Food spoilage, which causes 40–50% of all losses of root crops, fruits, and vegetables each year, is one of the biggest problems the world is currently experiencing. If the freshness or deterioration of a fruit can be determined before it is lost, the fruit waste problem may be mitigated. The goal of this work is to develop a simple model for tracking fruit quality using sensors and machine learning (ML). This model assists in determining the fruits that will ripen and require use earlier from the gases emitted by it. Two gas sensors (MQ3 & MQ7) and an Arduino Uno serve as the main processing components of the suggested system. Principal Component study (PCA) is a widely employed discriminating approach that has been utilized to differentiate between fresh and rotten apples based on sensed data. The study yields a cumulative variance of 99.1% over a span of one week. The data has also been evaluated using a linear Support vector machine (SVM) classifier, which has achieved an accuracy of 99.96%. The distinctive feature of the system is that it evaluates the levels of spoilage based on real-time data and deploys a low-cost, straightforward system that can be used anywhere to preserve any type of fruit.

A Safety Instrument Function (SIF) averts hazardous incidents that may arise due to diverse anomalies within a system. The SIF prevents potential dangers by comprising three integral components—the Sensing Element, the Logic Solver, and the Final Element. The 2oo3 architecture is the optimal configuration for each SIF component, employing both AND and OR logic designs for its voting mechanism. Type A devices, recognized for their passive nature, exemplify robustness and reliability. While these devices are acknowledged as the most dependable, semiconductor devices or microcontrollers, categorized as Type B, often find application in logic processing. This paper introduces the incorporation of memristors, inherently passive devices with memory attributes, into the system. The Logic Solver, which calculates confidence values, exhibited greater efficacy than Type B devices. Verification was conducted through LTspice circuit simulations. Result of the Memristor for Logic Solver in the safety Instrumentation function (SIF) IEC 61508/61511 standard: voter circuit has the lowest components, failure rate, and most mean time to failure. That is more reliable than the other voter.

Numbers of vehicles has increased tremendously on roads. Also, the number of roads that are constantly experiencing traffic jams during morning and evening peak hours has increased significantly, which calls for a better understanding of traffic stream characteristic and car-following models. Traffic stream macroscopic parameters (speed, flow and density) could be estimated through a number of traffic-flow theory models. In order to collect accurate data regarding fundamental of traffic stream parameters, a traffic monitoring system is needed to present the data from different roads. In this study, a real time traffic monitoring system is introduced for traffic macroscopic parameters estimation. The sensor network has been constructed using a set of linear fiber optic sensors. In order to validate the system for this study, the system was installed at MnROAD facility, Minnesota. Fiber optic sensor detects the propagated strains in highway pavement due to the vehicle movements through the changes of the laser beam characteristics. Traffic flow can be estimated by tracking the peak of each axle passed over the sensor or within the sensitivity area, Time Mean Speed (TMS) and Space Mean Speed (SMS) space mean speed can be estimated by the different time a vehicle arrived at the sensors. The density can be determined either by using fundamental traffic flow theory model or estimation the time that vehicles occupy the sensor layout. A real traffic was used to validate the sensor layout. The results show the capability of the system to estimate traffic stream characteristics successfully.

In this work, we present the design of an observer for Takagi-Sugeno fuzzy systems with unmeasurable premise variables. Moving away from Lipschitz-based and L₂ attenuation-based methods — which fall short in eliminating the mismatching terms in the estimation error dynamics — we leverage the differential mean value theorem. This approach not only removes these terms but also streamlines the factorization of the estimation error dynamics, making it directly proportional to the estimation error. To ensure the asymptotic convergence of the estimation error, we apply the second Lyapunov theorem, which provides sufficient stability conditions described as linear matrix inequalities. A numerical example applied on Three-tank hydraulic system is presented to demonstrate the observer's effectiveness.

The cognitive state of a person can be categorized using the Circumplex model of emotional states. The Circumplex model is a continuous model of two dimensions: arousal and valence, where arousal measures the energy level and valence measures the positivity level of a person's emotion. The arousal and valence values can be estimated via machine learning regression. We exploit the Remote Collaborative and Affective Interactions ‎‎(RECOLA) dataset which includes audio, video, and physiological recordings of online interactions between human participants. We previously succeeded to predict arousal and valence values using the physiological [1,2] and video [2,3] recordings of RECOLA. Features are attributes that describe the data. They can be predesigned or learned. Learned features are attributes that are automatically learned and utilized by deep machine learning solutions. On the other hand, predesigned features are attributes that are calculated before the machine learning process, and provided as input to the machine learner later on. Our previous work on the video recordings of RECOLA focused on learned features. In this paper, we expand our work to analyze and assess the predesigned visual features, extracted from the video recordings of RECOLA, for predicting the arousal and valence values of cognitive/emotional states. We process the visual features of RECOLA by applying time delay and sequencing, arousal and valence annotation labelling, and data shuffling and splitting. We then train and test machine learning regressors to predict the arousal and valence values. Our preliminary results outperform those from the literature. We have achieved a testing root mean squared error (RMSE), Pearson’s correlation coefficient (PCC), and concordance correlation coefficient (CCC) of ‎0.1033, 0.8498, ‎‎and 0.8001‎ on arousal predictions, respectively. We have achieved a testing RMSE, PCC, and CCC of ‎0.07016‎, 0.8473, and 0.8053 on valence predictions, respectively. These performances are obtained using an optimizable ensemble regressor.

The rapid advancement of agricultural technology has led to an increasing reliance on data-driven approaches for optimizing crop yields and resource management. In this research paper, we present a comparative study between two graph-based models, namely the Graph Convolutional Network (GCN) and the Graph Neural Network (GNN), for the task of crop recommendation based on various environmental factors. Our study focuses on leveraging a dataset encompassing critical parameters such as nitrogen level, potassium level, phosphorus level, temperature, local humidity, pH of soil, and rainfall, with the target being the selection of a suitable crop for a given season. To address the complexity and interdependencies of the provided dataset, we harness the power of graph-based models that are adept at capturing intricate relationships among features. Both the GCN and GNN are well-suited for such tasks due to their ability to process structured data represented as graphs. We adopt a supervised learning approach where the input features are organized as nodes in a graph, and edges represent potential associations between these features. The objective is to predict the most appropriate crop label for a given set of environmental conditions. Our experimentation involves pre-processing the dataset to construct an appropriate graph representation. We evaluate the performance of both models using metrics such as accuracy, precision, recall, and F1-score to ascertain their effectiveness in recommending crops. Additionally, we investigate the model's ability to generalize by employing techniques like k-fold cross-validation to mitigate overfitting concerns

In conclusion, this research contributes to the ongoing exploration of graph-based models in agricultural applications. By showcasing the comparative performance of GCN and GNN in the context of crop recommendation, we offer valuable insights into the potential of these models for aiding precision agriculture practices. Our findings underline the importance of choosing an appropriate graph-based model based on the nature of the dataset and its inherent relationships, leading to more informed decisions in crop management and resource allocation.

The present study introduces an innovative device designed to objectively record and quantify the daily standing profiles of individuals with lower limb motor impairment. The device is specifically developed to be seamlessly embedded onto the standing platform of an assistive standing mobile device, without compromising its structural integrity or functional capabilities. The primary objective of this device is to provide objective evidence of patients' standing activities within their home environment, thus facilitating the assessment of patient performance and usage. The embedded device captures and stores comprehensive data regarding the duration, frequency, and interval of patients' standing sessions. Furthermore, the device integrates wireless connectivity to facilitate data transfer and analysis. The development process involved close collaboration between rehabilitation engineers and physiotherapists to ensure optimal functionality, user-friendliness, and unobtrusiveness. Extensive testing and validation procedures were conducted to assess the reliability, validity, and feasibility of the device. Results demonstrate its high accuracy and reliability in capturing and quantifying standing profiles. The proposed device addresses a critical need within the field of rehabilitation, providing clinicians, researchers, and funding organizations with objective evidence of patients' standing abilities and adherence to rehabilitation protocols. This evidence-based approach has the potential to enhance clinical decision-making, improve treatment outcomes, and secure financial support for patients in need of assistive standing mobile devices. In conclusion, the embedded device presented in this study offers a novel and practical solution for quantifying and recording the daily standing profiles of individuals with lower limb motor impairment. By providing objective evidence of patients' standing activities, this device has the potential to advance the field of rehabilitation and facilitate improved access to assistive standing mobile devices for those in need.

The Internet of Things (IoT) has been developed using the current internet architecture. The IoT concept aims to increase productivity, accuracy, and financial gains. The purpose of this study is to evaluate how well the agricultural sector is using the Internet of Things (IOT). In this study, descriptive analysis approaches are used with qualitative methods. Reviews of the literature from numerous credible national and international periodicals are used in the data collection process. This study found that it is now possible to remotely monitor agricultural development, soil moisture, and crop risk thanks to the growth of the Internet of Things and the digital transformation of rural areas. The efficiency of agriculture and farming processes can be increased by automating human intervention, especially when using the Internet of Things.

We present a novel design for an electrochemical cell with a thickness of approximately 30 µm. The design incorporates a carbon fiber electrode (with a diameter of 10 µm) inserted through a bubble wall or stable surfactant film, which serves as the electrochemical medium. To establish a stable surfactant film, we utilized a solution filled with non-ionic Triton-X100 surfactants. This medium was employed for both electrochemical sensing and electrodeposition purposes.

For microelectroanalysis, a modified carbon microelectrode coated with nanosheet graphene oxide (serving as the sensing electrode) was positioned across a soap bubble wall. A 1 mm diameter silver wire was used as the counter/reference electrode, while the soap bubble contained dissolved nitrite ions. This approach enabled the proposed sensing system to successfully detect NO2-, both when present on a hand and when dissolved in the Triton-X100 surfactant film. This technique holds particular significance in criminal investigations, as the presence of NO2- ions on the hand indicates gunshot residue and can aid in suspect identification. Therefore, this sensing strategy offers rapid analysis with a low detection limit of 28 µM. It proves functional for on-site sensing, making it suitable for efficient police and criminal investigations. Notably, compared to the current method, it is simple, cost-effective, involves only one step, and eliminates the need for any sample preparation steps.

In the second part of the study, we evaluated the micro-electrodeposition process within a bubble wall. For electrodeposition, various ions, such as silver and palladium ions, were dissolved within the bubble film. A bare carbon microelectrode was positioned within the bubble wall, and an appropriate cathodic potential was applied. The resulting metallic film was analyzed using scanning electron microscopy (SEM) and energy-dispersive X-ray spectroscopy. The images obtained from the analysis revealed that the thickness of the bubble wall or electrochemical cell imposes limitations on the electrodeposition area at the microscale level. Furthermore, the lifespan of the bubble wall played a crucial role in controlling the duration and thickness of the deposited film, ranging from the nanoscale to the microscale.

Finally, we believe that this novel work can open new approaches in sensing and synthesis in electrochemical science.

Tin Sulfide (SnS)-based photodetector and photovoltaic devices emerged as potential candidates for low-cost and efficient eco-friendly photosensing and clean energy applications. The amazing optoelectronic properties of SnS-based devices, such as high optical absorption and tunable direct band-gap, are currently piquing the curiosity of researchers. However, the low recorded photoresponsivity is the major limitation that needs to be overcome without introducing toxic materials and increasing the elaboration cost of the solar cell. In this work, we propose a novel alternative design technique based on combined SCAPS numerical simulations and Machine Learning (ML) computation to improve the photocurrent performances for efficient eco-friendly photosensing photovoltaic applications. It is revealed that the proposed design framework can predict the better SnS photovoltaic configuration, and pave the way for the optoelectronic systems designers to identify the geometry and the appropriate material for each layer of the device. Moreover, the results of the proposed SnS-based heterostructure solar cell offers an innovative approach for elaboration of eco-friendly high-efficiency thin-film optoelectronics devices that is more promising than the previously reported designing techniques.