1st ICAI 2020

International Conference on Automotive Industry 2020

Mladá Boleslav, Czech Republic

Practical Examples of the Use of Machine Vision in the Automotive Industry Jaroslav Sláčala VŠB-Technical University of Ostrava Faculty of Materials Science and Technology, Department of Automation and Computing in Industry 17. listopadu 15, Ostrava, 708 00 Czech Republic e-mail: j.slacala@post.cz

Abstract The use of machine vision in the industry is not a matter of the past few years. Development of modern technology has only accelerated the arrival of new processors, high computing power and pressure on lower prices. These technologies include various types of camera systems called machine vision, which automatically determine the next step of the production process based on the information obtained by image analysis. Manufacturers from various industries have found machine vision as a tool to cut costs, increase quality, efficiency and thus increase their own competitiveness. One of the sectors where machine vision is successfully applied is the automotive industry. Car manufacturers integrate optical systems into their production lines to do a variety of tasks. For example, determination of vehicle types, automatic wheel mounting system, optical inspection of weld points and weld seams or robots automatically pick and place parts – bin picking. Keywords: automotive industry, machine vision, modern technology, optical systems JEL Classification: C88, L62, O32 1. Introduction The automotive industry provides many opportunities for vision systems. Demand for machinevisionsystems is increasingas the automotive sector requires improvingaccuracy in previously strict manual operations. For example, bin picking, assembly parts such as roofs, doors, car wheels or loading of stamped sheet metal parts on welding lines require enormous amounts of manpower due to the different car types and high requirements in assembling accuracy. In high quality vehicle manufacturing the best possible quality must be guaranteed while ensuring the highest rate of flexibility and the lowest possible costs. Customers demand a high level of production output and assembling accuracy. This can only be achieved using image processing systems that are matched exactly to the needs of the production process. The functions to be performed in manufacturing are becoming more complicated by the day and require a high degree of production output and assembling accuracy as well as flexibility (Isra Vision, 2019). Traditional sensors are often unable to solve the required detection and inspection tasks in a way that achieves

239