Project: Deep Learning and AI for LIDAR vs IFC consistency in BIM analysis
Description
Building Information Modelling (BIM) or Building Information Management, is a highly collaborative process that allows architects, engineers, real estate developers, contractors, manufacturers, and other construction professionals to plan, design, and construct a structure or building within one 3D model.
During the building phase it is highly desired to compare the actual (real) construction details with the 3D BIM design. For this purpose a LIDAR scan records 3D point clouds of the real situation, where engineers are interested in deviations from LIDAR and BIM. Comparison tools can show deviations between point cloud and 3D model. Such a tool enables visualization of deviations between scanned points and the model, with user-desired tolerances.
The goals of the project are to design and evaluate deep-learning algorithms for LIDAR versus IFC compatibility. In more detail:
- Select and process actual point cloud data and corresponding BIM models
- Neural network design and implementation for measuring deviations between selected point cloud data and BIM
- Train and test the neural network
- Assist in evaluating a possible demonstrator
- Reports and presentation of the results.
Detailed description
View PDFDetails
- Supervisor
-
Andrei Jalba
- Secondary supervisor
-
KKKubus
- Interested?
- Get in contact