Laser Cutting Layout Based on Particle Swarm Optimization and Graphic Interference Detecting

Laser cutting of thin metal parts involves packing 2-D arbitrary objects into rectangle work space. The problem is shown to be NP-complete. This paper presents a novel approach to solve the problem, by dividing the problem into 2 parts. The first part finishes packing minimum envelop rectangles of the objects by applying Particle Swarm Optimization (PSO) with the lowest line packing as its fitness function. The second part moves the objects to a most suitable place by detecting interference with neighbor objects. A complete interference detection method between two polygons, either convex or concave has been suggested. In addition, two approaches, both the scanning approach and the circumcirlcle approach, have been developed to find the neighbor objects of the current moving object. And the order to move the objects has also been found in a scanning manner. Computational experiments have been carried out both for the first and the second part and the results are inspiring.