Structural Health Monitoring (SHM) systems increasingly require efficient and scalable methods for identifying structural damage under dynamic loading. Traditional learning-based SHM models often rely on high-dimensional features or deep architectures, which may be computationally intensive and difficult to deploy in real-time applications, especially in scenarios with limited resources or bandwidth constraints. In this work, we propose a lightweight classification framework based on Hyperdimensional Computing (HDC) to detect structural damage using vibration-induced features, aiming to reduce complexity while maintaining detection performance. The proposed method encodes a rich feature set, including time-domain, frequency-domain, and autoregressive (AR) model features into high-dimensional binary vectors through a sliding window approach, capturing temporal variations and local patterns within the signal. A supervised HDC classifier is trained to distinguish between healthy and damaged structural states using these compact encodings. The framework enables fast learning and low memory usage, making it particularly suitable for edge-level SHM applications where real-time processing is required. To evaluate the feasibility and effectiveness of the proposed method, experiments are conducted on vibration data collected from controlled lateral impact tests on a concrete-filled steel tubular structure. The results validate the method ability to detect the damage-induced variations in modal frequencies and highlight its potential as a compact, robust, and efficient solution for future SHM systems based on modal data.

Wearable sensors enable continuous monitoring of physiological signals, offering opportunities for the early detection of metabolic dysfunction. In this study, we propose the use of cross-fuzzy entropy (X-FuzzEn) to quantify the dynamic coupling between wearable-derived time series, i.e., heart rate (HR), electrodermal activity (EDA), and body acceleration (ACC), across four clinically relevant glucose ranges. Analysis revealed differences in signal coordination across both metabolic and demographic groups. Prediabetic individuals exhibited elevated X-FuzzEn between HR and EDA during hypoglycemia compared to normoglycemic individuals, indicating potential autonomic dysregulation. Males showed lower X-FuzzEn compared to females, indicating more coherent and adaptive autonomic regulation. A similar pattern was observed in HR–ACC coupling, with lower X-FuzzEn in males during hypoglycemia. These findings suggest that cross-fuzzy entropy may serve as a sensitive, non-invasive biomarker of physiological resilience and autonomic stability in response to metabolic stress.

Underground coal mining is considered to be a highly dangerous activity and has been responsible for large amounts of accidents causing the death of many mine workers. One of the factors responsible for the fatal aspect of underground coal mining is the presence and accumulation of toxic gases during underground mining operations. This paper focused its investigation specifically on coal mine methane (CMM) which is released as a result of the extraction of coal and the disturbance inflicted to surrounding rocks’ formation during deep mining operations. Methane is considered a highly dangerous gas as it holds the capacity to cause explosions due to its high inflammable nature. It also can displace oxygen which eventually leads to asphyxiation. This research was based on the use of machine learning models to successfully predict dangerous concentrations of methane over the authorized threshold. Those predictions were made from a dataset containing information on the temperature, airflow, humidity, pressure and methane concentration at an underground coal mine. The temperature, airflow, humidity and pressure measurements were recorded by a series of sensors namely anemometers and component sensors THP2/93. Three machine learning classification models were implemented and compared with the objective towards finding the best model to predict and detect dangerous level of coal mine methane. The models that were investigated include: Naïve-Bayes, logistic regression and artificial neural networks (ANN). The paper concluded with an engineering decision matrix that illustrated the precision of these models towards predicting and detecting dangerous level of methane concentration in underground mines. Furthermore, recommendations for capacity improvement towards successfully predicting and detecting dangerous level of coal mine methane from an artificial intelligence’s perspective were provided.

Acetone necessitates reliable detection for the sake of both industrial and environmental safety. Metal oxides are widely used as functional materials for the development of gas sensors because techniques like nanostructure modification, doping, and solid solution formation can enhance their sensitivity and selectivity by tuning structural and electronic properties. This study developed PrFeTiO₅ nanostructures, synthesized via the solid-state reaction for acetone sensing. The sensor demonstrated a high response to acetone at an operating temperature of 400 °C, with a low influence of humidity, displaying outstanding selectivity towards acetaldehyde, NH₃, H₂, CO, and CO₂, making it suitable across various applications.

The recent synergistic explosion of Artificial intelligence and the world of machines, in a bid to make them smarter entities as a result of the fourth industrial revolution, has resulted in the concept of chatbots which has evolved over the years and gained heightened attention for the sustainability of most human corporates. Organisations are increasingly utilising chatbots towards enhancing customer engagement through the process of agent based autonomous sensing, interaction, and enhanced service delivery. The current state of the art in the chatbot technology is such that the system lacks the ability to conduct text-sensing in a bid to acquire new information or learn from the external world autonomously. This has limited the current chatbot systems to being system controlled interactive agents hence, strongly limiting their functionalities and posing a question on the purported intelligence. In this research, an integrated framework that combines the functionalities and capabilities of a chatbot and machine learning was developed. The integrated system was designed to accept new text queries from the external world and imported to the knowledge base using the SQL syntax and MySQL software. The search engine and decision-making cluster was built in the Python coding environment with the learning process, solution adaptation and inference, anchored using Reinforcement machine learning approach. This mode of chatbot operation, with an interactive capacity is known as the mixed controlled system mode, with a viable human-machine system interaction. The smart chatbot was assessed for efficacy using performance metrics (response time, accuracy) and user experience (usability, satisfaction). The analysis further revealed that several self-governed chatbots deployed in most corporate organisations, are system-controlled and significantly constrained hence, lacking the ability to adapt or filter queries beyond their predefined database when users employ diverse phrasing or alternative terms in their interactions.

Forest fires pose a serious threat to the environment with the potential of causing ecological harm, financial losses, and human casualties. While research suggests that climate change will increase the frequency and severity of these fires, recent developments in deep learning and convolutional neural networks (CNN) have greatly enhanced fire detection techniques and capability. These models can be leveraged by unmanned aerial vehicles (UAVs) to automatically monitor burning areas. However, drones can carry only limited computational and power resources, therefore on-board computing capabilities are constrained by hardware limitations. This work focuses on the design of segmentation models to identify and localize active burning areas from aerial RGB images processed on limited computing resources. To achieve this goal, the research compares the performance of different variants of the DeepLabv3 neural network model for fire segmentation when trained and tested with the FLAME dataset using a k-fold cross validation approach. Experimental results are compared with U-Net, a benchmark model used with the FLAME dataset, by implementing this model in the same codebase as the DeepLabv3 model. This work demonstrates that a refined version of DeepLabv3, with a MobileNetv2 backbone using pretrained layers and a simplified atrous spatial pyramid pooling (ASPP) module, yields a similar performance to U-Net with a precision of 87.8% and a recall of 83.2% while only requiring 20% of the number of parameters involved with the U-Net topology. This significantly reduces memory and power consumption, enabling longer UAV flight duration and reducing the processing overhead associated with sensor input, making it more suitable for deployment on unmanned aerial vehicles. The model’s compact architecture implemented using TensorFlow and Keras for model design and training, along with OpenCV for image preprocessing, makes it portable and easy to integrate with edge devices such as NVIDIA Jetson boards.

This research is framed within a larger project, whose general objective is the implementation of a SMG (Smart Micro-Grid) for distributed generation from renewable energy sources, with hydrogen as a backup. This project in-corporates an energy management system to optimize the operation of each of the systems involved while ensuring energy demand. Additionally, a hydrogen management strategy is included to maximize performance in its production, storage, and consumption within the SMG. Specifically, this work focuses on the design and implementation of a monitoring and optimization system for a SMG composed of a photovoltaic generator, a short-term energy storage system using a lithium battery, and a system for the generation, storage, and use of hydrogen produced in a fuel cell. The objective is the development of an HMI (Human Machine Interface) based on a touch operator panel KTP700 by the manufacturer Siemens, which runs in parallel with the existing SCADA (Supervisory Control and Data Acquisition) application implemented using the graphical programming software LabVIEW. The purpose of this HMI is to complement the SCADA system in such a way that it allows for direct, simple, and immediate interaction with all the equipment that comprises the SMG. This will provide quick and secure access to the monitoring of relevant variables and the parameterization of the hydrogen generator. Furthermore, due to the robustness and reliability of the industrial operator panels, the aim is to establish a supervision system with continuous and permanent operation, similar to the industrial plants automation and management systems.

Environmental conditions like temperature, humidity, light, and gas levels directly affect human health, agriculture, and industrial processes. Monitoring these factors in real time is necessary for detecting dangerous situations early and making informed choices. This work presents a compact, mobile, IoT-enabled device that measures environmental data and sends it wirelessly for remote access. The system uses the ESP32 microcontroller, cho-sen for its low power use, built-in Wi-Fi, and ease of connecting with sensors and cloud services. Key sensors include the DHT22 for temperature and humidity, MQ135 for am-monia and gas detection, and an LDR for checking light intensity. An infrared (IR) sensor identifies obstacles, and a buzzer alerts users to dangerous conditions. The collected data appears on a 16x2 LCD for local monitoring. It is also transmitted to the ThingSpeak cloud platform for long-term storage and visualization. Users can view this data in real time through the Blynk mobile application, which also enables remote control of the device. The system is built for mobility. Itoperates with DC motors powered by an L298N motor driver. This lets it navigate different environments and collect data from various locations. This feature gives more flexibility and improves the system’s effectiveness compared to tradi-tional stationary monitoring units. The innovative part of this project is the mix of re-al-time sensing, autonomous movement, and cloud connectivity in a low-cost, portable setup. The system was tested in controlled environments and consistently provided relia-ble readings. Its practical uses include smart agriculture, urban air quality monitoring, and industrial safety.

A new construction of a custom-made, low-cost electronic nose applying six TGS-type gas sensors manufactured by Figaro Inc. was assembled. The gas sensors were used to collect response signals caused by changes in gas composition from clean air to the studied odor, to which the sensors were exposed. In addition, modulation of sensor heater temperature was implemented in order to register complementary information useful for differentiation between the studied odor categories. An automatic mechanism was to open the gas sensor chamber, allowing sensors exposure to the studied gas and cleaning of sensors in the condition of a closed chamber. Sensor cleaning was conducted by forcing a clean air current through the application of a pneumatic pump. 3D printing was used to manufacture the sensor chamber. The Raspberry PI microcomputer was used for control of the measurement procedure and data collection. The operation of the device could be controlled by a web-based interface from a connected laptop or smartphone. The device was applied to the monitoring of the development of spoilage of soft fruits like strawberries. Periodic measurements were performed in an automatic manner. A dedicated system of separation of the measured sample from the gas sensor array, preventing heat flow, was designed. Technical challenges encountered during the measurement are presented and discussed in the presentation.

In automated laboratories, ambient monitoring and precise object tracking are essential for safety and system reliability. In this paper, we present the development and evaluation of a low-cost, portable sensor node for environmental sensing and ultrawideband (UWB) based localization. The sensor node integrates a set of commercial gas sensors for measuring environmental parameters and an ultra-wideband unit for object tracking. The device has an IoT microcontroller that can efficiently process the data from both environmental sensors and the location information from the UWB module and transmit it wirelessly to the cloud/monitoring server via Wi-Fi user datagram protocol (UDP). A custom Python application was developed for real-time monitoring, implementing trilateration and least-squares algorithms for accurate indoor positioning. Experimental results showed a location accuracy better than 50 cm under line-of-sight conditions.