Artificial intelligence in the automotive industry
New software from Fraunhofer IGD accelerates and enhances sorting processes
The solution is right there in front of your eyes! ARRANGE software in combination with augmented reality (AR) glasses provides digital support for sorting processes. Employees on the production line see components that belong together overlaid in color directly in their field of vision. The innovative solution from Fraunhofer IGD is helping users in the automotive sector to reduce costs by accelerating processes and minimizing error rates.
Darmstadt: Red for order A, yellow for order B, green for order C… ARRANGE uses color coding to assist in the sorting of components. A prime example is sheet metal parts in the automotive industry, where solutions of this kind are particularly welcome. Using AR glasses, employees on the production line can see which customer the punched-out items are intended for. The potential of the software package from the Fraunhofer Institute for Computer Graphics Research IGD is especially apparent when applied to sheet metal consisting of up to a hundred elements.
ARRANGE works with purely synthetic data, i.e. it recognizes the real objects solely on the basis of 3D models of the parts. This solution shortens the learning process of the artificial intelligence enormously; up to now, real photographs were necessary.
Software performs target-actual comparison
ARRANGE offers users multiple benefits. The AI-based software increases the speed of the sorting process because the workers have the color-coded parts directly in their field of vision and no longer have to compare the sheet metal with images on an external display. “There is no need for abstraction, as the elements are directly overlaid in color,” explains Fabian Rücker, a research assistant and doctoral student at Fraunhofer IGD. This reduces the susceptibility to errors in the sorting process. If an action is performed incorrectly, the software provides feedback.
Equally relevant is the performance of a concomitant quality assurance task, namely target-actual comparison. If parts do not correspond to the 3D models visualized, for example because they are uneven or have a faulty shape, this also generates an error message. Taken together, the acceleration of the work steps and the minimization of errors help the industrial user to reduce costs.
Basic technology for full automation
Rücker developed ARRANGE with his team as part of a Software Campus project which received funding of 100,000 euros from the German Federal Ministry of Education and Research. He was given the opportunity to design his own research project in a leading capacity together with a well-known tool manufacturer and automotive supplier as industrial partner.
Further solutions have already emerged from the ARRANGE project, such as the MARQUIS software package from Fraunhofer IGD which additionally recognizes complex three-dimensional objects such as brake calipers as well as flexible parts such as hoses. “This technology is a step toward automation,” explains Rücker. “It can be combined with robotic applications in the future.”