Motion control for laser machining via reinforcement learning

• DOI: 10.1364/oe.454793
• Publication Date: 2022-04-20T15:00:08Z
• Authors: Yunhui Xie; Matthew Praeger; James A. Grant-Jacob; Robert W. Eason; Ben Mills
• Keywords: Machine tool; Laser cutting; Laser beam machining

Laser processing techniques such as laser machining, marking, cutting, welding, polishing and sintering have become important tools in modern manufacturing. A key step these processes is to take the intended design convert it into coordinates or toolpaths that are useable by motion control hardware result efficient with a sufficiently high quality of finish. Toolpath can require considerable amounts skilled manual labor even when assisted proprietary software. In addition, blind execution predetermined unforgiving, sense there no compensation for machining errors may compromise final product. this work, novel approach demonstrated, utilizing reinforcement learning (RL) supervise process. This autonomous RL-controlled system machine arbitrary pre-defined patterns whilst simultaneously detecting compensating incorrectly executed actions, real time.