An improved PBIL algorithm for path planning problem of mobile robots

Abstract

The path planning problem of mobile robots is a NP-Hard problem often solved by evolutionary approaches such as Genetic Algorithm (GA) and Ant Colony Optimization (ACO). However, the algorithm's performance is often influenced heavily by the determination of the operators and the choice of related parameters. In this paper, a permutation code PBIL is proposed to solve the path planning problem. First, a free space model of the mobile robot is constructed by the MAKLINK graph; second, a sub-optimal path is generated by the Dijkstra algorithm; then global optimal path is constructed by the permutation code PBIL based on the sub-optimal path. Simulation results show that the PBIL can get satisfied solutions more simply and efficiently with fewer operators and parameters. © 2013 Springer-Verlag.