Monte Carlo-based improved ant colony optimization for path planning of welding robot

For spot-welding process, rational path planning of weld points can improve productivity for welding robot. In basic ant colony optimization, at the beginning of the iteration, the pheromone concentration is only related to the path length, but at this time, the path often contains a lot of redundant parts. Therefore, the pheromone in the initial iteration is almost difficult to reflect the advantages and disadvantages of the node, and it requires multiple iterations to slowly show the advantages and disadvantages of the path. For this shortcoming, a Monte Carlo-based improve ant colony optimization (MC-IACO) is proposed to solve the path planning of welding robots. The MC-IACO algorithm samples the path results generated in each generation and converts them into Monte Carlo indexes of the nodes. By substituting the Monte Carlo index of the feasible node into the sigmoid function, the probability that the corresponding node may constitute the optimal path, namely the Monte Carlo factor, is calculated to help ants make decisions. The new node transfer formula constructed with this factor can effectively improve the path finding efficiency of ants and the efficiency of the algorithm to complete the path planning task. The feedback adjustment factor is introduced to make the judgment of the advantages and disadvantages of feasible nodes also refer to the advantages and disadvantages of subsequent nodes. The simulation results show that compared with the basic ant colony optimization and other improved ant colony optimization, the MC-IACO algorithm in this paper improves significantly in solving the path planning problem of the welding robot, and its overall performance is better.