Learning Motion Skills with Adaptive Assistive Curriculum Force in Humanoid Robots

2025-06-29

AI Summary

To address the low learning efficiency and poor robustness in acquiring complex motor skills for humanoid robots, this paper proposes the Adaptive Assistive Control Framework (A2CF). A2CF establishes a state-aware dual-agent collaborative system that dynamically generates assistive forces and implements a progressive assistance reduction scheme, mimicking infant motor development under physical guidance. The method integrates reinforcement learning, state-dependent force control, and curriculum learning to achieve efficient policy optimization in high-dimensional action spaces. Evaluated on bipedal walking, choreographed dancing, and backflip tasks, A2CF improves training convergence speed by 30% and reduces failure rates by over 40%, yielding fully autonomous and robust locomotion policies—validated on a real robot platform. The core contribution lies in modeling developmental physical assistance as a learnable closed-loop control problem, enabling, for the first time, online adaptive modulation of assistive forces and concurrent evolution of control policies.

Technology Category

• Intelligent Robots: Behavior Learning & Control
• Humans and AI: Human-Aware Planning and Behavior Prediction
• Machine Learning: Imitation Learning & Inverse Reinforcement Learning

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Search and Retrieval-Augmented AI: Agentic search
• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems

Abstract

Learning policies for complex humanoid tasks remains both challenging and compelling. Inspired by how infants and athletes rely on external support--such as parental walkers or coach-applied guidance--to acquire skills like walking, dancing, and performing acrobatic flips, we propose A2CF: Adaptive Assistive Curriculum Force for humanoid motion learning. A2CF trains a dual-agent system, in which a dedicated assistive force agent applies state-dependent forces to guide the robot through difficult initial motions and gradually reduces assistance as the robot's proficiency improves. Across three benchmarks--bipedal walking, choreographed dancing, and backflip--A2CF achieves convergence 30% faster than baseline methods, lowers failure rates by over 40%, and ultimately produces robust, support-free policies. Real-world experiments further demonstrate that adaptively applied assistive forces significantly accelerate the acquisition of complex skills in high-dimensional robotic control.

Problem

• Learning complex humanoid motion skills efficiently
• Reducing failure rates in robotic motion tasks
• Adaptive assistive forces for skill acquisition

Innovation

• Dual-agent system with assistive force
• State-dependent adaptive force guidance
• Gradual assistance reduction for proficiency