在多目标跟踪问题中,观测站的有效机动可以提高观测信息的质量,从而提升目标跟踪的精度。对此,文中提出一种基于高斯混合概率假设密度(GM-PHD)滤波器的观测站最优机动马尔可夫决策方法。首先,用Fisher信息矩阵(FIM)行列式建立代价函数;然后,计算出马尔可夫链的转移矩阵,利用马尔可夫决策过程(MDP)来获得观测站最优机动策略。其中,利用GM-PHD滤波器来估计目标的实际位置和为每一决策周期提供概率假设密度(PHD)。通过实验仿真,验证了该机动策略在提高多目标跟踪精度方面的有效性。

In the multi-target tracking, the effective maneuver of the observer can improve the quality of the obstrvation information and the accuracy of the target tracking. In this paper, an optimal maueuvering Markov decision-making method for observer based on Gaussian mixture probability hypothesis density(GM-PHD) filter is proposed. Firstly, the Fisher information matrix (FIM) determinant is used to establish the reward function, then the transition matrix of the Markov chain is calculated, and the optimal maneuver strategy of the observer is obtained by using the Markov decision process (MDP). GM-PHD filter is used to estimate the actual location of the target and to provide probability hypothesis density (PHD) for each decision cycle. The numerical simulations illustrate the effectiveness of this strategy in improving the tracking accuracy of multi-target.

TN957.51

国家自然科学基金资助项目