe safety simplified solutions through type uncertain walking