臺灣博碩士論文加值系統

在追蹤多個目標時，常因為感應器的誤判造成追蹤上的錯誤。假使只用一個感應器來感測，通常會因量測範圍過大，造成掃描速度變慢使得量測資料被遺失掉，所以常會使用多個感應器來感測目標，以降低量測資料的錯誤率，使得追蹤目標誤差變的更小。因此感應器之間資料結合就變的非常重要。本篇論文主要是探討集中式逼近估測法，來處理資料結合和目標追蹤之間的相關問題。本文將結合不同位置感應器的資料，來解決追蹤判斷上的問題，進而產生一個總體的估測值。

In multi-target tracking, there will have some problems happened to track several targets. These also usually cause the sensor to make mistakes in judgment or errors in tracking. If only using one sensor to do the detection, it will usually slow down the simple rate because of the large measurement range, which may lead the lost of some measurement data. Generally, by using several sensors to detect the target data declines the rate of error and makes errors of tracking target smaller. Therefore, it is very important to combine with the data of the sensors. The main purpose of this research is focus on how centralized approaching estimation algorithm handles the related problems of data association and target tracking. This research will associate the data of sensors in different positions in order to solve the measuring problems in target tracking and further to get the overall estimate of integer.

目錄
封面內頁
簽名頁
授權書................................................iii
中文摘要...............................................iv
英文摘要………………………………………………………………v
誌謝 …………………………………………………………………vi
目錄…………………………………………………………………vii
圖目錄 ………………………………………………………………ix
表目錄 ………………………………………………………………xi
第一章　緒論…………………………………………………………1
1.1 目標追蹤問題…………………………………………1
1.2 研究動機………………………………………………1
1.3 研究方法………………………………………………2
第二章　座標轉換……………………………………………………5
2.1 簡介……………………………………………………5
2.2 原理介紹………………………………………………5
2.3 速度的座標轉換……………………………………11
第三章 卡門濾波器…………………………………………………13
3.1 卡門濾波器基本簡介………………………………13
3.2 卡門濾波器之系統狀態定義………………………13
3.3 卡門濾波器之數學推………………………………16
3.4 卡門濾波器之性質…………………………………19
3.5 擴展式卡門濾波……………………………………21
第四章 追蹤系統之數學模式和集中式感應器……………………26
4.1 模型定義……………………………………………26
4.2 集中式感測器 ………………………………………29
第五章 適應性Gate之介紹………………………………………31
5.1 簡介…………………………………………………31
5.2 目標追蹤程序………………………………………32
5.2.1 軌跡起始…………………………………………33
5.2.2 追蹤相互關係……………………………………33
5.2.3 軌跡更新…………………………………………34
5.2.4 軌跡預估…………………………………………35
5.2.5 軌跡刪除…………………………………………35
5.3 資料相關結合技術…………………………………36
5.3.1 資料相關機率……………………………………36
5.3.2 One-Step Conditional Maximum Likelihood理論推演 ………………………………………………39
5.4 適應性Gate的追蹤程序……………………………41
第六章 程式模擬分析 ……………………………………………44
6.1 初始假設定義 ………………………………………44
6.2 模擬方法與結果 ……………………………………45
6.3 模擬結果分析 ………………………………………56
第七章 結論 ………………………………………………………58
參考文獻 ……………………………………………………………59
圖目錄
圖1.1 工作流程圖……………………………………………………4
圖2.1 n個感應器示意圖 ……………………………………………6
圖2.2 單個感應器示意圖……………………………………………6
圖2.3 卡式座標系……………………………………………………8
圖3.1 卡門濾波器之系統方塊圖 …………………………………15
圖3.2 卡門濾波器的數學運算流程圖 ……………………………19
圖3.3 卡門濾波器之動態系統模型 ………………………………25
圖4.1集中式感測資料融合 ……………………………………… 30
圖5.1目標追蹤系統的工作流程圖 ……………………………… 33
圖5.2 追蹤初始相互關係圖 ………………………………………34
圖5.3 追蹤程序基本流程圖 ………………………………………36
圖6.1 3D單一感測器+卡門濾波器之追蹤軌跡圖…………………46
圖6.2 3D單一感測器+卡門濾波器誤差與時間的分佈圖……… 47
圖6.3 3D單一感測器+卡門濾波器之位置、速度、角度誤差圖
……………………………………………………………………… 47
圖6.4 卡門濾波器+座標轉換 之追蹤軌跡圖 ……………………48
圖6.5 卡門濾波器+座標轉換之誤差與時間的分佈圖 …………49
圖6.6卡門濾波器+座標轉換之位置、速度、角度誤差圖 ………49
圖6.7 集中式多感測器+卡門濾波器 之追蹤軌跡圖 ……………50
圖6.8 集中式多感測器+卡門濾波器之誤差與時間的分佈圖 …51
圖6.9 集中式多感測器+卡門濾波器之位置、速度、角度誤差圖…51
圖6.10 3D擴展式卡門濾波器 之軌跡追蹤圖 ……………………52
圖6.11 3D擴展式卡門濾波器 之位置與速度誤差圖……………53
圖6.12 3D擴展式卡門濾波器+多目標追蹤 之位置與速度誤差圖
………………………………………………………………………54
圖6.13 3D擴展式卡門濾波器+多目標追蹤之誤差圖……………55
表目錄
表6.1 初始條件表…………………………………………………46
表6.2 多目標的初始狀態…………………………………………54
表6.3 多目標的誤差值……………………………………………55
表6.4 各種方法的誤差表…………………………………………57

參考文獻
[1] LaViola, J.J., Jr, “ A comparison of unscented and extended Kalman filtering for estimating quaternion motion,” IEEE American Control Conference, Volume: 3 , 4-6 June 2003 Pages:2435 - 2440 Vol.3
[2] Gad, A.; Majdi, F.; Farooq, M. “A comparison of data association techniques for target tracking in clutter,” IEEE Information Fusion , Conference on , Volume: 2 , 8-11 July 2002 ,Pages:1126 - 1133 vol.2
[3] Hongshe Dang; Chongzhao Han. “ Multi-target data association approach for vehicle tracking in road situation,” Intelligent Transportation Systems, 2003 IEEE , Volume: 1 , 12-15 Oct. 2003
Pages:379 - 383 vol.1
[4] Lin, L.; Kirubarajan, T.; Bar-Shalom, Y. “New assignment-based data association for tracking move-stop-move targets,” IEEE Information Fusion Conference on , Volume: 2 , 8-11 July 2002
Pages:943 - 950 vol.2
[5] Blasch, E.; Lang Hong. “ Data association through fusion of target track and identification sets,” IEEE Information Fusion Conference , Volume: 1 , 10-13 July 2000 Pages:TUD2/17 - TUD2/23 vol.1
[6] Brumback, B.D. and M.D. Strinath, “A Fault-Tolerant Multisensor Navigation System Design,” IEEE Trans. Aerosp. Electron. Syst.,Vol. AES-23, No.6, pp.738-755 (1987)
[7] Chang, C.B. and J.A. Tabaczynski, ’’Application off State Estimation to Target Tracking,’’ IEEE Trans. Automat. Contr. Vol. AC-29. No. 2. pp.98-109 (1984).
[8] Chang. K.C., C.Y. Chong, and Y. Bar-Shalom.’’Joint Probabilistic Data Association in Distributed Sensor Netowrks,’’ IEEE Trans. Automat. Contr., Vol. AC-31, No. 10. pp. 889-897 (1986).
[9] Chang, K.C. and Y. Bar-Shalom, ‘’Distributed Adaptive Estimation with Probabilistic Data Association,’’ Automatica, Vol. 25, No. 3, pp. 359-369 (1989).
[10] Chang, W.T. and S.A. Lin,’’Incremental Maneuver Estimation Model for Target Tracking,’’ IEEE Trans.Aerosp. Electron. Syst., Vol. AES-28, No. 2, pp. 439-452 (1992).
[11] Chung, Y.N., D.L. Gustafson, and E. Emre, ‘’Ex-tended Solution to Multiple Maneuvering Target Tracking,’’ IEEE Trans. Aerosp. Electron. Syste., Vol.AES-25, pp. 876-887 (1990).
[12] Chung, Y.N., E. Emre, and D. Gustafson, ‘’A Global Modeling Approach for Multisensor Problems,’’ in Proc. Signal and Data of Small Targets FL, pp. 306-314(1991).
[13] Emre, E. and J. Seo, ‘’A Unifying Approach to Multi-Target Tracking,’’ IEEE Trans. Aerosp. Electron. Syst., Vol. AES-25, No. 4, pp. 520-528 (1990).
[14] Castella, F.R “Multisensor,multisite trackung filter,” Radar, Sonar and Navigation, IEE Proceedings - , Volume: 141 , Issue: 2 , April (1994)
[15] Gish,H. and R. Mucci, ‘’Target State Estimation in A Multitarget Environment Using Multiple Sensor,’’ IEEE Trans. Aerosp. Electron. Syst., Vol. AES-23,No. 1, pp.60-72 (1987).
[16] Tugnai, J.K., ‘’Detection and Estimation for Abruptly Changing System,’’ Automatica, Vol. 18, No. 5, pp. 607-615 (1982).
[17] Willsky, A.S., J.J. Deyst, and B.S. Crawford, ‘’Two Self-Test Methods Applied to An Inertial System Problem,’’ J. Spacecr. Rockets, Vol. 12, No. 7, pp.434-437 (1975).