运动方向约束的自适应SRUKF目标跟踪算法

An adaptive square root unscented Kalman filter with motion direction constraints for target tracking

针对非线性目标跟踪中,单一传感器观测冗余不足及模型噪声时变的问题,提出一种附加运动方向约束的自适应平方根无迹卡尔曼滤波(SRUKF)算法:在传统SRUKF基础上,利用新息向量构建自适应因子调节动力学模型时变误差;并引入载体运动方向作为虚拟观测量,增加观测冗余,以提高滤波性能。实验结果表明,所提出的利用运动方向约束的自适应SRUKF算法可有效提高目标跟踪精度,削弱动力学模型误差影响,特别在不良跟踪条件时具有较强的稳健性。

Aiming at the problems of insufficient observation redundancy of single sensor and time-varying model noise in nonlinear target tracking,the paper proposed an adaptive square root unscented Kalman filter(SRUKF)algorithm with additional motion direction constraints:based on the traditional SRUKF algorithm,the innovation vector was used to construct an adaptive factor to adjust the time-varying error of the dynamic model;and the vector motion direction was introduced as a virtual quantity to increase the observation redundancy to improve the filter performance.Experimental result showed that the proposed algorithm could effectively improve the accuracy of target tracking and weaken the influence of dynamic model errors;especially,the proposed algorithm would have strong robustness under poor tracking conditions.