Path Planning Based on Improved Ant Colony Optimization in Congestion

Abstract

Automated guided vehicle (AGV) is widely used in intelligent warehouse systems. The ant colony optimization is a classical swarm intelligence optimization algorithm. It can achieve good path planning results for AGV. But the use of uniformly distributed initialization pheromone concentration often leads the algorithm to fall into a local optimum solution. And under congestion, the common strategy of the ant colony optimization is to wait or re-plan the path, resulting in reduced efficiency of the AGV. This paper proposes a static path planning method considering the congestion factor. The congestion factor is introduced into the ant colony algorithm. Firstly, the congestion degree is analyzed according to the traffic flow. And the congestion factor is added to the transfer rule so that the AGV can choose autonomously whether to avoid congested road sections when planning the path. Then, nonuniform distribution of initialization pheromone concentration is used instead of uniform distribution of initialization pheromone concentration, avoiding blind search in the early stage of the algorithm. Finally, simulation experiments are conducted on the AGV in a 20×20 map. And the results prove that the method can effectively avoid congested road sections, increase the flexibility of the algorithm and improve the convergence speed of the algorithm.