Audi is testing software using artificial intelligence to recognise and identify tiny cracks in sheet metal. The deep learning system, deployed in the press shop at Ingolstadt, is said to detect the finest cracks in door panels, hoods, fenders and other sheet metal components in an automated inspection process.
A database has been built up of several million test images from the seven presses at Ingolstadt and from some of the Volkswagen facilities, to ‘train’ the artificial neural network to spot the defects. The sample images – amounting to several terabytes of information – have been marked down to the pixel level to give the highest possible degree of accuracy, says Audi. The software is said to have been mainly developed in-house, in a collaboration between Audi IT’s innovation department and the product technology division of the equipment and metal forming technology competence center.
The machine learning system will replace the current optical crack detection process, which uses smart cameras with image-detection software in addition to visual inspections. This has to be reconfigured for every new component, and also results in regular ‘false detections’, since its simple algorithms are said to be highly dependent on ambient factors such as lighting and surface properties. The new, more advanced system could also be deployed in paint shops or on assembly lines, once a sufficiently large dataset of labelled images for these processes has been collected.
Jörg Spindler, head of Audi’s competence center for equipment and forming technology, said in a statement: “This method supports our employees and is another important step for Audi in the transformation of its production plants into modern smart factories.”
Frank Loydl, Audi CIO, added: “Artificial intelligence and machine learning are key technologies for the future at Audi. With their help, we will continue to sustainably drive the digital transformation of the company. In this cross-divisional project, we are jointly developing a production-ready solution that Audi will use exclusively in the company and which is unique in the market.”