Sim-to-Real Learning for Bipedal Locomotion Under Unsensed Dynamic Loads

TL;DR

This paper investigates reinforcement learning-based bipedal locomotion under dynamic loads using only proprioceptive feedback, demonstrating successful sim-to-real transfer and analyzing load-specific gait adaptations.

Contribution

It introduces load-aware RL policies for bipedal robots and compares single versus load-specific training, highlighting the challenges in sim-to-real transfer under dynamic loads.

Findings

Load-aware policies outperform unloaded policies under dynamic loads.

Training for loads improves real-world robustness.

Sim-to-real transfer is successful but has a wider gap than unloaded scenarios.

Abstract

Recent work on sim-to-real learning for bipedal locomotion has demonstrated new levels of robustness and agility over a variety of terrains. However, that work, and most prior bipedal locomotion work, have not considered locomotion under a variety of external loads that can significantly influence the overall system dynamics. In many applications, robots will need to maintain robust locomotion under a wide range of potential dynamic loads, such as pulling a cart or carrying a large container of sloshing liquid, ideally without requiring additional load-sensing capabilities. In this work, we explore the capabilities of reinforcement learning (RL) and sim-to-real transfer for bipedal locomotion under dynamic loads using only proprioceptive feedback. We show that prior RL policies trained for unloaded locomotion fail for some loads and that simply training in the context of loads is enough to result in successful and improved policies. We also compare training specialized policies for each load versus a single policy for all considered loads and analyze how the resulting gaits change to accommodate different loads. Finally, we demonstrate sim-to-real transfer, which is successful but shows a wider sim-to-real gap than prior unloaded work, which points to interesting future research.