基于机动转弯目标的自适应交互式多模型算法

摘要

To solve turn maneuver target tracking, an adaptive computing algorithm of model set utilizing adaptive grid method was discussed and proposed. We expected to match the current state of motion for targets by utilizing adaptive grid and designing initial grid to regulate the turn rate of turn models adaptively, and combined with interacting multiple model algorithm (IMM) to conduct filter calculation. Thus we can track the targets. Simulation of the proposed algorithm was carried out together with other three common IMM algorithms. We compared those results and came to a conclusion that the proposed algorithm is effective with high precision which indicates that it is close to the idealized model set.%在车辆转弯性能优化控制问题的研究中,针对转弯机动目标跟踪,由于控制模型为单模型,造成跟踪性能差.为解决上述问题,设计了一种利用自适应网格方法对模型集合进行自适应计算的方法.利用自适应网格,通过设定初始粗略网格,将转弯模型的转弯速率作为网格值进行自适应的调整,以期能够符合目标当前时刻的运动状态.然后与交互式多模型算法相结合,对模型进行滤波计算,以达到跟踪目标的目的.最后,通过仿真比较自适应网格交互式多模型算法与三种常规交互式多模型算法的跟踪效果,验证了算法的优越性,证明改进算法跟踪精度高、速度快,能够接近理想的模型设计.