分布式伪线性卡尔曼滤波纯方位跟踪

Distributed Pseudolinear Kalman Filter for Bearing-only Tracking

针对纯方位跟踪(BOT)的非线性滤波和距离可观测性较差问题,提出了一种新的分布式多传感器辅助变量伪线性卡尔曼滤波器(DM-IVPLKF)。该滤波器利用辅助变量伪线性卡尔曼滤波器(IVPLKF)独立处理目标测量值,通过偏差补偿伪线性卡尔曼滤波器(偏差补偿PLKF)解决由于量测向量与伪线性噪声相关而产生的偏差,将递归辅助变量估计方法嵌入偏差补偿PLKF中,对目标状态估计和协方差进行修正。所提算法利用多传感器最优信息融合准则,对目标状态进行融合估计。然后,推导了多传感器BOT的克拉默-拉奥下界(CRLB)。通过仿真实验,将所提算法与传统算法进行对比,仿真结果证明了所提算法具有较高的跟踪精度。

A new distributed multi-sensor instrumental variable pseudolinear Kalman filter(DM-IVPLKF) is proposed to solve the problems of nonlinear filtering and poor range observability of bearing-only tracking(BOT). The filter processes the target measurement independently by using an instrumental variable pseudolinear Kalman filter(IVPLKF), tackled the bias arising from the correlation between measurement vector and pseudolinear noise by bias compensated pseudolinear Kalman filter(PLKF). The recursive instrumental variable estimation method is embedded into bias compensated PLKF to modify the target state estimation and covariance. The multi-sensor optimal information fusion criterion is used to estimate the target state. Then, the Cramer-Rao lower bound(CRLB) of multi-sensor BOT is derived. Through simulation experiments, the proposed algorithm is compared with the traditional algorithm, and the simulation results show that the proposed algorithm has high tracking accuracy.