Safe Whole-Body Loco-Manipulation via Combined Model and Learning-based Control

TL;DR

This paper presents a combined model-based and learning-based control framework for safe, compliant whole-body locomotion and manipulation in legged robots, validated through simulation and hardware experiments.

Contribution

It introduces a novel integrated control approach combining admittance control, reinforcement learning, and safety guarantees for legged robots with manipulators.

Findings

Accurate velocity tracking during interactions

Demonstrated safe and compliant behavior in dynamic scenarios

Validated on hardware with real-time performance

Abstract

Simultaneous locomotion and manipulation enables robots to interact with their environment beyond the constraints of a fixed base. However, coordinating legged locomotion with arm manipulation, while considering safety and compliance during contact interaction remains challenging. To this end, we propose a whole-body controller that combines a model-based admittance control for the manipulator arm with a Reinforcement Learning (RL) policy for legged locomotion. The admittance controller maps external wrenches--such as those applied by a human during physical interaction--into desired end-effector velocities, allowing for compliant behavior. The velocities are tracked jointly by the arm and leg controllers, enabling a unified 6-DoF force response. The model-based design permits accurate force control and safety guarantees via a Reference Governor (RG), while robustness is further improved by a Kalman filter enhanced with neural networks for reliable base velocity estimation. We validate our approach in both simulation and hardware using the Unitree Go2 quadruped robot with a 6-DoF arm and wrist-mounted 6-DoF Force/Torque sensor. Results demonstrate accurate tracking of interaction-driven velocities, compliant behavior, and safe, reliable performance in dynamic settings.