Project website: http://www.ce.cmu.edu/~ITR

THE PROJECT

The asdmCon project was a collaboration of three disciplines — Architecture, Robotics, and Civil and Environmental Engineering — to investigate ways of integrating suites of emerging evaluation technologies to find, record, manage, and limit the impact of construction defects.

This project is about early defect detection and management at construction sites using integrated project models, laser scanners and embedded sensor systems. Frequent accurate assessment of the status of work-in-place, identifying critical spatio-temporal and quality related deviations, and predicting the impacts of these deviations during a construction project are necessary for active project control and for accurate project history.

Advances in generating 3D environments using laser scanning technologies, and collecting quality information about built environments using embedded and other advanced sensors, makes feasible the collecting quality as-built data. The then current trends in the A/E/C industry for the use of integrated project models showed that a semantically rich integrated project database could support various project management and facility management functions.

FLEXIBLE REPRESENTATION for design and construction

Construction projects commonly engage participants from various domains. Integrating different views into a single project model, and supporting information exchange between alternative representations was a focus of our research. We developed a representation structure that included a model of the as-built information, and provided a flexible decomposition of the product model. We implemented a system to provide: A general view of the project model; A predefined view generated from a predefined representational schema for a defined user; and A user-defined view for an unspecified user.

		This work, based in sorts, offers a description through representational structures from formal compositions over primitive data types. Sorts allows for dynamic information entities, enabling creative design by supporting reinterpretations of existing design descriptions through emergence. This work was developed with Rudi Stouffs and his researchers at TU Delft. Through case studies, we were able to capture dynamic changes in construction, including update, addition, and removal of data from the project model, and present an effective representation for the specific needs of the various experts. The idea of individual needs, a complex project model and its representation emphasized the important of user interaction. Our research into flexible representation focused on database form, queries, and user interaction, and included how to integrate and maximize the use of collected data sets from embedded sensors and laser scanning.
		Case study example; as-designed model, site condition, and as-built model

OBJECT RECOGNITION and VISUALIZATION

We applied object recognition techniques to detect defects

construction defect visualization environment

Using three-dimensional modeling and recognition techniques, construction defects are detected whilst a building is in construction. A time-varying 3D reconstruction of the as-built condition of the construction site is compared to a 3D model derived from the design plan; defects are presented through a visual user interface. The visualization environment allows user interaction and analysis of the integrated project model. It incorporates functionalities for viewing major components of the project model (as-built model, as-designed-model, etc.), manual analysis of potential defects (e.g., deviation measurement), and defect management. The visualization environment provides a central interface for accessing and controlling different tasks in the processing cycle.

An early example of as-designed based object recognition

PUBLICATIONS

• Rudi Stouffs, Ramesh Krishnamurti, Kuhn Park. Sortal Structures: Supporting Representational Flexibility for Building Domain Processes. Computer-aided Civil and Infrastructure Engineering, 22(2), 98-116, 2007. Special Issue on Product Models (Eds. A Watson and C M Eastman)


• Kui Yue and Ramesh Krishnamurti. Extracting building geometry from range images of construction sites. Digitization and Globalization: CAADRIA 07 (eds. Yu Gang, Zhou Qi and Dong Wei), pp 459-466, Nanjing, China, April 2007


• Rudi Stouffs, Ramesh Krishnamurti, Albert ter Haar. A sortal building model supporting interdisciplinary design communication. Building on IT: Joint International Conference on Computing and Decision Making in Civil and Building Engineering (Eds. H Rivard, E Miresco and H Melhem) pp 2056-2065, Montréal, Canada, 2006


• Kui Yue, Daniel Huber, Burcu Akinci, Ramesh Krishnamurti. The ASDMCon project: The challenge of detecting defects on construction sites. Poster Paper, Third International Symposium on 3D Data Processing, Visualization and Transmission (3DPVT’06), Chapel Hill, NC, June 2006


• Kuhn Park, Viraj Srivastava, Ramesh Krishnamurti. SmartBIM: The progression of integrated building information model over the lifecycle of a building. ACADIA 2005: Smart Architecture - Integration of Digital and Building Technologies, Savannah, Georgia, October 2005


• Kuhn Park, Ramesh Krishnamurti. Diary of a building. Poster Paper, CAADfutures2005, Vienna, June 2005


• Kuhn Park, Ramesh Krishnamurti. Digital diary of a building. CAADRIA’05 (Ed. A Bhatt), vol 2, pp 15-25, TVB School of Habitat Studies, New Delhi, India, April 2005


• Rudi Stouffs, Ramesh Krishnamurti. Data views, data recognition, design queries and design rules. Design Computing and Cognition'04 (Ed. JS Gero), pp 219-238, Kluwer Academic Publishers, Dordrecht, 2004


• Rudi Stouffs, Ramesh Krishnamurti, Michael Cummings. Mapping design information by manipulating representational structures. Generative CAD Systems (Eds. Ö Akin, R Krishnamurti and K-P Lam), pp 387-400, School of Architecture, Carnegie Mellon University, Pittsburgh, July 2004


• Kuhn Park, Ramesh Krishnamurti. Flexible design representation for construction. CAADRIA'04 (Eds. Hyun Soo Lee and Jin Won Choi), pp 671-680, Yonsei University Press, Seoul, Korea, 2004


• Chris Gordon, Frank Boukamp, Daniel Huber, Ed Latimer, Kuhn Park, Burcu Akinci. Combining Reality Capture Technologies for Construction Defect Detection: A Case Study. EIA9: E-Activities and Intelligent Support in Design and the Built Environment, 9th EuropIA International Conference, pp 99-108, Istanbul, Turkey, 2003

The asdmCon project was funded by a five-year grant (NSF-CMS-0121549) from the National Science Foundation under the Information Technology Research initiative.

ramesh krishnamurti, January 2011