Learning-Based Strategy for Composite Robot Assembly Skill Adaptation

2026-04-08

AI Summary

This work addresses the challenge of robustly executing high-contact insertion tasks—such as peg-in-hole assembly—under stringent geometric tolerances, highly variable friction, and uncertain contact dynamics in industrial robotics. To this end, the authors propose a modular adaptive strategy based on composite skills, wherein each skill is explicitly defined with preconditions, postconditions, and invariants. Residual reinforcement learning (RRL) is employed to locally optimize individual skill phases while preserving the overall hierarchical structure. Implemented within the SAC algorithm and JAX framework, the approach is trained in MuJoCo simulations using a UR5e robotic arm equipped with a Robotiq gripper. Experimental results demonstrate that the method significantly improves sample efficiency and generalization while ensuring safety, thereby achieving robust, reusable, and high-precision assembly performance.

Technology Category

• Intelligent Robots: Manipulation
• 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
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Contact-rich robotic skills remain challenging for industrial robots due to tight geometric tolerances, frictional variability, and uncertain contact dynamics, particularly when using position-controlled manipulators. This paper presents a reusable and encapsulated skill-based strategy for peg-in-hole assembly, in which adaptation is achieved through Residual Reinforcement Learning (RRL). The assembly process is represented using composite skills with explicit pre-, post-, and invariant conditions, enabling modularity, reusability, and well-defined execution semantics across task variations. Safety and sample efficiency are promoted through RRL by restricting adaptation to residual refinements within each skill during contact-rich interactions, while the overall skill structure and execution flow remain invariant. The proposed approach is evaluated in MuJoCo simulation on a UR5e robot equipped with a Robotiq gripper and trained using SAC and JAX. Results demonstrate that the proposed formulation enables robust execution of assembly skills, highlighting its suitability for industrial automation.

Problem

• contact-rich manipulation
• peg-in-hole assembly
• industrial robotics
• position-controlled manipulators
• contact dynamics

Innovation

• Residual Reinforcement Learning
• Composite Skills
• Peg-in-Hole Assembly
• Skill Adaptation
• Contact-Rich Manipulation