The Eurecat technology centre has developed a new automated system for dismantling and sorting electric vehicle battery components using collaborative robotics and artificial intelligence systems to simplify recycling at the end of their useful life with a view to cutting processing costs and times, ensuring worker safety and achieving utmost efficiency and sustainability.
The main challenge currently facing lithium-ion battery recycling is the wide variability between different models which differ in shape and size. Eurecat has therefore put in place a raft of new processing technologies to achieve a scalable and modular solution.
It has targeted dismantling the battery and sorting components and materials for recycling or reuse.
In the case of battery dismantling, it has identified the tasks which pose the greatest risk to operators and then developed innovative solutions for the selected operations, namely removing screws and taking off the top cover.
The outcome is a collaborative robotic solution which can be applied regardless of the battery model and allows the machine and operator to share the workspace. This means “the worker can perform tasks with greater added value in the disassembly workspace since it does not pose a risk,” says Óscar Palacín, a researcher in Eurecat’s Robotics and Automation Unit.
The robot is fitted with an industrial screwdriver and multiple foam clamps and equipped with a system which identifies the parts of the battery and works with a machine learning algorithm, You Only Look Once (YOLO), to enable the machine to disassemble it.
The robot’s behaviour has been developed using BehaviorTree, a C++library designed by Eurecat which creates behaviour trees for planning robotic system tasks.
Eurecat has also implemented an artificial intelligence-based visual inspection system that allows lithium battery components to be sorted for subsequent reuse.
The system can analyse the battery in detail, identifying each component and spotting any visual imperfections such as cracks, bulges, burn marks or corrosion. “We have trained the algorithms with a set of images of battery modules to accurately sort the components into categories such as critical, non-critical or safe,” points out Néstor García, a researcher in Eurecat’s Robotics and Automation Unit.
Special emphasis has also been placed on defining safe temperature ranges for battery components, which vary depending on the model type. The algorithm is thus able to screen for thermal anomalies during the sorting process, ensuring accurate and safe inspection.
“This solution not only reduces the risk of human exposure to hazardous sources but also optimises the choice of modules suitable for reuse, thus contributing to a more efficient and safer recycling process,” comments Daniel Serrano, director of Eurecat’s Robotics and Automation Unit.
These innovative solutions have been delivered as part of the European BatteReverse project, funded by the European Union’s Horizon Europe programme, which seeks to develop technologies to build a reverse logistics value chain for lithium-ion batteries from electric vehicles. The idea is to enhance recycling by ensuring utmost safety, efficiency and sustainability while also cutting processing costs and time.
