SteadyTray: Learning Object Balancing Tasks in Humanoid Tray Transport via Residual Reinforcement Learning

TL;DR

This paper presents SteadyTray, a hierarchical reinforcement learning framework that effectively stabilizes payloads during humanoid tray transport, ensuring robustness and zero-shot sim-to-real transfer in dynamic environments.

Contribution

The paper introduces ReST-RL, a modular hierarchical reinforcement learning architecture that decouples locomotion from payload stabilization for humanoid robots.

Findings

Achieved 96.9% success in variable velocity tracking.

Attained 74.5% robustness against external force disturbances.

Demonstrated reliable zero-shot sim-to-real transfer on hardware.

Abstract

Stabilizing unsecured payloads against the inherent oscillations of dynamic bipedal locomotion remains a critical engineering bottleneck for humanoids in unstructured environments. To solve this, we introduce ReST-RL, a hierarchical reinforcement learning architecture that explicitly decouples locomotion from payload stabilization, evaluated via the SteadyTray benchmark. Rather than relying on monolithic end-to-end learning, our framework integrates a robust base locomotion policy with a dynamic residual module engineered to actively cancel gait-induced perturbations at the end-effector. This architectural separation ensures steady tray transport without degrading the underlying bipedal stability. In simulation, the residual design significantly outperforms end-to-end baselines in gait smoothness and orientation accuracy, achieving a 96.9% success rate in variable velocity tracking and 74.5% robustness against external force disturbances. Successfully deployed on the Unitree G1 humanoid hardware, this modular approach demonstrates highly reliable zero-shot sim-to-real generalization across various objects and external force disturbances.