Spatial data acquisition, integration, and modeling for real-time project life-cycle applications

Current methods for site modeling employs expensive laser range scanners that produce dense point clouds which require hours or days of post-processing to arrive at a finished model. While these methods produce very detailed models of the scanned scene, useful for obtaining as-built drawings of existing structures, the associated computational time burden precludes the methods from being used onsite for real-time decision-making. Moreover, in many project life-cycle applications, detailed models of objects are not needed. Results of earlier research conducted by the authors demonstrated novel, highly economical methods that reduce data acquisition time and the need for computationally intensive processing. These methods enable complete local area modeling in the order of a minute, and with sufficient accuracy for applications such as advanced equipment control, simple as-built site modeling, and real-time safety monitoring for construction equipment. This paper describes a research project that is investigating novel ways of acquiring, integrating, modeling, and analyzing project site spatial data that do not rely on dense, expensive laser scanning technology and that enable scalability and robustness for real-time, field deployment. Algorithms and methods for modeling objects of simple geometric shape (geometric primitives from a limited number of range points, as well as methods provide a foundation for further development required to address more complex site situations, especially if dynamic site information (motion of personnel and equipment). Field experiments are being conducted to establish performance parameters and validation for the proposed methods and models. Initial experimental work has demonstrated the feasibility of this approach