基于高斯和滤波的高轨双星高速目标跟踪

发布时间：2018-03-05 13:11

本文选题：高超音速目标跟踪　切入点：高斯和滤波　出处：《江南大学》2017年硕士论文　论文类型：学位论文

【摘要】：无源定位跟踪技术是现代电子战的重心之一,其通过截获地面、海上、空中和太空等各类辐射源发射的电磁波信号,测量信号的特征参数,从而实现辐射源的定位过程。其中,临近空间是一个蓄势开发的重要战略空间,特别是临近空间飞行器(即高超音速飞行器)。其具有较高的速度,并能对全球任一地点出现的事件在短时间内做出反应;具有突防能力强、生存力高等优点,因此,高超音速飞行器的出现给对空预警、防御提出了新的挑战。针对高超音速飞行器的无源定位问题,论文考虑使用两颗位于地球同步轨道的高轨卫星,获取高超音速目标信号到达双星的时间差(Time Difference of Arrival,TDOA)和频率差(Frequency Difference of Arrival,FDOA),实现处于等高程巡航状态的高超音速目标的无源跟踪。主要研究工作如下:首先,论文针对高超音速巡航目标在WGS-84椭球下的非线性运动特征,考虑将三维空间场景转换成二维运动状态模型,并建立其离散时间运动状态方程。随后,根据目标运动方程和量测方程,推导出高速目标定位跟踪精度的克拉美罗下线(Cramer Rao Lower Bound,CRLB),并通过实验仿真研究不同因素对CRLB的影响。计算机仿真结果表明:频差观测噪声的取值对目标状态估计的CRLB无显著影响;而时差观测量、运动速度和运动方向对CRLB影响较大。其次,论文提出基于高斯混合模型替代扩展卡尔曼滤波(GMM-Alternative Extennded kalman Filter,GMM-AEKF)的高超音速目标跟踪算法。算法基于二分法的思想,引入时差观测量的均匀GMM表示,以进一步提升算法的估计性能,同时研究了AEKF实现频差状态更新对目标轨迹收敛速度的影响。此外,算法基于Kullback-Leibler散度(KL Divergence)原理,给出了有效抑制高斯分量指数增长的管理方法,保持算法的计算复杂度基本恒定。最后,论文通过蒙特卡洛仿真实验对比分析了GMM-AEKF、GMM-EKF和GS-5CKF三种算法的跟踪性能。仿真实验表明:论文提出的GMM-AEKF算法,在目标位置和速度估计精度上具有收敛较快的优势,显示出GMM-AEKF在跟踪高超音速目标问题中的适用性。此外,论文还分析了多种因素对GMM-AEKF算法跟踪性能的影响。
[Abstract]:Passive location and tracking technology is one of the center of gravity of modern electronic warfare. It detects the characteristic parameters of the signal by intercepting electromagnetic wave signals from various kinds of emitter sources such as ground, sea, air and space, so as to realize the localization process of emitter. Near-space is an important strategic space for potential development, especially for the near space vehicle (i.e. hypersonic vehicle), which has high speed and can respond to events occurring anywhere in the world in a short time. It has the advantages of strong penetration ability and high survivability. Therefore, the appearance of hypersonic aircraft poses a new challenge to air warning and defense. The paper considers the use of two high-orbit satellites in geosynchronous orbit, The time difference between hypersonic target signals and the frequency difference Difference of ArrivalTDOAA and the frequency difference frequency Difference of Arrival-FDOAA are obtained to achieve passive tracking of hypersonic target in the condition of constant altitude cruising. The main research work is as follows: first of all, In this paper, considering the nonlinear motion characteristics of hypersonic cruise target under WGS-84 ellipsoid, the three-dimensional space scene is transformed into a two-dimensional motion state model, and its discrete-time motion state equation is established. According to the target motion equation and the measurement equation, The Cramer Rao Lower Boundary CRLB of high speed target positioning and tracking accuracy is deduced, and the influence of different factors on CRLB is studied by experimental simulation. The computer simulation results show that the frequency difference observation noise has no significant effect on the CRLB of target state estimation. Secondly, a hypersonic target tracking algorithm based on Gao Si hybrid model instead of extended Kalman filter (GMM-alternative Extennded kalman filter) is proposed. The algorithm is based on the idea of dichotomy. The uniform GMM representation of time-difference observations is introduced to further improve the estimation performance of the algorithm. The influence of frequency difference state updating by AEKF on the convergence rate of target trajectory is also studied. In addition, the algorithm is based on the principle of Kullback-Leibler divergence and KL divergence. This paper presents a management method to effectively suppress the growth of Gao Si component exponent, keeping the computational complexity of the algorithm basically constant. Finally, The tracking performance of GMM-AEKFU GMM-EKF and GMM-EKF is analyzed by Monte Carlo simulation experiment. The simulation results show that the proposed GMM-AEKF algorithm has the advantage of fast convergence in the accuracy of target position and velocity estimation. The applicability of GMM-AEKF in tracking hypersonic target is demonstrated. In addition, the influence of many factors on the tracking performance of GMM-AEKF algorithm is analyzed.
【学位授予单位】：江南大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN95;TN713

【参考文献】