Abstract
A Digital Twin (DT) of a building is marked by a regular inflow of spatial and visual data to keep its digital representation up to date. A seamless integration of this data with the existing geometric semantic DT is key. To link the two data sources on building element-level, existing methods use 1. a proximity-based approach which relies on good spatial registration, or, 2. a key-point matching approach using characteristic geometry for specific elements. Positional and geometric deviations between the two data sources can be a challenge. In this paper, a method is proposed to link semantic instances in point cloud data to their DT equivalent using high-level geometric features and topological relationships in the building. The method is demonstrated using a small-scale case study that contains positional and geometric deviations. It is demonstrated that the method is robust to positional deviations of doors and deviating geometry.
Original language | English |
---|---|
State | Published - 2023 |
Event | 30th International Conference on Intelligent Computing in Engineering 2023, EG-ICE 2023 - London, United Kingdom Duration: 4 Jul 2023 → 7 Jul 2023 |
Conference
Conference | 30th International Conference on Intelligent Computing in Engineering 2023, EG-ICE 2023 |
---|---|
Country/Territory | United Kingdom |
City | London |
Period | 4/07/23 → 7/07/23 |