被动跟踪中模糊变结构交互多模型算法

A Fuzzy Variable Structure IMM Algorithm Used in Passive Tracking

在对机动目标进行被动跟踪时,为了提升跟踪效果,提出了一种基于最小二乘的模糊变结构交互多模型算法。首先,对于被动跟踪中状态与量测之间存在的非线性关系,算法采用最小二乘原理对角度量测进行预处理,降低非线性量测方程的线性化误差。然后,针对交互多模型算法中固定结构的模型集带来的局限性,算法引入模糊推理规则以进行模型集自适应,减小模型之间的竞争,确保跟踪精度。在相同实验条件下,分别用新算法和传统跟踪算法对同一设定轨迹进行估计,仿真结果表明,新算法的跟踪效果优于传统算法。

To improve the effect of passive tracking to maneuvering targets,a least-square based fuzzy variable structure Interacting Multiple Model（IMM） algorithm was proposed.Firstly,for the nonlinear relation between the state variables and measurements in passive tracking,the algorithm used least square principle to pretreat the measured angles,and thus could decrease the error caused by the linearization process of nonlinear measure equation.Then,considering the limitation of the fixed structure model set of IMM,fuzzy inference was introduced to make the model set self-adaptive,which could reduce the competition among models and ensure the accuracy of tracking.The new algorithm and traditional algorithm were used to estimate the definite flight track under the same experiment condition,and the simulation result showed that the improved algorithm has higher accuracy of tracking.