To solve the problem of dynamic route planning for multi-rotor Unmanned Aerial Vehicle (UAV) formation, an improved hybrid algorithm combining ant colony algorithm with Rapidly-exploring Random Tree (RRT) algorithm was proposed.Firstly, the ant colony algorithm was used to search the original global route with the lowest cost, and the “cooperative avoidance-reconstruction” strategy was proposed in the local route planning.Then the improved RRT algorithm was used to modify the geometric route in real time, thus the UAV formation could avoid the static threats and unexpected obstacles and fly to the destination while satisfying the time constraints.The simulation results show that:1) The improved hybrid algorithm and strategy can implement the dynamic route planning for UAVs effectively;and 2） Compared with the common RRT algorithm, the optimality of the overall route and the online search rate of the local route are improved significantly.