The Research Of Path Planning Based On FNN For AUV

Autonomous Underwater Vehicles is a kind of economical vehicle for detecting underwater and exploring ocean resource. Research on path planning is one of the key technologies for AUV to accomplish a task safely in uncertain environment. The paper aims at the application of the fuzzy theory,artificial neural network and fuzzy neural network technology in the movement control; And it proposes separately path planning methods for Autonomous Underwater Vehicle based on the former three theories.First, the developing history and the actuality of the AUV are introduced. And the principle of the path planning for AUV and the several familiar path planning methods is analyzed in detail. Besides, it makes research on the structure of AUV control system. The movement state and the measure of getting environment information are thoroughly explored.Then, it describes briefly the fuzzy control theory and the fuzzy control system, explores AUV path planning method based on the principle of the path planning and fuzzy control system, and simulates it and finds out the merits and defects. Researching on artificial neural network and BP neural network is made.Making use of its self-studying and self-adapting ability, the AUV path planning control system is designed and simulated.Finally, it is found that fuzzy logic system can show knowledge easily and has stronger fuzzy logic inference ability, and artificial neural networks have particular superiority in self-studying respect. In this paper, fuzzy logic system and neural networks are integrated, and a fuzzy neural network AUV path planning controller is constructed. Besides, it is validated by simulation.