Integrated Improved Ant Colony and DWA Algorithm for Robot Dynamic Path Planning

Addressing the limitations of the ant colony algorithm in global path planning, such as its blind search tendencies, susceptibility to local optima, and slow convergence, as well as the Dynamic Window Approach (DWA) algorithm’s inadequacies in effectively avoiding dynamic obstacles in local path planning, this study proposes a hybrid strategy that integrates an improved ant colony algorithm with the DWA algorithm. This strategy enhances the heuristic function of the ant colony algorithm by combining the actual distance from the current node to adjacent nodes with the relative distance from adjacent nodes to the target point, making the heuristic information more aligned with the real-world search environment. The pheromone update strategy integrates local and global optimal path information with meticulous control over pheromone evaporation, balancing exploration and exploitation and preventing premature convergence of the algorithm. Furthermore, the DWA algorithm’s local path planning capabilities are strengthened by incorporating target-oriented evaluation functions, speed evaluation functions, and obstacle avoidance evaluation functions, enabling the robot to more flexibly respond to changes in its environment. This study implemented the algorithm optimization through the following steps: initializing algorithm parameters and the environment map; employing the improved Ant Colony Optimization (ACO) for global path searching; utilizing DWA for local path decisions and obstacle avoidance; and continuously optimizing global and local avoidance strategies through iterative updates and feedback mechanisms. A series of simulation experiments demonstrate that the proposed integrated strategy significantly enhances path planning efficiency, reduces path length, and improves obstacle avoidance capabilities, providing an effective solution for path planning issues in complex environments.