The contactless digital AI coworker takes digital and robotic collaboration to a new level.

Kimoknow, a spinoff from the Karlsruher Institut of Technology (KIT) in Germany takes this intelligent technology even a step further. It has successfully developed a technology that automates the training, not of humans but of KI-algorithms themselves! The automated object recognition system can be placed into diverse applications, inter alia, AR glasses. AI enables machines to detect objects for which a large imagery is necessary and for which algorithms are usually trained manually. The new technology automatizes this painstaking, time consuming, extensive and expensive training process of the AI algorithms.

A digital mounting assistant serves as company Kimoknow’s first use case and is already engaged in contactless coworking scenarios next to humans and machines. “We use CAD and PDM data as basis for our use case since they are generated anyways while creating objects”, emphasizes Kimoknow founder Lukas Kriete. These data inform about material, geometry and position of each object. The subsequent process is the same as the one already in use at SaarGummi. When combined with already existing systems it might bolster further cost saving solutions through automated algorithm trainings and therefor be shine in the SaarGummi application spotlight again. AR coupled with AI offers the perfect solution for remote application, saves travelling costs and supports production in the time of pandemics where human distancing is a necessity.