摘要
由于目标运动的复杂性和不确定性,传统的非线性滤波算法难以得到较好的滤波精度,为了提高目标的跟踪性能,提出了一种改进后的非线性滤波算法。该算法利用BP神经网络,对非线性滤波的滤波误差进行了修正,然后将修正后的滤波误差补偿给滤波估计值,进而得到新的状态更新。实验以扩展卡尔曼滤波算法（EKF）和不敏卡尔曼滤波算法（UKF）为例,对运动的单目标进行了蒙特卡洛仿真,结果表明,该算法具有更快的收敛性和更高的滤波精度,能够有效改善目标的跟踪性能。
Due to the complexity and uncertainty of the target motion,the traditional nonlinear filtering algorithm can not obtain good filtering accuracy.In order to improve the tracking performance of the target,an improved nonlinear filtering algorithm is proposed.The algorithm uses the BP neural network to correct the filtering error of the nonlinear filtering,then the corrected filtering error is compensated to the filtering estimation value,and then gets a new state.In this experiment,Monte Carlo simulation of single target is carried out by using extended Kalman filter（EKF）and unscented Kalman filter（UKF）as examples.The results show that the proposed algorithm has faster convergence and higher filtering accuracy,can effectively improve the tracking performance of the target.
作者
李松
汪圣利
Li Song;Wang Shengli(Nanjing Research Institute of Electronics Technology,Nanjing 210039,China)
出处
《电子测量技术》
2018年第12期34-39,共6页
Electronic Measurement Technology
