Article
Engineering, Manufacturing
Ali Maghami, Meshkat Salehi, Matt Khoshdarregi
Summary: The study proposes a novel and fully autonomous system for detecting and segmenting damages and cracks around drilled holes in CFRPs. The system includes an automated multi-light imaging end-effector, image processing steps, and a deep Fully Convolutional Network (FCN) with the U-Net architecture for pixel-wise semantic segmentation.
CIRP JOURNAL OF MANUFACTURING SCIENCE AND TECHNOLOGY
(2021)
Review
Materials Science, Multidisciplinary
Ana Rita Diogo, Bruno Moreira, Carlos A. J. Gouveia, Joao Manuel R. S. Tavares
Summary: This article provides an overview of signal processing techniques used for filtering signals, isolating modes, and identifying and localizing defects in ultrasonic guided wave testing (UGWT). The techniques are summarized and grouped based on the geometry of the tested structures. Although satisfactory results have been achieved, there is still room for improvement, especially by integrating machine learning algorithms to enhance defect identification.
Article
Chemistry, Multidisciplinary
Hossam A. Gabbar, Abderrazak Chahid, Md. Jamiul Alam Khan, Oluwabukola Grace Adegboro, Matthew Immanuel Samson
Summary: In this paper, a new toolbox called CT-Based Integrity Monitoring System (CTIMS-Toolbox) is proposed for automated inspection of CT images and volumes in non-destructive testing for industrial tool quality and safety control. The toolbox consists of three main modules: database management, pre-processing, and defect inspection, utilizing computer vision and deep learning techniques.
APPLIED SCIENCES-BASEL
(2022)
Article
Construction & Building Technology
Cheng Zhang, Feng Wang, Yang Zou, Johannes Dimyadi, Brian H. W. Guo, Lei Hou
Summary: This paper proposes a framework to address the challenge of gaining a fast and holistic understanding of as-is building conditions from fragmented images by registering UAV images into a Building Information Model (BIM) in three stages. The proposed approach includes extracting position and optical parameters from UAV images to configure virtual cameras in BIM, employing an improved generalised Hough transform (GHT) to extract building fasade components with arbitrary shapes, and projecting UAV images onto an orthophoto and incorporating them into 2D and 3D views. The developed Dynamo prototype automates the process and the results show a mean image-to-BIM registration error of within 21 mm.
AUTOMATION IN CONSTRUCTION
(2023)
Article
Computer Science, Information Systems
Hideya So
Summary: This paper proposes a migratory UAV (MiUAV) system that enables UAVs to automatically reach infrastructure facilities, allowing pilots to conduct inspections remotely and significantly improving inspection efficiency. The proposed system includes multiple ports for accommodating and charging UAVs, and UAVs can travel between these ports to reach remote infrastructure. The route decision scheme autonomously sets a route for each UAV based on weights considering port utilization ratio and inclement weather probability, resulting in reduced travel time by avoiding high-utilization ports and bad weather areas, as shown by computer simulations.
Article
Construction & Building Technology
Min-Yuan Cheng, Riqi Radian Khasani, Kent Setiono
Summary: This study proposes a novel HybridGAN that combines ESRGAN and DeblurGANv2 to improve the resolution and blurriness of dynamically acquired railway inspection images. Experimental results show that HybridGAN consistently improves mAP scores across multiple resolution levels and performs significantly better on low-quality dataset.
AUTOMATION IN CONSTRUCTION
(2022)
Article
Environmental Sciences
Dominik Merkle, Alexander Reiterer
Summary: This study proposes a method for measuring the performance of SLAM in indoor and outdoor GNSS-denied areas using a terrestrial scanner and a tachymeter. The method is independent of time synchronization and works on sparse SLAM point clouds. The evaluation results show that the proposed method is efficient and MA-LIO performs superiorly compared to other SLAM algorithms.
Article
Materials Science, Characterization & Testing
Xiuyuan Yang, Wenjuan Sun, Claudiu L. Giusca
Summary: This paper investigates the critical step of surface determination in X-ray Computed Tomography, demonstrating the effectiveness of the marker-controlled watershed algorithm in automating the process and improving its performance.
NDT & E INTERNATIONAL
(2022)
Article
Computer Science, Information Systems
Pi Ko, Samuel A. Prieto, Borja Garcia de Soto
Summary: Inspection of cracks is a crucial task for building maintenance, but it is time-consuming and risky. With the advancement of AI, UAVs, and smartphone cameras, an automated software called ABECIS has been developed for efficient and accurate crack detection. It uses state-of-the-art segmentation algorithms to identify concrete cracks and provide detailed reports.
Article
Materials Science, Characterization & Testing
Lucas Kling E. Silva, Gustavo Almeida, Creison Nunes, Gabriela Ribeiro Pereira, Daniel Kadoke, Werner Daum
Summary: High quality tubular products are crucial for the oil and gas industry, with quality control focusing on non-destructive detection of surface defects. The study demonstrates the potential of structured light technique for real-time defect detection and evaluation, with automatic processing offering advantages over current methods based on individual operator assessments.
Article
Engineering, Aerospace
Jonas Aust, Sam Shankland, Dirk Pons, Ramakrishnan Mukundan, Antonija Mitrovic
Summary: In the field of aviation, maintenance and inspections of engines are important for safe aircraft operation. This study aimed to develop methods to automatically detect defects on engine blade edges and support inspector decision-making. The combined system achieved an 83% accuracy in detecting and locating defects, showcasing the potential value of image-processing approaches for defect detection in small datasets.
Article
Engineering, Multidisciplinary
Sebastian Meister, Mahdieu Wermes, Jan Stueve, Roger M. Groves
Summary: This research proposes an approach for analyzing the classification procedure of fiber layup defects, with smooth integrated gradients and DeepSHAP identified as particularly suitable for visualizing such classifications. Maximum-sensitivity and infidelity calculations confirm the effectiveness of these methods in future applications.
COMPOSITES PART B-ENGINEERING
(2021)
Article
Computer Science, Interdisciplinary Applications
Dayi Zhang, William Jackson, Gordon Dobie, Graeme West, Charles MacLeod
Summary: This paper presents a post-processing approach based on Structure-from-Motion (SfM) for unwrapping and stitching inspection images in small-bore pipe inspections. The method does not rely on image features and is less sensitive to image quality, leading to improved accuracy and coverage of the reconstructed area.
COMPUTERS IN INDUSTRY
(2022)
Review
Construction & Building Technology
Cheng Zhang, Yang Zou, Feng Wang, Enrique del Rey Castillo, Johannes Dimyadi, Long Chen
Summary: This study conducts a systematic review of 115 journal articles to understand the level of automation in UAV-enabled bridge inspection (UBI) and highlight challenges and future research opportunities.
CONSTRUCTION AND BUILDING MATERIALS
(2022)
Article
Materials Science, Composites
Qing-Qing Ni, Jun Hong, Ping Xu, Zhenzhen Xu, Kirill Khvostunkov, Hong Xia
Summary: This study proposed a new nondestructive testing method EMW-NDT using electromagnetic wave technique, showing good detection sensitivity to damages such as delamination size and thickness in CFRP composites. With huge potential, this contactless method could be widely used in the field of damage detection for CFRP composites.
COMPOSITES SCIENCE AND TECHNOLOGY
(2021)
Article
Engineering, Aerospace
Dmitry I. Ignatyev, Hyo-Sang Shin, Antonios Tsourdos
Summary: This paper discusses sparse online Gaussian Processes (GP) adaptive augmentation for Incremental Backstepping (IBKS) flight control. The proposed indirect adaptation significantly reduces model dependency, and global uniform ultimate boundness is proved for the resultant GP adaptive IBKS. The research conducted shows that if the input-affine property is violated, the IBKS can lose stability, but the proposed sparse GP-based estimator provides fast online identification and the resultant controller demonstrates improved stability and tracking performance.
AEROSPACE SCIENCE AND TECHNOLOGY
(2023)
Article
Automation & Control Systems
Runqi Chai, Antonios Tsourdos, Senchun Chai, Yuanqing Xia, Al Savvaris, C. L. Philip Chen
Summary: This article studies the problem of trajectory optimization for autonomous ground vehicles with the consideration of irregularly placed on-road obstacles and multiple maneuver phases. It proposes a novel desensitized trajectory optimization method to provide an effective alternative for addressing the complexity of the mission formulation.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2023)
Article
Computer Science, Information Systems
Dingde Jiang, Feng Wang, Zhihan Lv, Shahid Mumtaz, Saba Al-Rubaye, Antonios Tsourdos, Octavia Dobre
Summary: This article proposes a user-oriented content distribution scheme for satellite-terrestrial networks (STN) to improve content distribution efficiency. The scheme includes algorithms for network division, caching satellite deployment, cache node selection, and content updating mechanism. Simulation results demonstrate that the scheme can reduce propagation delay and network load under different network conditions and has stability and self-adaptability.
IEEE TRANSACTIONS ON MOBILE COMPUTING
(2023)
Article
Automation & Control Systems
Dongyang Li, Dmitry Ignatyev, Antonios Tsourdos, Zhongyuan Wang
Summary: In this study, a cost-effective approach for ROA estimation is proposed based on Lyapunov theory and shape functions. The proposed method significantly improves the accuracy of ROA estimation compared to existing methods.
Article
Engineering, Aerospace
Ruifan Liu, Hyo-Sang Shin, Antonios Tsourdos
Summary: This paper proposes a specific drone delivery problem with recharging (DDP-R) characterized by directional edges and stochastic edge costs subject to wind conditions. A Markov decision process is utilized to address the problem, with an edge-enhanced attention model (AM-E) suggested to map the optimal policy via deep reinforcement learning (DRL). Extensive simulations demonstrate that the proposed DRL method outperforms state-of-the-art heuristics for solving DDP-Rs, especially at large sizes.
JOURNAL OF AEROSPACE INFORMATION SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Chen Li, Schyler C. C. Sun, Zhuangkun Wei, Antonios Tsourdos, Weisi Guo
Summary: Increased drone proliferation poses new threats to airports and national infrastructures, with estimated economic damages of millions of dollars per day. Training accurate drone detection algorithms under scarce data is challenging. We propose a method using GAN and TDA to understand the general data distribution and acquire under-represented data, achieving a significant improvement in accuracy compared to existing methods.
NEURAL COMPUTING & APPLICATIONS
(2023)
Article
Engineering, Aerospace
Gabriele Dessena, Marco Civera, Dmitry I. Ignatyev, James F. Whidborne, Luca Zanotti Fragonara, Bernardino Chiaia
Summary: This study assesses the computational performance of the Loewner framework for mechanical system identification. The framework is compared to two well-established methods, LSCE and N4SID, using a hybrid numerical and experimental dataset. The results show that the Loewner framework achieves better accuracy than LSCE and better computational performance than N4SID.
Proceedings Paper
Engineering, Electrical & Electronic
Huw Whitworth, Saba Al-Rubaye, Antonios Tsourdos
Summary: The fifth generation (5G) technology is expected to have a significant impact on the development of urban air mobility (UAM). 5G networks can provide high-speed, low-latency connectivity necessary for UAM vehicles to communicate effectively. This paper investigates the requirements for UAM connectivity and analyzes the performance of the communication data link using 3rd Generation Partnership Project (3GPP) standard. The integration of 5G technology with UAM has the potential to revolutionize urban transportation, making it faster, more efficient, and safer.
2023 IEEE 24TH INTERNATIONAL SYMPOSIUM ON A WORLD OF WIRELESS, MOBILE AND MULTIMEDIA NETWORKS, WOWMOM
(2023)
Proceedings Paper
Computer Science, Artificial Intelligence
Hasan Karali, Gokhan Inalhan, Antonios Tsourdos
Summary: This paper introduces an intelligent conceptual design framework for choosing the configuration of aerial vehicles. By utilizing AI-driven analysis models, quantitative data is incorporated at the earliest stage of design to select the most suitable configuration. Through design optimization, more accurate initial dimensions of key components are provided, enabling better design point selection through the design iteration process. The paper demonstrates the capabilities of the proposed model through a generic use case focusing on a high-performance combat UAV design study.
2023 IEEE CONFERENCE ON ARTIFICIAL INTELLIGENCE, CAI
(2023)
Proceedings Paper
Automation & Control Systems
Bhaskar Biswas, Dmitry Ignatyev, Argyrios Zolotas, Dmitry Ignatyev
Summary: Control Lyapunov function (CLF) is crucial for designing a certified controller with a known stable region in control systems. Existing methods for constructing the CLF with a stable region known as region of attraction (ROA) often yield conservative results. This paper proposes a new approach based on the Union Theorem in sum-of-squares optimization, which utilizes multiple variable size regions generated by positive functions called Shape Functions. Numerical simulations demonstrate the effectiveness of the proposed method, which outperforms existing methods and provides a significantly enhanced ROA.
2023 31ST MEDITERRANEAN CONFERENCE ON CONTROL AND AUTOMATION, MED
(2023)
Proceedings Paper
Engineering, Aerospace
Cheng Huang, Ivan Petrunin, Antonios Tsourdos
Summary: Tactical conflict management is crucial for time-sensitive urban air mobility operations. A graph-based approach is proposed to generate multiple subgraph views representing candidate actions, which are assessed based on a global cost metric to determine the final action.
2023 INTERNATIONAL CONFERENCE ON UNMANNED AIRCRAFT SYSTEMS, ICUAS
(2023)
Article
Computer Science, Information Systems
Huw Whitworth, Saba Al-Rubaye, Antonios Tsourdos, Julia Jiggins
Summary: The advent of Fifth Generation (5G) technology has brought about a new era of development in the aviation industry. However, the introduction of smart infrastructure has increased the vulnerability of airports to cyber threats. This research paper proposes a deep learning methodology that uses a CNN-GRU architecture to detect various types of cyber threats using tabular-based image data.
Article
Engineering, Electrical & Electronic
Weisi Guo, Zhuangkun Wei, Oscar Gonzalez, Adolfo Perrusquia, Antonios Tsourdos
Summary: Autonomous systems (ASs) cooperate for safe navigation by using centralized or distributed coordination mechanisms that consist of observations, unobservable states, and control variables. The security of data transfer between ASs is crucial for safety, and both cryptography and physical layer security (PLS) methods are employed to secure communication surfaces, each with their own limitations and dependencies.
IEEE VEHICULAR TECHNOLOGY MAGAZINE
(2023)
Proceedings Paper
Engineering, Aerospace
Junjie Zhao, Christopher Conrad, Quentin Delezenne, Yan Xu, Antonios Tsourdos
Summary: The HADO project proposes a Digital Twin system for mixed-reality tests, which allows virtual obstacles to be injected into physical test environments, blurring the boundaries between virtual environments and reality for safe, flexible, efficient, and effective testing of UAS operations.
2023 INTEGRATED COMMUNICATION, NAVIGATION AND SURVEILLANCE CONFERENCE, ICNS
(2023)
Article
Engineering, Multidisciplinary
Gabriel Dessena, Dmitry I. Ignatyev, James F. Whidborne, Luca Zanotti Fragonara
Summary: This paper introduces an enhanced version of the Efficient Global Optimization technique, called rEGO, for finite element model updating of large or complex structures. The method expands the global search capability using a two-step refinement and selection technique, achieving good precision and computational performance. It has practical applications in model updating.
COMPUTER METHODS IN APPLIED MECHANICS AND ENGINEERING
(2024)
