Mobile Robot Path Planning Based On Fuzzy Neural Network Technology Research

Path Planning for Mobile Robot is one of the most critical technologies among all of the Robotics researches, whose chief task is to seek one optimal path under the robot's running space, along which Mobile Robot can get to the goal point from start point without collision and the cost assigned caused by robot moving to the end point along this way is smallest. Based on mobile robot AS-R, the paper researches Path Planning for Mobile Robot.First, the paper expatiates the theories and algorithms of Global Path Planning and Local Path Planning. On the base of AS-R, the paper introduces the kinematic model of mobile robot driving by the difference of the two wheels. Then on the basis of kinematic model, the paper calculates the position and angle of the robot, and carries on the experimentation, the experiment confirms the method is effective.To solve the Path Planning when the working environment is known, the paper advances one modified artificial potential field approach, using which robot can get away from the partial minimum point, and the result of simulate confirms the modified approach is feasible.At last, to solve the problem of path planning under fully unknown obstacles space, the paper researches the method based on fuzzy neural network. With the help of fuzzy logic, we transform the exact value of environment information into fuzzy value ,then builds the fuzzy neural network. The result of simulate experiment confirms the algorithm can adapt itself to many kinds of obstacles space successfully.