Non-linear Multi-target Tracking Based on SIS Framework and Ant Colony Optimization
基于SIS框架和蚁群算法的非线性多目标跟踪

A new method based on ACA (Ant Colony Algorithm) is proposed for data association in multi-target tracking. Firstly, the ACA is modified according to specific data association rule, in which the path of ACA and the length of the path are redefined by considering the effect of target moving characteristics on the association possibility. Then the ACA could be applied to find the best tour to the data association problem. Since SIS (Sequential Importance Sampling) performs well in non-linear tracking system, this paper employs it to track targets after achieving the association result with ACA. In computer simulation, examples for multi-target tracking in one-dimension and two-dimension situation are presented. Experimental results show the feasibility and effectiveness of the proposed method.