Improved ant colony optimization algorithm for UAV path planning

Traditional Unmanned aerial vehicles (UAV) path planning methods have poor practical properties as they rarely take mission constraints like terminal angle constraint into consideration. A bidirectional searching ant colony optimization algorithm was proposed to solve above problem without losing path searching efficiency. The workspace of UAV was modeled by applying grid method and each grid was labeled. Then ant colonies start searching from two positions near the starting point and destination point simultaneously following the predetermined directions. A novel path selecting method was used to combine the paths and choose the optimal ones as the final path when the two paths from different points. Pheromone updating rules and successive points selecting method were also improved to increase algorithm convergence speed and avoid local optima. Simulations were made in two grid maps and the results showed that the modified path planning algorithm could find the qualified paths if the one exists with higher efficiency.