基于ST-SRCKF的超高速强机动目标跟踪算法

针对超高速强机动目标运动模型难以准确建立且观测数据易出现不良量测而导致滤波发散的问题，提出一种适用于超高速强机动目标的跟踪算法。该算法根据正交性原理推导了一种新的强跟踪平方根容积卡尔曼滤波（ST-SRCKF）结构，并引入多重渐消因子，渐消因子求解方法和作用位置均不同于已有的ST-SRCKF。根据新息的统计学特性，即新息协方差矩阵的迹服从卡方分布，建立了一种改进的CS-Jerk模型，该模型对目标机动的描述更准确，它与改进ST-SRCKF算法的结合实现了对超高速强机动目标的高精度跟踪。仿真结果表明，改进算法对超高速强机动目标的跟踪性能更佳。

Highly maneuvering hypervelocity-target tracking algorithm based on ST-SRCKF

The movement model of highly maneuvering hypervelocity-target is difficult to construct accurately, and the existence of bad measurements in tracking process may lead to filtering divergence. In order to deal with these problems, a tracking algorithm applicable to highly maneuvering hypervelocity-target is proposed. This algorithm derives a new strong tracking square-root cubature Kalman filter (ST-SRCKF) structure from the orthogonality principle, and introduces multiple fading factors. The solution and function position of fading factors are both different from original ST-SRCKF. According to the statistical characteristics of innovation that the trace of innovation covariance matrix is in a chi-square distribution, a modified CS-Jerk model is constructed. The model describes target movement more accurately. When the modified CS-Jerk model is combined with the modified ST-SRCKF, highly maneuvering hypervelocity-target is tracked with high precision. Simulation results show that the modified algorithm has better tracking performance for highly maneuvering hypervelocity-target.