A team of European researchers has developed adaptable, AI-driven robots that could transform the way electronic waste is recycled — offering major benefits for both the environment and the economy.
At Electrocycling GmbH, one of Europe’s largest e-waste recycling facilities located near Goslar, Germany, up to 80,000 metric tons of electronic waste are processed annually. Despite modern equipment, over half of the staff still manually dismantle discarded electronics. This often involves removing dangerous lithium batteries from increasingly compact and complex devices — a process that is repetitive, labor-intensive, and potentially hazardous.
Many of these tasks could be performed by robots, but traditional industrial robots struggle to adapt. Each time a new product or process is introduced, both the hardware and software need to be re-engineered — a costly and time-consuming undertaking.
To solve this challenge, the ReconCycle project was launched, bringing together researchers from Slovenia, Germany, and Italy between 2020 and 2024. Coordinated by the Jožef Stefan Institute in Slovenia, the team developed self-reconfiguring robotic systems designed specifically for e-waste environments. These AI-supported robots are capable of adapting to different types of electronic devices, such as smoke detectors and radiator heat meters — common household items that contribute significantly to electronic waste.
Unlike typical industrial robots that repeat a single, pre-programmed task in controlled environments, these new robots are built for flexibility. They are embedded within robotic work cells — modular systems consisting of robotic arms, tools, controllers, and sensors — that can autonomously adapt to a variety of dismantling tasks using AI algorithms and modular hardware. The inclusion of soft robotic components, such as the SoftHand (a dexterous, human-like robotic hand), allows these machines to handle delicate and irregularly shaped parts with care and precision.
Working in close collaboration with Electrocycling, the researchers successfully integrated these systems into real-world recycling operations. The robots also include built-in safety mechanisms such as collaborative functions and emergency stop features, making them safe for human-robot interaction.
The urgency of improving e-waste management is growing. The EU generates nearly 5 million metric tons of electronic waste annually — roughly 11 kilograms per person — yet less than 40% of it is properly recycled. Globally, the situation is even more alarming. In 2022, an estimated 62 million metric tons of e-waste were generated, and the amount is increasing five times faster than the rate of recycling. The European Parliament has responded with policies such as the Waste from Electrical and Electronic Equipment (WEEE) Directive, and the new robotics development aligns with the EU’s broader digital strategy, which promotes AI in manufacturing to boost efficiency and support climate goals.
Beyond the environmental impact, there are significant economic losses tied to e-waste. The UN estimates that €84 billion worth of valuable materials — including copper, iron, gold, silver, and palladium — is discarded each year instead of being recovered. Electrocycling already recovers around 80% of the materials from its incoming waste stream, and the addition of advanced robotics could improve that yield even further.
The potential applications for this technology go far beyond recycling. The adaptable design and intelligent systems could be repurposed for tasks in housekeeping, healthcare support, or any environment where conditions frequently change. As AI and robotics continue to evolve, the vision is to bring automation into domains where robots have yet to make a practical impact.
With continued development and collaboration between research institutions and industry leaders, these adaptable robots could be the key to solving one of the fastest-growing waste problems — and in doing so, help pave the way toward a more circular economy.
By Impact Lab
