用于RAM目标跟踪数据滤波的模型参数自适应CMUKF算法

A CMUKF Algorithm with Adaptive Model Parameters for RAM Target Tracking Data Filtering

C-RAM系统对RAM目标跟踪数据滤波算法有精度高、收敛快的要求。弹道系数未知造成模型不准确会导致传统滤波方法难以满足要求,针对不能准确辨识弹道系数的情况,提出了一种模型参数自适应的RAM目标跟踪数据滤波方法。该方法使用量测转换无迹卡尔曼滤波算法(CMUKF)对系统状态进行估计,在无迹卡尔曼滤波(UKF)的框架下使用最小二乘递推辨识算法(RLS)在线辨识弹道系数,形成了模型参数自适应,并对辨识出的弹道系数进行二次滤波,提高了弹道系数的辨识精度。将本文算法与传统C-RAM中的跟踪数据滤波算法比较,仿真结果表明该算法提高了估计精度,且具有良好的鲁棒性和较少的执行时间。

C-RAM systems have high precision and fast convergence requirements for RAM target tracking data filtering algorithms.In view of the inaccuracy of the model caused by the unknown ballistic coefficient,which makes the traditional filtering methods difficult to meet the requirements and unable to accurately identify the ballistic coefficient,a RAM target tracking data filtering method with adaptive model parameters is proposed.In this method,the measurement transformation unscented Kalman filter algorithm(CMUKF)is used to estimate the system state.Under the framework of the unscented Kalman filter(UKF),the least squares recursive identification algorithm(RLS)is used to identify the trajectory coefficients online,making model parameter adaptive.And the identified ballistic coefficients are filtered twice to improve identification accuracy.Compared with the traditional tracking data filtering algorithm used by C-RAM systems,the simulation results show that the algorithm improves estimation accuracy,has good robustness and less execution time.