Sim-to-Real of Humanoid Locomotion Policies via Joint Torque Space Perturbation Injection
This addresses the challenge of deploying simulated control policies in real-world robotics, offering a novel method for enhanced robustness in humanoid locomotion.
The paper tackles the sim-to-real transfer problem for humanoid locomotion policies by injecting state-dependent perturbations into joint torque during simulation, which improves robustness against unseen reality gaps without extra training.
This paper proposes a novel alternative to existing sim-to-real methods for training control policies with simulated experiences. Unlike prior methods that typically rely on domain randomization over a fixed finite set of parameters, the proposed approach injects state-dependent perturbations into the input joint torque during forward simulation. These perturbations are designed to simulate a broader spectrum of reality gaps than standard parameter randomization without requiring additional training. By using neural networks as flexible perturbation generators, the proposed method can represent complex, state-dependent uncertainties, such as nonlinear actuator dynamics and contact compliance, that parametric randomization cannot capture. Experimental results demonstrate that the proposed approach enables humanoid locomotion policies to achieve superior robustness against complex, unseen reality gaps in both simulation and real-world deployment.