A Comprehensive Optimization for Path Planning: Combining Improved ACO and Smoothing Techniques

The ant colony algorithm is an approach for path planning that is used in multiple industries. This paper proposes an improved robot path planning method, referred to as Improved-ACO. First, the heuristic information calculation is optimized to increase algorithm efficiency and shorten convergence time. Secondly, an enhanced Tanh function is included into the heuristic information, allowing dynamic modifications during the search period and preventing the algorithm's convergence to local optima. Then, a novel pheromone update strategy is employed to accelerate convergence. Next, a novel pheromone diffusion mechanism is proposed to strengthen the ants' search capability. Additionally, a collision avoidance system and improved B-spline curves are included for path smoothing, guaranteeing that the optimized pathways conform to the robot's kinematic limitations. Simulation results indicate that the improved ant colony algorithm decreases the average number of turns by 37.5% and accelerates convergence time by 39.45% relative to existing methods across diverse map dimensions. The experiments confirm that Improved-ACO achieves rapid convergence and constructs smooth curves that adhere to the robot's kinematic constraints. Consequently, Improved-ACO is confirmed as an efficient and adaptable route planning method for robotic navigation under complicated situations.