Learning Terrain-Specialized Policies for Adaptive Locomotion in Challenging Environments
This work addresses robust locomotion for legged robots in challenging environments, representing an incremental improvement with specific gains in simulation.
The paper tackled the problem of enabling legged robots to navigate diverse, unstructured terrains without prior terrain information, achieving up to 16% higher success rates and lower tracking errors compared to a generalist policy in simulations.
Legged robots must exhibit robust and agile locomotion across diverse, unstructured terrains, a challenge exacerbated under blind locomotion settings where terrain information is unavailable. This work introduces a hierarchical reinforcement learning framework that leverages terrain-specialized policies and curriculum learning to enhance agility and tracking performance in complex environments. We validated our method on simulation, where our approach outperforms a generalist policy by up to 16% in success rate and achieves lower tracking errors as the velocity target increases, particularly on low-friction and discontinuous terrains, demonstrating superior adaptability and robustness across mixed-terrain scenarios.