Revisiting Reward Design and Evaluation for Robust Humanoid Standing and Walking

A necessary capability for humanoid robots is the ability to stand and walk while rejecting natural disturbances. Recent progress has been made using sim-to-real reinforcement learning (RL) to train such locomotion controllers, with approaches differing mainly in their reward functions. However, prior works lack a clear method to systematically test new reward functions and compare controller performance through repeatable experiments. This limits our understanding of the trade-offs between approaches and hinders progress. To address this, we propose a low-cost, quantitative benchmarking method to evaluate and compare the real-world performance of standing and walking (SaW) controllers on metrics like command following, disturbance recovery, and energy efficiency. We also revisit reward function design and construct a minimally constraining reward function to train SaW controllers. We experimentally verify that our benchmarking framework can identify areas for improvement, which can be systematically addressed to enhance the policies. We also compare our new controller to state-of-the-art controllers on the Digit humanoid robot. The results provide clear quantitative trade-offs among the controllers and suggest directions for future improvements to the reward functions and expansion of the benchmarks.

Learning Perceptive Bipedal Locomotion over Irregular Terrain

In this paper we propose a novel bipedal locomotion controller that uses noisy exteroception to traverse a wide variety of terrains. Building on the cutting-edge advancements in attention based belief encoding for quadrupedal locomotion, our work extends these methods to the bipedal domain, resulting in a robust and reliable internal belief of the terrain ahead despite noisy sensor inputs. Additionally, we present a reward function that allows the controller to successfully traverse irregular terrain. We compare our method with a proprioceptive baseline and show that our method is able to traverse a wide variety of terrains and greatly outperforms the state-of-the-art in terms of robustness, speed and efficiency.