基于当前统计模型的机动目标自适应跟踪算法

Adaptive tracking algorithm of maneuvering targets based on current statistical model

当前统计模型及其自适应卡尔曼滤波算法对强机动目标具有很好的跟踪效果,但当机动目标为弱机动和非机动时算法跟踪性能较差。针对这一问题,提出了采用铃形函数作为模糊隶属函数对模型中加速度极值进行修正的自适应滤波算法,调整加速度稳定时的系统过程噪声方差,提高算法的跟踪精度。同时,借鉴强跟踪滤波算法的渐消自适应滤波因子思想,针对加速度突变的情况引入渐消因子对修正的加速度极值进行调节,提高算法在加速度突变情况下的跟踪速度。仿真实验结果表明,算法对弱机动目标和非机动目标的跟踪具有良好的效果。

The current statistical model and adaptive Kalman filter algorithm have a good performance on strong maneuvering targets tracking, but poor on weak and non-motorized maneuvering targets. To solve this problem, a bell shape function is utilized as fuzzy membership function to adjust the upper and lower limits of target acceleration. Then the algorithm can adjust the process noise variance of stable acceleration adaptively and improves the tracking accuracy effectively. By using the idea of fading factor of the strong tracking filter, a fading factor is proposed to adjust revised extreme value of acceleration. The delay time of tracking can be shortened obviously when there is a sudden maneuver or the acceleration changed greatly. Simulation results show that the algorithm has a good performance on tracking weak and non-maneuvering maneuvering targets.