Sorting out recycled batteries is now possible by means of Artificial Intelligence (AI), thanks to research conducted at the University of Gothenburg, Sweden and Chalmers University of Technology, Sweden.
Claes Strannegård, whose brilliant idea led to this striking innovation, is an AI researcher at the University of Gothenburg and Chalmers University of Technology. Collaboration with a publicly owned recycling company Renova in Gothenburg, Strannegård was able to construct a machine that uses computerized optical recognition to sort as many as 10 batteries per second.
The interesting component of the machine is its neural network, also known as its artificial nervous system, which should be trained how to act. In this particular instance, the machine takes pictures of more than 2,000 types of batteries from all angles, which helps it recognize them.
Once fed into the machine, a visual inspection takes place. This process involves comparison of the current battery with any one of the previously photographed ones, regardless of whether the batteries are dirty or damaged. When the battery is identified, it gets sent to a container through the process of separation by compressed air. Each container collects batteries based on their chemical content.
Hans-Eric Melin, CEO of the Gothenburg-based company Optisort, which has developed the machine, explains that the information about each battery, including its brand and model, is stored in the system, which is very commercially beneficial to recyclers besides being environmentally friendly.
This new machine sets the grounds for a new market for battery waste. Two machines have already been sold- one in Sweden and one in the UK.
Strannegård is proud that his invention has been accepted so well and hopes that this will encourage new research towards using artificial intelligence for sorting out all types of waste.