Research on path planning for UAV based on improved ant colony algorithm
Wenbo Wei, Maoyong Cao, Fengying Ma, Peng Ji, X. G. Wang · 2023
Abstract: The field of unmanned aerial vehicle (UAV) research places significant emphasis on path planning, especially intelligent optimization method. In order to address issues with classical ant colony optimization (ACO) algorithm, such as its tendency to become stuck in local optima and limited search capabilities, this paper proposes a novel algorithm, IACO, which improves upon the typical ACO algorithm. This algorithm is designed for static hilly environments and incorporates a comprehensive heuristic function along with self-updating rules. Referred to A* algorithm, IACO utilizes a heuristic function that considers the position of initial and target nodes respectively, using Manhattan distance to simplify calculation compared with Euclidean distance. On the process of global updating of pheromone, IACO improves the accumulation by considering the average length of path and the number of iterations, while also taking into account the worst path during pheromone volatilization. Additionally, the decay parameter is replaced with the normal distribution that is related to the number of iterations. The elitist strategy is also used to prevent local optima and accelerate iteration rate. Simulation experiments show that the entire path is above the hill, enabling the UAV to fly at a safe distance. IACO significantly reduces path length by 24.69% and iteration time by 20.32%, which has been demonstrated its superiority over ACO in terms of optimization effectiveness and search efficiency.