Grinding is a crucial manufacturing process at the final stage of the machining chain, involving removing material from the surface of machined parts using an abrasive grinding wheel. One of the primary challenges in this process is determining the appropriate time for dressing the grinding wheel, which is essential to restore its cutting efficiency. The previous study entitled "In-Dressing Acoustic Map by Low-Cost Piezoelectric Transducer" introduced an innovative technique for diagnosing the surface integrity of the grinding wheel, employing a methodology for generating acoustic images from an acoustic emission sensor and a piezoelectric diaphragm. However, acquiring sharp acoustic maps remained challenging due to intense noise and interferences typical of harsh industrial environments. The present work further investigates this issue by applying digital image processing techniques to enhance the acoustic maps, utilizing tools including Google Colab and libraries like OpenCV, NumPy, and Matplotlib.pyplot. These techniques include smoothing, equalization, and edge filtering, using methods such as Sobel, Canny, and Prewitt. Examination of the treated acoustic maps revealed more detailed and relevant features, allowing a more accurate assessment of dressing conditions. The results demonstrate the efficacy of digital image processing techniques in improving the evaluation of the grinding wheel's cutting condition, contributing significantly to the efficient management of dressing cycles. This improvement can be applied to other machining processes, such as drilling and milling, and integrated into IoT (Internet of Things) sensor systems for applications in Industry 4.0.

Agrivoltaics is defined as “the dual use of land for solar energy production and agriculture”. On this topic, a number of issues are still to be properly addressed, to understand e.g. how the shading effect of the solar panels affects crop growth. In this work, the development of a large-scale digital twin model to predict crop yield under a varying solar panel coverage is discussed. A framework is proposed to exploit Internet of Things (IoT) concepts, with a sensor network to collect data on the field, merged with sensor fusion to also handle information gathered by satellite images. The aim of the entire work being related to the synergic optimization of energy production and crop yield, data analytics based on artificial intelligence tools are to be extensively developed. Results are reported of an experimental activity, currently under way at the Fantoli laboratory of Politecnico di Milano. Wooden panels, placed above the crop with varying orientation and pattern, are used to study the aforementioned shading effect with a specific target on conditions typical of Northern Italy. The laboratory facility is equipped with a comprehensive sensor network, to acquire the data necessary to build the targeted large-scale digital twin of the agrivoltaic system.

Introduction

Proper nutrition for pets is crucial for their well-being. This project addresses feeding issues by developing an automatic pet feeder with Internet of Things (IoT) technology and deep-learning (DL) techniques. The device is designed to dispense appropriate food portions, enhancing pet nutrition management and overall health.

Methods

The automatic pet feeder integrates multiple sensors and deep learning algorithms, specifically convolutional neural networks. The sensor network includes a weight sensor for precise measurement, a camera for pet identification, an ultrasonic sensor for detecting proximity, and a servomotor for controlled food dispensing. Data from these sensors are processed using a microcontroller with Wi-Fi capabilities, facilitating real-time monitoring. The DL model was trained using a dataset of images of dogs and cats to ensure accurate identification and customized feeding plans.

Results

Testing has shown that the DL system can identify pets with precision and accurately dispense appropriate food portions based on weight, enabling species-specific feeding and providing real-time monitoring. The integration and fusion of sensors provided reliable data on food consumption and pet weight, optimizing feeding quantities. Alerts and notifications were successfully transmitted to pet owners via an application, ensuring continuous monitoring and adjustment of feeding patterns and reassuring them about their pets' safety and well-being.

Conclusions

The automatic pet feeder achieved its objectives of providing a convenient, reliable, and adaptable solution for managing pet nutrition. Its ability to customize feeding based on individual pet needs, combined with IoT and DL technologies, highlights its potential for improving pet health and owner convenience. Future enhancements will focus on refining sensor accuracy, expanding functionalities, and further incorporating more advanced DL techniques to personalize pet care and feeding routines. This project demonstrates the potential of IoT-based and DL solutions in promoting pet well-being through precise and automated nutritional management.

The metric system deployed in identifying, sensing and measuring the level of competitiveness associated with a production system such as the agro-seed processing industry, where grains are nurtured and developed, is largely non-objective due to the chain of embedded and interconnected non-metric qualitative tasks and activities. However, in this research the diverse pool of activities inherent in the agro-grain nurturing system of systems (SoSs) network of functional and operational activities and the associated management centric tasks referred to as the System of Systems Engineering Management (SoSEM) processes, were sensed, analysed and measured. This was done in a bid to quantitatively understand the performing levels of a range of dynamic variables, including the skill level of employees assigned to specific tasks and how all of these culminated into decision making on the level of the SoSs competitiveness on a percentile scale.

Traditionally, the procedures available for the identification, sensing and measurement of operational activities in system of subsystems (SoSubs) or system of systems (SoSs) are often limited to verbal articulations, physical observations, bench marking of tasks with desired targets at the end of the tasks, amongst others. In this research, a holistic and integrated framework depicting a SoSs network in the agricultural grain culturing industry was developed. Furthermore, a metric system for the identification, sensing and measurement of tasks performance, skillset rating of employees and overall quantitative evaluation of the SoSEM was presented.

The proposed solution was developed by first designing and architecting a holistic framework that depicts a homogeneous SoSs in the agro-grain industry. Following this, a metric system comprising the sensing and measurement systems was developed. This metric system was premised on a developed time variant quantitative approach, and a continuous weight assignment based on a prioritisation model that utilised dependency theory of the tasks.

In this research, an innovative smart wheelchair designed to enhance mobility and health monitoring for individuals with disabilities is introduced. The wheelchair incorporates a tri-wheel mechanism, enabling seamless navigation over various terrains, including stairs. This design addresses the common limitations of traditional wheelchairs by providing increased autonomy and flexibility to users. Central to the design is the integration of advanced sensor networks and Internet of Things (IoT) technology. The wheelchair is equipped with an Electroencephalography (EEG) system that allows users to control movements using neural impulses, providing a hands-free operation mode. This feature is particularly beneficial for users with severe physical impairments, enabling them to navigate more independently. In addition to mobility enhancements, the smart wheelchair features comprehensive health monitoring capabilities through continuously monitoring vital signs such as blood oxygen levels (SpO2), and electrocardiogram (ECG) data. These health metrics are regularly transmitted to healthcare providers via a secure IoT platform. In emergency situations, the system is programmed to automatically send alerts, including the patient’s location, to caregivers and emergency services. The study demonstrates that the smart wheelchair not only improves mobility for users but also significantly enhances their quality of life by integrating health monitoring and emergency response features. This innovation represents a step forward in developing assistive technologies that support independent living and proactive healthcare management in smart cities.

KM3NeT is an underwater neutrino detector currently under construction. Since the installation of its first Detection Unit in 2015, it has been continuously collecting data. Due to its complex design, a 3D array of sensors, an Acoustic Positioning System (APS) was developed to monitor the position of each sensor. Given the increasing number of acoustic sensors used for the APS, both receivers and emitters, a solution was implemented to check their status. In this contribution, a monitoring tool for this instrumentation is presented, capable of evaluating its status at both the data and operational levels. For effective monitoring, it is crucial to associate the signal recorded by a receiver with the corresponding transmitter. The Acoustic Data Filter (ADF) performs a cross-correlation between the signals retained in a buffer and those emitted by each installed emitter. It saves the maximum peak value and its associated time of arrival for each expected signal. However, the growing number of beacons complicates the differentiation of corresponding transmitters due to the huge amount of data recorded by the ADF, needing post-processing. To address this challenge, a monitoring tool that analyzes the internal clock of each emitter to distinguish and filter the data collected by the ADF is developed. This tool has tested highly effective in verifying the correct operation of all acoustic devices deployed at sea. The acoustic monitoring graphical output produced for each data slot facilitates quick failure detection, enabling a swift response. Last but not least, the tool is modular and scalable, adapting to the addition or removal of sensors from the detector.

For applications such as home surveillance systems and assisted living for elderly care, sensing capabilities are essential for tasks such as locating, determining the approximate position of a person or identifying the status of a person (static or moving), since the effects caused by the presence of people can be captured in the power received of signals in an infrastructure deployed for these purposes. Human interference on the Received Signal Strength Indicator (RSSI) measurements between different pairs of wireless nodes can vary depending on whether the target is moving or static. To test these ideas, an experiment was conducted using four nodes equipped with the ZigBee protocol in each corner of an empty 6.9m x 8.1m x 3.05m room. These nodes were configured as routers communicating to a coordinator outside of the room that instructed the nodes to send back their pairwise RSSI measurements. The coordinator is connected to a computer in order to log the measurements, as well as the time at which the measurements are generated. The code was run for every iteration of the experiment, whether the target was static, moving or when the number of targets was increased to five. The data was then statistically analyzed to extract the pattern and other target relational parameters. There was also a correlation between the change of the pairwise RSSI and the path described by the target when moving through the room. The data presented by the results can aid with algorithms for device-free localization and crowd classification with a low infrastructure cost for both and shed light into relevant characteristics correlated to path and crowd size in indoor settings.

Stress has multiple effects on human health. Sensors designed to measure stress and indicate health status by recognizing illnesses or other conditions (e.g., heart problems and blood pressure) have been widely utilized to monitor and characterize this physiological phenomenon. Stress has two response mechanisms: the autonomic nervous system (ANS) and hypothalamic-pituitary-adrenal (HPA) axis. ANS can affect heart rate, breathing rate, skin conductance, blood pressure, and other hemodynamic parameters. Continuous non-invasive blood pressure (cNIBP) measurement, pulse volume, cardiac output, and other hemodynamic parameters are important for stress measurement and health indicators. There is still room for research and development of different approaches to measurement in this area. Very few sensor systems associated with cNIBP have been developed or are currently in progress. Photoplethysmography (PPG), impedance plethysmography (IPG), and ultrasound imaging were performed along with other non-invasive sensors, such as electrocardiography (ECG), cardioseismography (CSG), and ballistocardiography (BCG), to measure hemodynamic parameters. In the HPA axis, stress hormones are the most important measurement from the perspective of cortisol levels. This measurement is also important in general for the health of the subject, especially for good functioning of the axis itself (HPA axis). Sensors have been developed to detect cortisol levels for academic and research purposes. Cortisol levels can be measured in two ways: direct and indirect hormone measurements. Non-invasive direct hormone measurement uses a sensor to evaluate the cortisol levels in sweat. In contrast, indirect measurement uses an increase or decrease in cortisol levels in relation to other substances such as sodium or potassium. Therefore, in the present study, we investigated technologies, methods, and wearable sensors for continuous hemodynamic measurements at the ANS level and cortisol measurements at the HPA axis level. These sensors and measurements are crucial for improving healthcare applications.

Structures are exposed to aging and extreme events that can decrease the relevant safety margins or even lead to (partial) collapse mechanisms under the foreseen loading conditions. Structural health monitoring (SHM) looks therefore compulsory to avoid accidents, by tracking the evolution of the state of the system and sending out warnings as soon as critical conditions are met or drifts from the response of the undamaged structure are identified. One of the approaches to online SHM rests on Kalman filtering, which is able to build the time evolutions of the structural state upon the Bayes’ rule. In a customary joint version of the filtering procedure, state variables and health parameters are joined together in an extended state vector: while state variables, like e.g. lateral displacement of shear buildings, can be observed thanks to pervasive sensor networks, the health parameters usually linked to the structural stiffness cannot, leading to possible divergence issues characterized by biases in the estimates. This is further enhanced by epistemic uncertainties and related difficulties in setting the covariance terms allowing for modelling strategies not in perfect agreement with reality. In this work, we propose the use of an adaptive strategy to the online tuning of the aforementioned covariance terms, so that the accuracy of filtering outcomes is improved without issues linked to filter instability. Results are proposed for the SHM of shear buildings, showing that the proposed method outperforms other strategies available in the literature, both in terms of accuracy of the estimations and readiness to track their time evolutions.

Hybrid Electric Vehicles (HEVs) functions using a combination of electric power and fossil fuels like petrol or diesel. HEVs are operated with their dual power sources, which presents unique challenges and opportunities in terms of thermal management. Lithium-ion batteries on the other hand, are sensitive to operating temperature and performs best within a certain temperature range. Consequently, to guarantee the safe operation of HEVs, a Battery Thermal Management System (BTMS) is required. This paper presents an IoT-based thermal monitoring system for hybrid vehicle where the sensor data is uploaded to cloud and controlled by developing the code on Arduino IDE-Embedded C environment. This work uses DS18B20 and DHT11 temperature sensors and MQ-135 gas sensor which records the physical parameters and uploads it to the cloud. The NODE-MCU ESP8266 is the core component of this work, which reads sensor data and receives control signals from the Blynk IoT app for motor control. The ESP8266 then transmits this data back to the app for visualization. The developed code is dumped into the Arduino IDE environment which compares the sensor measured value with the threshold temperature and gas value stored already in the cloud for decision making. Proportional Integral Derivative (PID) controllers and fractional PID controllers are used for speed control of HEV-DC motors. The L298N, a dual H-bridge motor driver Integrated Circuit (IC) allows the module to control the speed of motors, which directly controls the speed of the vehicle. In addition to this, use of monitoring circuit in BTMS will monitor the key parameters of the battery like voltage, current, temperature during charging and discharging situation. The developed moule is found to be improved in the areas of flexibility, scalability, complexity and performance.