Learning Robust, Agile, Natural Legged Locomotion Skills in the Wild
This work addresses the problem of improving robot locomotion for real-world applications by making it less conservative and more natural, though it appears incremental as it builds on existing sim-to-real and teacher-student methods.
The paper tackles the problem of learning legged locomotion skills that are robust, agile, and natural over challenging terrain, achieving results where a quadruped robot can traverse stairs, rocky ground, and slippery floors with only proprioceptive perception, while showing more agile, natural, and energy-efficient gaits compared to baselines.
Recently, reinforcement learning has become a promising and polular solution for robot legged locomotion. Compared to model-based control, reinforcement learning based controllers can achieve better robustness against uncertainties of environments through sim-to-real learning. However, the corresponding learned gaits are in general overly conservative and unatural. In this paper, we propose a new framework for learning robust, agile and natural legged locomotion skills over challenging terrain. We incorporate an adversarial training branch based on real animal locomotion data upon a teacher-student training pipeline for robust sim-to-real transfer. Empirical results on both simulation and real world of a quadruped robot demonstrate that our proposed algorithm enables robustly traversing challenging terrains such as stairs, rocky ground and slippery floor with only proprioceptive perception. Meanwhile, the gaits are more agile, natural, and energy efficient compared to the baselines. Both qualitative and quantitative results are presented in this paper.