Visual data analytic and robotics for automated construction performance monitoring

The construction industry is one of the major economic sectors in most countries, with 9% to 15% of total Gross Domestic Product (GDP) being allocated to their built environment. Despite this economic importance, this industry is suffering from low productivity and inefficiencies. During the past few decades, the productivity rate in many sectors has been steadily increasing; however, this rate in the construction industry has barely increased, and it may have even decreased. Automated systems and robotics are known as technologies that have the potential to revolutionize the construction industry by addressing productivity challenges while improving quality. Accordingly, the overall goal of this research is to increase the degree of automation in construction applications. To achieve this goal, this research addresses three specific objectives: (1) developing and deploying multi-agent autonomous systems for automated data collection, (2) proposing an automated registration of as-built video sequence (e. g. , video captured during the first objective) to as-planned Building Information Model (BIM) in real-time to facilitate as-built and as-planned data comparison, and (3) developing robotic construction workers capable of performing different construction activities. Regarding the first objective and to characterize the limitations of current techniques on realtime performance and identify challenges in integration and implementation for construction applications, this research focuses on developing and validating autonomous systems for construction applications, especially data collection and construction monitoring. Over the past few years, camera-mounted unmanned vehicles have received significant popularity for developing vision-based data acquisition and construction monitoring. However, there are still numerous open problems for further research, such as autonomous navigation in building construction and a low computational complexity platform for streamlining the data collection process. The first objective of the dissertation is to develop a cooperative UAV/UGV system that dynamically collects and analyzes data with minimal human input. This system has the potential to be used in several construction applications, such as site surveying, progress and safety monitoring, and structural health monitoring. Having accomplished the first objective, the next objective uses big visual data from the first objective as as-built data to facilitate as-built and as-planned data comparison. Big visual data analytics used in conjunction with BIM facilitates as-built and as-planned data comparison, which enables early detection of potential schedule delays and facilitates the communication of progress information accurately and quickly. The second objective proposes an automated registration of a video sequence (i. e. , a series of image frames) to an as-planned BIM in realtime. As part of this effort, the camera poses of image frames are discovered in the 3D model coordinate system by performing an augmented monocular Simultaneous Localization and Mapping (SLAM) and perspective detecting and matching between the image frames and their corresponding BIM views. This system automates visual data localization with respect to BIM in real-time, which can facilitate communication on job sites by associating quality and progress with visuals (being collected by robotics systems in the first objective) that are in the BIM coordinate system. This is particularly useful in addressing limitations of state-of-the-art localization algorithms and has the potential to be used for the localization of any entity (e. g. , worker, equipment, etc), especially in indoor environments where other localization algorithms such as GPS are inefficient.