Flying Welding Path Planning Based on Improved Two-chromosome Genetic Algorithm

Aiming to address the issues of moving speed and overall machining efficiency of a robot arm in path planning for laser flying welding, a method is proposed. This method involves using the non-central point of the trajectory to be welded as the welding position point and is further extended to the generalized traveling salesman problem. Through the analysis of the optimization model, the two-chromosome genetic algorithm is utilized to solve the path planning. Based on this, path intersection elimination and neighbor node replacement are carried out for every iteration in order to enhance the quality of the best solution. Finally, the path planning based on the general genetic algorithm for central welding position points, and the path planning based on the two-chromosome genetic algorithm and the improved two-chromosome genetic algorithm for non-central welding position points are simulated. The results show that the improved two-chromosome genetic algorithm can calculate the optimal solution of higher quality and a more reasonable machining path for non-central point welding positions.