Akademik Veri Yönetim Sistemi
4TH INTERNATIONAL THALES CONGRESS ON LIFE, ENGINEERING, ARCHITECTURE AND MATHEMATICS, Cairo, Mısır, 20 - 22 Temmuz 2025, ss.104-111, (Tam Metin Bildiri)
The world’s electrification powerhouse Electric vehicle (EV) penetration in the world has been surging and the growth has accelerated the call for sustainable solutions to recycle Li-ion batteries. Millions of EVs will soon begin retiring, and traditional manual disassembly is incredibly difficult due to safety concerns, labor intensity, and the greatly varying nature of battery designs. In this paper, a semi-automated HRC framework for the safety, scalability, and adaptability of the EV battery disassembly process is proposed. The proposed architecture combines collaborative robots (cobots), teleoperation interfaces supplemented with haptic feedback, and AI-driven perception systems, in order to optimize the task sharing between humans and robots. Monotonous and dangerous handling – for instance, unscrewing or removal of battery covers – are carried out by robots, human beings are used for decision-critical activities, like cutting of cables and inspection and working with design variability. This feature minimises human contact with toxic and high voltage devices The haptic feedback technology enables fine manipulation of objects in the ultra-deepWater environment even at a distance. To accommodate the heterogeneous battery structures, deep learning-based object recognition (e.g., YOLO, ResNet) and 3D sensing (LiDAR, thermal imaging) are used in this framework. These are enabling technologies for real-time task reconfiguration and enhanced system flexibility without the necessity of rigid fixtures. While the system is far from being completely autonomous, limited robotic dexterity and multi-agent coordination suggest a pragmatic path toward intelligent system-level disassembly at industrial scales. This work provides not just a viable and scalable solution to EV battery recycling, but also paves the way for the use of robots in other hazardous sectors such as medical instrument decommissioning or nuclear appliance decommissioning. We will explore robotic end-effectors improvement and AI for task planning prediction in the future as well as pilot study in real world environment.