一种稳健的被动目标跟踪方法

A robust passive target tracking method

针对海洋环境中估计目标方位的误差而导致多传感器被动目标跟踪算法效果不佳、稳定性差的问题,提出一种稳健的多传感器被动目标跟踪方法。该方法首先将各个传感器的观测信息从角度转化为位置,从而有效地线性化了系统模型,然后运用卡尔曼滤波,在滤波过程中根据滤波预测值剔除通信过程中因误码率产生的失真角度测量值,并用预测值代替,从而有效地保证了目标跟踪方法的稳健性。最后进行了数值仿真,结果表明该方法在信息误差较大时也可以提供较为稳健的跟踪性能。

Because of the poor reliability and high BER （bit error ratio） of the acoustic communication in ocean, the algorithm of multi-sensor passive target tracking does not work weU. A robust multi-sensor passive target tracking me- thod is proposed. Firstly, the information observed by each sensor is converted from direction of arrival into target＇s position so as to linearize the nonlinear system modal. Secondly, Kalman filtering is used. During filtering, the measured angle value, which is not true due to BER, is eliminated when the error between the predicted angle and the measured one is over the threshold. Finally, the untrue angle value is replaced by the predicted angle to guarantee the robusmess of the target tracking method. Numerical simulations demonstrate the effectiveness of proposed scheme, even under the circumstances of high BER in the acoustic wireless communication.