The traditional ant colony algorithm is prone to problems such as long search time and many turning points in the path w hen searching on the grid map. In this paper, a map modeling method is proposed, which combines Voronoi diagram to c onstruct feasible paths and threat circle to describe obstacle areas. At the same time, the initial pheromone concentration of nodes is set differently, and the path is optimized by the point-by-point method. The experimental results show that the proposed method can effectively improve the ant colony algorithm and make it more suitable for UAV flight in battlefiel d environment.