To solve the path planning problem of high-subsonic unmanned aerial vehicles （UAVs） in air combat scenarios， an improved artificial bee colony algorithm （IABC） is proposed. Firstly， by comprehensively considering the obstacles in three-dimensional space and the coordination problem of UAV’s path planning， a combat scenario model and an objective function are established. Secondly， in the employed bee stage， the particle swarm optimization （PSO） algorithm is introduced to reduce the blindness while searching and enhance the search ability of the algorithm. Finally， in the onlooker bee stage， local smoothing processing is carried out on the food sources in the early stage of iteration based on the dynamic greedy criterion， which further improves the convergence speed of the algorithm. In order to verify the effectiveness of the algorithm， a simulation comparison experiment on the algorithm is conducted. The simulation experiment shows that the IABC algorithm inherits the search advantages of the ABC and PSO algorithms. Compared with the ABC algorithm， the average convergence speed of the algorithm is increased by 47.83%， and the average convergence accuracy of the algorithm is increased by 53.49%.