The Application Of CMAC In Autonomous Underwater Vehicle

The thesis is for the identification and control of an open-frame remotely operated underwater vehicle named by "GDROV". The vehicle is designed for the purpose of the inspection of cracks, crevice and potential problem of dikes . It is a high-tech underwater vehicle equipped with many advanced inspecting sensors. The two main inspecting sensors equipped are shallow-dissection sonar and high-frequency sonar. The movement of the GDROV has to be stable for the sake of the regular performance of these two sonars .Therefore ,The research on how to acquire stable movement of the vehicle is crucial to the real application of the vehicle.The computer simulation environment is very important to the design and regulation of the underwater vehicle motion control system when the filed-test is not feasible for some reasons. To the "GDROV", the author acquires its hydrodynamic coefficients through theory computation and bases the simulation environment on these coefficients.Therefore, the thesis can be divided into two main part:1. the research on the building of the "GDROV" simulation system2. the research on the motion control of the vehicleFor the first part , a experiential equation of the vehicle is first considered on the integrated consideration of the equipped sensors, the physical constraint of identification tank and the structure and motion features of the vehicle.For the second part ,a control method-CMAC is proposed and tested. The primary thory of the NN(Artificial Neural Networks)and AM(Associate Memory) are introduced simply.Then the various structures of the NN are analyzed.Considering of the strong non-linearity and coupling of the vehicle system.the author proposed a control method based on the CMAC.Furthermore the simulation result and pool-test result are given.Simulation result show that the CMAC controller trained exhibit good robustness faced with noise and unmodeled dynamics.