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

TL;DR

This paper introduces Heterogeneous Meta-Control (HMC), a framework that combines multiple control modalities to improve contact-rich loco-manipulation tasks in robots, demonstrating significant performance gains over baselines.

Contribution

The paper presents a novel HMC framework with an interface for blending control modalities and a heterogeneous policy architecture for robust, force-aware manipulation learning.

Findings

Over 50% improvement over baselines on real robot tasks

Effective blending of control modalities in torque space

Successful learning from large-scale position and force demonstrations

Abstract

Learning from real-world robot demonstrations holds promise for interacting 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 vs. baselines on challenging tasks such as compliant table wiping and drawer opening, demonstrating the efficacy of HMC.