The Application Of Information Fusion Based On Fuzzy Neural Network In Mobile Robot

In recent years, the application of mobile robot is becoming wider and wider. One important object of mobile robot is that it can self-navigate and avoid all kinds of obstacle. In this paper, the application of information fusion based on fuzzy neural network in mobile robot is given. By this way, mobile robot can move without collision in dynamical surroundings with static and moving obstacles. In order to realize self-navigation, global position system (GPS) and photocode plate position system are applied to orient in this task, and many kinds of outer sensors are used to sense surroundings' information in order to avoid obstacles.In the back of Tianjin's Natural Science Foundation "the Study of Mobile Robot System Technology Faced to Complicated Task" (task No.003601211) and Hebei Province's Tackling Key Problem Plan " the Study of AGV Accurate Navigation Used in Material Automation Transportation" (task No.00547001D-18), this dissertation developed application study of information fusion based on fuzzy neural network in mobile robot's self-navigation. In this dissertation, the main work includes:1. Vehicle structure and driving system of HEBUT- I mobile robot that is designed by Intelligent Researching Room was introduced, and analyzed kinematics character.2. The principle of fuzzy neural network is analyzed and introduced, and a fuzzy neural network suitable of mobile robot's information fusion is proposed. This fuzzy neural network is based on back propagation neural network and takes fuzzy variable as its input and output. After training, the fuzzy neural network can memorize the fuzzy control rules and can be used in intelligent navigation of mobile robot.3. Global position system's composing and principle were introduced. Absolute position and relative position were given. And discussing position method of mobile robot in local coordinate system. Photocode plate is used in local position.4. Do experiment according as theory above. Information fusion based on fuzzy neural network was applied in mobile robot. According to result of experiment, this network is valid. HEBUT- I mobile robot can move well in avoiding obstacles.