This work deals with an adaptive disturbance rejection controller which combines the equivalent control based integral sliding mode control (ISMC) strategy and a Mamdani fuzzy inference system (FIS) for trajectory tracking of a quadrotor vehicle subject to Dryden-type wind gusts. The FIS carries out an on-line estimation of the time constant of a low-pass filter (LPF) to extract the equivalent control from the discontinuous control signal. This strategy dynamically estimates the disturbances and uncertainties acting on the plant. The estimated disturbances are compensated via a feedback control law making the system insensitive to these undesired effects. Finally, numerical simulations to evaluate the proposed controller performing a simulated quadrotor flying inspection task are presented.
This paper presents a control strategy for blood glucose regulation in the presence of an unknown time-varying delayed input. The proposed control scheme is based on an observer-based linear parameter varying (LPV) control with an $H$ criterion, designed for the Bergman minimal model (BMM) for type 1 diabetes mellitus patients. This control strategy regulates glucose levels in response to meal intake. A novel LPV system representation is used to mitigate the inherent conservativeness of the nonlinear system. The observer employs an event-triggered (ET) strategy to avoid unnecessary signal updates, enabling the estimation of the current state vector despite the delayed input, even when the delay is relatively long. Consequently, the control law ensures good performance despite a long unknown delay. The observer-based control scheme is validated under various meal disturbances and unknown delay.
Anemia, characterized by a deficiency in red blood cells (RBCs) or hemoglobin (HGB), poses significant global public health challenges. Accurate diagnosis and classification of anemia types, crucially dependent on hemoglobin (HGB) level estimation, is essential for effective management and treatment planning. This study investigates the prevalence, severity, and demographic associations of various anemia types among 364 adult patients using publicly available data from complete blood count (CBC) tests. The dataset includes 11 CBC attributes pivotal for anemia diagnosis, encompassing HGB levels and RBC counts. To address the complexities inherent in CBC data, such as variations in normal ranges based on age and gender, and challenges in data interpretation across diverse patient demographics, machine learning techniques are employed. Specifically, projected layers of long short-term memory networks (PLSTM) are utilized. These techniques require robust methodologies for data preprocessing, feature selection, and model optimization based on coefficient of determination R2, prioritizing realism and the preservation of critical medical information. This approach ensures accurate predictive modeling of anemia types based on hemoglobin (HGB) levels, classified as mild, moderate, or severe according to World Health Organization (WHO) guidelines. Insights derived from this study, leveraging PLSTM over traditional LSTM networks, contribute to enhancing diagnostic accuracy and developing personalized treatment strategies for anemia. This underscores the role of machine learning in tailoring approaches to complex medical datasets, thereby advancing healthcare outcomes.
The estimation of three abdominal tissue thicknesses, namely subcutaneous fat (SAT), muscle tissue (MT), and visceral fat (VT), was performed using radio frequency (RF) signal measurement and machine learning algorithms. The scattering parameter data were collected from simulations of a planar multilayered model of the human abdomen, with parametric analysis on each layer's thickness. Various models, incorporating combinations of several frequency bands and 120 data points, were investigated. The results showed that polynomial regression (PR) and ridge polynomial regression (RiPR) using scattering parameters from frequency bands of 300 MHz, 400 MHz, 500 MHz, and 1 GHz performed the best among other models. PR (300 MHz, 400 MHz, 500 MHz, and 1 GHz) achieved RMSE values of 0.71 mm for SAT estimation, 1.2 mm for MT estimation, and 1.17 mm for VT estimation. RiPR (300 MHz, 400 MHz, 500 MHz, and 1 GHz) achieved RMSE values of 0.35 mm for SAT estimation, 1.57 mm for MT estimation, and 1.7 mm for VT estimation. However, these results indicated a high error percentage for VT estimation, as VT had a mean thickness of only 3 mm. The results also showed that the deeper the tissue, the more prone the estimation was to error. This is due to the nature of electromagnetic reflection and transmission within a multilayered medium, which affects the scattering parameter measurement. While the models perform well for upper tissue estimation, further improvements in the model, feature selection, and physical solutions are necessary to enhance the accuracy of deeper tissue estimation.
The increase in information technologies has generated the need to implement new methodologies to secure the exchange of information. This paper proposes a cryptographic algorithm based on the use of quantum differential geometry operators. These operators exhibit mathematical behaviors similar to quantum behavior, with the advantage that specialized equipment is not required to perform the encryption process. The algorithm is composed of two phases: In the first phase, a subkey is generated through an addition and multiplication process in Z_16 based on a user-defined key. In the second phase, three types of fundamental quantum operators are employed: creation, annihilation, and crossing, which are sequentially connected. The specific selection and configuration of these operators are based on their ability to avoid generating correlation between the data. The method proposed in this work is implemented using grayscale images, obtaining the Peak Signal-to-Noise Ratio (PSNR), the Structural Similarity Index Measure (SSIM), and the Correlation Value, in order to evaluate the fidelity and integrity of the recovered images.
Cardiovascular diseases (CVDs) remain the leading cause of global mortality, with significant contributions from ischemic heart disease and stroke. This study explores the multifaceted risk factors for heart disease hospitalizations in Mexico City, focusing on socioeconomic status (SES) and air pollution. Using a dataset of 11,031 hospitalization records from 2015 to 2020, we employ negative binomial regression and Gradient Boosting Machine (GBM) models to analyze the effects of economic and social SES components, as well as air pollution indices, on the frequency and severity of hospitalizations. Our results reveal that both economic and social aspects of SES, alongside exposure to particulate matter and carbon monoxide, significantly impact hospitalization rates for various heart disease categories. The study highlights that socioeconomic disparities and environmental pollution are crucial determinants of cardiovascular health, emphasizing the need for targeted public health policies and interventions in urban settings. By integrating detailed SES indicators and pollution data, our findings provide a nuanced understanding of the complex relationships influencing heart disease hospitalizations.
This paper proposes using the Forward-Forward (FF) algorithm to train hybrid neural networks. We implement the FF algorithm in Morphological-Linear Neural Networks (MLNNs), which consist of two layers. The first layer is made up of morphological neurons and the second consists of perceptron-type neurons. For this we adapt the FF training algorithm in the calculation of the goodness and loss functions derived from the morphological neuron layer. The calculation of the goodness and the loss functions was not modified for the perceptron-type neuron layer. The experimental results show that it is possible to train morphological-linear neural networks with the FF algorithm, obtaining results similar to those obtained with the original FF algorithm. The comparison between the original training algorithm and the proposed one is carried out with low-dimensionality and binary classification datasets from the IMDB natural language processing dataset. Likewise, the ability of the FF algorithm is analyzed and compared in terms of its ability to disentangle patterns for classification tasks.
The preparation of p+ hole transport layers (HTLs) with high transparency and conductivity is an important step for the development of thin film electronic devices such as solar cells, thin film transistors, and photodetectors; especially, when the films are prepared by low-cost techniques such as spin-coating from a solution. In this study, the deposition of CuI thin films with appropriate properties, using a spin-coating technique, is explained. Single and double layers were obtained (thicknesses up to 75 nm). The material was formed by nanocrystals (27 nm) with a single plane orientation along the diagonal of the zinc-blende cubic lattice (FCC). The bandgap in average was 2.98 eV, assuring high transparency in the visible range. The conductivity was in the range from 0.05 to 0.1 (S/cm), with hole concentrations in the range from 2 to $\mathbf{4} \times \mathbf{10}^{\mathbf{19}}\ \mathbf{cm}^{-\mathbf{3}}$ , and mobilities around $\mathbf{3}\ \mathbf{cm}^{\mathbf{2}}/\mathbf{Vs}$ , which correspond to high values for this material when prepared by spin-coating, achieved previously only when doping it with iodide. In addition, the morphological (coverage and uniformity) characteristics for the double layers are much better than those reported by other researchers.
In this paper, a new control strategy for drying process was developed to control the product temperature in a type tunnel dryer. A two-stage control was developed to track the temperature trajectories of product and air using a Bezier curve at different setpoints, i.e., 50 °C for Spirulina platensis and 65 °C for Red Chili. The results demonstrate that the proposed control strategy leverages the heat supplied to the first product to enhance the heating performance of the second product. Furthermore, the use of a nonlinear model ensures that the tracking error tends to zero. Additionally, inherent disturbances in the model were considered in the second stage of the control to track the trajectory of the air temperature. Even when accounting for these disturbances, the control strategy is capable of effectively tracking the trajectory.
This work presents results from examining thin films of Zinc Oxide (ZnO) and Copper-doped Zinc Oxide (ZnO:Cu) as photocatalysts for degrading Methylene Blue (MB) in aqueous solutions. Films were deposited on soda-lime glass substrates at temperatures from 450 to 525°C and thicknesses from 200 to 1370 nm using the Pneumatic Pyrolytic Spray technique. Doping with Cu aimed to enhance photocatalytic efficiency. Structural, optical, morphological and compositional properties were analyzed using X-ray Diffraction (XRD), UV-Vis Spectrophotometry, Scanning Electron Microscopy (SEM), and secondary ion mass spectrometry (SIMS). ZnO and Cu-doped ZnO films displayed a hexagonal wurtzite-like structure and good homogeneity. Photocatalytic degradation of MB was tested under UV light at 254 nm for 5 $h$ , obtaining a degradation of 99.97% for Cu-doped ZnO films compared to 90% for pure ZnO. The study carried out in this work confirms the efficacy of Cu-doped ZnO films for degradation of organic contaminants in water by low-cost deposition techniques.
Parametric identification in robotic systems involves determining key parameters within a system's model to enhance performance. This paper presents a novel convolutional neural network (CNN) architecture designed for a parametric identification of four parameters in the dynamic model of a Cartesian robot with three degrees of freedom, which it was designed by the authors of this paper. The dynamic model is derived by the lumped parameter methodology using the Euler-Lagrange equations of motion. Detailed explanations of the neural network's configuration and training methodology are provided, aiming to maximize its effectiveness in real-world applications, thereby improving overall performance and robustness. The results show that the parametric identification of the CNC robot achieves a similarity percentage of approximately 85 % with the experimental data.
This article aims to present experimental results of nMOSFETs from a 180 nm commercial CMOS technology, with different channel lengths, operating at temperatures ranging from 80 K to 300 K. The source-drain resistance $(\mathbf{R}_{\mathbf{SD}})$ , threshold voltage $(\mathbf{V}_{\mathbf{TH}})$ , subthreshold slope (SS), low-field mobility $(\boldsymbol{\mu}_{\mathbf{0}})$ , and the linear $(\boldsymbol{\theta}_{\mathbf{1}})$ and quadratic $(\boldsymbol{\theta}_{\mathbf{2}})$ mobility degradation factors were extracted. The extraction of $\mathbf{R}_{\mathbf{SD}}$ yielded an average value of $\mathbf{61.90} \pm \mathbf{3.83}\ \mathbf{\Omega}$ all devices and temperatures. Comparing all the devices, as the temperature decreased, $\mathbf{V}_{\mathbf{TH}}$ showed an increase in its value between 22.3 % and 33.9 %; SS showed a decrease between 59.7 % and 62.1 % of its value; $\boldsymbol{\mu}_{\mathbf{0}}$ increased 162.4 % for the shorter device and 243.2 % for the longer device; $\boldsymbol{\theta}_{\mathbf{1}}$ varied less for longer devices; and $\boldsymbol{\theta}_{\mathbf{2}}$ showed more variations in its values, being less intense in shorter devices. Analyzing the channel shortening, the rate of change $\mathbf{dV}_{\mathbf{TH}}/\mathbf{dT}$ showed, in magnitude, a decrease from 0.696 mV/K to 0.561 mV/K, the rate of change dSS/dT showed an increase from $\mathbf{0.197}\ \mathbf{mV} /(\mathbf{dec} \cdot \mathbf{~K})$ to $\mathbf{0.223}\ \mathbf{mV}/(\mathbf{dec}\cdot \mathbf{K})$ , and the ZTC was from 0.67 V to 0.77 V.
In this work the progression of damage is illustrated by reducing the preload in a bolted joint and subsequently losing one or more joints in a prototype of the assembly between the main beam of a semi-wing with the torsion box of an airplane using the Natural Frequencies Vector Assurance Criterion (NFVAC) and the Damage Natural Frequencies Vector Assurance Criterion (DNFVAC) supported by experimental tests applying Experimental Modal Analysis (EMA) and simulation by means of the Finite Element Method (FEM) for applications in Structural Health Monitoring (SHM).
Intelligent transportation systems (ITS) have gained significant traction since the 1980s and 1990s, driven by technological advancements and increasing urbanization, which have caused intense transportation challenges. Drone systems, with their superior imaging capabilities, offer critical solutions for swift traffic surveillance, surpassing traditional monitoring systems. In this context, object recognition is crucial, and the YOLO algorithm stands out for its speed and efficiency. This study conducts a detailed performance evaluation of the YOLOv9 and YOLOv10 networks for motor vehicle classification through aerial images captured by drone platforms. Datasets were created from these UAV-based traffic images, and the performances of both algorithms were measured and compared. The results were analyzed to highlight YOLOv9 and YOLOv10's strengths and drawbacks. Additionally, the study discusses qualitative aspects, including advantages, disadvantages, and potential improvements for both algorithms in aerial traffic monitoring.
This paper presents a method to build workflow nets (WFN), a subclass of Petri nets, from sets of event traces (called event logs) based on event precedence structures. An event log $\boldsymbol{\lambda}$ is partitioned into sub-logs $\boldsymbol{\lambda}_{\mathbf{i}}$ containing few traces that have common sub-sequences of length two or more events. Event precedence structures are defined to represents the sub-logs; they are structures similar to partial order relations that are derived from event precedence relations drawn from traces. The method includes several steps regarding the building of the event structures and a technique for the synthesis of the WFN.
In integrated circuit manufacturing, a wafer map represents a pattern of defective dies or chips on a wafer. In order to identify different defective wafer maps or patterns by a deep learning model, it is essential that a balanced number of samples of different patterns of wafer maps is used to train the model. In this paper, the Enhanced CycleGAN generative network is used to generate realistic defective wafer maps for which adequate numbers of samples do not exist. The identification of defective wafer maps is then carried out without and with such a data augmentation. Based on the public domain wafer map dataset of WM-811K, it is shown that our data augmentation improves the overall identification accuracy by 6% and 38% for the two test cases examined compared to the no data augmentation cases.
Video games are used as simulations in training for activities of real life, with that as a base, this work presents a comparative study between seven different architectures of convolutional neural networks that perform semantic segmentation. These networks are applied to segment individuals in first-person shooters, four of them aim to return a result as fast as possible and the other three are short versions of networks focused on the quality of the segmentation.
Ship detection from Synthetic Aperture Radar (SAR) images plays a crucial role in maritime surveillance and safety. This study focuses on evaluating the performance of the latest state-of-the-art YOLO algorithm, YOLO11, for ship detection, particularly because it has not been tested on SAR images prior to this research. YOLO11 was selected for its recent release and potential improvements over previous iterations. To assess its effectiveness, the algorithm is compared with earlier YOLO versions using SAR imagery. The dataset is categorized into two subsets: open ocean images and coastal images, where distinguishing ships from coastal structures presents a significant challenge. The advantages and limitations of YOLO11 are thoroughly examined through a comparative analysis with its predecessors. Results indicate that YOLO11 outperforms earlier versions in most scenarios, particularly excelling in open ocean environments. Although ship detection from SAR images is inherently difficult, YOLO11 achieves promising Precision, Recall, and mAP values of 0.865, 0.813, and 0.792, respectively. Its performance in open ocean images exceeds these average values, highlighting YOLO11's efficacy in maritime surveillance. However, performance in coastal images is lower, with YOLOv10 performing closely to YOLO11 in these cases. The findings underscore YOLO11's effectiveness for ship detection from SAR images, showcasing its enhanced ability to detect ships in challenging environments and emphasizing its relevance for future maritime applications.
In this paper, we introduce an innovative design approach centered on optimized source/channel/drain structure using Silicon-Tin (SiSn) binary alloys to enhance the Subthreshold Swing (SS) factor of nanoscale ultra-thin film Double Gate (DG) Tunnel-FET design. In this context, accurate numerical models taking into account tunneling effects and quantum transport are developed. The influence of the Sn mole fraction on the device's switching characteristics and resulting current capability is thoroughly examined. Additionally, the influence of high-k dielectric materials on device performance is analyzed. The optimized SiSn DG TFET device demonstrates a very low Subthreshold Swing factor of 25.5 mV/dec, while maintaining a reduced ambipolar behavior. Therefore, the proposed design framework strategy paves not only to identify the appropriate binary alloys associated with the optimized mole fraction values, but also to develop efficient ultra-low power multigate transistors.
Mechanosynthesis, particularly through high-energy ball milling, offers a potent method for the fabrication of nanohybrids. This study explores the characterization of TiO 2 and graphene oxide (GO) nanohybrids, focusing on their optical and electrical properties, as well as their photocatalytic performance. Optical measurements showed a reduction in the bandgap from ≈3.27 eV in pristine TiO 2 to ≈3.02 eV in milled TiO 2 , while electrical conductivity increased from 5.59 × 10 -9 to 2.48 × 10 -8 S/cm. Despite these improvements, the addition of GO did not significantly impact the bandgap or electrical properties of the nanohybrids. Photocatalytic experiments using methylene blue (MB) under visible light irradiation demonstrated a dye degradation of ≈30-32% in all hybrid samples, indicating consistent photocatalytic activity regardless of GO oxidation degrees.
