为解决城市客运交通枢纽场景中行人仿真存在的冲突次数过多和不合理的震荡现象，基于社会力模型，引入主动避让力，针对行人的避让行为进行研究。因不同行人的微观特征不同，其期望速度存在较大差异，在避让力模型中引入期望速度差，将期望速度作为判断行人避让程度的关键指标，突出避让时行人的差异化选择。通过Matlab进行仿真实验，将仿真结果与真实通道内行人的移动路径进行对比。研究结果表明，期望速度更大的强势个体会更主动地进行避让，改进后的模型可以更好地还原差异化的避让行为，对行人路径的模拟误差较未引入避让力的模型降低了43.4%，改进的社会力模型仿真中不合理的碰撞现象减少了49.3%，且可以再现同向行人之间的超越行为。这表明改进后的模型能更好地还原行人对冲突的避让行为，可用于人群较为复杂场所的行人仿真。

In order to solve the problems of too many conflicts and unreasonable oscillation in pedestrian simulation in urban passenger transport hubs, the active avoidance force was introduced to study the avoidance behavior of pedestrians based on the social force model. Considering that there are great differences in pedestrians′ expected speeds due to their different microscopic characteristics, the expected speed difference was introduced into the avoidance force model and was taken as the key index for judging the degree of pedestrian avoidance, to highlight the differentiated choices of pedestrians in avoidance. Finally, a simulation experiment was conducted using Matlab to compare the simulation results of the model constructed with the movement paths of pedestrians in real channels. The results showed that the stronger individuals with higher expected speed would be more active in avoiding, and the improved model could better restore the differentiated avoidance behavior; the simulation error of pedestrian path was reduced by 43.4% compared with the model without avoidance force; the unreasonable collision phenomenon in the simulation of improved social force model was reduced by 49.3%. It means that the improved model can better demonstrate pedestrians' avoidance behaviors toward conflicts, and can be used for pedestrian simulation in places with complex crowds.