The Challenges of Using Robots to Automate the Recycling of Electronic Devices
Ale� Ude, Mihael Simonič, Boris Kuster, Matija Mavsar, Martin Bem, Sebastian Ruiz, Minija Tamosiunaite, Manuel Giuseppe Catalano, Vinicio Tincani, Antonio Bicchi, Kübra Karacan, Hamid Sadeghian, Sami Haddadin, Riccardo Persichini, Hannes Fröhlich, Florentin Wörgötter
AI summary
Problem
Automating battery removal from highly variable and damaged small electronic devices remains a critical bottleneck in e-waste recycling due to safety risks and the inflexibility of traditional robotic systems.
Approach
The authors integrate Vision-Language Models for real-time adaptive planning with tactile force control, soft robotic grippers, and modular hardware to dynamically handle diverse device designs and damage states.
Key results
- VLM-based adaptive disassembly planning with retrieval-augmented generation
- Unified force-impedance control for precise tactile manipulation
- Modular, reconfigurable robotic workcell enabling seamless device adaptation
- Laboratory validation demonstrating higher efficiency and reduced manual intervention
Why it matters
It provides a scalable, safe pathway for automating hazardous e-waste preprocessing, directly supporting circular economy goals and recycling facility safety.
Abstract
This paper tackles the challenges of automating battery removal from small electronic devices, such as heat cost allocators and smoke detectors. Safe and efficient removal is essential to mitigate fire hazards posed by lithium batteries in recycling facilities and to support a circular economy. We present advanced methodologies and robotic technologies to address hurdles arising from diverse device designs, complex battery compartments, and varying states of damage. Our approach integrates Vision-Language Models (VLMs) for real- time, adaptive disassembly planning, computer vision, tactile skills, soft robotics, and reconfigurable robotic workcells to enhance perception, dexterity, and adaptability. The resulting workcell with modular hardware and standardized interfaces enables seamless adaptation across device types. Laboratory tests demonstrate higher efficiency and reduced manual intervention, underscoring the potential of AI-driven, reconfigurable robotics for scalable and sustainable e-waste recycling.