Dense clutter multi-target tracking algorithm based on improved fuzzy clustering data association

In order to improve the accuracy and rapidity of multi-target tracking in clutter environment, the fuzzy clustering algorithm is used to obtain the relative degrees of different measurement target memberships, and the fuzzy correlation probabilities of measurement values and targets are calculated in combination with public measurement influence factors.  Then using fuzzy association information combined with Kalman filtering, each target is separately tracked through the weighted fusion method to achieve the state update of the target. The simulation results show that the combination of fuzzy clustering and joint probability data improves the fastness of the target clustering in the relatively dense environment of clutter, and the accuracy of the target tracking is ensured by the combination of Kalman filter and the weighting of the influence factor.

Fuzzy clustering, multi-target tracking, data association, kalman filter

Xuchao Kang, Guangjun He, Dense clutter multi-target tracking algorithm based on improved fuzzy clustering data association. 2019 5th International Conference on Advanced Computing, Networking and Security (ADCONS 2019). 2019: 68-76.