一种带机动检测的权值约束多新息修正算法

发布时间：2018-05-01 03:29

本文选题：目标跟踪 + 卡尔曼滤波　；参考：《电光与控制》2017年10期

【摘要】：目标运动状态的突变会导致跟踪算法精度大幅降低。为了提高对目标机动阶段的跟踪性能,提出了一种带机动检测的权值约束多新息修正算法。首先,为了准确判断机动时机,提出了一种双误差椭圆的机动检测算法,通过设置双边门限,加强算法对机动的敏感度;然后,为了降低因延迟修正造成的机动误差,以预测量测与真实量测间的欧氏距离为基础,建立距离与权值间的映射关系,从而获得之前修正信息的权值以加大对之前隐含信息的利用率;最后,通过3种场景下的仿真分析说明所提算法的有效性,并经过与标准卡尔曼滤波及自适应渐消卡尔曼滤波的对比,证明所提算法在跟踪强机动目标及弱机动目标情况下均具有较高的费效比。
[Abstract]:The sudden change of the moving state of the target will result in a significant decrease in the accuracy of the tracking algorithm. In order to improve the tracking performance of target maneuvering phase, a weighted constrained multi-innovation correction algorithm with maneuvering detection is proposed. First of all, in order to judge the timing of maneuver accurately, a dual-error ellipse maneuver detection algorithm is proposed. The sensitivity of the algorithm to maneuvering is enhanced by setting bilateral threshold, and then, in order to reduce the maneuvering error caused by delay correction, Based on the Euclidean distance between the predicted and real measurements, the mapping relationship between distance and weights is established to obtain the weights of the previously modified information in order to increase the utilization rate of the previously implied information. The effectiveness of the proposed algorithm is demonstrated by simulation analysis in three scenarios, and compared with standard Kalman filter and adaptive fading Kalman filter. It is proved that the proposed algorithm has a high cost effectiveness ratio in tracking strong maneuvering targets and weak maneuvering targets.
【作者单位】：中国民用航空飞行学院;
【基金】：中国民用航空飞行学院面上项目(J2016-68) 中国民用航空局科研项目(MHRD20150228) 国家自然科学基金民航联合研究项目(U1433126)
【分类号】：E91;TN713

【相似文献】