RL-Augmented MPC for Non-Gaited Legged and Hybrid Locomotion

2026-03-11

AI Summary

This work addresses the challenge of locomotion control for legged and wheeled-legged hybrid robots in unstructured environments, where complex contact sequencing complicates motion planning. The authors propose an explicit contact-aware hierarchical architecture that decouples contact timing planning from execution: a high-level reinforcement learning agent generates gait and navigation commands, while a low-level model predictive controller carries out the motion. Notably, this approach achieves zero-shot sim-to-real transfer without domain randomization and demonstrates emergent capabilities—including aperiodic gait generation and adaptive contact timing—across multiple robotic platforms weighing 50–120 kg. Experimental validation on the Centauro wheeled-legged humanoid confirms the method’s effectiveness, and the associated software framework along with evaluation results has been publicly released.

Technology Category

• Intelligent Robots: Behavior Learning & Control
• Humans and AI: Human-Aware Planning and Behavior Prediction
• Planning, Routing, and Scheduling: Learning for Planning and Scheduling

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Search and Retrieval-Augmented AI: Agentic search
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

We propose a contact-explicit hierarchical architecture coupling Reinforcement Learning (RL) and Model Predictive Control (MPC), where a high-level RL agent provides gait and navigation commands to a low-level locomotion MPC. This offloads the combinatorial burden of contact timing from the MPC by learning acyclic gaits through trial and error in simulation. We show that only a minimal set of rewards and limited tuning are required to obtain effective policies. We validate the architecture in simulation across robotic platforms spanning 50 kg to 120 kg and different MPC implementations, observing the emergence of acyclic gaits and timing adaptations in flat-terrain legged and hybrid locomotion, and further demonstrating extensibility to non-flat terrains. Across all platforms, we achieve zero-shot sim-to-sim transfer without domain randomization, and we further demonstrate zero-shot sim-to-real transfer without domain randomization on Centauro, our 120 kg wheeled-legged humanoid robot. We make our software framework and evaluation results publicly available at https://github.com/AndrePatri/AugMPC.

Problem

• non-gaited locomotion
• contact timing
• legged robots
• hybrid locomotion
• combinatorial complexity

Innovation

• Reinforcement Learning
• Model Predictive Control
• Non-gaited Locomotion
• Zero-shot Sim-to-Real Transfer
• Contact-explicit Hierarchical Control