基于线性约束H_∞滤波的道路目标跟踪算法研究

Road Targets Tracking Methods Research Based on Linear Constraints H_∞ Filtering

针对地面道路目标跟踪噪声未知统计特性,以及道路约束条件,为提高跟踪精度,提出线性约束鲁棒H∞滤波跟踪算法。根据H∞滤波理论,给出无约束条件下H∞滤波目标跟踪算法流程;建立道路网路段的数学模型,在H∞滤波算法流程中添加线性约束条件,首先对过程噪声进行修正,然后构造拉格朗日方程,求出线性约束条件下H∞滤波最优估计值,并根据道路网参数修正状态协方差矩阵。蒙特卡洛仿真结果证明,在有色噪声和道路约束条件下,相对于传统的目标跟踪算法,提出的方法为跟踪性能的提高提供了依据。

A robust H∞ filtering algorithm with linear constraints was proposed to deal with ground moving target tracking with road constraints and noise whose statistic property is unknown. A moving target model was set up. The H∞ filtering algorithm without constraint was proposed based on H∞ filtering theory. Then a road segment model was described. With the constraints, Lagrangian equation was formed to get the optimal estimation of H∞ filtering under linear constraints. State covariance matrix was also modified according to the road parameter. The Monte Carlo simu- lation results show that the proposed algorithm has an obviously better performance in target tracking with colored noise and road constraints.