机动目标跟踪的S修正无迹卡尔曼滤波算法

The algorithm of S-amended UKF in maneuvering target tracking

针对非线性观测条件下的非线性机动目标跟踪问题,借鉴线性滤波中卡尔曼滤波器的S修正防发散思想,对基本无迹卡尔曼滤波算法进行改进,提出S修正无迹卡尔曼滤波(SUKF)方法.对二维机动目标跟踪的仿真结果表明,该算法与基本UKF算法相比,跟踪精度大幅提高,但计算时间略有增加;与SPPF算法相比,跟踪精度提高,且计算复杂度大幅降低,计算时间大幅缩减.

Aiming at the problem of maneuvering target track- ing under nonlinear observation, an S-amended unscented Kalman filtering （SUKF） was developed by using the idea of S-amended anti-divergent method for Kalman filter in linear filtering and improving the performance of unscented Kalman filtering algorithm. Two-dimensional maneuvering target track- ing simulations were carried out, and results show that tracking accuracy is significantly improved comparing with the bas- ic UKF algorithm, but calculation time increases slightly. Comparing with the basic UKF algorithm, tracking accuracy is improved, and calculation time reduces significantly.