A Cellular Ant Colony Algorithm for Path Planning Using Bayesian Posterior Probability

Abstract

In order to solve the problem of slow convergence rate in traditional ant colony algorithm for UAV path planning, a new cellular ant colony algorithm is proposed. First, we construct a sector prediction area in grid environment map. Then we build heuristic and obstacle repulsion functions of target nodes in the prediction area. Using these functions, we can get the Bayesian conditional probability and posterior estimation of target nodes. In the end, we select the node with the largest posterior estimation as the next path node. The simulation results show that the new algorithm has better global search ability. Furthermore, using the sector prediction area makes the planned path more consistent with the UAV flight characteristics. And the designed functions in the sector prediction area speeds up the path search process.