Cyber-physical Systems Research Articles

A common problem in the development of Internet-of-Things (IoT) and cyber–physical system (CPS) applications is the complexity of these domains, due to their hybrid and distributed nature at multiple layers (hardware, network, communication, frameworks, etc.). This complexity often leads to implementation errors, some of which result in undesired states of the application and/or the system. The current work focuses on low-code development of behavior verification processes for IoT and CPS applications, in order to raise productivity, minimize risks (due to errors) and enable access to a wider range of end-users to create and verify applications for state-of-the-art domains, such as smart home and smart industry. Model-Driven Development (MDD) approaches are employed for the implementation of a Domain-Specific Language (DSL) that enables the evaluation of IoT and CPS applications, among others. The proposed methodology automates the development of behavior verification processes, allowing domain experts to focus on the real problem, instead of struggling with technical and technological breaches. Through comparative scenario-based analysis and 43 detailed use cases, we illustrate how the proposed methodology automates the development of behavior verification processes, allowing end-users to focus on the verification definition, instead of technical and technological intricacies.

In industrial process installations, the improper operation or misconfiguration of safety-critical components, such as manually operated ball valves, can seriously compromise both process performance and plant safety. This work proposes a sensorless Edge AI method to estimate hand-operated ball valves states without the use of physical position sensors. Using multivariate time-series data collected from a PLC-based pilot plant, a benchmark evaluation is conducted comparing four Deep Learning (DL) and four classical Machine Learning (ML) models for classification and regression tasks. The models are deployed on an embedded platform, enabling real-time inference at the edge with a minimum latency of 500ms. Results show Decision Tree (DT) and Random Forest (RF) achieve high regression accuracy (R2 >0.98, MAE < 0.5), while all eight model reach high classification accuracy. Additionally, the computational efficiency metric that combines model accuracy, latency, and size, confirming DT as the most efficient model (1.83/(ms.KB) for edge deployment. This work contributes a cost-effective and scalable monitoring strategy, particularly suitable for complex industrial environments where physical sensing and visual inspection are limited, offering a viable path toward early anomaly detection and intelligent supervision within cyber-physical systems. • Key Words: Edge AI, machine learning, deep neural networks, cyber-physical systems, industrial valves, PLC, embedded inference.

Joint Estimation of Deception Attacks on Sensors and Actuators in Cyber–Physical Systems

Guest Editorial Special Issue on Monitoring and Control in Cyber-Physical Systems: Security, Resilience, and Privacy

An innovative framework for driving behavior Analysis: Privacy-Preserving, explainable, and adaptive Artificial Intelligence in cyber-physical systems

Sequential recovery of cyber-physical power systems considering cyber-attacks

Enhanced knowledge graph cascade learning model for cyber–physical systems

Fault localization of AI-enabled cyber–physical systems by exploiting temporal neuron activation

Design of Switching-Type Zonotopic Unknown Input Observers for Attack Reconstruction and Detection in Uncertain Cyber–Physical Systems

Switching time-/event-triggered control of cyber–physical systems under hybrid attacks and finite bit rates

Edge-AI empowered Cyber-Physical Systems: A comprehensive review on performance analysis

A Traffic Normalization Location Attention Network for cyber attack detection in Industrial Cyber-Physical Systems

A Model Predictive Control Strategy Under Partial State Availability for Resilience and Maintenance Operations of Cyber-Physical Systems

This paper presents a systematic methodology for identifying and integrating safety and security requirements in autonomous driving systems, demonstrated through the case of an autonomous intersection. The study focuses on modeling the intelligent intersection using the MBSE Grid Framework, the SysML modeling language, and the Cameo Systems Modeler tool. Two specific use cases are modeled to illustrate the system’s functionality. A multidisciplinary approach is developed to incorporate safety and security requirements into the system model, combining theoretical foundations with practical implementation techniques. The methodology includes both a generalizable framework and domain-specific strategies tailored to autonomous driving. The proposed approach is applied and critically evaluated using the intelligent intersection as a case study. By extending SysML to systematically address safety and security concerns, the work contributes to the development of safer and more efficient autonomous transportation systems. The results provide a foundation for future research and practical applications in the field of intelligent mobility and cyber–physical systems.

The U.S. agricultural sector is undergoing a paradigm shift driven by the convergence of smart agriculture practices and Industry 4.0 technologies. Rising demands for food security, sustainability, and resource efficiency are compelling stakeholders to adopt advanced tools that integrate data-driven decision-making with traditional agricultural management. This paper explores how industrial engineering tools, such as process optimization, lean methodologies, predictive analytics, and systems modeling, can be combined with smart agriculture and Industry 4.0 technologies to significantly improve agricultural productivity in the United States. Key enabling technologies include the Internet of Things (IoT), robotics, artificial intelligence (AI), big data analytics, blockchain, and cyber-physical systems, which collectively allow for real-time monitoring, precision farming, predictive maintenance of agricultural machinery, and supply chain optimization. By applying industrial engineering methods such as value stream mapping, simulation modeling, and queuing theory, agricultural operations can be systematically streamlined to minimize waste, reduce downtime, and optimize input usage (e.g., water, fertilizer, energy). Case studies and simulation results presented in this paper demonstrate that integrating Industry 4.0 frameworks with industrial engineering tools in U.S. farms can increase crop yields by up to 18%, reduce resource wastage by 25%, and enhance overall operational efficiency by 20%. Furthermore, the adoption of smart agriculture practices supported by data-driven MIS (Management Information Systems) can improve resilience to climate variability and labor shortages. While challenges remain in terms of high upfront costs, interoperability of digital platforms, and farmer training, the proposed framework offers a structured roadmap for modernizing U.S. agriculture and enhancing food security. The findings contribute to the growing body of knowledge on agricultural digital transformation and highlight the critical role of industrial engineering tools in accelerating smart agriculture adoption.

The growing demand for sustainable, efficient, and resilient water treatment systems has driven increasing interest in the integration of Industry 4.0 technologies. This review explores the practical applications of key digital innovations including the Internet of Things (IoT), artificial intelligence (AI) and machine learning, big data analytics, and automation with cyber-physical systems in modern water treatment. These technologies enable real-time monitoring, predictive maintenance, process optimization, and data-driven decision-making, transforming conventional facilities into adaptive, smart systems. A literature search was conducted across peer-reviewed publications and technical reports from 2015 to 2025, with data extracted on study areas, methodologies, outcomes, and practical implications. The analysis highlights successful case applications in water quality monitoring, wastewater treatment, and infrastructure management, while also identifying challenges related to cost, interoperability, and regulatory frameworks. Future perspectives emphasize the need for low-cost and scalable solutions, seamless integration with existing infrastructure, supportive policies, and collaborative partnerships across research, industry, and governance. Emerging opportunities include the convergence of Industry 4.0 with advanced biosensors, blockchain, and autonomous robotics, paving the way for fully automated and self-optimizing treatment plants.

In this article, we investigate fault-tolerant and resilient control approaches for cyber-physical systems within a unified control and detection framework. Particularly, a novel strategy is presented to simultaneously detect and accommodate anomalies in cyber-physical systems subject to multiplicative physical faults and additive integrity cyberattacks. An observer-based cyber-secure system configuration is first analyzed by means of the coprime factorization technique, wherein multiplicative faults are characterized by coprime factor uncertainties. It is revealed that fault- and cyberattack-induced variations possess distinct attributes with respect to the closed-loop dynamics. This observation motivates a collaborative detection scheme to distinguish both types of anomalies. Specifically, a performance-based fault detector is implemented on the plant side, delivering fault detection results to the monitoring and control (MC) side, where an observer-based attack detector operates collaboratively. Subsequently, the local and remote controllers are reconfigured to enhance the fault tolerance and attack resilience against faults and cyberattacks. To provide more independent design freedoms, the residual signal derived from the controller dynamics is incorporated into the Youla parameterization-based stabilizing controller. Finally, the proposed scheme is verified on a leader-follower robot system.

The object of the study is the processes of formation, transmission and processing of service and useful traffic in cyber-physical systems of Smart Manufacturing Ecosystem multi-level architecture type, vulnerable to cyberattacks aimed at compromising control data, authentication and coordination. In modern computer networks, service traffic determines the stability and security of the infrastructure, since any distortion or interception of service traffic can lead to disruption of the system as a whole. In smart systems, industrial Internet of Things and critical infrastructure, the volume of service messages reaches significant scales, because it is they that support the synchronism of thousands of systems in real time. The paper investigates the problem of protecting service traffic in Smart Manufacturing Ecosystem cyber-physical systems. A mathematical model of service and useful traffic segmentation is proposed, which takes into account the criteria of stability (access segmentation, integrity and authenticity control) and security (probability of compromise, channel criticality, level of trust in the transmission medium). To construct an integral risk indicator, the convolution method is used, which allows combining different types of parameters and determining the feasibility of dividing traffic for target analysis. The study was conducted using industrial protocols Modbus, DNP3, OPC UA, MQTT and HTTP, which are widely used in production networks. It was shown that the use of the model allows reducing the integral risk of attacks on service traffic by an average of 15–20% compared to approaches without segmentation. The developed model forms a scientific basis for creating methods and practical cyber protection solutions that ensure increased resilience of the Smart Manufacturing infrastructure and are able to withstand current and future challenges in the field of cybersecurity

The integration of autonomous robots with intelligent electrical systems introduces complex energy management challenges, particularly as microgrids increasingly incorporate renewable energy sources and storage devices in widely distributed environments. This study proposes a quantum-inspired multi-agent reinforcement learning (QI-MARL) framework for energy-aware swarm coordination in smart microgrids. Each robot functions as an intelligent agent capable of performing multiple tasks within dynamic domestic and industrial environments while optimizing energy utilization. The quantum-inspired mechanism enhances adaptability by enabling probabilistic decision-making, allowing both robots and microgrid nodes to self-organize based on task demands, battery states, and real-time energy availability. Comparative experiments across 1500 grid-based simulated environments demonstrated that when benchmarked against the classical MARL baseline, QI-MARL achieved an 8% improvement in path efficiency, a 12% increase in task success rate, and a 15% reduction in energy consumption. When compared with the rule-based approach, improvements reached 15%, 20%, and 26%, respectively. Ablation studies further confirmed the substantial contributions of the quantum-inspired exploration and energy-sharing mechanisms, while sensitivity and scalability analyses validated the system’s robustness across varying swarm sizes and environmental complexities. The proposed framework effectively integrates quantum-inspired AI, intelligent microgrid management, and autonomous robotics, offering a novel approach to energy coordination in cyber-physical systems. Potential applications include smart buildings, industrial campuses, and distributed renewable energy networks, where the system enables flexible, resilient, and energy-efficient robotic operations within modern electrical engineering contexts.

As Machine Learning (ML) becomes integral to Cyber-Physical Systems (CPS), there is growing interest in shifting training from traditional cloud-based to on-device processing (TinyML), for example, due to privacy and latency concerns. However, CPS often comprise ultra-low-power microcontrollers, whose limited compute resources make training challenging. This paper presents RockNet , a new TinyML method tailored for ultra-low-power hardware that achieves state-of-the-art accuracy in timeseries classification, such as fault or malware detection, without requiring offline pretraining. By leveraging that CPS consist of multiple devices, we design a distributed learning method that integrates ML and wireless communication. RockNet leverages all devices for distributed training of specialized compute efficient classifiers that need minimal communication overhead for parallelization. Combined with tailored and efficient wireless multi-hop communication protocols, our approach overcomes the communication bottleneck that often occurs in distributed learning. Hardware experiments on a testbed with 20 ultra-low-power devices demonstrate RockNet ’s effectiveness. It successfully learns timeseries classification tasks from scratch, surpassing the accuracy of the latest approach for neural network microcontroller training by up to 2x. RockNet ’s distributed ML architecture reduces memory, latency and energy consumption per device by up to 90 % when scaling from one central device to 20 devices. Our results show that a tight integration of distributed ML, distributed computing, and communication enables, for the first time, training on ultra-low-power hardware with state-of-the-art accuracy.