HMC: Learning Heterogeneous Meta-Control for Contact-Rich Loco-Manipulation

Lai Wei, Xuanbin Peng, Ri-Zhao Qiu, Xuxin Cheng, Xiaolong Wang

Keywords: Loco-Manipulation, Contact-Rich Manipulation, Heterogeneous Learning, Teleoperation

TL;DR: A meta controller framework to combine position-based learning and force-aware learning evaluated on a humanoid robot.

Abstract: Learning from real-world robot demonstrations hold promises for inter acting with complex real-world environments. However, the complexity and variability of interaction dynamics often cause purely positional controllers to struggle with contacts or varying payloads. To address this, we propose a Heterogeneous Meta-Control (HMC) framework for Loco-Manipulation that adaptively stitches multiple control modalities: position, impedance, and hybrid force-position. We first introduce an interface, HMC-Controller, for blending actions from different control profiles continuously in the torque space. HMC-Controller facilitates both teleoperation and policy deployment. Then, to learn a robust force-aware policy, we propose HMC-Policy to unify different controllers into a heterogeneous architecture. We adopt a mixture-of-experts style routing to learn from large-scale position-only data and fine-grained force-aware demonstrations. Experiments on a real humanoid robot show over 50% relative improvement over baselines on challenging tasks such as compliant table wiping and drawer opening, demonstrating the efficacy of HMC.

Submission Number: 14