The study proposes a new methodology based on time-frequency analysis for the indirect monitoring of the dressing operation of conventional grinding wheels. Through a low-cost piezoelectric diaphragm (PZT), acoustic signals are captured during the process. The analysis is based on the coefficient of variation of the Short-Time Fourier Transform (STFT). The results indicate that the signal instability is high in the first passes but progressively decreases, reaching stability between passes 10 and 15. This suggests that the surface of the grinding wheel is regularized and ready for grinding. The methodology can serve as an objective indicator to assist the operator in interrupting the dressing process at the optimal moment, thereby optimizing grinding quality and reducing operational costs.

The construction industry is increasingly adopting digital technologies to enhance productivity and efficiency, in alignment with the principles of Construction 4.0 (C4). The progress and advances recorded thus far are largely due to advancements in cyber-physical systems (CPS), computational processing power, deep learning solutions, robotics, and other related technologies. However, a major challenge in this research space is the lack of an integrated solution for both the interior and exterior construction environments, which has led to fragmented data, hindering efficiency. Several researchers have proposed frameworks in recent years that focused on either indoor or outdoor construction environments; this approach has resulted in the creation of siloed information, to the detriment of the C4 ideals and principles. In this study, a comprehensive system architecture for raw data captured using sensors and other inputs to provide useful insight for the construction team and stakeholders was mapped out. This study presents an integrated framework of various technologies for both indoor and outdoor construction environments. The solution provided for localisation algorithms and technologies such as Simultaneous Localisation and Mapping (SLAM), odometry, and inertial measurement unit (IMU) devices. The unified 5-level Cyber-Physical Systems (CPS) architecture was used as the primary architecture, and it was compared with the IoT Architecture layers in terms of data analytics and management perspectives. The Digital Twin (DT), which sits at the cyber level of the architecture, warehouses and tracks in real-time the dynamic complexities of the construction site throughout the project life cycle, serving as the single source of truth for the project. This system architecture and framework presented in this research contributed towards advancing the field of construction automation by offering a scalable solution for efficient construction in project management.

With the increasing demand for the safety monitoring of offshore engineering structures, the traditional single-modality sensing and centralized data-processing models face challenges such as insufficient real-time performance and weak anti-interference ability in complex marine environments. This research proposes an intelligent monitoring system based on multimodal sensor fusion and edge computing, aiming to achieve high-precision real-time diagnosis of offshore structure damage. The research plans to construct multimodal sensors through sensors such as stress change sensors, vibration sensors, ultrasonic sensors, and fiber Bragg grating sensors. A distributed wireless sensor network will be adopted to realize the transmission of sensor data, re-duce the complexity of wiring, and meet the requirements of high humidity and strong corrosion in the marine environment. At the edge computing layer, lightweight deep-learning models (such as multi-branch Transformer) and D-S evidence theory fusion algorithms will be deployed to achieve real-time feature extraction of multi-source data and damage feature fusion, supporting the intelligent identification of typical damages such as cracks, corrosion, and deformation. Experiments will simulate the coupled working conditions of wave impact, seismic load, and corrosion to verify the real-time performance and accuracy of the system. The expected results can provide a low-latency and highly robust edge-intelligent solution for the health monitoring of offshore engineering structures and promote the deep integration of sensor networks and artificial intelligence in Industry 4.0 scenarios.

There are various non-destructive techniques for determining the internal properties of materials in fluids, semi-solids, solids, and biological tissue. One of these techniques is low intensity ultrasonic testing. In this proceeding, a study on the architecture of a piezoelectric acoustic detector (PAD) is presented, from which the analysis for design, development and construction of the acoustic wave detector in the ultrasonic spectrum has emerged. Its aim is to apply it to soft matter and tissue. The 110 μm thick polyvinylidene fluoride (PVDF) piezoelectric element was used as the active element in the thickness mode configuration. Piezoelectric constitutive equations were applied to a one-dimensional model for the analysis. A cylindrical iron-nickel backing was used, and the parts were bonded with silver conductive epoxy glue. The results are presented. The equation for the output voltage of the piezoelectric acoustic detector is described. Functional testing of the PAD is demonstrated using the pulse-echo technique, in which an acoustic wave generator excited an ultrasonic immersion sensor in emission configuration and the DAP was connected to a digital oscilloscope to observe the received signal. Finally, pulsed photoacoustic spectroscopy was applied to a biological tissue emulator and yielded significant results in detection of a ruby sphere embedded in the emulator. It is proposed to further investigation the DAP models in multilayer structural configurations to increase their sensitivity.

Background: Detection is an important component of precision agriculture because accurate weed identification and treatment have a direct impact on crop yield and resource efficiency. Recent breakthroughs in artificial intelligence (AI) have enabled automatic weed recognition systems; nevertheless, standard centralised machine learning models present substantial obstacles such as high communication overhead, privacy issues, and limited scalability in remote farming contexts. To address these restrictions, federated edge learning (FEL) combined with deep learning and multimodal sensor fusion provides a viable solution by allowing for distributed model training while maintaining data privacy. Objective: In this work, our goal is to develop a privacy-preserving distributed weed detection and management system. The proposed work is integrated with FEL (Federated Learning), Deep learning with multi-modal sensor fusion to enhance the performance of the model and simultaneously minimize the data transfer, latency, and energy consumption. Materials and Methods: In this study, we used Multimodal sensors, such as LiDAR (Light Detection and Ranging), RGB (Red-Green-Blue) cameras, multispectral imaging devices, and soil moisture sensors placed in controlled agricultural plots. Each modality gave complementary information for weed identification: RGB provided texture and colour cues, multispectral collected spectral reflectance patterns, LiDAR delivered structural depth information, and soil sensors supported contextual environmental conditions. For robustness, three sensor fusion techniques were used: Early Fusion (feature-level concatenation), Mid Fusion (intermediate feature aggregation), and Late Fusion (decision-level integration). Deep learning models, such as Convolutional Neural Networks (CNNs), LSTM-CNN hybrids, and Vision Transformers, were trained using standardised parameters. A proposed Federated CNN (FedCNN) was deployed across multiple edge devices, each locally trained on sensor data without exchanging raw data, using FedAvg and FedProx algorithms. Validation was performed using a stratified 80/20 train-test split combined with 5-fold cross-validation to ensure model generalization. Model performance was assessed using accuracy, precision, recall, F1-score, AUC, latency, and energy consumption, enabling a holistic evaluation of both predictive quality and computational efficiency. The DL models, including CNNs, LSTM-CNN hybrids, and Vision Transformers, are used. A FedCNN model is distributed across many edge nodes, allowing for decentralized training without exchanging raw data. For model performance measures, we used different metrics like accuracy, precision, recall, F1-score, AUC, latency, and energy. Result: The experimental work reveals that the model FedCNN performs well in comparison to other models and achieved the highest accuracy of 94.1%, precision is 94.3%, recall is 93.9% and F1-score is 94.1%, AUC is 94.1% during hybrid fusion strategies. We compared the centralized and federated learning performance. The FEL (Edge) accuracy is 94.1%, the Latency is 120 ms, the energy consumption is 300 (mWh), and the privacy risk level is low. Conclusion: The combination of FEL and multi-modal sensor fusion provides a reliable and scalable approach for weed detection in precision agriculture. By processing data locally and collaboratively at the edge, the system achieves high accuracy, decreases response time, lowers energy consumption, and preserves data privacy.

This study advances prior work on the remote control of Advanced Driver Assistance Sys-tems (ADAS) by introducing a full manual teleoperation mode that enables remote control over both longitudinal and lateral vehicle dynamics via accelerator, brake, and steering inputs. The core contribution is a flexible, dual-mode teleoperation architecture that allows seamless switching between assisted ADAS control and full manual operation, depending on driving context or system limitations. While teleoperation has been explored primarily for autonomous fallback or direct remote driving, few existing systems integrate dynamic mode-switching in a unified, real-time control framework. Our system leverages a wireless game controller and a Robot Operating System (ROS)-based vehicle software stack to translate remote human inputs into low-latency vehicle actions, supporting robust and adaptable remote driving. This design maintains a human-in-the-loop approach, offering improved responsiveness in complex environments, edge-case scenarios, or during au-tonomous system fallback. The proposed solution extends the applicability of teleopera-tion to a broader range of use cases, including remote assistance, fleet management, and emergency response. Its novelty lies in the integration of dual-mode teleoperation within a modular architecture, bridging the gap between ADAS-enhanced autonomy and full re-mote manual control.

In recent years, advancements in technologies related to hydrogen have facilitated the exploitation of this energy carrier in conjunction with renewable energies, to meet the energy demands of diverse applications. This paper describes a pilot plant within the framework of a Research and Development (R&D) project aimed at utilizing hydrogen in both industrial and domestic sectors. To this end, this facility is comprised of six subsystems. Initially, a photovoltaic (PV) generator consisting of 48 panels is employed to generate electrical current form solar radiation. This PV array powers a Proton Exchange Membrane (PEM) electrolyzer, which is responsible for producing green hydrogen by means of water electrolysis. This produced hydrogen is subsequently stored in a bottling storage system for late use in a PEM fuel cell that reconverts it into electrical energy. Finally, a programable electronic load is utilized to simulate the electrical consumption patterns of various profiles. These physical devices exchange operational data with an open-source supervisory system integrated by a set of Industry 4.0 (I4.0) and Internet of Things (IoT)-framed environments. Initially, Node-RED acts as middleware, handling communications, and collecting and processing data from the pilot plant equipment. Subsequently, this information is stored in MariaDB, a structured relational database, enabling efficient querying and data management. Ultimately, Grafana environment serves as a monitoring platform, displaying the stored data by means of graphical dashboards. The system deployed with such I4.0/IoT applications places a strong emphasis on the continuous monitoring of the power inverter that serves as the backbone of the pilot plant, both from an energy flow and communication standpoint. This device ensures the synchronization, conversion and distribution of electrical energy, while simultaneously stands as a primary data source for the supervisory system. The results presented describe the design of the system and provide evidence of its successful implementation.

Industry 5.0 represents the next stage in the industrial evolution, with a growing impact in the shipbuilding sector. In response to these challenges, Navantia, a leading international player in the field, is transforming its shipyards towards the creation of a Shipyard 5.0 through the implementation of digital technologies that enable human-centered, resilient and sustainable processes. This approach gives rise to Cognitive Cyber-Physical Systems (CCPS) in which the system can learn and where the generated data are integrated into a digital platform that support operators in decision-making. In this scenario, different smart elements (e.g., IoT-based tows, trucks) are used to transport key components of ship like pipes or steel plates, which are present in a large number, representing a strategic opportunity to enhance traceability in shipbuilding operations. The accurate tracking of these elements, from manufacturing to assembly, helps to improve operational efficiency and enhances safety within the shipyard environment. Considering the previous context, the proposed paper will describe a CCPS that enables tracking and real-time data visualization through portable interfaces adapted to the shipyard workspace. Following the Industry 5.0 foundations, the presented solution is focused in providing human-centric interfaces, tackling issues like information overload, poor visual organization and accessibility of the control panels. Thus, to address such issues, an iterative human-centred redesign process was performed. This approach incorporated hands-on testing with operators at each development stage and implemented specific adjustments to improve interface clarity and reaction speed. The obtained results demonstrate that progressive adaptation of the interface, guided by usability principles, significantly enhances system effectiveness: it reduces errors, it improves response times, and it supports the seamless integration of real-time traceability into the shipyard industrial workflows.

This paper presents the design and implementation of a custom-built LiFePO₄ battery monitoring system that offers real-time visibility into the status of individual battery cells. The system is based on a Battery Management System (BMS) architecture while it is implemented the measuring of voltage, current, and temperature for each cell in a multi-cell pack. These key parameters are essential for ensuring safe operation, prolonging battery life, and optimizing energy usage in off-grid or mobile power systems.The system architecture is based on an ESP32 microcontroller that interfaces with INA219 and DS18B20 sensors to continuously measure individual cell voltage, current, and temperature. Data is transmitted wirelessly via Wi-Fi to a remote time-series database for centralized storage, analysis, and visualization. Experimental validation, conducted over a 15-day period, demonstrated stable system performance and reliable data transmission. Analytically, the findings indicate that utilizing an advanced smart charger for precise cell balancing and improving the physical layout for cooling led to superior thermal performance. Even with load currents nearly tripling to 110 mA, the system maintained a stable cell operating temperature range of 29.8 °C to 30.3 °C. This result confirms significantly reduced cell stress compared to previous iterations, which is critical for enhancing battery health and lifespan. The application of this project is aimed to demonstrates how a combination of open hardware components and lightweight network protocols can be used to create a robust, cost-effective battery monitoring solution suitable for integration into smart energy systems or remote IoT infrastructures.

This study proposes a novel approach to monitoring the grinding wheel during the dressing operation by using acoustic emission (AE) signals and the statistical Counts method. AE signals were acquired during the dressing passes and processed in MATLAB^®. The Counts matrices were segmented according to the grinding wheel rotation, and the metric termed z-ratio, which combines mean and standard deviation statistics, was calculated for each subwindow. The vectors were then filtered, normalized, and represented in polar coordinates. The results demonstrate the method’s ability to track the evolution of dressing and detect grinding wheel eccentricity, offering a promising tool for the indirect monitoring of the surface conditions of the grinding tool during the dressing operation of conventional grinding wheels.