請用此 Handle URI 來引用此文件:
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/85796
標題: | 自主式水面船隊圓形勢場法編隊及其應用於海洋聲層析之研究 Autonomous Surface Vehicles Formation for Ocean Acoustic Tomography Using Circular Potential Field |
作者: | Yu-Hsuan Ying 應雨軒 |
指導教授: | 郭振華(Jen-Hwa Guo) 郭振華(Jen-Hwa Guo | jguo@ntu.edu.tw | ), |
關鍵字: | 欠驅動無人載具,動力模型,編隊控制,反步控制,比例-微分控制器,移動海洋深層析,圓形勢場法, under-actuated marine vehicle,dynamic modeling,formation control,PD controller,moving vehicle tomography,circular potential field, |
出版年 : | 2022 |
學位: | 碩士 |
摘要: | 本文探討基於「圓形勢場法」提出一種適用於自主式水面載具之編隊控制演算法以實現移動海洋聲層析。由於移動海洋聲層析的資料收集需使用裝載在載具上成對的聲層析收發器以繞圓的隊形於待測海域移動並進行掃描並依據收集的聲線資料反算流場。此外,考量到聲音訊號須完整傳遞因此繞圓半徑需要足夠大使成對的聲層析系統之間的距離足夠遠。圓形勢場法之演算法會計算出合力,並使無人載具沿著合力方向移動,而此合力由兩種作用力組成。第一種為虛擬導引點所提供的吸引力。虛擬導引點為一組永遠保持於無人載具前方的目標點。本研究設計一種根據所有相鄰無人載具之間的相位角差、基本虛擬導引點位置及理想的相位差值之誤差來決定修正虛擬導引點位置的決策過程。第二種則是載具相互之間根據距離的作用力,延伸合力方向得到參考點,各載具追蹤其參考點最終使相位差收斂到理想的差值。接著,追蹤法可計算無人載具追蹤參考點之目標速度與方向,而載具運動控制則根據反步控制器設計一種追蹤控制器使欠驅動無人載具得以達到目標角度與方向。 本文透過設計與建造三台無人水面載具來驗證此演算法,並於湖中進行實驗。由實驗的結果可知,編隊控制可產生理想的繞圓軌跡並且使載具之間相位角的誤差在過程中逐漸收斂。本文最後也將此算法實驗於基隆望海巷灣,藉由編隊控制,聲納系統收發的資料成功被收集。透過比較聲層析收發器傳遞訊號的走時及全球定位系統量測資料即可驗證資料的正確性。 This study investigates a formation control based on the Circular Potential Field method (CPF) for achieving moving vehicle tomography (MVT) with autonomous surface vehicles (ASVs). A pair of transmitters and receivers collect the acoustic ray data for the tomography inversion of the current field by surrounding a target sea area. Circular trajectories are designed for the MVT system to collect data. The radius of circular trajectories must be large enough to allow for sound transmission duration between a pair of transducers. The algorithm of CPF produces a resultant force vector for each vehicle to follow. One of the forces is an attraction to a virtual leader. Virtual leaders are a set of goal points which are always in front of ASVs. This work designs a decision process based on errors of phase difference, the initial positions of all basic virtual leaders, and the desired phase differences. The other force is the interaction forces of relative distances between vehicles. The resultant force provides an orientation command for the vehicle. According to the orientation command, reference positions for the vehicle are generated to ensure that phase differences converge to expected angles. A tracking law then controls the velocity and orientation. The motion tracking controller is designed by a backstepping-based controller for the heading and the velocity control using the underactuated surface vehicle model. Three ASVs were manufactured to demonstrate the guidance and control. The experiments were conducted in a lake. Experimental results illustrate that formation control can generate satisfactory circular trajectories for the ASVs’ controller, and the errors of phase difference between vehicles converge. Finally, the experiments were carried out in a Bay in Keelung City. With the formation control, the MVT system collected ray data successfully. The arrival patterns and the signal-to-noise ratio of the data are presented. Derived distance between pairs of transducers using the travel times are compared with the results of GPS data to verify the correctness of the travel time patterns of the ocean acoustic sensors. |
URI: | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/85796 |
DOI: | 10.6342/NTU202200677 |
全文授權: | 同意授權(全球公開) |
電子全文公開日期: | 2032-04-14 |
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 | |
---|---|---|---|
U0001-0304202218030100.pdf 此日期後於網路公開 2032-04-14 | 11.23 MB | Adobe PDF |
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。