Nilar AB is a Swedish SME that designs, develops and manufactures batteries for stationary energy storage. To ensure excellent performance, lifespan and safety of the batteries, the manufacturing process involves both optical quality inspection as well as a set of tests that the battery modules must pass through before they can be sent to the customers.  

Together with researchers at RISE, Nilar aims to improve optical inspection by using AI-based machine vision. This work is part of an ongoing Vinnova SiP PiiA funded project, “AI for increased process efficiency”, and with the acquired development access to VEGA in Slovenia with the help of ENCCS, it has been possible to speed up the training time of the AI models significantly, and thus greatly increase the number of experiments that can be done on the image datasets.