Multi-objective obstacle avoidance path planning of inspection robot integrating ant colony optimization and dynamic window

Substation inspection robots face issues including slow path planning convergence and weak dynamic obstacle avoidance capabilities in complex environments. This study proposes a multi-objective path planning method that considers global optimization and local real-time obstacle avoidance to improve the safety and efficiency of substation inspection. This study uses raster modeling to represent the substation hybrid environment, and accurately describes obstacle information through expansion processing and dynamic update mechanisms. An enhanced ant colony algorithm integrating artificial potential fields is constructed to optimize the global path, and dynamic pheromone update and triangle pruning methods are combined to improve path quality. It integrates and improves the dynamic window approach to achieve local obstacle avoidance, and proposes a global path fit evaluation item to coordinate global and local decision-making. Experiments show that when this method is iterated 100 times, the path length converges to 45.8 m, the standard deviation of the smoothness angle is 11°, the dynamic Obstacle Avoidance Response Time (OART) is as low as 52 ms, and the success rate reaches 98%. The success rate of dynamic obstacle avoidance reaches 98% in normal scenarios, the success rate of obstacle avoidance in extreme scenarios still exceeds 88%, and the path re-planning is time-consuming and controllable. This method effectively addresses the issue of multi-objective inspection path planning in complex substation environments and achieves a balance between path optimality, real-time obstacle avoidance, and motion smoothness. This simulation-based study provides theoretical reference for the efficient, safe, and stable operation of inspection robots, and has important application value for further practical deployment.