针对非线性系统中的目标跟踪问题,在机动转弯模型的基础上提出容积卡尔曼滤波(CKF)与自适应网格(AG)思想相结合的变结构多模型(VSMM)算法,即基于CKF的自适应网格交互式多模型算法(CKF-AGIMM)。该算法将CKF作为滤波器,利用网格中心和网格最小间距的调整对转弯模型集进行自适应变化以有效跟踪目标。与基于容积卡尔曼滤波的交互式多模型算法(IMMCKF)进行了仿真对比。实验结果表明,在未增加运行时间的情况下,该算法相比于IMMCKF算法有更高的跟踪精度与稳定性。
To the issue of target tracking in nonlinear system, a Variable Structure Multiple Model (VSMM) algorithm combining Cubature Kalman Filter (CKF) with Adaptive Grid (AG) is proposed based on the turning model, which is called CKF-AGIMM algorithm. The algorithm uses CKF as a filter, and adjusts the turning model set adaptively by using grid center and the minimum distance between grids, so as to effectively track the targets. Simulation is made for comparison with the IMM based on CKF (IMMCKF). The simulation results show that the proposed algorithm can improve the tracking accuracy and stability without increasing the running time.