Fuzzy logic fusion capabilities for efficient implementation of data association

摘要

In this paper, we propose a data association algorithm employing fuzzy logic based on multisensor and the multifeature of targets to solve the uncertainty of the data received from sensor measurements in high noised environment. The learning method to train fuzzy data association system with full-fuzzy model based on a steepest descent gradient is analyzed. The improvement of many primary sensors data fusion on data association is analyzed. The innovation lies to gain the improvement of data association performance by fusing many features of targets while at the same time unincreasing the computational structure of the tracking filter. The theoretical analysis and example demonstrate the feasibility of efficient fusing data of different forms using fuzzy logic system for data association in multisensor multitarget tracking.