摘要
针对机动目标跟踪中常见的量测转换问题,提出了一种基于球坐标系下最优线性无偏估计滤波的交互多模型算法。该算法的核心思想是将最优线性无偏估计滤波作为交互多模型中的基本滤波,完成对机动目标的跟踪。在仿真试验中,将该算法与基于扩展卡尔曼滤波的交互多模型算法进行比较,结果表明该算法有效地抑制了扩展卡尔曼滤波中常见的滤波发散问题,并且提高了跟踪的精度,具有较好的实用性。
Aiming at the problem of measurement conversion within the target tracking, a new algorithm combined best linear unbiased estimation with interacting multiple model methods is derived. The main idea of this paper was using the best linear unbiased estimate filter as the basis filter in interacting multiple model methods. Simulations are conducted to compare it with IMM based on EKF and the results indicates that the new approach is able to control the problem of exhale which was common in EKF, improve the precision of tracking and can be applied for engineering.
出处
《指挥控制与仿真》
2008年第5期52-54,共3页
Command Control & Simulation
