Global and local path planning of robots combining ACO and dynamic window algorithm.

As a key technology, robot path planning is of great significance in safely and efficiently completing tasks. However, faced with the challenges of static and dynamic obstacles in complex environments, traditional path planning methods have limitations in terms of efficiency and adaptability. Therefore, a global and local path planning method combining improved ant colony algorithm and improved dynamic window algorithm is proposed. The optimization ability, convergence efficiency, and obstacle avoidance performance are optimized in complex environments. The innovation is reflected in the proposed cone pheromone initialization, adaptive heuristic factor regulation, and ant colony division of labor strategies, which improve the global search ability and convergence speed of ant colony algorithm. In addition, the path direction angle evaluation function and dynamic velocity sampling optimization are introduced to enhance the obstacle avoidance stability of dynamic window algorithm. The research results showed that the optimization method had the greatest improvement on the basic ant colony algorithm, with an average path reduction of 30.18 and an accuracy increase of 98.46%. In a 2020 grid map, the improved strategy achieved convergence in the 23rd iteration, with an average path length of only 25.87 during convergence. In a 3030 grid map, the improved method converged in the 81st iteration, and the path length at convergence was 41.03. In the designed four environments, the smoothness was 0.94, 0.91, 0.79, and 0.65, all of which were better than comparison algorithms. The designed method based on improved dynamic window algorithm can effectively avoid dynamic obstacles. The research not only improves the efficiency and robustness of the path planning algorithm, but also improves the autonomous navigation ability of robots in complex environments, providing more adaptive path planning schemes for industrial automation, service robots, exploration robots and other fields.