Design and Implementation of a Multifunctional Screw Disassembly Workstation

The rapid growth of the electric vehicle industry has created a significant demand for the recycling of end-of-life electric vehicle batteries (EOL-EVB). Manual disassembly methods suffer from low efficiency, highlighting the urgent need for intelligent disassembly solutions for electric vehicle batteries. A major challenge in intelligent disassembly is dealing with uncertainty, especially when it comes to the disassembly of screws, which vary in shape, size, and rust level. To address this challenge, we present a multifunctional screw disassembly workstation specifically designed for the disassembly of screws, which constitutes a substantial portion of the EOL-EVB disassembly process. The workstation incorporates an automated sleeve replacement device that can seamlessly replace and disassemble sleeves during disassembly. Additionally, we propose a screw-type recognition method based on attributes, enabling the identification of various screw attributes to determine appropriate disassembly methods. This method exhibits scalability and requires only a small amount of data. By expanding the capabilities of our previous Neurosymbolic TAMP (Task and Motion Planning) work, we can support multiple types of screw disassembly and integrate it into the overall process of EOL-EVB disassembly, significantly reducing repetitive tasks such as screw disassembly during the disassembly process. Experimental results demonstrate the effectiveness of the workstation in disassembling multiple types of screws within a realistic disassembly environment.