Recycling-oriented collaborative waste sorting by continual learning

The goal of recAIcle is to support sorting workers in the manual sorting of non-hazardous waste (focus: plastics and batteries) through artificial intelligence (AI) and machine learning (ML). The AI continuously learns passively by observing the experienced sorting workers and supports them in the sorting decision by an augmentation system (light pulses).

The recAIcle system should increase the output of manual sorting with the same number of sorters. Monotony should also be reduced, and the training phase should be made easier for new employees. The machine learning model will continuously learn to adapt to the fluctuations of the input material and to new waste streams.

You can find more information about the project here.