Multi-UAV target tracking is a hot topic in recent years. In this paper, a multi-UAV target tracking model is constructed for a directed graph network composed of multiple UAVs, and a distributed fusion processing method is adopted.Based on the multi-UAV target estimation of distributed unscented information filtering, the thesis deduces the weighted average consistency algorithm based on directed graph network, and proposes two weighting matrix design methods. Based on this, a problem solving framework for distributed unscented information filtering consistency algorithm is constructed. The simulation analysis shows that the consistency algorithm based on directed graph network can track the target well under different network topologies, which has good estimation accuracy and robustness, and can improve the performance of multi-UAV fusion estimation.