基于“当前”模型的IMM-UKF机动目标跟踪融合算法研究

Exploring a Better IMM-UKF Fusion Algorithm Based on Current Statistical Model in Target Tracking

文章设计了一种基于"当前"统计模型的交互式多模型无迹卡尔曼滤波(IMM-UKF)融合算法。首先在交互式多模型算法框架内,计算"当前"统计模型的概率,自适应地调整"当前"统计模型中目标加速度,提高了"当前"统计模型的自适应性。其次,该算法结合了交互式多模型和无迹卡尔曼滤波算法,该算法具有交互式多模型具有对不同目标机动模式自适应跟踪的能力和无迹卡尔曼滤波滤波度高的优点。最后,采用分布式融合算法提高了系统抗干扰能力及对目标跟踪的有效性和跟踪精度。通过对三维机动目标进行仿真,结果表明文中所设计的IMM-UKF融合算法对于跟踪以多种机动策略实时机动的目标具有较好的跟踪性能,可以减小系统机动跟踪的误差均值和标准差。较之传统的交互式多模型算法,跟踪性能更加优越。

Aim. The introduction of the full paper reviews a number of relevant papers in the open literature and points out that there is an urgent need to explore; but, to our knowledge, there is no paper in the open literature that explores the effects of interacting multiple models and unscented Kalman filtering （IMM-UKF） fusion method. Sections 1, 2 and 3 explain the result of our exploration, which is the design of our IMM-UKF fusion algorithm based on the current statistical model and which we believe is better than previous ones. The rest of the core of sec- tions 1, 2 and 3 is that the probability of the current statistical model is calculated and that the calculation method combines the advantages of the current statistical model with those of IMM-UKF fusion algorithm so as to extend the applicability range of the current statistical model. The simulation results, given in Figs. 2 through 8, and their a- nalysis show preliminarily that our IMM-UKF fusion method can indeed effectively track the real - time maneuvering target and reduce the average value of errors and their standard deviation value and improve the convergence speed and tracking precision.