The Research Of Path Planning Based On BP Neural Network For AUV

Research of path planning is one of the key technologies for robot to accomplish a task safely in uncertain environment. How to find a non-obstacle path from start to target is the focus of this technology.In this paper, based on making a relatively deep study on path planning, the algorithm of BP neural network is put forward. The main work includes the following aspects:Firstly, the algorithm of BP neural network is studied. And a novel method of initial weights optimization is presented.Secondly, the paper presents the configuration of AUV, including sonar, pressure sensor and so on.At last, the method using BP neural network is used to fulfill the global path-planning and local path-planning of robot. The simulation experiment proves the method available.