An Ant Colony Optimization Parameter Tuning Method Based on Uniform Design for Path Planning of Mobile Robots

Abstract

Ant Colony optimization (ACO) has been widely used in mobile robot path planning in recent years. The parameters of ACO have great influence on the global search ability and convergence speed of the algorithm. The existing parameter tuning methods of the ACO mostly depends on experience or cannot evaluate the interaction between parameters, which make it very time-consuming and difficult to obtain excellent parameter combination. An experimental design method Uniform Design (UD) is applied for the ACO offline tuning of mobile robot path planning algorithm in this paper. The results show that the UD method can rapidly obtain excellent parameter combination from few of experiment simulation. The simulation results and experimental results show that the optimal parameter combination of ACO obtained by the UD method can rank in the top 10% of the valid results of the enumeration method sorted by the path length. In addition, the algorithm running time of the optimal parameter combination is shorter than that of most parameter combinations obtained by enumeration method.