Article
Engineering, Industrial
G. Cabour, E. Ledoux, S. Bassetto
Summary: This paper uncovers and models the complexity of industrial inspection, providing a deep understanding of inspectors' cognition and actions. It offers descriptive empirical models to depict inspectors' meaning-making and decision-making processes and highlights the design implications for future human-automation work situations.
APPLIED ERGONOMICS
(2022)
Article
Multidisciplinary Sciences
Muhammad Rakeh Saleem, Robert Mayne, Rebecca Napolitano
Summary: This research aims to understand how an expert interacts with a structure during a facade inspection to improve future inspections using autonomous robots. Eye tracking is used to map the inspector's gaze and identify their attention. Experiments were conducted on a facade to analyze visually-based features important for human-infrastructure interaction. The collected data helped predict the inspector's intent and understand human behavior during inspections. This method will support collaborative decision-making and information-sharing in human-robot teams, enabling unmanned aerial vehicles (UAVs) to conduct building inspections with artificial intelligence.
SCIENTIFIC REPORTS
(2023)
Article
Computer Science, Artificial Intelligence
Saideep Nannapaneni, Sankaran Mahadevan, Abhishek Dubey, Yung-Tsun Tina Lee
Summary: This paper discusses the application of a predictive control framework in cyber-physical manufacturing processes, real-time monitoring of manufacturing process performance, and implementation of control to improve product quality. It also explores the impact of uncertainty sources on product quality predictions and how estimation and robust prediction can be achieved using a dynamic Bayesian network approach.
JOURNAL OF INTELLIGENT MANUFACTURING
(2021)
Review
Construction & Building Technology
Peixian Li, Yujie Lu, Da Yan, Jianzhuang Xiao, Huicang Wu
Summary: The paper reviews the state-of-the-art smart building research through bibliometric analysis, content analysis, and qualitative review. It reveals that smart building is an interdisciplinary field with high international collaboration, focusing on IoT, cloud computing, and human needs. The transition from CPS to HCPS framework in smart building research is detected, leading to future research directions centered around occupants' needs.
AUTOMATION IN CONSTRUCTION
(2021)
Article
Energy & Fuels
Xiangyu Kong, Yuce Sun, Muhammad Ahmad Khan, Lin Zheng, Junda Qin, Xiaotong Ji
Summary: This paper proposes a CPS planning method for hierarchical VPPs considering multidimensional uncertainties to provide reliable FR capacity. By establishing a reliable capacity model for VPPs and quantifying the impact of information-physical uncertainty, a CPS planning model for VPPs with the objective of minimizing investment and the constraints of normal capacity, reliable capacity, and delay is constructed.
Article
Computer Science, Software Engineering
Ferhat Ozgur Catak, Tao Yue, Shaukat Ali
Summary: This paper presents a method for addressing the uncertainty in deep learning applied to Cyber-Physical Systems (CPSs). The method quantifies uncertainty using prediction-time Dropout-based Neural Networks and utilizes a support vector machine to build a prediction validator model based on uncertainty metrics. Experimental results show a negative correlation between uncertainty quantification and prediction performance in CPS deep learning models.
ACM TRANSACTIONS ON SOFTWARE ENGINEERING AND METHODOLOGY
(2022)
Article
Computer Science, Information Systems
Abdul Razaque, Fathi Amsaad, Musbah Abdulgader, Bandar Alotaibi, Fawaz Alsolami, Duisen Gulsezim, Saraju P. Mohanty, Salim Hariri
Summary: CPS has seen rapid development in intelligent and connected cities, however, the issue of efficient mobility remains unresolved. In this study, we propose a privacy-aware secure human-centric mobility-aware (SHM) model and validate it through testing. The model strikes a balance between energy efficiency and QoS requirements and is compared to other models. The testing results demonstrate the compelling features of the secure CPS provided by the SHM model.
IEEE INTERNET OF THINGS JOURNAL
(2022)
Article
Horticulture
Diego Caballero-Ramirez, Yolanda Baez-Lopez, Jorge Limon-Romero, Guilherme Tortorella, Diego Tlapa
Summary: This study evaluates the suitability of human visual inspection and deep learning (DL) techniques for defect identification in the floriculture industry. The results indicate that DL models demonstrate a similar performance to humans in terms of precision and accuracy, highlighting their potential in assisting humans with defect identification, making the inspection process more intelligent and reliable.
Article
Engineering, Electrical & Electronic
Tao Jiang, Xiaotong Shi, Shuang Cheng, Yueyue Dai, Guanghua Liu
Summary: The ternary-integration of human-cyber-physical (HCP) has the potential to enable beneficial applications. Intelligent communication is the key technology to ensure a deep ternary integration. A human-cyber-physical ubiquitous intelligent communication system (HCP-UICS) is proposed to cope with multi-dimensional variability throughout the communication process. The potential key technologies of this system are analyzed and discussed, including cross-modal information pre-processing, intelligent transmission, and multi-dimensional assessment.
IEEE COMMUNICATIONS MAGAZINE
(2022)
Article
Construction & Building Technology
Pengkun Liu, Ying Shi, Ruoxin Xiong, Pinbo Tang
Summary: This paper investigates observation behaviors in bridge inspection to quantify the reliability of structural defect localizations. It uses a behavioral process graph to assess inspectors' performance and predict their inspection reliability index. The findings show that aggregating inspection records based on reliability indices effectively filters out false alarms while retaining reliable defect records.
DEVELOPMENTS IN THE BUILT ENVIRONMENT
(2023)
Article
Chemistry, Analytical
Nazakat Ali, Manzoor Hussain, Jang-Eui Hong
Summary: The System of Cyber-Physical Systems (SoCPS) consists of multiple independent Cyber-Physical Systems (CPSs) that collaborate to achieve a common mission. However, the complexity and safety-critical nature of SoCPS require a proper safety analysis mechanism. This paper introduces a composite safety analysis approach called SafeSoCPS, which analyzes potential hazards for the whole network of CPS and traces faults among the participating systems using a fault propagation graph. The results show that the SafeSoCPS approach can identify more faults in a network of SoCPS.
Review
Computer Science, Hardware & Architecture
Mohammad Kamrul Hasan, A. K. M. Ahasan Habib, Zarina Shukur, Fazil Ibrahim, Shayla Islam, Md Abdur Razzaque
Summary: The smart grid system is an intelligent technology that integrates green technology and environmental aspects through a two-way communication system for information transformation, power generation, and distribution. The development and application of communication technology in the traditional power system have led to complex architecture and cyber-attacks for smart grid systems. Therefore, research on smart grid cyber-physical and cyber security systems, standard protocols, and challenges is crucial to protect against these threats and ensure the system's security.
JOURNAL OF NETWORK AND COMPUTER APPLICATIONS
(2023)
Article
Automation & Control Systems
Xirong Ning, Jin Jiang
Summary: This article presents a modular experimental platform to support cyber security assessment and enhancement of cyber-physical systems. The platform is designed with a hardware-based prototype and provides effective methods to identify vulnerabilities and evaluate different security enhancement strategies.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2022)
Article
Engineering, Electrical & Electronic
Hongjun Gao, Xiaodong Lyu, Shuaijia He, Lingfeng Wang, Cheng Wang, Junyong Liu
Summary: This paper proposes a bilevel integrated planning model to address the challenges posed by multidimensional uncertainties from both the cyber space and physical space in a cyber-physical active distribution system (CPADS). The model comprehensively considers factors such as location and selection of active-management elements, investment constraints, power flow constraints, and operation constraints. A novel uncertainty set is proposed to characterize cyber system failure, while box-like uncertainty sets are used for renewable energy and load demand. The proposed planning model is transformed into a multi-stage robust optimization model and solved using the column and constraint generation algorithm.
IEEE TRANSACTIONS ON SMART GRID
(2022)
Article
Automation & Control Systems
Jitao Li, Zhenhua Wang, Yi Shen, Lihua Xie
Summary: This article discusses the security issues of cyber-physical systems against stealthy attacks and proposes a vulnerability characterization method based on zonotopic model. By quantifying the potential impact and measuring the level of vulnerabilities, the effectiveness of the proposed method is demonstrated through numerical simulations.
INTERNATIONAL JOURNAL OF ROBUST AND NONLINEAR CONTROL
(2022)
Article
Environmental Sciences
Qian Wang, Jingjing Guo, Min-Koo Kim
Article
Construction & Building Technology
Jingjing Guo, Liang Yuan, Qian Wang
AUTOMATION IN CONSTRUCTION
(2020)
Article
Construction & Building Technology
Liang Yuan, Jingjing Guo, Qian Wang
AUTOMATION IN CONSTRUCTION
(2020)
Article
Construction & Building Technology
Jingjing Guo, Qian Wang, Jae-Hyuk Park
AUTOMATION IN CONSTRUCTION
(2020)
Article
Computer Science, Interdisciplinary Applications
Jingjing Guo, Qian Wang, Yiting Li, Pengkun Liu
COMPUTER-AIDED CIVIL AND INFRASTRUCTURE ENGINEERING
(2020)
Article
Computer Science, Interdisciplinary Applications
Jingjing Guo, Qian Wang, Yiting Li
Summary: The study introduces a semi-supervised learning algorithm that trains an accurate classifier using only a small amount of labeled data, and improves classification accuracy by selecting reliable unlabeled data with an uncertainty filter. Validation experiments show that the proposed method significantly enhances model accuracy compared to traditional supervised learning algorithms, while also reducing the time and cost for preparing labeled data.
COMPUTER-AIDED CIVIL AND INFRASTRUCTURE ENGINEERING
(2021)
Article
Construction & Building Technology
Pengkun Liu, Hung-Lin Chi, Xiao Li, Jingjing Guo
Summary: This study explores multi-source datasets and corresponding data acquisition methods suitable for fatigue detection among crane operators, providing guidelines for data collection. Analysis of variations in public datasets, as well as proposing a hybrid learning architecture for fatigue detection, were also conducted.
AUTOMATION IN CONSTRUCTION
(2021)
Article
Construction & Building Technology
Jingjing Guo, Qian Wang, Yiting Li
Summary: The study proposes a rule-based deep learning method for evaluation-oriented facade defects detection, which improves the performance of detection to meet the requirement of condition evaluation. The method is adaptable to various distance settings in the requirement.
AUTOMATION IN CONSTRUCTION
(2021)
Article
Computer Science, Interdisciplinary Applications
Jiajun Li, Qian Wang, Jun Ma, Jingjing Guo
Summary: Facade defects have significant impact on building owners, and traditional manual inspection methods are inefficient and unsafe. This study proposes a balanced copy-paste method using the Mask R-CNN model to automatically detect and segment facade defects, achieving high accuracy and computational efficiency.
COMPUTER-AIDED CIVIL AND INFRASTRUCTURE ENGINEERING
(2022)
Article
Construction & Building Technology
Qi Qiu, Mengjun Wang, Jingjing Guo, Ziwen Liu, Qian Wang
Summary: This study proposes an adaptive down-sampling method that can effectively preserve the geometric and semantic information of existing buildings, achieving better performance.
AUTOMATION IN CONSTRUCTION
(2022)
Article
Engineering, Industrial
Xiaoying Tang, Mengjun Wang, Qian Wang, Jingjing Guo, Jingxiao Zhang
Summary: This study provides a quantitative analysis of the time and cost benefits of TLS-based geometric QA/QC compared to the conventional approach. The results show that the TLS-based QA/QC approach is more efficient and cost-effective, and provide recommendations for both single projects and companies in the long term.
JOURNAL OF MANAGEMENT IN ENGINEERING
(2022)
Article
Construction & Building Technology
Zhiyan Cui, Qian Wang, Jingjing Guo, Na Lu
Summary: This study proposes a few-shot classification method based on an extensible classifier and contrastive learning to address the problem of insufficient data and the extension constraint of classifiers. Experimental results demonstrate that the proposed method significantly improves the few-shot classification accuracy on novel and unseen classes with limited samples, while maintaining a high accuracy on base classes.
AUTOMATION IN CONSTRUCTION
(2022)
Proceedings Paper
Computer Science, Interdisciplinary Applications
Jingjing Guo, Qian Wang
COMPUTING IN CIVIL ENGINEERING 2019: DATA, SENSING, AND ANALYTICS
(2019)