Master of Science
Unmanned aerial vehicles (UAV), commonly referred to as drones (Dynamic Remotely Operated Navigation Equipment), show promise for deploying regular, automated structural inspections remotely. Deep learning has shown great potential for robustly detecting structural faults from collected images, through convolutional neural networks (CNN). However, running computationally demanding tasks (such as deep learning algorithms) on-board drones is difficult due to on-board memory and processing constraints. Moreover, the potential for fully automating drone navigation for structural data collection while optimizing deep learning models deployed to computationally constrained on-board processing units has yet to be realized for infrastructure inspection.
Thus, an efficient, fully autonomous drone infrastructure inspection system is introduced. Using inertial sensors, mounted time-of-flight (ToF) and optical sensors to calculate distance readings for obstacle avoidance, a drone can autonomously track around structures. The drone can localize and extract faults in real-time on low-power processing units, through pixel-wise segmentation of faults from structural images collected by an on-board digital camera. Furthermore, proposed modifications to a CNN-based U-Net architecture show notable improvements to the baseline U-Net, in terms of pixel-wise segmentation accuracy and efficiency on computationally constrained on-board devices.
After fault segmentation, the fault points corresponding to the predicted fault pixels are passed into a custom fault tracking algorithm; based on a robust line estimation technique, modifications are proposed using a quadtree data structure and a smart sampling approach. Using this approach, the drone is capable of following along faults robustly and efficiently during inspection to better gauge the extent of the spread of the faults.
Summary for Lay Audience
Timely and high-quality structural inspections are necessary. However, manual inspection practices are still widely adopted, which have proven to be costly, time-consuming, and risky to inspectors who must manually assess these structures close-up. Technological advances in recent years have opened the possibility of automating parts of the inspection process: the data collection process and the analysis of collected data. Aerial vehicles called drones can be controlled by an offboard pilot and are beginning to be used to perform close-up structural inspections as opposed to humans. Instead of human senses and hand-held apparatuses respectively collecting qualitative and quantitative measurements, cameras and other sensors can be mounted on the drone to automatically collect this information during fly-by. However, processing this information is difficult on drones, due to their limited processing capabilities. Sensors also enable the possibility for fully autonomous navigation without the need for a human pilot. Yet, most current applications of drones for structural inspection require drones to be manually piloted.
Thus, proposed is a fully autonomous inspection system that uses a drone that can navigate on its own without the need for a manual pilot. This drone, mounted with a camera, can collect and process images during structural inspection in an efficient manner, to extract possible structural defects and faults (such as cracks) in live time, while also tracking along these faults during the inspection.
Manka, Marlin, "Developing an Efficient Real-Time Terrestrial Infrastructure Inspection System Using Autonomous Drones and Deep Learning" (2022). Electronic Thesis and Dissertation Repository. 8834.