A New Ant Colony Data Association method with Multi-feature Considered in Target Moving

Due to the advantages of ACO (Ant Colony Optimization) in solving complex problems, this paper proposes a new data association algorithm, which is based on Ant Colony Optimization in a cluttered environment, called DDG-ACDA（Distance Direction Gray-ACDA）. In the first instance, the concept for tour and the length of tour are redefined according to the specific requirement for data association problem in multi-target tracking. In the next place, the distance information, directional information and gray information are incorporated into the proposed method, since they are the very important factors that affect the performance of data association process. In the computer simulation of this paper, two targets which move in criss-cross motion are used to validate the performance of the proposed method, and at the same time, the traditional nonlinear filter, EKF, is employed to estimate the target states. Computer simulation results show that the proposed method could carry out data association process in an acceptable CPU time, which is less than other traditional methods, and the correct data association rate is higher than that obtained by the data association algorithm not combined with directional information and gray information. The proposed method is effective for data association in multi-target tracking.