Switch: Learning Agile Skills Switching for Humanoid Robots

2026-04-16

AI Summary

This work addresses the challenge of achieving real-time, safe, and smooth transitions among diverse motor skills in humanoid robots. To this end, the authors propose Switch, a hierarchical multi-skill system that innovatively constructs a skill graph based on kinematic similarity to model inter-skill transition relationships. By integrating whole-body motion tracking with an online scheduler, the system enables seamless switching between arbitrary skills at any time. The framework combines deep reinforcement learning with online graph search algorithms to achieve agile, high-success-rate transitions across a variety of complex motor skills while preserving high-fidelity motion imitation performance.

Technology Category

• Intelligent Robots: Multimodal Perception & Sensor Fusion
• 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
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Recent advancements in whole-body control through deep reinforcement learning have enabled humanoid robots to achieve remarkable progress in real-world chal lenging locomotion skills. However, existing approaches often struggle with flexible transitions between distinct skills, cre ating safety concerns and practical limitations. To address this challenge, we introduce a hierarchical multi-skill system, Switch, enabling seamless skill transitions at any moment. Our approach comprises three key components: (1) a Skill Graph (SG) that establishes potential cross-skill transitions based on kinematic similarity within multi-skill motion data, (2) a whole-body tracking policy trained on this skill graph through deep reinforcement learning, and (3) an online skill scheduler to drive the tracking policy for robust skill execution and smooth transitions. For skill switching or significant tracking deviations, the scheduler performs online graph search to find the optimal feasible path, which ensures efficient, stable, and real-time execution of diverse locomotion skills. Comprehensive experiments demonstrate that Switch empowers humanoid to execute agile skill transitions with high success rates while maintaining strong motion imitation performance.

Problem

• humanoid robots
• skill switching
• locomotion skills
• transition flexibility
• whole-body control

Innovation

• skill switching
• hierarchical control
• skill graph
• deep reinforcement learning
• humanoid locomotion