中国科学院光电技术研究所机构知识库
Advanced  
IOE OpenIR  > 光电技术研究所博硕士论文  > 学位论文
题名:
光电跟踪测量系统多传感器数据融合技术研究
作者: 张进
学位类别: 博士
答辩日期: 2009-06-05
授予单位: 中国科学院光电技术研究所
授予地点: 光电技术研究所
导师: 吴钦章
关键词: 光电跟踪测量系统 ; 多传感器数据融合 ; 机动目标跟踪 ; 航迹起始 ; 航迹关联 ; 多特征融合
其他题名: Multi-sensor Data Fusion based on Electro-Optic Tracking and Measuring System
学位专业: 信号与信息处理
中文摘要: 光电跟踪测量设备是目标跟踪测量领域中的重要设备,为了保证对目标稳定可靠的跟踪以及提高设备的作用距离,现代光电跟踪测量系统通常具有可见光电视、红外电视等多种传感器。在实际应用中,如何有效可靠的完成对机动目标的稳定跟踪,如何充分利用多个传感器数据提高跟踪精度和跟踪稳定性,不仅是光电跟踪测量系统的研究重点,也是现代目标跟踪系统中的热点问题。本文针对这两个问题,对多传感器数据融合的基本理论、光电跟踪测量系统中的机动目标跟踪、目标关联以及多传感器航迹融合进行了分析与研究,主要研究内容包括以下几个方面: 对多传感器数据融合的基本原理进行了介绍,对JDL功能模型的组成和各部分的功能进行了详细的描述,并且分析了集中式、分布式和混合式三种融合结构的优缺点和应用范围,为深入分析光电跟踪测量系统中多传感器数据融合方式的提供了坚实的基础。 多传感器航迹融合的成功应用,同样需要单传感器获得精度较高的目标航迹,因此首先对单传感器滤波预测算法进行了分析与研究。针对光电跟踪目标的特点,研究了光电跟踪中常用的滑窗最小二乘滤波器和Kalman滤波器,针对常用滤波算法解决非线性问题的不足,提出了适合光电目标跟踪的扩展Kalman算法。对于光电跟踪中机动目标跟踪问题,研究了Singer算法、当前统计模型算法以及交互多模型算法等自适应滤波算法,分析和比较了它们在实际应用中的跟踪效果。 对目标进行跟踪预测是建立在获得准确的目标航迹基础上,而目标航迹的建立主要是通过航迹起始和航迹关联实现。针对目标航迹起始问题,提出了适合光电跟踪测量系统的逻辑起始算法,它能够很好的在密集目标和复杂背景下进行航迹起始。针对航迹关联问题,在研究了最近邻关联算法、概率数据关联算法等目标关联算法的基础上,根据光电跟踪测量设备可以获得目标灰度、面积等多种特征的特点,基于数据融合的思想,采用多特征融合的方式充分利用各种特征信息,结合概率数据关联算法,提出了基于多特征融合的概率数据关联算法。 对于光电跟踪测量系统的多传感器航迹融合,深入研究了各种航迹融合算法,包括早期的等权值融合算法、常用的先验精度极大似然算法、适合动态调整的协方差组合融合算法以及最优分布式融合算法,分别对其融合精度进行了分析和比较。然后针对光电跟踪测量系统的实际情况,提出采用联合Kalman滤波器来进行航迹融合,并深入分析了联合滤波器的关键技术、工作流程和主要结构。最后,针对光电跟踪测量系统在跟踪时无法获得距离信息,量测模型无法准确建立的问题,提出将外引导数据和光电跟踪测量系统的航迹进行融合处理,提高量测模型的精度,从而提高系统的跟踪预测精度。 针对光电跟踪测量系统的工程实现问题,提出将数据融合处理系统作为数据处理核心,负责系统中所有分系统数据的输入输出和处理,获取对系统各个部分的状态及控制,实现了数据处理的模块化和流程化。对整个融合系统的工程实现包括基本流程以及功能模块进行了全面的介绍。 对于各种滤波算法和多传感器数据融合算法而言,算法参数的调整和测试是实现稳定高精度跟踪的关键,这需要大量数据进行算法效果的验证。为了便于算法的验证,研究了算法测试中常用的轨迹仿真方法,并在工程中软件实现,提高了仿真算法的测试效率。
英文摘要: The electro-optic (EO) tracking and measuring device is one of the most important devices in the target tracking field. In order to meet the requirements of stability and the distance of tracking, the modern EO tracking and measuring system usually includes several sensors such as the television sensor and the infrared sensor etc. In the engineering applications, how to track maneuvering target and utilize the multi-sensor characteristic to get higher stability and accuracy of target track become the focus of the EO tracking and measuring system even modern tracking system. In order to utilize multiple sensors in the EO tracking and measuring system effectively, the basic theory of multi-sensor fusion and some algorithms are researched in this PhD dissertation, and the algorithms include maneuvering target tracking algorithms, target association algorithms and multi-sensor track fusion algorithms. In addition, the dissertation describes the main works and results as follows: The dissertation presents the basic theory of multi-sensor fusion, and describes the composite and function of JDL function model in details. Then it analyzes the advantages, disadvantages and applications of the three fusion structures including central-level tracking, distributed-level tracking and hybrid tracking by comparing. The results include their advantages, disadvantages, and their application field. The structures mentioned above to lay a foundation for discuss the multi-sensor data fusion method in EO tracking and measuring system. The successful application of multi-sensor data fusion also requires high precision track of a single sensor. Therefore, the target tracking of a single sensor is introduced ahead. In the field of EO target tracking, common filters such as the slide windows least square filter and the Kalman filter are researched and compared. In addition, in order to resolve the nonlinear problem of EO target tracking, the extended Kalman filter fit for EO tracking is presented. Then some adaptive filter model such as the Singer model, the current status model and the interacting multiple model are compared in the practical tracking application. The target tracking and prediction are based on achievement of correct target track. The target track is constructed by tracking initialization and tracking association. For the case of tracking initialization problem in clutter and complex background, a logic-based tracking initialization algorithm suitable to EO tracking and measuring system was proposed. Additionally, for the tracking association problem, the dissertation analyses conventional association algorithms such as the nearest neighbor algorithm (NN) and probability data association algorithm (PDA). Afterwards,an algorithm based on information fusion of multi-feature and combines with PDA is proposed, which makes full use of the characteristic of sensor and adopts multi-feature of target effectively to improve the success probability of association and tracking precision. For track fusion algorithms, the dissertation introduces some algorithms such as equal weight fusion, max likelihood fusion (ML), cross covariance fusion and optimal track fusion. The fusion precision of these fusion algorithms are compared through simulations. Then the federate filter is applied to fuse the track data from each sensor. The key techniques, process flow and primary structures of the federate filter are described detailed. After EO sensor track fused, the exterior guide data is used to fuse with EO fusion track in order to construct a more precise measurement model. For the engineering application of EO tracking and measuring system, it is proposed to consider fusion process system as a core of the data process with responsibility for input, output and data process of all subsystems and acquirement of statue and control signals of all modules of the whole system. Thus, it realizes the modularization of flow process. The dissertation describes the implementation of the fusion system includes basic process flow and function module. Before filter algorithms and multi-sensor data fusion algorithms are applied in the engineering system, it is important to adjust parameters and test of algorithms. All kinds of experiments of the target tracking are needed to conduct for validation of the algorithms. Therefore,the dissertation introduces and applies some track emulation methods. It can improve the testing efficiency of algorithm simulation.
语种: 中文
内容类型: 学位论文
URI标识: http://ir.ioe.ac.cn/handle/181551/353
Appears in Collections:光电技术研究所博硕士论文_学位论文

Files in This Item:
File Name/ File Size Content Type Version Access License
10001_200618015127034张进_paper.pdf(2176KB)----限制开放View 联系获取全文

Recommended Citation:
张进. 光电跟踪测量系统多传感器数据融合技术研究[D]. 光电技术研究所. 中国科学院光电技术研究所. 2009.
Service
Recommend this item
Sava as my favorate item
Show this item's statistics
Export Endnote File
Google Scholar
Similar articles in Google Scholar
[张进]'s Articles
CSDL cross search
Similar articles in CSDL Cross Search
[张进]‘s Articles
Related Copyright Policies
Null
Social Bookmarking
Add to CiteULike Add to Connotea Add to Del.icio.us Add to Digg Add to Reddit
文件名: 10001_200618015127034张进_paper.pdf
格式: Adobe PDF
此文件暂不支持浏览
所有评论 (0)
暂无评论
 
评注功能仅针对注册用户开放,请您登录
您对该条目有什么异议,请填写以下表单,管理员会尽快联系您。
内 容:
Email:  *
单位:
验证码:   刷新
您在IR的使用过程中有什么好的想法或者建议可以反馈给我们。
标 题:
 *
内 容:
Email:  *
验证码:   刷新

Items in IR are protected by copyright, with all rights reserved, unless otherwise indicated.

 

 

Valid XHTML 1.0!
Copyright © 2007-2016  中国科学院光电技术研究所 - Feedback
Powered by CSpace