UKF参数辨识的T-S模糊多模型目标跟踪算法

针对非线性系统中机动目标动态模型不确定性问题,提出了一种新的基于UKF参数辨识的T-S模糊多模型机动目标跟踪算法。在提出的算法中,用多个语义模糊集对目标特征信息进行模糊表示,构建一个通用的T-S模糊语义多模型框架。在T-S模糊语义多模型中,使用模糊C回归聚类算法实现对前件参数的辨识,同时,为了实现系统的非线性特征,引入无迹卡尔曼算法辨识后件参数。仿真结果表明,提出的算法跟踪性能优于传统的交互多模型算法和交互多模型无迹卡尔曼滤波算法,在被跟踪目标突然发生方向改变或目标的动态先验信息不精确等复杂情况时,能够有效地对目标进行精确跟踪。

T-S Fuzzy Multiple Model Target Tracking Algorithm with UKF Parameter Identification

A novel T-S Fuzzy Multiple Model Target Tracking Algorithm with UKF Parameter Identification(TS-UKF) is proposed to solve the uncertainty problem of maneuvering target dynamic model in nonlinear systems. Firstly, the target feature information is represented by multiple semantic fuzzy sets, and a general T-S fuzzy semantic multiple model framework is constructed. Then, the fuzzy C regression clustering algorithm is used to identify the premise parameters of the T-S fuzzy semantic multiple model. Meanwhile, to realize the nonlinear characteristics of the system, the unscented Kalman filtering algorithm is introduced to identify the consequence parameters. Simulation results show that the proposed algorithm has better tracking performance than the traditional interacting multiple model algorithm and interacting multiple model unscented Kalman filter. When the direction of the target is changed suddenly or the dynamic prior information of the target is not accurate, it can effectively track the target.