几种面向弹道目标跟踪算法的性能评估

Performance evaluation of several methods for tracking a ballistic object

提出一种非线性滤波器性能的评估方法.在导弹系数未知的情况下,研究二维弹道目标动力学模型,推导相应的Cramer-Rao下界.分析扩展卡尔曼滤波器、无迹卡尔曼滤波器、积分卡尔曼滤波器和容积卡尔曼滤波器,基于推导出的Cramer-Rao下界,通过仿真实验比较这4种非线性滤波的性能.理论与实验结果表明,扩展卡尔曼滤波的实时性能最好,但跟踪精度较差;容积卡尔曼滤波器在跟踪的速度和精度方面都有较好的表现.

A new evaluation method was proposed to solve the selection of nonlinear filters. The relevant Cramer-Rao lower bounds （CRLBs） were derived based on ballistic motion modal with an unknown ballistic coefficient in two-dimensional space. The performance of extended Kalman filter, unscented Kalman filter, quadrature Kalman filter and cubature Kalman filter was analyzed in theory and compared through simulation based on the concluded theoretical CRLBs. Both the theoretical analysis and the simulation results have indicated that the performance of extended Kalman filter is fast but inaccurate, and the cubature Kalman filter is desirable due to a better trade-off between the tracking accuracy and the computational load.