Based on local information interaction, a new multi-target tracking method is proposed in this paper, for the problem of lack of target information and target assignment in a multi-agent tracking. First, for the lack of target information in the initial state of an agent, a messenger individual with information dissemination capability is created to help the agents acquire random target information. Second, based on the acquisition of multi-agent information, a target selection strategy is designed in order to make a multi-agent group track each target. Then, the multi-target tracking process stability is analyzed by combining Lyapunov stability theorem and LaSalle invariance principle. Finally, the effectiveness of the above proposed method is demonstrated by simulations, and it is shown that the multi-agent has an ability of dynamic obstacle avoidance.