Thruster-Enhanced Locomotion: A Decoupled Model Predictive Control with Learned Contact Residuals

TL;DR

This paper introduces a decoupled control architecture for a quadruped robot with thrusters, combining a position-based leg controller with a learned residual-enhanced MPC for thrusters, enabling stable narrow-path walking and impact handling.

Contribution

It proposes a novel decoupled control framework with learned contact residuals to improve thruster-assisted quadruped locomotion stability.

Findings

Enhanced stability in push recovery and gait control

Effective leg-ground impact modeling with learned residuals

Successful validation through simulation and hardware experiments

Abstract

Husky Carbon, a robot developed by Northeastern University, serves as a research platform to explore unification of posture manipulation and thrust vectoring. Unlike conventional quadrupeds, its joint actuators and thrusters enable enhanced control authority, facilitating thruster-assisted narrow-path walking. While a unified Model Predictive Control (MPC) framework optimizing both ground reaction forces and thruster forces could theoretically address this control problem, its feasibility is limited by the low torque-control bandwidth of the system's lightweight actuators. To overcome this challenge, we propose a decoupled control architecture: a Raibert-type controller governs legged locomotion using position-based control, while an MPC regulates the thrusters augmented by learned Contact Residual Dynamics (CRD) to account for leg-ground impacts. This separation bypasses the torque-control rate bottleneck while retaining the thruster MPC to explicitly account for leg-ground impact dynamics through learned residuals. We validate this approach through both simulation and hardware experiments, showing that the decoupled control architecture with CRD performs more stable behavior in terms of push recovery and cat-like walking gait compared to the decoupled controller without CRD.