- All Journals
- Al-Qadisiyah University
- Basrah University
- Kirkuk University
- Mosul University
- The Iraqi Borad for Medical Specialization
- University of Anbar
- University of Technology
- Al Mustafa University
- Al-Zahraa University for Women
- Alnoor University
- Kunooz University
- Misan University
- Ninevah University
- Tikrit University
- University of Al-Hamdaniya
- University of Kerbala
Development of An AI-Driven UAV System with Cascade PID Control for Infrastructure Defect Detection Using YOLOv8 | ||
| Al-Qadisiyah Journal for Engineering Sciences | ||
| Articles in Press, Accepted Manuscript, Available Online from 21 July 2026 | ||
| Document Type: Research Paper | ||
| DOI: 10.30772/qjes.2026.171718.1998 | ||
| Author | ||
| Rizauddin Ramli* | ||
| Department of Mechanical and Manufacturing Engineering, Faculty of Engineering and Built Environment, Universiti Kebangsaan Malaysia, 43600 Bangi, Selangor, Malaysia | ||
| Abstract | ||
| In recent years, drone technology has become very significant in many industrial applications especially for civil infrastructure inspection. In this paper, we present a cascade Proportional Integral Derivative (PID) controller supported by an Artificial Intelligence (AI)-based defect detection system by using You Only Look Once (YOLOv8) algorithm using a drone for civil infrastructure. In conventional inspections, the inspectors are facing difficulties in identifying defects because of old and hazardous buildings, unseen small defects and having substantial safety risks while inspecting high-risk building or bridge structures. This study developed a low-cost Do It Yourself (DIY) drone which was outfitted with a flight controller and First Person View (FPV) camera to capture image in real-world scenarios. The developed AI-based DIY drone system utilized a high-resolution imaging camera which enables rapid and remote data acquisition. During flight testing, the camera captured 60 frames per second of real-time footage while maintaining stable navigation and carrying a payload of up to 800 grammes. The YOLOv8 model trained, validated and tested 6,998 annotated photos, which showed a high mean Average Precision ([email protected]) of 99.4%, precision of 96.9%, and recall of 98.4%. The results ascertained that the system can reliably identify structural flaws like cracks, corrosion, peeling paint, and water seepage in difficult environmental conditions. | ||
| Keywords | ||
| Drone technology; Proportional Integral Derivative; You Only Look Once (YOLOv8); Civil infrastructure; Image Processing | ||
|
Statistics Article View: 84 |
||