Article
Computer Science, Artificial Intelligence
Zdzislaw Kowalczuk, Michal Czubenko
Summary: Concepts based on psychology can be applied to robotics and artificial intelligence research. The article proposes an operational cybernetic model developed with psychological knowledge to build an autonomous robot system with decision-making capabilities. The model still requires further validation and improvement.
ARTIFICIAL INTELLIGENCE REVIEW
(2023)
Article
Engineering, Civil
Kai Yang, Boqi Li, Wenbo Shao, Xiaolin Tang, Xiaochuan Liu, Hong Wang
Summary: This paper proposes a risk-aware decision-making (RADM) framework to handle the epistemic uncertainty of prediction models due to insufficient training data. The framework utilizes a multi-agent prediction network with uncertainty quantification and model predictive control technique to process the prediction results and consider the uncertainty. Experimental results demonstrate that RADM can reduce driving risk and improve safety.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2023)
Article
Business
Shavneet Sharma, Nazrul Islam, Gurmeet Singh, Amandeep Dhir
Summary: This article investigates customers' adoption of AI-based autonomous decision-making processes by analyzing 454 customer responses. The results reveal that effort expectancy, performance expectancy, facilitating conditions, and social influence are positively associated with customers' adoption of autonomous decision-making processes. Collectivism strengthened the positive association of social influence with customer attitude, whereas uncertainty avoidance dampened the associations of other factors with attitude.
IEEE TRANSACTIONS ON ENGINEERING MANAGEMENT
(2022)
Article
Psychology, Multidisciplinary
Alessandra Talamo, Silvia Marocco, Chiara Tricol
Summary: The application of artificial intelligence in the financial field is creating a new area of study called financial intelligence, aimed at assisting in complex decision-making processes. This is particularly crucial for venture capitalist organizations where different actors with varying decision-making behaviors are involved. This study proposes a modeling approach for financial AI-based services and suggests the integration of human/AI systems for better decision-making support.
FRONTIERS IN PSYCHOLOGY
(2021)
Article
Multidisciplinary Sciences
Mustafa Karatas, Keisha M. Cutright
Summary: Research shows that when people think about God, they are more likely to accept recommendations based on Artificial Intelligence (AI). Eight preregistered experiments involving 2,462 participants demonstrate that people are more willing to consider AI-based recommendations when God is salient. This effect is observed across various contexts and is driven by a heightened feeling of smallness and recognition of human fallibility when God is salient.
PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA
(2023)
Article
Psychology, Multidisciplinary
Cindy Candrian, Anne Scherer
Summary: Delegation is crucial for organizational success, as it allows individuals to overcome personal shortcomings and draw upon the expertise of others. The emergence of AI as an option for delegation brings new opportunities and challenges. Research shows that people prefer to delegate decisions to AI, especially when the decisions involve losses.
COMPUTERS IN HUMAN BEHAVIOR
(2022)
Editorial Material
Computer Science, Artificial Intelligence
Yang Tang, Gary Yen, Juergen Kurths
Summary: This special section of papers focuses on the use of computational intelligence for perception and decision making in autonomous systems. Autonomous systems have shown great potential in efficiently completing complex tasks that humans cannot achieve, thanks to their powerful capabilities in environmental perception, real-time computing, and intelligent decision-making. However, traditional algorithms may struggle with the high-dimensional, heterogeneous, unstructured, and unpredictable data from different sensors. Recent advancements in computational intelligence algorithms, such as deep neural networks and evolutionary algorithms, have the ability to extract useful information from diverse data sources and have been successfully applied in computer vision and natural language processing. Therefore, integrating advanced computational intelligence algorithms with autonomous systems is promising for achieving high-level environmental perception and decision-making.
IEEE TRANSACTIONS ON EMERGING TOPICS IN COMPUTATIONAL INTELLIGENCE
(2022)
Article
Automation & Control Systems
Yingxu Wang, Ming Hou, Konstantinos N. Plataniotis, Sam Kwong, Henry Leung, Edward Tunstel, Imre J. Rudas, Ljiljana Trajkovic
Summary: This paper explores the intelligent and mathematical foundations of autonomous systems, focusing on the structural and behavioral properties that constitute their intelligent power, as well as the evolution of system intelligence from reflexive to cognitive intelligence.
IEEE-CAA JOURNAL OF AUTOMATICA SINICA
(2021)
Article
Ethics
Jakob Mokander, Jessica Morley, Mariarosaria Taddeo, Luciano Floridi
Summary: The article examines the feasibility and efficacy of ethics-based auditing (EBA) as a governance mechanism to validate claims made about ADMS. EBA can promote procedural regularity and transparency, with seven criteria proposed for successful design and implementation of EBA procedures. The article also identifies and discusses the constraints associated with EBA in various aspects.
SCIENCE AND ENGINEERING ETHICS
(2021)
Review
Computer Science, Information Systems
Anindya Pradipta Susanto, David Lyell, Bambang Widyantoro, Shlomo Berkovsky, Farah Magrabi
Summary: This study aims to summarize the research literature evaluating machine learning (ML)-based clinical decision support (CDS) systems in healthcare settings. The findings suggest that ML-based CDS systems have been applied successfully in assisting clinical tasks, but their effects on decision-making, care delivery, and patient outcomes are mixed.
JOURNAL OF THE AMERICAN MEDICAL INFORMATICS ASSOCIATION
(2023)
Article
Engineering, Civil
Yuchao Feng, Wei Hua, Yuxiang Sun
Summary: In recent years, the advancement of deep-learning technologies has greatly promoted the research progress of autonomous driving. However, the lack of explainability in deep neural networks makes it unsafe to deploy them in unseen or unexpected environments. To address this issue, we propose a deep neural network that predicts decision-making actions along with natural-language explanations based on semantic scene understanding. We also provide a large-scale dataset with hand-labelled ground truth for evaluation.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2023)
Article
Chemistry, Multidisciplinary
Turgay Ertekin
Summary: In the process of making critical decisions in reservoir engineering, intuition alone is not enough and missing knowledge and understanding must be generated to make informed decisions based on facts. Fast algorithmic protocols are necessary to ensure a wide search domain to generate desired solutions.
APPLIED SCIENCES-BASEL
(2021)
Article
Business
Yash Raj Shrestha, Vaibhav Krishna, Georg von Krogh
Summary: The growing field of artificial intelligence in management studies has reinvigorated research on decision-making in organizations, particularly with the use of deep learning algorithms. The concept of deep learning-augmented decision-making (DLADM) is illustrated through two case studies involving image recognition and sentiment analysis tasks. Challenges and recommendations for managers in addressing these challenges are also discussed.
JOURNAL OF BUSINESS RESEARCH
(2021)
Article
Information Science & Library Science
Omar A. Nasseef, Abdullah M. Baabdullah, Ali Abdallah Alalwan, Banita Lal, Yogesh K. Dwivedi
Summary: This research investigates the effects of using an artificial intelligence-driven public healthcare framework to enhance the decision-making process in healthcare organizations in Saudi Arabia. The study validates the extended model of Shaft and Vessey's cognitive fit model and finds that it has a significant impact on improving decision-making related to COVID-19. The research contributes to expanding the theoretical horizon of cognitive fit models and offers insights for future research directions.
GOVERNMENT INFORMATION QUARTERLY
(2022)
Article
Energy & Fuels
Eric Cayeux, Benoit Daireaux, Adrian Ambrus, Rodica Mihai, Liv Carlsen
Summary: The complex drilling process is prone to vibrations and lacks uniform monitoring due to the long and slender system. To achieve autonomous drilling operations, a method capable of estimating internal states and making safe decisions is necessary. The developed solution optimizes the time to reach total depth, adapting to various drilling conditions for effective and cautious operations.
Article
Chemistry, Analytical
Michal Czubenko, Zdzislaw Kowalczuk
Summary: The article discusses the practical aspect of collision detection with the use of a simple neural architecture in collaborative robots. Tested on the CURA6 robot prototype, the MC-LSTM architecture was found to be the most effective for collision detection with a regression level set at 12 samples. This virtual sensor based on a neural network can be used to detect various types of collisions in cobots or other systems operating based on human-machine interaction.
Article
Energy & Fuels
Zdzislaw Kowalczuk, Marek Sylwester Tatara
Summary: This article discusses the optimal discretization grid for simulating fluid flow through a pipeline. By analyzing the relationship between the numerically set coefficient mu(opt) and the physical and technological parameters of flow mechanics, the optimal Courant number for pipeline flow is determined. The study also examines four cases for determining the coefficient mu(opt) and emphasizes the importance of considering physical laws and numerical methods for accurate model simulation.
JOURNAL OF NATURAL GAS SCIENCE AND ENGINEERING
(2021)
Article
Mathematics, Interdisciplinary Applications
Piotr Kotuszewski, Krzysztof Kukielka, Pawel Kluk, Andrzej Ordys, Karol Bienkowski, Jan Maciej Koscielny, Michal Syfert, Pawel Wnuk, Jakub Mozaryn, Bartlomiej Fajdek
Summary: This laboratory set-up is designed to test the concept and components of Industry 4.0 enabled mechatronic system, including a mock manufacturing lane connected to a collaborating robot and various PLCs. Information exchange through remote programming and cloud service allows for resilience testing of the control structure.
Article
Energy & Fuels
Anna Sibilska-Mroziewicz, Andrzej Ordys, Jakub Mozaryn, Pooyan Alinaghi Hosseinabadi, Ali Soltani Sharif Abadi, Hemanshu Pota
Summary: This article introduces two controllers, LQR and FLC, for a three-area power system to compare their performance in power stabilization. The model and control strategies are tested in MATLAB software under various simulation scenarios, considering disturbances and faulty tie-lines between areas. The comparison leads to proposed metrics and advice on the application of each controller.
Article
Mechanics
Anna Sibilska-Mroziewicz, Jakub Mozaryn, Ayesha Hameed, Maria Molina Fernandez, Andrzej Ordys
Summary: This paper presents a simulation and evaluation framework for the controller design of snake robots, as well as the application of mechanical design and control algorithms. The framework allows for testing new solutions and strategies, as well as analyzing position tracking and fault-tolerant control of the robot.
MULTIBODY SYSTEM DYNAMICS
(2022)
Article
Computer Science, Artificial Intelligence
Zdzislaw Kowalczuk, Michal Czubenko
Summary: Concepts based on psychology can be applied to robotics and artificial intelligence research. The article proposes an operational cybernetic model developed with psychological knowledge to build an autonomous robot system with decision-making capabilities. The model still requires further validation and improvement.
ARTIFICIAL INTELLIGENCE REVIEW
(2023)
Article
Computer Science, Artificial Intelligence
Zdzislaw Kowalczuk, Michal Czubenko, Weronika Zmuda-Trzebiatowska
Summary: The aim of this article is to present a neural system based on stock architecture for recognizing emotional behavior in dogs. The system was developed by compiling a set of photographic data representing five classes of emotional behavior in dogs of one breed. The system was compared with other systems and showed promising results, with VGG16 and VGG19 identified as the most suitable backbone networks.
COMPUTATIONAL INTELLIGENCE
(2022)
Article
Energy & Fuels
Jakub Filip Mozaryn, Michal Fratczak, Krzysztof Stebel, Tomasz Klopot, Witold Nocon, Andrzej Ordys, Stepan Ozana
Summary: This paper focuses on studying the workflow of a detection centre for detecting stealthy attacks on industrial installations that cause an increase in energy consumption. These long-lasting, undetected attacks affect the competitiveness and long-term integrity of the industrial facilities. The authors propose a remote detection system that analyzes monitored signals from a PLC-controlled installation, identifying discrepancies using Control Performance Assessment indices. Experimental results support the effectiveness of their approach.
Review
Chemistry, Multidisciplinary
Ayesha Hameed, Andrzej Ordys, Jakub Mozaryn, Anna Sibilska-Mroziewicz
Summary: Collaborative robots assist humans in various fields by undertaking tasks ranging from simple to complex. They are essential in the Industry 4.0 revolution, which sets new manufacturing standards and product organization. While incorporating collaborative robots in the workspace improves efficiency, safety measures are crucial to ensure safe and robust interactions.
APPLIED SCIENCES-BASEL
(2023)
Article
Chemistry, Analytical
Anna Sibilska-Mroziewicz, Ayesha Hameed, Jakub Mozaryn, Andrzej Ordys, Krzysztof Sibilski
Summary: This article presents a new approach to engineering simulation in an interactive environment. A synesthetic design approach is used to gather information and facilitate interaction with the simulated system. The simulation of a snake robot moving on a flat surface is realized in dedicated engineering software, which exchanges information with 3D visualization software and a VR headset. Several scenarios comparing different visualization methods are presented, illustrating how immersive experiences in VR can facilitate system analysis and design in the engineering context.
Article
Energy & Fuels
Michal Syfert, Andrzej Ordys, Jan Maciej Koscielny, Pawel Wnuk, Jakub Mozaryn, Krzysztof Kukielka
Summary: This paper focuses on the diagnostics of process faults and the detection of cyber-attacks in industrial control systems. It proposes merging these two into one comprehensive anomaly detection system and presents test results to support the approach.
Proceedings Paper
Automation & Control Systems
Jan Koscielny, Michal Syfert, Andrzej Ordys, Pawel Wnuk, Jakub Mozaryn, Bartlomiej Fajdek, Vicenc Puig, Krzysztof Kukielka
Summary: This paper explores enhancing cyber-attack detection by leveraging information and methods related to fault detection. Through an experimental platform and simulator, combined cyber attacks and faults in industrial processes were tested to potentially differentiate between them. The study presented, analyzed, and successfully detected cyber attacks using this method.
2021 EUROPEAN CONTROL CONFERENCE (ECC)
(2021)