Scholarlink

In the interests of efficient remanufacturing, this paper provides an overview of the most effective approach for disassembling end-of-life products through a comparison of available methods, involving the use of computer vision and robots. It also outlines a very flexible approach to locating hardware components in electronic waste (e-waste). This case study focuses on the process of extracting components with a view to recycling or reusing them. Using computer hardware as an illustration, the aim is to find the right disassembly method and to allocate tasks appropriately. Prioritizing the HRC method, is particularly effective for disassembling the computer, as both the human and the robot can work simultaneously on it. In addition, the object detection system connected to the robot was trained using a dataset of computer hardware. The YOLO (You Only Look Once) has been chosen as the most appropriate algorithm. This enables our robot to detect the components it needs at the extraction station.

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

Enhancing Human-Robot Collaboration in Disassembly Processes Through Vision-Based Hardware Detection

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Ameur Soufiane,
  • Tabaa Mohammed,
  • Hamlich Mohamed,
  • Hidila Zineb,
  • Bearee Richard

摘要

In the interests of efficient remanufacturing, this paper provides an overview of the most effective approach for disassembling end-of-life products through a comparison of available methods, involving the use of computer vision and robots. It also outlines a very flexible approach to locating hardware components in electronic waste (e-waste). This case study focuses on the process of extracting components with a view to recycling or reusing them. Using computer hardware as an illustration, the aim is to find the right disassembly method and to allocate tasks appropriately. Prioritizing the HRC method, is particularly effective for disassembling the computer, as both the human and the robot can work simultaneously on it. In addition, the object detection system connected to the robot was trained using a dataset of computer hardware. The YOLO (You Only Look Once) has been chosen as the most appropriate algorithm. This enables our robot to detect the components it needs at the extraction station.