Robotic Agentic Platform for Intelligent Electric Vehicle Disassembly

2026-03-19

AI Summary

This work addresses the challenge of scalable recycling of electric vehicle battery packs, whose structural heterogeneity necessitates labor-intensive disassembly. The authors propose a perception-driven embodied agent framework that integrates an industrial robotic arm, RGB-D vision, and an automated fastening tool, enabling reliable mapping from large language models (LLMs) to robotic actions through open-vocabulary object detection (mAP50 = 0.9757) and a structured tool interface. Built upon the SmolAgents framework, the system deploys GPT-4o-mini and Qwen models on edge devices to support intelligent disassembly of full-scale battery packs. Experiments demonstrate a 97% success rate using a teach-in pose strategy and 100% task completion via the structured API—significantly outperforming automatic ROS service discovery, which exhibits a 43.3% failure rate. The platform has been open-sourced to advance research in human-robot collaboration and autonomous disassembly.

Technology Category

• Intelligent Robots: Embodied AI
• Machine Learning: Large Multimodal Models (LMMs)
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Search and Retrieval-Augmented AI: Agentic search

Abstract

Electric vehicles (EV) create an urgent need for scalable battery recycling, yet disassembly of EV battery packs remains largely manual due to high design variability. We present our Robotic Agentic Platform for Intelligent Disassembly (RAPID), designed to investigate perception-driven manipulation, flexible automation, and AI-assisted robot programming in realistic recycling scenarios. The system integrates a gantry-mounted industrial manipulator, RGB-D perception, and an automated nut-running tool for fastener removal on a full-scale EV battery pack. An open-vocabulary object detection pipeline achieves 0.9757 mAP50, enabling reliable identification of screws, nuts, busbars, and other components. We experimentally evaluate (n=204) three one-shot fastener removal strategies: taught-in poses (97% success rate, 24 min duration), one-shot vision execution (57%, 29 min), and visual servoing (83%, 36 min), comparing success rate and disassembly time for the battery's top cover fasteners. To support flexible interaction, we introduce agentic AI specifications for robotic disassembly tasks, allowing LLM agents to translate high-level instructions into robot actions through structured tool interfaces and ROS services. We evaluate SmolAgents with GPT-4o-mini and Qwen 3.5 9B/4B on edge hardware. Tool-based interfaces achieve 100% task completion, while automatic ROS service discovery shows 43.3% failure rates, highlighting the need for structured robot APIs for reliable LLM-driven control. This open-source platform enables systematic investigation of human-robot collaboration, agentic robot programming, and increasingly autonomous disassembly workflows, providing a practical foundation for research toward scalable robotic battery recycling.

Problem

• electric vehicle disassembly
• battery recycling
• design variability
• automation
• robotic manipulation

Innovation

• agentic AI
• open-vocabulary object detection
• vision-guided robotic disassembly
• LLM-driven robot control
• battery recycling automation