A Multiobjective Hybrid Optimization Algorithm for Path Planning of Coal Mine Patrol Robot

In the complex underground environment, the paths planned for coal mine patrol robot are often too long and unsmooth under the influence of low visibility and poor road conditions. To solve the problems, this paper improves the hybrid algorithm between the improved artificial fish swarm algorithm (AFSA) and the dynamic window algorithm (DWA) for global path planning of coal mine patrol robot and introduces the improved genetic algorithm (GA) to enhance the path planning accuracy. Based on the global optimal path, the improved DWA was adopted to design a new adaptive trajectory evaluation function, which improves the ability of the patrol robot to avoid local obstacles. The proposed optimization algorithm was proved feasible through simulations. In addition, a simulation platform for the control of coal mine patrol robot was established, using the software development platform for coal mine patrol robot and robot operating system (ROS). The simulation results show that the improvement shortened the path length by 0.12 m, reduced the time by 3.14 s, and removed many turning points and redundant points. Therefore, the proposed improved hybrid path planning algorithm is effective and superior.