In this paper, a method using building information modeling (BIM) data to correct LiDAR erroLrs is proposed to solve the indoor precise positioning problem of building robots. Firstly, probabilistic modeling was carried out for the sampled radar data, and then the significance level of uncertainty variables affecting the positioning accuracy of the robot was studied by using the analysis of variance. After the sensitive variables were determined, BIM information was introduced and combined with the maximum likelihood estimation method to correct the radar data. Finally, the position accuracy of the object scanned by LiDAR is evaluated. The results show that the material error variance across different types is on the millimeter scale, and the proposed correction method effectively mitigates these errors.