An optimal measurement data association algorithm

作者：Jiang Zhijian^*; Xu Shumei

来源：Journal of Computational Information Systems, 2012, 8(18): 7653-7660.

• Data association
• Fusion
• Multi-source information
• Multi-target tracking

In the light of the disadvantage of worse tracking performance of the optimal assignment (OA) algorithm based on dynamic information in dense target and clutter scenario, a new optimal assignment algorithm fusing multi-source information (MFIOA) is proposed. The new algorithm firstly fuses multi-source information by using combination rule of D-S evidence theory, and gets the association degree between fusion information of single sensor and target track. Secondly, the inconsistent measure factors between the fusion measurements from every two sensors are calculated by using combination rule of D-S evidence theory again. And they are used to construct the fusion measurement data association cost matrix of multi-sensor. Lastly, the update state of target is determined according to the optimal point-track assignment result. Simulation results show that MFIOA algorithm has an advantage over of OA algorithm. And in comparison with sequential processing multi-sensor generalized probabilistic data association (SMS-GPDA) algorithm, the new algorithm not only improves the multi-target tracking accuracy, but also decreases the time spent significantly.

• 出版日期2012
• 单位北京理工大学