为优化无人机影像采集基站布设方式，以提高工程结构三维逆向建模效率。提出了控制参数BHR（breadth height ratio），设置了总计25组对比分析工况；然后设计了一个能够有效模拟实际工程结构三维重建情景的实验，对比了各工况在点定位，移动轨迹和线形变化3类特征测量的误差，从工作效率和模型精度两方面对基站的合理布设方式进行了评估；最后对一个装配式钢框架结构进行三维重建，以验证提出的最优布设方法在工程应用中的有效性。结果表明，25组工况在3类特征测量中表现出相似的规律，即误差随着BHR和拍摄距的下降而减小；25组工况的误差变化率有显著差异，其中拍摄距离>0.8 m的工况绝大多数表现出了负贡献；结合建模时间，综合评估最合理的布设方式为BHR = 0.18，H = H_eff/2的工况，将其应用到实际工程中得到了良好的效果。所提出的无人机基站布设方式可以在保证三维重建精度的前提下大幅缩减建模时长。

To optimize the layout of UAV (unmanned aerial vehicle) base station and improve the efficiency of 3D reverse modelling of engineering structures, a control parameter named BHR (breadth height ratio) was proposed and a total of 25 comparison sets were set up; then, an experiment was designed to effectively simulate the three-dimensional reconstruction scenario of actual engineering structures, comparing the errors of point positioning, movement trajectory, and linear change measurement under different comparison sets. The reasonable layout of base station was evaluated from two aspects: work efficiency and model accuracy. Finally, a three-dimensional reconstruction of a prefabricated steel frame structure was conducted to verify the effectiveness of the proposed optimal layout method in engineering applications. The results show that 25 comparison sets reveal similar patterns in the measurement of three types of features, that is, the error decreases with the decrease of BHR and observation distance. There is a significant difference in the error change rate of 25 comparison sets, with majority of comparison sets with a shooting distance > 0. 8 m showing a negative contribution. According to the modelling time, the most reasonable layout method is comprehensively evaluated as BHR = 0. 18 and H = H_eff/2, and it has been applied to practical engineering and achieved good results. The proposed UAV base station layout method can significantly reduce the modelling time while ensuring the accuracy of 3D reconstruction.