Abstract: Group members generally have a purpose in common and tend to share information within a certain scope in biological community migration. Vicsek model and artificial potential field method are the very results of inspirations from biological community movement and have found wide application in the field of UAV swarm intelligence. This study, by integration of the advantages of the two models, established a vertical force obstacle-avoiding model. This model, by improving the repulsion force function of traditional artificial potential fields, characterizes the repulsion force of the obstacle, whose main force is always perpendicular to the direction of group movement. Through theoretical derivation and experimental comparative analysis, it can be concluded that the new model effectively improves the efficiency of group obstacle avoidance. a simulation experiment is carried out under the condition of a single obstacle and the simulation results show that compared with the traditional artificial potential field method, the new model not only shortens the obstacle avoidance time, but also displays a better adaptability and effectiveness of group obstacle avoidance. At the same time, compared with the traditional limit cycle obstacle avoidance model, in the entire process of group obstacle avoidance, the number of collision individuals and obstacle avoidance time of the new model are greatly improved.