SHaRe-RL: Structured, Interactive Reinforcement Learning for Contact-Rich Industrial Assembly Tasks

2025-09-17

AI Summary

Reinforcement learning (RL) faces challenges in high-mix, low-volume (HMLV) industrial assembly—including low sample efficiency, unsafe exploration, and poor adaptability to contact-rich tasks. Method: We propose a safe and efficient learning framework integrating structured operational primitives, human demonstration guidance, online human-robot interaction for correction, and multi-axis compliant force constraints. This reduces reliance on expert programming and enables domain-expert process engineers—not robotics specialists—to participate in deployment. Skill structuring and force-feedback-driven compliance enable long-horizon, high-precision assembly decisions (0.2–0.4 mm gap tolerance). Results: Evaluated on Harting connector insertion, the system achieves reliable autonomous learning within limited training time, demonstrating safety, generalization across variants, and engineering practicality in real-world HMLV settings.

Technology Category

• Intelligent Robots: Learning & Optimization for ROB
• Machine Learning: Adversarial Learning & Robustness
• Planning, Routing, and Scheduling: Learning for Planning and Scheduling

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

High-mix low-volume (HMLV) industrial assembly, common in small and medium-sized enterprises (SMEs), requires the same precision, safety, and reliability as high-volume automation while remaining flexible to product variation and environmental uncertainty. Current robotic systems struggle to meet these demands. Manual programming is brittle and costly to adapt, while learning-based methods suffer from poor sample efficiency and unsafe exploration in contact-rich tasks. To address this, we present SHaRe-RL, a reinforcement learning framework that leverages multiple sources of prior knowledge. By (i) structuring skills into manipulation primitives, (ii) incorporating human demonstrations and online corrections, and (iii) bounding interaction forces with per-axis compliance, SHaRe-RL enables efficient and safe online learning for long-horizon, contact-rich industrial assembly tasks. Experiments on the insertion of industrial Harting connector modules with 0.2-0.4 mm clearance demonstrate that SHaRe-RL achieves reliable performance within practical time budgets. Our results show that process expertise, without requiring robotics or RL knowledge, can meaningfully contribute to learning, enabling safer, more robust, and more economically viable deployment of RL for industrial assembly.

Problem

• Enabling flexible automation for high-mix low-volume industrial assembly tasks
• Addressing poor sample efficiency in contact-rich reinforcement learning
• Ensuring safe exploration during robotic assembly learning processes

Innovation

• Structured skills into manipulation primitives
• Incorporated human demonstrations and corrections
• Bounded interaction forces with per-axis compliance