Research On Mobile Robot Local Path Planning

Path-planning is a very important task in robot control system, and local path-planning problem is a basic part on research of mobile robot. How to find a non-obstacle path from start to target is the focus of this paper. When the precision problem in robot hardware system can't be solved in short time, the algorithm research appears particularly significant.Firstly, considering mobile robot in uncertain environment, this paper designs a double fuzzy controller with coordinating control behaviors for moving towards a target and obstacles avoidance behaviors to control next movement of mobile robot. The simulation tests show the validity and feasibility of the proposed method.Secondly, this paper adopts improving BP neural network algorithm, so it can overcome the shortage in path-planning with conventional BP neural network algorithm. The simulation results indicate that the BP neural network with accessional momentum term can efficiently improve the convergent speed of algorithm. At last, mobile robot dynamic path planning simulation is carried out in the environment with immobile and mobile obstacles, the simulation results indicate that the method is feasible.