Automated complex path extraction and optimization based on scalable vector graphics and genetic algorithm for industrial robot applications

Advances in industrial automation and image processing technology have enabled the utilization of industrial robots more efficiently in intelligent manufacturing and modern automation systems. In robotics applications, the simple path of the tool attached to the end effector is easily online programmed by the teaching and playback method using a teach pendant. However, programming complex tool paths, such as those found in cracks for tasks like welding, deburring, and adhesive application remains a significant challenge. This study proposes an automated offline programming (AOLP) method for industrial robots, utilizing Scalable Vector Graphics (SVG) and Genetic Algorithms (GA) to extract and optimize complex paths from raster images. The method involves capturing a raster image of the workpiece, converting it to SVG format, splitting branched paths, and optimizing the sequence using GA. The optimized path is simulated in RoboDK and executed on a KUKA KR6 R700-2 robot. The simulation and experimental results show that the presented method can be utilized successfully and accurately to automatically extract complex paths in two-dimensional robotics applications.