Full metadata record
| DC Field | Value | Language |
|---|---|---|
| dc.contributor | Department of Building and Real Estate | en_US |
| dc.contributor.advisor | Li, Heng (BRE) | en_US |
| dc.creator | Chang, Siwei | - |
| dc.identifier.uri | https://theses.lib.polyu.edu.hk/handle/200/12576 | - |
| dc.language | English | en_US |
| dc.publisher | Hong Kong Polytechnic University | en_US |
| dc.rights | All rights reserved | en_US |
| dc.title | Development of a robotic control system for automated building crack inspection | en_US |
| dcterms.abstract | To monitor the quality and health of structures and make sure they are constructed in accordance with regulatory requirements, construction inspection is essential. In most cases, professional inspectors are engaged to inspect the quality defects using visual inspection or measurement devices such as rulers. However, with the fast development of construction industry, the drawbacks of conventional inspection techniques—such as a shortage of skilled labor, high costs, and poor efficiency and accuracy—become increasingly serious. The challenges grow much worse when inspecting building cracks because cracks occur the most frequently. They can exist in any type of building component, such as slabs, walls, or beams, as well as during any stage of construction, such as when a building is being built or demolished. The urgent practical needs, therefore, leave the motivations and opportunities for advanced building crack inspection techniques. | en_US |
| dcterms.abstract | To alleviate the aforementioned concerns, computer vision techniques, such as the convolutional neural network (CNN), are increasingly integrated to achieve the automated building crack inspection. In the process of using CNN to inspect cracks, the images should be captured first to build the datasets. The datasets are then imported into the pretrained CNN model to predict and demonstrate whether there are cracks in the images. To improve the speed and accuracy of CNNs, researchers have been focusing on developing or modifying their architectures. Although CNNs do contribute to automatic crack inspection to some extent, they are less efficient and inconsistent when compared to the fully automated inspection techniques. | en_US |
| dcterms.abstract | This research developed a robotic control system for the automated building crack inspection with considering the mentioned limitation into account. The robotic control system is fully automated, flexible, robust, and user-friendly in comparison to the current computer vision-based crack inspection method. Controlled by the developed robotic control system, the inspection robot can assist or even replace manual works by automatically, remotely, smoothly, and continuously inspecting building cracks. To build the robotic control system, the following research was explored: 1) The development trend of construction inspection robotics was investigated to target the supporting technologies for the development of the robotic inspection system. 2) A lightweight CNN model was designed for the development of robotic vision. 3) A fuzzy logic controller enabled wall follower algorithm was designed for the robotic navigation. 4) A web user interface was designed for the robotic visualization. A series simulation and on-site validation were carried out to validate the feasibility. It has been proven that the developed robotic control system can be successfully employed in robot platforms to conduct building crack inspection works, including video stream capture, CNN-based crack inspection, autonomous navigation in building environments, and the demonstration of inspection outcomes, without human intervention. Consequently, to reduce the reliance on skilled inspectors and increasing productivity and accuracy of building crack inspection. | en_US |
| dcterms.extent | xvi, 173 pages : color illustrations | en_US |
| dcterms.isPartOf | PolyU Electronic Theses | en_US |
| dcterms.issued | 2023 | en_US |
| dcterms.educationalLevel | Ph.D. | en_US |
| dcterms.educationalLevel | All Doctorate | en_US |
| dcterms.LCSH | Building inspection | en_US |
| dcterms.LCSH | Building failures -- Investigation | en_US |
| dcterms.LCSH | Concrete—Cracking | en_US |
| dcterms.LCSH | Concrete—Defects | en_US |
| dcterms.LCSH | Hong Kong Polytechnic University -- Dissertations | en_US |
| dcterms.accessRights | open access | en_US |
Copyright Undertaking
As a bona fide Library user, I declare that:
- I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
- I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
- I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.
By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.
Please use this identifier to cite or link to this item:
https://theses.lib.polyu.edu.hk/handle/200/12576