In this thesis, we discuss the modeling and control issues surrounding the development of a highly maneuverable autonomous underwater vehicle (AUV-HM1).

       A numerical motion simulation system of an autonomous underwater vehicle is developed. In the equations of motions, the effects of trimming weight subsystem, deballast subsystem, control surfaces and main propulsion subsystem are included. The added mass terms are computed by the similar ellipsoids method, and the damping terms as well as the control surfaces effects are estimated by the data base “DATCOM”. The estimations of AUV-HM1 are compared with those obtained by Planar Motion Mechanism(PMM) testing system and Free Running Tests.

      The control of AUVs has been a challenge to control engineers due to combined non-linear nature of both the vehicle itself and the environment in which they operate. The thesis presents an experimental research on the adaptive controller of an AUV test-bed in which the controller architecture is composed of multi-layered neural networks. The problem considered is that of designing a controller for an AUV to provide directional control. A Fiber Optic Gyro(FOG) is used to measure the yaw angle and yaw rate. Directional control performed by two thrusters in the horizontal plane. Weight adaptation of the neural network is achieved by minimize an objective function that is the weighted sum of tracking errors and control input rates. According to the experiments on various command trajectories, we show that when the learning process is kept active through the control operation, the neural network adapts to time-varying plant dynamics as well as disturbance upsets.

        本論文完成建立水下載具運動操控模擬系統﹔其包含移重次系統、拋載次系統、控制面及推進次系統在內之六自由度非線性運動方程式。針對本所發展中“高操控性自主式水下載具(AUV-HM1)”﹐其附加質量及附加慣性矩以近似橢圓體估算法計算﹐流體黏性阻尼力以“DATCOM”為基礎之翼面近似法估算﹐並配合平面運動機構(PMM)試驗分析系統及自航試驗討論理論估算法之適用性。

        由於自主式水下載具的系統本身及所處環境都存有非線性與不確定性，使得自主式水下載具控制器的設計變的複雜與困難。本論文提出以類神經網路為架構之適應型控制器，用以達成自主式水下載具之方向控制。由光纖陀螺儀測得方向角及方向角速率，再利用左右螺槳之推力差以達到方向角的控制。類神經網路的適應學習是藉由減少由誤差量及控制器輸出變化率加權而成之成本函數來達。實驗結果證明隨訓練次數及時間的增加，控制器能很快適應系統及環境的特性，達到控制的要求。本法不需對載具及所處環境有太多的了解，因此本法可充分應付當載具配載不同儀器所造成的動態變化或在不同海域工作所面臨的環境變化等問題。實驗以本所自行發展之自主式水下載具試驗機為驗證的對象。