高频地波雷达目标航迹跟踪滤波算法应用研究

发布时间：2018-11-14 14:51

【摘要】：高频地波雷达(HFSWR,High Frequency Surface Wave Radar)是一种重要的海上舰船目标探测手段,能够进行大范围、全天候的海面监视监测。航迹跟踪是目标探测的重要过程,不仅可以实现连续探测并剔除虚假目标,还可以利用滤波作用提高探测精度。高频地波雷达目标航迹跟踪涉及航迹起始、数据关联、航迹跟踪滤波等多个方面,其中,航迹跟踪滤波是一个重要环节。目前,用于高频地波雷达航迹跟踪的滤波方法主要有Kalman滤波和α-β滤波两种。由于地波雷达数据是在极坐标下获取的,直接在极坐标下进行目标跟踪可以避免复杂的坐标转换,因此本文对极坐标下基于α-β滤波和基于Kalman滤波的高频地波雷达目标航迹跟踪方法进行研究,并结合实测数据进行跟踪效果的评价。文章主要工作如下:(1)为了开展航迹跟踪仿真实验,提出一种极坐标下的高频地波雷达航迹模拟模型;针对作为实测地波雷达数据跟踪结果评价手段的AIS(Automatic Identification System)数据与地波雷达数据时、空不一致问题,对AIS数据进行了插值处理,得到了时空匹配的评价数据;针对航迹跟踪整体流程中的不同环节,选取了对应的方法。(2)针对α-β滤波在不同雷达数据航迹跟踪处理中,恒定滤波系数不具有普适性的问题,本文提出了一种自适应α-β滤波方法。该方法分别利用速度误差、距离误差和角度误差三个参量建立了各参量滤波系数的函数模型,实现了各个参量滤波系数的自适应获取,并结合实测数据中高频地波雷达测速精度高、测角精度差的特点,分配了三个参量滤波系数确定的优先级。仿真实验结果和与AIS信息对比的实测数据处理结果均表明,本文方法运算时间短、航迹跟踪精度高,且能够跟踪雷达远端航迹。(3)极坐标下Kalman滤波增益的确定需利用量测噪声方差,在实际航迹跟踪中噪声方差不易得到,为此本文提出一种噪声确定方法。该方法利用目标滤波值和测量值来近似求取噪声方差,使每个具体的测量时刻都具有更加精确的噪声方差值,在此基础上求取更加准确的滤波增益值,最终获得更好的跟踪结果。仿真实验和与AIS信息对比的实测数据处理结果都表明,本文方法能够得到精度更高的航迹结果,而且能够较好地跟踪到远端航迹。最后比较了自适应α-β滤波和Kalman滤波的航迹跟踪性能,结果表明,前者优势在于时间资源消耗小,后者优势在于跟踪精度高,在实际应用中可以根据对运算速度的要求高还是对跟踪精度的要求高来选择采用哪种滤波方法。
[Abstract]:High frequency ground wave radar (HFSWR,High Frequency Surface Wave Radar) is an important method for marine ship target detection. Track tracking is an important process of target detection. It can not only realize continuous detection and eliminate false targets, but also improve the detection accuracy by using filter function. High frequency ground wave radar target track tracking involves many aspects, such as track initiation, data association, track tracking filtering and so on, among which, track tracking filtering is an important part. At present, Kalman filter and 伪-尾 filter are two main filtering methods for track tracking of HF ground wave radar. Since the ground wave radar data is acquired in polar coordinates, tracking targets directly in polar coordinates can avoid complex coordinate conversion. Therefore, in this paper, the tracking method of high frequency ground wave radar based on 伪-尾 filter and Kalman filter in polar coordinates is studied, and the tracking effect is evaluated based on the measured data. The main work of this paper is as follows: (1) in order to carry out the track tracking simulation experiment, a high-frequency ground wave radar track simulation model under polar coordinates is proposed; In order to solve the problem of spatial inconsistency between AIS (Automatic Identification System) data and ground wave radar data, the AIS data are interpolated, and the spatio-temporal matching evaluation data are obtained. According to the different links in the whole process of track tracking, the corresponding methods are selected. (2) the constant filtering coefficient is not universal in different radar data track tracking processing. In this paper, an adaptive 伪-尾 filtering method is proposed. In this method, three parameters, velocity error, distance error and angle error, are used to establish the functional model of the filtering coefficients of each parameter, and the adaptive acquisition of the filtering coefficients of each parameter is realized. Combined with the characteristics of high frequency ground wave radar velocity measurement accuracy and poor angle measurement accuracy in the measured data, the priority of three parameter filtering coefficients is assigned. Both the simulation results and the data processing results compared with the AIS information show that the proposed method has the advantages of short operation time and high track tracking accuracy. And can track the far end track of radar. (3) the determination of Kalman filter gain in polar coordinates needs to use measurement noise variance which is difficult to get in the actual track tracking. Therefore a noise determination method is proposed in this paper. The method uses the target filter value and the measurement value to approximate the noise variance, so that each specific measurement time has a more accurate noise variance value, and based on this, a more accurate filter gain value is obtained. Finally, better tracking results are obtained. The simulation results and the actual data processing results compared with AIS information show that the proposed method can obtain more accurate track results and can track the far end track better. Finally, the track tracking performance of adaptive 伪-尾 filter and Kalman filter is compared. The results show that the former has the advantage of less time and resource consumption, and the latter has the advantage of high tracking accuracy. In practical application, we can choose which filtering method to adopt according to the high requirement of operation speed or the high requirement of tracking precision.
【学位授予单位】：中国石油大学(华东)
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN953

【相似文献】