在水下目标纯方位角跟踪研究中,针对传统扩展Kalman滤波(EKF)方法需要通过Taylor展开实现非线性项的线性化而导致计算过程繁琐且存在奇点的问题,本文提出一种基于一阶差分滤波器(DDF1)的纯方位角跟踪算法,建立了目标的运动学模型及无源声纳的测量模型;利用Stirling插值法实现非线性项的线性化,进而通过DDF1滤波方法实现对目标位置及速度的跟踪。仿真结果与一致性分析表明,与基于EKF的目标纯方位角跟踪方法相比,论文所提出的算法具有较高的跟踪精度和较好的跟踪性能。
In the research of underwater target bearing-only tracking, for solving the problems of complicated calculations and singular points during linearization, which arise in traditional extended Kalman filter (EKF) because of Taylor expansion, a target tracking algorithm based on first order divided difference filter (DDF1) is proposed. Kinematic model of underwater target and measurement model of passive sonar are built. Stifling interpolation is used to the linearization of nonlinear terms. Then tracking position and velocity of the target is realized with DDF1. Both simulation results and consistency test indicate that, under same condition, the proposed beating-only target tracking algorithm based on DDF1 has better precision and performance than the one based on EKF.