Improved Ant Colony Optimization for Ground Robot 3D Path Planning

Abstract

Path planning is an important part in the navigation control of mobile robot in a 3D environment. We proposed an improved ant colony algorithm to address the problems of falling into local optimum easily and long search time in 3D path planning. We redesigned pheromone update and heuristic function. New search mode is designed to solve the problem of searching time. We used a number of 3D terrains to carry out experiments, and set different starting and end points in each terrain. By comparing the results of improved ant colony algorithm and traditional ant colony algorithm, the improved one can reduce the shortest path length by an average of 8.164%.