基于强跟踪滤波器的交互式多模型算法

Interacting multiple models algorithm using strong tracking filter

针对传统的EKF-IMM算法鲁棒性较差等问题,提出了一种基于强跟踪滤波器(STF)的交互式多模型算法。该算法通过引入强跟踪滤波器(STF)的渐消因子,实现了对滤波器增益的实时调节,从而提高了系统对机动目标的自适应跟踪能力和跟踪精度。仿真结果表明,在目标不发生机动时,该算法和EKF-IMM算法的跟踪效果相近,在目标发生强机动时,该算法在径向速度和方位角的跟踪精度要优于EKF-IMM算法;提出的算法具有更优的机动目标跟踪性能。

Aiming at the problem of the robustness of EKF-IMM is below average, an interacting multiple models algorithm using Strong Tracking Filter （STF） is proposed. Through introducing a fading factor of strong tracking filter, this algorithm realizes the realtime adjusting the gain of the filters, and updating the adaptive tracking performance and tracking precision for maneu- vering targets accordingly. The Monte Carlo simulation result shows that this algorithm has the same tracking effect for non- maneuvering target as EKF-IMM, and the tracking performance for maneuvering target is superior to EKF-IMM on radial velocity and azimuth. The simulation results verify that this algorithm has better performance than EKF-IMM in tracking maneuvering targets.