基于UKF的自适应网格交互式多模型算法

摘要

针对非线性观测条件下的机动目标跟踪问题，基于机动目标的协同转弯（ CT）模型，采用无迹卡尔曼滤波（ UKF）方法和自适应网格（ AG）的模型集自适应策略，研究了一种变结构交互式多模型算法，即基于UKF的自适应网格交互式多模型（ UKF-AGIMM）算法。对二维机动目标跟踪的仿真结果表明，本算法与相应的固定结构交互式多模型算法相比，可以解决固定结构多模型算法存在的问题，有效提高多模型算法的精度和费效比，缩短计算时间，且适合工程应用。%Devoted to the problem of maneuvering target tracking under nonlinear observation, a variable structure multiple model ( VSMM) algorithm called unscented Kalman filtering and adaptive grid interacting multiple model( UKF-AGIMM) is developed in this paper, which is based on the coordinated turn ( CT) models, unscented Kalman filtering ( UKF) method and model set adaptive strategy of adaptive grid ( AG) . Simulation results show that this algorithm can solve the problems of fixed structure multiple model ( FSMM) algorithm, effectively improve the accuracy and cost-efficiency ratio of the multiple model algorithm, reduce the computation time in contrast to corresponding fixed structure algorithm, and is suitable for engi-neering applications.