1st ICAI 2020

International Conference on Automotive Industry 2020

Mladá Boleslav, Czech Republic

satisfactory results, especially not when multiple features of a specific product need to be considered. Due to the demographic trends that affect the numbers of skilled workers in the labour market and the pressure to reduce car prices, car manufacturers are increasingly focusing on automation in their production processes, which includes machine vision. 2. Problem Formulation and Methodology In the time of the Fourth Industrial Revolution, vision systems in conjunction with industrial robots became important players in industrial automation. You will not find industries that are unaffected by one of the machine vision techniques. Machine vision is increasingly being applied to the processes that people have performed so far. Examples include quality control, surface defects detection (David, 2015), safety, robot guidance systems and many more. Many applications with different final solutions can be found in the literature. The suitability of the machine vision depends on its end use. This paper is dedicated exclusively to the commercial use of vision systems for robotic guidance. The aim is not to describe the various types of machine vision in detail, and does not describe software algorithms for image processing. Some recent practical cases of implementation of visual systems of car manufactures in practice rather than academic research are describe below. This paper only pointed out a few applications that are actually applied in the automotive industry, but each of the above examples can be successfully applied to different processes. 3. Problem Solution Non-contact measurements play a significant role in automation. Wide portfolio of faster and more precise cameras is available on the market. None of the solutions is the best choice, it always depends on the specific type of application. It is necessary to consider the advantages and disadvantages of the chosen solution. Furthermore, the paper briefly describes several optical techniques used for robot guidance such as structured light, time of flight and laser triangulation. The combination of a measurement system and robot is an optimal solution as it makes use of the robot ability for precise movement and overcomes the accuracy deficiencies. 3.1 Time of flight (ToF) An active vision technique acquires 3D information by projecting a visible or infrared pattern onto the object. Time of flight camera uses light pulses which are reflected from the objects back into the optic. Plane optic sensor captures incoming light depending on delay caused by different flight distance and calculate according to the Equation (1) where t D is delay, D is distance and c is the speed of light.

240