Robot path planning based on PSO and D∗ algorithmsin dynamic environment

Abstract

This paper proposes two new approaches of robot path planning in dynamic environments based on D∗ algorithm and particle swarm optimization. Generally speaking, the grid method is used to decompose two-dimensional space to build class node which contains the information of the space environment. D∗ algorithm analyze the environment from the goal node and computes the cost for each node to the start node. In the first approach, Lbest PSO algorithm is used to move the robot from the start node through dynamic environment which contains dynamic obstacle moving in free space by finding and displaying the optimal path. In the second approach a method is developed to manipulate the gate raise state where the robot cannot pass this node unlike D∗ algorithm. Some experimental results are simulation in different dynamic environments, show that in second approach the robot reaches its target without colliding obstacles and finds the optimal path with minimum iterations, minimum total arc cost and minimum time occupy.