It is required to plan the best route for the military UAV when carrying out accompanying penetration assignment, and the restraints of threat and arm performance should be considered seriously. An improved ant colony algorithm is proposed. Firstly, the threat effect is imported to correct the calculation of transfer visibility factor. The transfer probability is influenced sequentially and the entire security of the route is improved. Then, the pheromone is increased around the target node in reason to reduce the searching space and obtain the optimum route faster. The simulation results show the feasibility of the route planning method finally. The optimum routes are generated by weight change in transfer visibility factor calculation and can satisfy the commander needs for different backgrounds and tasks.