3D Laser Scanning Spatial Change Detection in Building Systems
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for automated spatial change detection in construction projects using 3D laser scanning data face challenges in accurately associating as-designed models with as-built conditions, especially in large-scale building systems like ductworks, due to computational complexity and errors in nearest neighbor searching algorithms, leading to misalignments and inefficiencies in change management.
Innovation Solution
A hybrid approach integrating nearest neighbor searching and relational graph matching, with preprocessing steps like data-model registration and subnetwork isolation, to efficiently detect and classify spatial changes in large-scale building systems, using ClearEdge's Edgewise 3D Plant Suite for segmentation and CloudCompare for geometric primitive extraction, and constrained ICP registration for alignment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If nearest neighbor searching algorithm is used to associate as-designed models with as-built conditions, then the process is simple and fast, but the accuracy deteriorates leading to misalignments and errors
Solution Approach 1:
The system segments the building system into multiple subnetworks based on spatial relationships and connectivity. Each subnetwork contains a localized group of building elements (e.g., ducts, pipes, equipment) that are spatially connected. This segmentation allows the complex global association problem to be divided into smaller, more manageable local problems, improving both accuracy and computational efficiency.
Solution Approach 2:
The system applies different association strategies to different parts of the building system. Within each subnetwork, relational graph matching is used to capture local spatial relationships and contextual information. This local quality approach allows the system to maintain high association accuracy in each localized region while managing computational complexity through segmentation.
2Measurement precision
If relational graph matching is used to accurately associate building elements, then the association accuracy improves, but the computational complexity increases exponentially with the number of building elements
Solution Approach 1:
The system divides the entire building system into multiple smaller subnetworks based on spatial proximity and connectivity relationships. Instead of performing relational graph matching on all building elements simultaneously (which would result in exponential computational complexity), the system performs matching within each smaller subnetwork independently. This segmentation reduces the computational burden from O(n²) for the entire system to O(Σnᵢ²) where nᵢ is the size of each subnetwork, significantly reducing overall complexity.
Solution Approach 2:
The system performs relational graph matching on a subset of building elements (those within the same subnetwork) rather than attempting to match all elements globally. This partial action approach focuses computational resources on local associations where they are most needed, achieving sufficient accuracy for construction monitoring without the prohibitive cost of complete global matching.
3Reliability
If manual updates of as-designed BIM are performed to track field changes, then change coordination can be maintained, but errors increase and time consumption grows
Solution Approach 1:
The system enables automated self-updating of the as-designed BIM model by comparing laser scan data with the original design model. The system automatically detects spatial changes, associates them with corresponding building elements using relational graph matching, and updates the BIM model without requiring manual intervention. This self-service capability eliminates the time-consuming and error-prone manual update process while maintaining reliable change coordination.
Solution Approach 2:
The system replaces the manual mechanical process of updating BIM models with an automated computational process. Instead of construction engineers manually comparing field observations with design drawings and updating models, the system uses laser scanning technology combined with automated image processing and relational graph matching algorithms to perform the same function electronically, significantly reducing time and human error.
4Loss of information
If design changes and field adjustments are tracked manually, then change coordination efforts can be managed, but reworks and wastes increase due to improper change management
Solution Approach 1:
The system implements continuous feedback by automatically comparing the as-built conditions (captured by laser scanners) with the as-designed model at multiple stages during construction. This feedback mechanism provides real-time information about deviations between design and construction, allowing project stakeholders to identify and correct issues before they propagate through the construction process or result in rework during commissioning and operations.
Solution Approach 2:
The system performs preliminary detection and documentation of spatial changes during the construction phase itself, rather than waiting until later stages. By identifying deviations early in the construction process, the system enables timely corrective actions before completed work needs to be demolished and rebuilt, thereby preventing rework and reducing material waste.
Data Source
AI summary
Systems and methods for automated spatial change detection and control of buildings and construction sites using three-dimensional laser scanning data are disclosed.