雷达目标跟踪算法研究

【作者】 王莹；

【导师】 王建新；

【作者基本信息】 南京理工大学 ， 通信与信息系统， 2008， 硕士

【摘要】 在众多领域,比如军事上的战场监视、防空系统,民用上的交通管制、机器智能、医疗器械,目标跟踪都是一个基本的或重要的问题。雷达跟踪系统中的关键就是目标跟踪算法的设计。随着应用的推广,各种新的技术被应用到雷达目标跟踪中来以适应更加复杂的环境。而目标跟踪中的一个核心部分就是滤波算法,本文重点研究了现阶段应用比较广泛的几种滤波算法,并讨论了这些跟踪算法目前所存在的问题。本文首先概述了目标跟踪模型的建立,给出了几种常见的机动目标运动模型和量测模型坐标系的选择,在此基础上介绍了α-β滤波和α-β-γ滤波,改进型α-β滤波,重点分析了kalman滤波(包括扩展kalman滤波)跟踪技术,通过建立相应的模型仿真了解到,标准kalman滤波对于匀速直线运动跟踪效果较好,而基于EKF的跟踪方法对于线性或者弱机动性的目标有很精确的跟踪性能,但是,对于多维状态的跟踪系统,EKF中的雅克比矩阵计算会很复杂也很困难。其次,本文还详细介绍了基于kalman滤波的交互多模型(IMM:Interacting Multiple Model)目标跟踪技术,阐述了该算法的原理和滤波步骤,分析了其性能和优点,提出了一种优化的2模型IMM算法,并通过建立相同目标模型的matlab仿真,和其他算法的性能作了对比,结果表明改进的IMM算法具有良好的跟踪性能,是一种可行的算法。最后对其他几种常见滤波进行了扼要的介绍,结论部分对未来的研究方向、发展趋势以及滤波发散问题作了展望和建议。更多还原

【Abstract】 In many fields, like in the field of the battlefield surveillance, antiaircraft system in a war, and the civil traffic control, machine intelligence, medical equipment, target tracking is a basic or an important problem. The key of radar tracking system is design of target tracking algorithm. As its application in more fields, many new techniques are introduced to the radar target tracking for more complicated situation. While, one core part in target tracking is filtering algorithm. This dissertation studies several usually ways in filter arithmetic and advances some problems.Firstly, this dissertation summarizes the establishment of the target tracking model, and gives some of familiar maneuvering models and introduces the choice of coordinate. And then,α-βfilter arithmetic,α-β-γfilter arithmetic and improvedα-βfilter arithmetic are introduced. The principle analyzes the classical target tracking method based on kalman filter. Through a simulation, it is known that traditional kalman method is the most useful in a linear system and target tracking method based EKF performs well when tracking a target in a linear system or with small maneuver. Also, the Jacobians matrix of EKF is very hard to calculate in tracking system with multidimensional state. Secondly, the algorithm of Interacting Multiple Model (IMM) basing on kalman for the maneuvering target and its merits are mostly studied. And a new method: 2-model IMM algorithm is proposed. Through making matlab experiment, we can draw the conclusion that this algorithm obtains a better performance in target tracking. At last, we introduce other filter algorithms and summary this paper and prospect the future research directions and development trend.更多还原