Detection of both surface and subsurface defects is a vital task for maintaining the structural health and reliability of airport runways. We report the automated data collection and analysis for airport runways based on our novel robotic system, which employs a camera and a GPR (Ground Penetrating Radar) to inspect the surface and subsurface conditions, respectively. To perform the automated data analysis, we propose a novel crack detection algorithm based on the images, and a subsurface defect detection method with GPR data. Additionally, to create a composite global view of a large airport runway span, a camera/GPR data sequence from the robot is aligned accurately to create a continuous mosaic for visualization. We combine these algorithms into a software to perform automated on-site analysis. We have put our robot and software into engineering practice over 20 airports in China, achieving the performance of 70% and 67% F1-measure for crack detection and subsurface defect detection, respectively. More importantly, the results of our algorithms can satisfy the requirement of applications.
The detection and restoration of subsurface defects are essential for ensuring the structural reliability of airport runways. Subsurface inspections can be performed with the aid of a robot equipped with a Ground Penetrating Radar (GPR). However, interpreting GPR data is extremely difficult, as GPR data usually contains severe clutter interference. In addition, many different types of subsurface defects present similar features in B-scan images, making them difficult to distinguish. Consequently, this makes later maintenance work harder as different subsurface defects require different restoration measures. Thus, to automate the inspection process and improve defect identification accuracy, a novel deep learning algorithm, MV-GPRNet, is proposed. Instead of traditionally using GPR B-scan images only, MV-GPRNet utilizes multi-view GPR data to robustly detect regions with defects despite significant interference. It originally fuses the 3D feature map in C-scan data and the 2D feature map in Top-scan data for defect classification and localization. With our runway inspection robot, a large number of real runway data sets from three international airports have been used to extensively test our method. Experimental results indicate that the proposed MV-GPRNet outperforms state-of-the-art (SOTA) approaches. In particular, MV-GPRNet achieves F1 measurements for voids, cracks, subsidences, and pipes at 91%, 69%, 90%, and 100%, respectively.
Purpose
This paper aims to present a novel robotic system for airport pavement inspection tasks.
Design/methodology/approach
The cloud-edge-terminal-based distributed architecture is designed for the proposed robotic system. Then, the following three major parts are designed and deployed, respectively: Terminal: the wheeled-robot-based data collection system. Edge: remote monitoring and data analysis system. Cloud: shared database center of the inspection data and knowledge.
Findings
Validation and application results show that the proposed system satisfies the demands of automated airport pavement inspection tasks and saves the cost of manpower and time.
Originality/value
The proposed system provides a novel solution for the full process of airport pavement inspection. Compared with the traditional manual method, the robotic system can guarantee complete coverage and provide high-precision pavement inspection results with less time and labor costs.
The wall-climbing mobile platform (MP) of a robot for repairing a hydraulic turbine blade onsite is developed. The MP is equipped with ferromagnetic adhesive devices and can work on a spatial curved surface. The contradiction between mobility and load-bearing ability is analyzed, and the problem of self-adaptation to the curved face is solved using differential-driven wheeled locomotion with ferromagnetic adhesive devices. The platform adheres to the blade surface through the force provided by the ferromagnetic devices, and a certain gap exists between the magnetic devices and the blade's surface. A mechanism of three revolution degrees of freedom, which connects the magnetic devices with the platform's chassis, is developed to make the platform self-adapt to the complex curved surface of the turbine blade. A proofof-principle prototype has been manufactured, and experiments prove the success of the MP. The payload of the zero-turn-radius MP with excellent maneuverability exceeds 80 kg. The platform can automatically adapt to complex spatial surfaces, which satisfy the requirements of a hydraulic turbine blade in-situ repair robot.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.