基于粒子滤波的三维转弯目标跟踪方法

3D Turning Target Tracking Method Based on Particle Filter

为了提升对转弯目标的跟踪精度,本文提出了一种基于粒子滤波的三维转弯目标跟踪方法。首先,针对在三维空间中做HGB机动的目标提出了一种三维转弯模型,并建立了目标拦截过程中合理、可信的导弹动力学模型。然后,分别用粒子滤波(PF)、扩展卡尔曼滤波(EKF)、无迹卡尔曼滤波(UKF)对三维转弯模型进行匹配滤波,通过对各滤波方法的仿真对比分析,选用PF作为三维转弯模型的匹配滤波方法。最后,将机动目标跟踪问题转化为粒子滤波的求解,通过抑制粒子退化和增加粒子多样性的方法,提高了非线性滤波的估计精度。仿真结果表明,基于粒子滤波的三维转弯模型可以对做HGB机动的目标实现稳定可靠的跟踪,对基于三维转弯模型的非线性滤波问题,相较于EKF和UKF,PF的估计精度至少可提升30%。

In order to improve the tracking accuracy of turning targets,a 3D turning target tracking method based on particle filter is proposed.Aiming at the HGB maneuvering target in 3D space,this article first proposes a 3D turning motion model and establishes a reasonable and credible missile dynamics model during the interception process.Then,particle filter(PF),extended Kalman filter(EKF)and unscented Kalman filter(UKF)are used to match the 3D turning model,and PF is selected as the matching filtering method of the 3D turning model through the simulation comparison and analysis of each filtering method.Finally,the maneuvering target tracking problem is transformed into the solution of particle filter,and the estimation accuracy of nonlinear filtering is improved by suppressing particle degradation and increasing particle diversity.Simulation results show that the 3D turning model based on particle filter can achieve reliable tracking for HGB maneuvering targets.Compared with EKF and UKF,the tracking accuracy of PF is improved by at least 30%for nonlinear filtering problems based on 3D turning model.