在高斯噪声环境下,为了解决扩展卡尔曼滤波(EKF)在目标跟踪应用中精度低和滤波发散的问题,将无迹卡尔曼滤波(UKF)应用于非线性系统的目标跟踪。研究了无迹卡尔曼滤波估计方法,对采样策略进行了比例修正。通过UKF在目标跟踪中的应用仿真结果表明,与EKF相比较,UKF有更好的跟踪性能、收敛快、对噪声有更强的适应能力,算法实现简单。
The Extended Kalman Filter (EKF) is widely applied to target tracking in Gaussian noise environment. However its drawbacks are poor filtering precision and filtering disconvergence. In order to overcome those the Unscented Kalman Filter(UKF) is applied to target tracking in nonlinear system, and sampling strategy is proportionably modified. UKF is compared with EKF for target tracking. Simulation results show that UKF is superior in filtering precision, and provides stronger ability to suppress noise in lower complex.