水下導航定位──多被動感測器網路系統之資訊迴授應用研究

Applications of Information Feedback in Multi-Passive-Sensor Systems for Underwater Navigation Positioning

「潛艦國造」計畫，是國家近期內最大宗的軍事投資。在開發潛艦或水下無人載具（Unmanned Underwater Vehicles，UUV）（前述兩者以下合稱為水下潛航器）當中，水下導航為其中重要的一個環節，而要實現水下導航的關鍵在於具有準確的水下潛航器定位估計。為了保持隱身性，水下潛航器使用慣性導航系統。然而，隨著時間的增長，該導航系統會累積誤差，導致導航存在偏差，需要外部校準。若不進行定期的校準，即使是最好的慣性導航系統也不能應用。使用GPS信號可以校準水下潛航器，但是電磁訊號在水中衰減嚴重，水下潛航器需上浮至水面接收GPS信號。水聲定位需要有聲源發射特定的聲信號，定位系統通過接收來自信標（Beacon）的聲源信號進行定位。信標是一種聲、光或電磁波源，信標位置需固定，可產生特定的信號，用以判定方向、方位或位置。信標可作為水下GPS應用於水下潛航器導航。水下潛航器在民用或軍事用途越來越廣泛，隱蔽的導航一直是亟待解決的問題。因此，本文提出一個動態被動定位的方法，是利用混合座標系濾波技術以及遞迴式資訊迴授估計融合演算法，估計水下潛航器自主導航動態位置與速度。該方法考量水下潛航器利用被動感測器網路，其中組成包含多個信標，一個被動陣列聲納接收來自於信標的訊號，以及利用多源資訊融合技術處理僅有角度的量測數據。本文利用兩座信標，運用蒙地卡羅電腦模擬技術，重複實驗，以驗證本方法之有效性。通過資訊迴授的演算法與無資訊迴授的演算法比較，資訊迴授演算法使得每一個區域處理器的收斂速度大幅提升，該算法也顯著提高了區域處理器導航定位精確度。從模擬結果可知，本文所建議的演算法，適用於水下潛航器導航定位。

'Submarine-made' project is the country's largest military investment in near future. In the development of submarines or underwater unmanned vehicles (UUV), underwater navigation is one of the most important issues. In order to realize the navigation, the underwater vehicle must be capable of accurate position estimate. In order to maintain stealth, underwater vehicles use inertial navigation system. However, as time increases, the navigation system will accumulate errors, resulting in navigation bias, requiring external calibration. No regular calibration, even the best inertial navigation system cannot be applied. The use of GPS signals can be calibrated underwater submersible, but the electromagnetic signal in the water attenuation is serious, underwater vehicles need to float to the surface to receive GPS signals. The underwater acoustic location requires a sound source to transmit a specific acoustic signal, which is positioned by receiving the beacon source signal from the beacon. Beacon is a sound, light or electromagnetic wave source, the beacon position must be fixed, can produce a specific signal to determine the direction, orientation or position. Beacon can be used as underwater GPS for underwater navigation. Underwater vehicles in civilian or military use more and more widely, hidden navigation has been an urgent problem to be solved. Therefore, this paper proposes a dynamic passive positioning method, which is based on the hybrid coordinate system filtering technique and the recursive feedback estimation fusion algorithm to estimate the dynamic position and velocity of the underwater vehicle autonomous navigation. The method considers an underwater vehicle using a passive sensor network, wherein the composition comprises a plurality of beacons, a passive array sonar receives signals from the beacons, and utilizes multi-source information fusion techniques to process angle-only measurement data. In this paper, the use of two beacons and the use of Monte Carlo computer simulation technology, repeat the experiment to verify the effectiveness of this method. Compared with the algorithm of information feedback, the information feedback algorithm makes the convergence speed of each local processor greatly improved, and the algorithm also significantly improves the accuracy of local processor in navigation positioning. From the simulation results, we can see that the algorithm proposed in this paper is suitable for navigation positioning of underwater vehicles.