MuJoCo MPC for Humanoid Control: Evaluation on HumanoidBench

TL;DR

This paper improves humanoid robot control in MuJoCo MPC by introducing regularization to achieve realistic behaviors, leading to higher benchmark scores and smoother control signals.

Contribution

The paper proposes regularization terms for MuJoCo MPC that enhance stability and realism in humanoid control, addressing issues with sparse reward functions.

Findings

Regularization improves stability and realism in humanoid control.

Achieves highest scores on HumanoidBench with realistic postures.

Code availability facilitates rapid prototyping.

Abstract

We tackle the recently introduced benchmark for whole-body humanoid control HumanoidBench using MuJoCo MPC. We find that sparse reward functions of HumanoidBench yield undesirable and unrealistic behaviors when optimized; therefore, we propose a set of regularization terms that stabilize the robot behavior across tasks. Current evaluations on a subset of tasks demonstrate that our proposed reward function allows achieving the highest HumanoidBench scores while maintaining realistic posture and smooth control signals. Our code is publicly available and will become a part of MuJoCo MPC, enabling rapid prototyping of robot behaviors.