請用此 Handle URI 來引用此文件:
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/92387
標題: | 水下音響目標偵測與追蹤研究 Study of detection and tracking of underwater acoustic target |
作者: | 洪靖唐 Ching-Tang Hung |
指導教授: | 陳琪芳 Chi-Fang Chen |
關鍵字: | 到達時間差,接收訊號的強度指示,水下目標偵測定位,水下聲學,貝式定理, TDOA,RSSI,Underwater localization,Underwater acoustic,Bayes' theorem, |
出版年 : | 2024 |
學位: | 博士 |
摘要: | 本研究發展一套結合到達時間差(Time Difference of Arrival,TDOA)、接收訊號的強度指示(Received Signal Strength Indication,RSSI)、傳輸損失(Transmission Loss)與貝式定理(Bayes'' theorem)之水下目標定位技術,藉由計算聲源強度分佈、接收強度分佈與機率的不確定性使得傳統之定值結果擴展成機率分佈結果,並帶入貝式定理計算條件機率關係與更新偵測結果信心程度,讓研究人員能更方便的了解目標分佈情形。
本文主要分為三大核心部分,分別為使用單支水下麥克風時偵測方法、使用水下陣列時偵測方法以及建構聲學傳播與定位模擬環境。單支水下麥克風偵測方法的偵測目標為裝有水下發報器之美洲擬鰈,利用低噪音自動無人帆船範圍式偵測與本文演算法計算出波士頓港灣內區域之美洲擬鰈分佈情形。水下陣列偵測方法的偵測目標為飛航紀錄器的水下發報器,與洋聲股份有限公司、海洋委員會國家海洋研究院、國家運輸安全調查委員會合作,利用拖曳方式於臺灣小琉球南方海域試驗,並使用本文演算法推估出水下發報器位置以及信心權重。聲學傳播與定位模擬環境使用Bellhop、Gazebo、ROS與MOOS-IvP開源軟體在Ubuntu 20.04作業系統上架設,其程式語言主要為C++與Python,此模擬環境中利用Bellhop聲線模擬模組計算聲線與轉移函數並做摺積得出模擬接收訊號,且環境中透過兩支水下麥克風計算到達時間差估計聲源方位。 雖說本文提出的兩個偵測方法使用不同的偵測目標做試驗,但兩者偵測目標的本質均相同-被動音響目標定位,因此,無論為單數或複數水下麥克風做水下目標定位,只要在了解聲源特性的情況下均能使用本文發展的水下定位演算法,在生物族群的偵測或是水下發報器定位上達到預期目標。 This study develops an underwater target localization technique that integrates Time Difference of Arrival (TDOA), Received Signal Strength Indication (RSSI), Transmission Loss, and Bayes'' theorem. By calculating the distribution of sound source strength, received strength, and the uncertainty of probabilities, this method extends traditional determine results to probabilistic distributions, allowing researchers to better understand target distributions. The paper is divided into three sections: detection methods using a single hydrophone, detection methods using an array of hydrophones, and an acoustic propagation and localization simulation environment. The single hydrophone detection targets winter flounder tagged with underwater transmitters, using low-noise automated unmanned sailboats for range-based detection and the algorithms presented in this paper to map the distribution of winter flounder in the Boston Harbor area. The hydrophone array detection targets the underwater transmitters of flight recorders. In collaboration with OceanSound CO., LTD., National Academy of Marine Research (OAC), and Taiwan Transportation Safety Board, experiments were conducted south of Xiao Liuqiu, Taiwan. The location and confidence weight of the underwater transmitters were estimated using the algorithms in this paper. The acoustic propagation and localization simulation was set up on Ubuntu 20.04 using open-source software such as Bellhop, Gazebo, ROS, and MOOS-IvP. The main programming languages are C++ and Python. The Bellhop acoustic ray simulation module calculates the eigen ray and transfer functions, and convolution is performed to obtain simulated received signals. With two hydrophones in the simulation environment, TDOA can be calculated to determine the bearing angle. Although the two detection methods proposed in this study detect different targets, the nature of both targets is the same - passive acoustic target localization. Therefore, whether using a single or multiple hydrophones for underwater target localization, the algorithm developed in this paper can be utilized, provided the characteristics of the sound source are understood. This approach can achieve the expected goals in detecting biological populations or locating underwater beacons. |
URI: | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/92387 |
DOI: | 10.6342/NTU202400690 |
全文授權: | 同意授權(限校園內公開) |
顯示於系所單位: | 工程科學及海洋工程學系 |
文件中的檔案:
檔案 | 大小 | 格式 | |
---|---|---|---|
ntu-112-1.pdf 目前未授權公開取用 | 15.08 MB | Adobe PDF | 檢視/開啟 |
系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。