This site uses cookies and related technologies for site operation, analytics and third party advertising purposes as described in our Privacy and Data Processing Policy. You may choose to consent to our use of these technologies, or further manage your preferences. To opt-out of sharing with third parties information related to these technologies, select "Manage Settings" or submit a Do Not Sell My Personal Information request.
AI is currently a trending topic across a variety of industries and the automotive sector is no different. Škoda Auto is now using an AI-based image recognition system identify maintenance issues on the assembly line as quickly as possible at its main plant in Mladá Boleslav.
Cameras are placed on the assembly line’s overhead conveyor and it images equipment and parts that are subject to wear to detect any irregularities. This could be girders, bolts, or cabling, and the AI system will flag a need for maintenance in real time, saving time and resources in the long run.
Magic Eye instantly compares the high-precision photos that it already has against stored images which then enables it to detect any departures from the “optimal baseline conditions” and identify the source of the error.
It uses blue light to make sure that the AI can tell the difference between scratch and cracks to make the correct diagnosis every time and it is always expanding its knowledge base as it learns from each part that it examines. Škoda uses the example that if it finds a worn bolt it will mark the spot as error free as soon as the part is replaced and rechecked and then will use this information to detect the same irregularity going forward.
Currently the technology is only being used on the assembly line for the Enyaq iV and the Octavia but to allow for further optimisation and integration of the system Škoda has simulated a section of the assembly line. This allows them to experiment away from the real time production line with imagery, camera settings, system parameters, and simulate damage to be identified.
