Scholarlink

The automation of disassembly processes for electric vehicle (EV) batteries is critical to achieving sustainable recycling, reuse and resource recovery. This paper presents a vision-based robotic system for the automated detection and removal of screws in EV battery packs. The system integrates a YOLOv8 deep learning model for screw detection with an Intel RealSense depth camera, mounted on a six-degree-of-freedom robotic arm, to perform precise and efficient screw removal. Experimental validation demonstrates the system’s high precision and recall in detecting screws, with an overall screw removal success rate of 89%. By automating the disassembly process, the system reduces hazardous manual labor, improves recycling throughput, and contributes to multiple Sustainable Development Goals (SDGs). The paper concludes with a discussion on the system’s potential for future improvements in robustness and scalability for industrial applications.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Automated Screw Detection and Removal for Sustainable EV Battery Recycling: A Vision-Based Approach Aligned with SDGs

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Markus Schmitz,
  • Robin Heitz,
  • Daniel Gossen,
  • Burkhard Corves

摘要

The automation of disassembly processes for electric vehicle (EV) batteries is critical to achieving sustainable recycling, reuse and resource recovery. This paper presents a vision-based robotic system for the automated detection and removal of screws in EV battery packs. The system integrates a YOLOv8 deep learning model for screw detection with an Intel RealSense depth camera, mounted on a six-degree-of-freedom robotic arm, to perform precise and efficient screw removal. Experimental validation demonstrates the system’s high precision and recall in detecting screws, with an overall screw removal success rate of 89%. By automating the disassembly process, the system reduces hazardous manual labor, improves recycling throughput, and contributes to multiple Sustainable Development Goals (SDGs). The paper concludes with a discussion on the system’s potential for future improvements in robustness and scalability for industrial applications.