Multi-Robot Path Planning Based on Multi-Objective Particle Swarm Optimization

摘要

In this paper, a new method is proposed for the path planning of multi-robots in unknown environments. The method is inspired by multi-objective particle swarm optimization (MOPSO) and is named multi-robot MOPSO. It considers shortness, safety, and smoothness. Due to the obscurity of the environment, the robots should decide the moving direction based on the information gathered by sensors only such that the optimal path between the start and goal positions can be found at the end of the algorithm. Sharing knowledge among the navigating robots is necessary to achieve this aim. So, a new concept, named the probabilistic window, is introduced in this paper. It combines the current information obtained through the robot sensors and experiences of the previous robots to select the paths that seem more likely to achieve higher fitness in the mentioned objectives. The proposed method has an outstanding performance on different complex benchmarks, and the results have shown that it is more effective and efficient compared with the classic and the state-of-the-art methods.