Adaptive System Research Articles

Model for motivating learners with personalized learning objects in a hypermedia adaptive learning system

<span lang="EN-US">A number of weaknesses were demonstrated in the E-learning platforms during the Covid-19 pandemic despite the efforts invested. This has negatively influenced learners' motivation and consequently their performance. With the proliferation of technology and the revolution of information and communication technologies (ICT), learning objects have become new epitomes widely used, accessible, and implemented with educational resources and technological support. The integration of learning objects into E-learning has enhanced educational progress, but during critical periods, it is crucial to ensure pedagogical continuity and learner motivation. Based on this observation, we will propose architecture of a personalized learning object model in the context of an adaptive hypermedia learning system (AHS). The objective of our model is to increase the motivation factor which is a determining element in the success of E-learning, our model aims to improve the performance of the learners in order to avoid the abounding of learning and to promote the attendance of the learners. This will be useful later for any design or development of learning objects in hypermedia learning systems that are adaptive to the needs of the learners and in line with their preferences and profiles throughout the learning process offered by the system. </span>

Smart traffic forecasting: leveraging adaptive machine learning and big data analytics for traffic flow prediction

The issue of road traffic congestion has become increasingly apparent in modern times. With the rise of urbanization, technological advancements, and an increase in the number of vehicles on the road, almost all major cities are experiencing poor traffic environments and low road efficiency. To address this problem, researchers have turned to diverse data resources and focused on predicting traffic flow, a crucial issue in Intelligent Transportation Systems (ITS) that can help alleviate congestion. By analyzing data from correlated roads and vehicles, such as speed, density, and flow rate, it is possible to anticipate traffic congestion and patterns. This paper presents an adaptive traffic system that utilizes supervised machine learning and big data analytics to predict traffic flow. The system monitors and extracts relevant traffic flow data, analyzes and processes the data, and stores it to enhance the model's accuracy and effectiveness. A simulation was conducted by the authors to showcase the proposed solution. The outcomes of the study carry substantial implications for transportation systems, offering valuable insights for enhancing traffic flow management.

Learner’s attention detection in connected smart classroom using internet of things and convolutional neural networks

Detecting learner attention is an essential part of learning assessment. Consequently, it becomes an essential requirement for adaptive intelligent teaching systems, to identify specific needs and anticipate orientations. In this article, we propose a new model of a connected smart classroom, based on the internet of things, artificial intelligence and machine learning to detect in real time learners' attention and marking their presence during the execution of a teacher-assisted pedagogical activity, as well as to adapt the most suitable learning objects to these learners. The proposed model is based on head position, gaze direction, yawning and eye-state analysis as facial landmarks detected by cameras connected via the Bluetooth low energy network and transmitted to a developed convolutional neural network. In addition, a series of experiments have been conducted to evaluate the performance and efficiency of the model developed. The findings demonstrate that the model developed can be used to precisely capture the status of learners in the classroom in terms of attention and identification. In this way, these interesting findings can be used to adapt teaching activities to the individual needs of learners, and to identify areas where they have difficulties and needs extra help.

Do domain-specific protein language models outperform general models on immunology-related tasks?

Deciphering the antigen recognition capabilities by T-cell and B-cell receptors (antibodies) is essential for advancing our understanding of adaptive immune system responses. In recent years, the development of protein language models (PLMs) has facilitated the development of bioinformatic pipelines where complex amino acid sequences are transformed into vectorized embeddings, which are then applied to a range of downstream analytical tasks. With their success, we have witnessed the emergence of domain-specific PLMs tailored to specific proteins, such as immune receptors. Domain-specific models are often assumed to possess enhanced representation capabilities for targeted applications, however, this assumption has not been thoroughly evaluated. In this manuscript, we assess the efficacy of both generalist and domain-specific transformer-based embeddings in characterizing B and T-cell receptors. Specifically, we assess the accuracy of models that leverage these embeddings to predict antigen specificity and elucidate the evolutionary changes that B cells undergo during an immune response. We demonstrate that the prevailing notion of domain-specific models outperforming general models requires a more nuanced examination. We also observe remarkable differences between generalist and domain-specific PLMs, not only in terms of performance but also in the manner they encode information. Finally, we observe that the choice of the size and the embedding layer in PLMs are essential model hyperparameters in different tasks. Overall, our analyzes reveal the promising potential of PLMs in modeling protein function while providing insights into their information-handling capabilities. We also discuss the crucial factors that should be taken into account when selecting a PLM tailored to a particular task.

Performance evaluation of BLDC motor drive mounted in aerial vehicle (drone) using adaptive neuro-fuzzy

The development of autonomous drones equipped with cameras and various sensors has paved the way for their application in agriculture and perimeter security. These aerial drones require specific power, acceleration, high torque, and efficiency to meet the demands of agricultural tasks, utilizing built-in brushless DC (BLDC) motors. However, a common challenge drone’s face is maintaining the desired speed for extended periods. Enhancing the performance of BLDC motors through advanced controllers is crucial to address this issue. This research paper proposes optimizing the size and speed of brushless DC motors for aerial vehicles using an adaptive fuzzy inference system and supervised learning techniques. When these drones carry loads, the BLDC motors must dynamically adjust the drone's speed. During this phase, the motors must control their speed and torque using artificial intelligence controllers like adaptive neuro-fuzzy inference systems (ANFIS) to enhance the drone's functionality, resilience, and safety. This research has conducted analyses focused on improving the performance of BLDC motors explicitly personalized for unmanned aerial vehicle (UAVs). The proposed method will be implemented using MATLAB/Simulink, expecting to significantly enhance the BLDC motor's performance compared to conventional controllers. Comparative analyses will be conducted between traditional and ANFIS controllers to validate the effectiveness of the proposed approach.

A modified sliding mode controller based on Fuzzy logic to control the longitudinal dynamics of the autonomous vehicle

This article delves into the intricate world of controlling the longitudinal dynamics of autonomous vehicles. In the first part, we studied two distinct controllers: the Super Twisting Sliding Mode Control and its modified version enriched with the integration of fuzzy logic, applied to the longitudinal dynamics of the autonomous automobile to follow a desired speed longitudinal profile, the two controllers are compared with a Neural Network-Based Non-singular Terminal Sliding-Mode Control, the system takes the throttle and brake as its inputs and delivers speed and acceleration as outputs. The overarching objective is to ensure that the controlled vehicle maintains a close and precise alignment with the desired speed profile. The second part of our research is dedicated to the development of adaptive cruise control systems and cruise control according to the safety conditions. This controller consists of two blocks low and upper controller, in upper controller the inputs are the speeds of the automobile in front and of the autonomous automobile itself, the safety distance, the measured distance. The output is the desired acceleration. The objective is to maintain a distance between the front vehicles, greater than or equal to the safety distance. For this, to achieve this task, we have implemented a control system known as the Proportional Integral Derivative (PID) controller in the adaptive cruise control system to control this system. In the low controller block, the same controllers used in the first part: the Super Twisting Sliding Mode Control and its modified version based on fuzzy logic, are applied, the system inputs are throttle and brake, and the outputs are speed and acceleration. This system is processed by MATLAB code, we obtained a better result with our proposed controller such that the maximum absolute speed error is equal to 0.0144 m/s in the first case of speed tracking, and to 0.006 m/s in the second case of the using adaptive cruise control, the illustrations below show the efficiency and robustness of these controllers.

Kernel density estimation of Tsalli’s entropy with applications in adaptive system training

Information theoretic learning plays a very important role in adaption learning systems. Many non-parametric entropy estimators have been proposed by the researchers. This work explores kernel density estimation based on Tsallis entropy. Firstly, it has been proved that for linearly independent samples and for equal samples, Tsallis-estimator is consistent for the PDF and minimum respectively. Also, it is investigated that Tsallis-estimator is smooth for differentiable, symmetric, and unimodal kernel function. Further, important properties of Tsallis-estimator such as scaling and invariance for both single and joint entropy estimation have been proved. The objective of the work is to understand the mathematics behind the underlying concept.

Application of AI in the Field of Education: Dynamic Course Generation

Abstract: This research article examines the evolving relationship between Artificial Intelligence (AI) and Education, exploring its impact and potential. It covers adaptive learning systems, intelligent tutoring platforms, and emerging technologies, analyzing real-world applications and addressing ethical and implementation challenges. By synthesizing research findings, it contributes to understanding AI's role in transforming learning paradigms

Waste-Heat Renewable Gasifier Design through Taguchi's Method and MANFIS

Ensuring a global average temperature increase of no more than 1.5°C is crucial for aligning with the objectives of the Paris Agreement. The Intergovernmental Panel on Climate Change (IPCC) estimates that a 45% reduction in carbon dioxide emissions by 2030 or a 25% reduction by the same year is necessary to cap the temperature rise at 2°C. Failure to achieve this target by 2030 would necessitate substantial subsequent reductions to compensate for the delayed progress toward net zero emissions, potentially incurring higher costs. Elevating the focus on enhancing energy efficiency has emerged as a key strategy in expediting progress toward these climate goals. Hydrogen, renowned for its high efficiency and emission-free combustion, is anticipated to play a pivotal role in this endeavor. Traditionally, gasifiers have served as the conduit for transforming inexpensive fuels like coal or biomass into hydrogen. However, their energy-intensive nature poses a challenge.Addressing this inherent issue, an innovative approach known as the external-heating gasifier has been proposed and designed. The design of the external-heating gasifier represents a quintessential nonlinear and multiple-input multiple-output (MIMO) problem. To achieve optimal performance, the integration of Taguchi's method and inverse-model technology has been employed, with a simulation platform developed using the COMSOL Multiphysics software.Initially, Taguchi's method was utilized to determine the minimum number of experiments required for a full-factorial design and identify the most critical factors influencing product performance. Subsequently, an adaptive neuro-fuzzy inference system (ANFIS) was developed with a MIMO–ANFIS architecture. This system was employed to train an inverse model, facilitating the mapping of relationships between each input and output set. In essence, this approach enables the identification of manufacturing parameters that result in optimal performance. To illustrate the efficacy of the proposed method, a simulation study on an existing gasifier is presented.

Sensorless speed controller of an induction motor with MRAS-based model predictive control

Speed sensors play critical role in induction motor (IM) control systems. However, the use of these sensors cause several drawbacks such as increasing the overall cost, decreased reliability on control and increased noise problems in data acquisition. The sensorless control methods also have been proposed in control systems of IM to increase the overall performance. On the other hand, model predictive controllers (MPCs) which is one of the most recently used controller methods require the speed measurement to perform speed and torque control of an IM. In addition to control method, the topology of voltage source inverter (VSI) affects the overall efficiency of the IM drive because of its losses and total harmonic distortion (THD) ratios delivered by the topology. The main proposal of this study is to implement an IM drive built with a three-level active neutral point clamped (3L-ANPC) voltage source inverter (VSI) and to propose an MPC that eliminates sensor requirement in speed control of IM. The proposed IM drive takes benefits of multilevel inverter topology, which reduces the torque ripple, enables low switching frequency and provides decreased harmonic distortion. The switching pulses of ANPC VSI are generated by the proposed sensorless MPC method for controlling the torque of IM. The modelling and experimental verification of the ANPC based IM drive have been carried out and the sensorless MPC controller has been validated in this study. There two types of model reference adaptive system (MRAS) have been proposed to ensure precise estimation of IM speed that should be supplied to MPC controller. These MRAS methods are titled as the rotor flux based model reference adaptive system (MRASF), and the stator current based model reference adaptive system (MRASCC) according to estimation references. The proposed MPC-based IM drive has been analyzed under variable speed and torque circumstances to detect the efficiency and reliability of MRASF and MRASCC algorithms. The efficiency indicators such as flux and torque responses, and speed control of the suggested MPC IM drive are validated with experimental analysis. The results of revealed that the IM effectively tracks the reference speed and produces the required torque depending to the precise control of proposed MPC algorithm.

Increasing the raw contrast of VLT/SPHERE with the dark hole technique

Context. Imaging exoplanetary systems is essential to characterizing exoplanet atmospheres and orbits as well as circumstellar disks and to studying planet-disk interactions to understand the planet formation processes. Imaging exoplanets or circumstellar disks in the visible and near-infrared is challenging, however, because these objects are very faint relative to their star, even though only fractions of an arcsecond away. Coronagraphic instruments have already allowed the imaging of a few exoplanets, but their performance is limited by wavefront aberrations. Adaptive optics systems partly compensate for the Earth’s atmosphere turbulence, but they cannot fully control the wavefront. Some of the starlight leaks through the coronagraph and forms speckles in the astrophysical image. Focal plane wavefront control, used as a second stage after the adaptive optics system, has been proposed to minimize the speckle intensity within an area called the dark hole. Aims. We previously demonstrated the on-sky performance of dark hole techniques, pairwise probing coupled with electric field conjugation, using the apodized pupil Lyot coronagraph of the VLT/SPHERE instrument. In this paper, we probe their performance using the SPHERE four-quadrant phase mask coronagraph, and we demonstrate the interest of combining dark hole techniques and reference differential imaging. Methods. We used these dark hole techniques on-sky to create a dark hole in the narrow band around 1.7 |j.m observing HR 4796. We then recorded broadband images of HR 4796 and a reference star at the H band. Results. The dark hole techniques improved the H-band detection limit by a factor of three. The dark hole was stable from one star to a nearby star enabling reference differential imaging. Conclusions. This stability offers two new strategies of observation. First, one can quickly create a dark hole observing a bright star before pointing to a faint target star. Furthermore, one can couple dark hole techniques and reference differential imaging. A very interesting point is that the performance of these methods does not depend on the astrophysical signal.

Vital health indicator based state of health estimation of lithium-ion battery by adaptive neuro-fuzzy inference system

ABSTRACT The battery management system (BMS) in electric vehicles (EVs) relies on State of Health (SOH) assessment to monitor battery health and ensure performance. BMS can detect and prevent cell balance, thermal runaway, and extend its lifespan. Methodical charging and discharging under different circumstances reduce battery degradation. Grid Partitioning (GP), Subtractive Clustering (SC), and Fuzzy C-Means (FCM) from the Adaptive Neuro-Fuzzy Inference System (ANFIS) have been employed to predict battery SOH in this work. This study examines charge-discharge cycles of 8 lithium-ion pouch cells from the Oxford Battery Degradation Dataset. Charging cycle data extracts the five vital health indicators (VHIs), whereas discharging cycle data calculates cell capacity. For analyzing the effects of VHIs on SOH, a correlation matrix is formed. Three instances are defined depending on the individual VHIs impacts. Instance 1 utilizes VHI 1, VHI 2, and VHI 5, instance 2 utilizes VHI 3 and VHI 4, and instance 3 uses all VHIs as inputs. Due to enormous data points, only even-numbered cell data is utilized for training the ANFIS model, while odd-numbered cell data is used for validation and testing. In the best-case scenario, instance 1 gives the lowest root mean square error (RMSE) from all the methods.

Bioconjugated Antibody-Trojan Immune Converter Enhance Cancer Immunotherapy with Minimized Toxicity by Programmed Two-Step Immunomodulation of Myeloid Cells.

Current immune checkpoint blockade therapy (ICBT) predominantly targets T cells to harness the antitumor effects of adaptive immune system. However, the effectiveness of ICBT is reduced by immunosuppressive innate myeloid cells in tumor microenvironments (TMEs). Toll-like receptor 7/8 agonists (TLR7/8a) are often used to address this problem because they can reprogram myeloid-derived suppressor cells (MDSCs) and tumor-associated M2 macrophages, and boost dendritic cell (DC)-based T-cell generation; however, the systemic toxicity of TLR7/8a limits its clinical translation. Here, to address this limitation and utilize the effectiveness of TLR7/8a, this work suggests a programmed two-step activation strategy via Antibody-Trojan Immune Converter Conjugates (ATICC) that specifically targets myeloid cells by anti-SIRPα followed by reactivation of transiently inactivated Trojan TLR7/8a after antibody-mediated endocytosis. ATICC blocks the CD47-SIRPα ("don't eat me" signal), enhances phagocytosis, reprograms M2 macrophages and MDSCs,and increases cross-presentation by DCs, resulting in antigen-specific CD8+ T-cell generation in tumor-draining lymph nodes and TME while minimizing systemic toxicity. The local or systemic administration of ATICC improves ICBT responsiveness through reprogramming of the immunosuppressive TME, increased infiltration of antigen-specific CD8+ T cells,and antibody-dependent cellular phagocytosis. These results highlight the programmed and target immunomodulation via ATICC could enhance cancer immunotherapy with minimized systemic toxicities.

A verified low-level implementation and visualization of the adaptive exterior light and speed control system

A verified low-level implementation and visualization of the adaptive exterior light and speed control system

Data stream solution for decision-making processes: a general and adaptive system for decision support

ABSTRACT With the rise of new data generation processes and availability rates, new challenges appear linked with the information and pattern extraction for decisional processes. In these contexts, decision support systems (DSS) need to incrementally integrate the data flows to provide readable, understandable and meaningful conclusions. This research presents a novel adaptive DSS that offers updated and explainable knowledge for decision makers, especially in uncertain decision scenarios. Also, this work explores applications of this technology in some social fields, aiming to analyse the potential contribution of machine learning, combined with adaptive decision systems, in society.

Hematological Markers in Children and Adults with Atopic Dermatitis: A Retrospective Cohort Study.

Background Atopic dermatitis (AD) is a common chronic skin disease with an inflammatory pathophysiology that includes the activation of the innate and adaptive immune systems. Objectives We aimed to investigate the neutrophil-to-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR), eosinophil-to-lymphocyte ratio (ELR) and eosinophil-to-neutrophil ratio (ENR) in AD patients, according to age and disease severity. Methods This is a retrospective, population-based cohort study conducted between the years 2005 and 2020, comparing hematological markers of AD patients and sex-age-ethnicity matched controls. AD patients were further divided by age and disease severity (mild, moderate-to-severe AD). We created a decision tree to predict moderate-severe AD. Results A total of 13,928 patients with AD were included in this study: 6,828 adults and 7,100 children, with 13,548 controls. NLR, PLR, and ELR were lower in children compared to adults (p-values < 0.001). NLR, PLR, ELR, and ENR were increased in moderate-severe AD patients compared to mild AD patients (p-values < 0.001). PLR, ELR, and ENR were increased in AD patients versus controls (p-values < 0.001), with an additional increase in the NLR of moderate-to-severe AD patients. Patients with an ELR < 0.21, a PLR > 161, and ENR  0.016 should be considered high risk for developing severe AD, as well as patients with an ELR > 0.21 and age at diagnosis < 30 or age > 30 years and Mean Platelet Volume (MPV)  9. Conclusion Hematological ratios were significantly higher in moderate-to-severe AD patients, compared to mild AD patients. Hematological markers were lower in children with AD compared to adults, except for ENR, likely reflecting age-related changes in blood count parameters. These markers can assist in the management and follow-up of AD patients.

Decoding the commutable first hyperpolarisability of keto–enol tautomer using DFT and AI-based soft computing method

ABSTRACT Non-linear optical (NLO) parameter of a series of donor–acceptor substituted N-salicylidene-4-benzenesulfonylaniline (SBSA) derivatives with a contrast of first hyperpolarisability has been investigated using density functional-based theory. The derivative of SBSA with both the donor and acceptor substitution (NMe2-SBSA-NO2) exhibits the highest first hyperpolarisability ( ∼ 3 × 103 a.u in gas phase) value. Frequency and solvent polarity responsive behaviours of first hyperpolarisability are utilised to demonstrate the variation of NLO as switchable electrical parameter. Each of the studied compounds displays very high NLO response at frequency corresponding to the absorption maxima of their electronic state. Likewise, the increase of solvent polarity induces a substantial increase of first hyperpolarisability. The computation of NLO responsive first hyperpolarisability by varying the frequency and solvent dielectric constant within a wide domain is very tedious and time-consuming. To overcome the lacuna, we implemented machine learning tools such as artificial neural networks (ANNs), fuzzy logic (FL) and adaptive neuro-fuzzy inference system (ANFIS) to predict the computation data of the NMe2-SBSA-NO2 and moderately good results were obtained using ANFIS as soft computing tool.

Adaptive Network-Based Fuzzy Inference System–Proportional–Integral–Derivative Controller Based on FPGA and Its Application in Radiofrequency Ablation Temperature Control

The radiofrequency ablation temperature system is characterised by its time-varying, non-linear, and hysteretic nature. The application of PID controllers to the control of radiofrequency ablation temperature systems has a number of challenges, including overshoot, dependence on high-precision mathematical models, and difficulty in parameter tuning. Therefore, in order to improve the effectiveness of radiofrequency ablation temperature control, an adaptive network-based fuzzy inference system combined with an incremental PID controller was used to optimise the shortcomings of the PID controller in radiofrequency ablation temperature control. At the same time, the learning rate at the time of updating the consequence parameters was set by segmentation to solve the problem of poor control accuracy when the ANFIS-PID controller is implemented based on FPGA fixed-point decimals. Based on FPGA-in-the-loop simulation experiments and ex vivo experiments, the effectiveness of the ANFIS-PID controller in the temperature control of radiofrequency ablation was verified and compared with the PID controller under the same conditions. The experimental results show that the ANFIS-PID controller has a superior performance in terms of tracking capability and stability compared with the PID controller.

Expected performance of the Pyramid wavefront sensor with a laser guide star for 40 m class telescopes

Context. The use of artificial laser guide stars (LGS) is planned for the new generation of giant segmented mirror telescopes in order to extend the sky coverage of their adaptive optics systems. The LGS, being a 3D object at a finite distance, will have a large elongation that will affect its use with the Shack–Hartmann (SH) wavefront sensor. Aims. In this paper, we compute the expected performance for a Pyramid WaveFront Sensor (PWFS) using an LGS for a 40 m telescope affected by photon noise, and also extend the analysis to a flat 2D object as reference. Methods. We developed a new way to discretize the LGS, and a new, faster method of propagating the light for any Fourier filtering wavefront sensors (FFWFS) when using extended objects. We present the use of a sensitivity model to predict the performance of a closed-loop adaptive optic system. We optimized a point-source-calibrated interaction matrix to accommodate the signal of an extended object by computing optical gains using a convolutional model. Results. We find that the sensitivity drop, given the size of the extended laser source, is large enough to make the system operate in a low-performance regime given the expected return flux of the LGS. The width of the laser beam is identified as the limiting factor, rather than the thickness of the sodium layer. Even an ideal, flat LGS will have a drop in performance due to the flux of the LGS, and small variations in the return flux will result in large variations in performance. Conclusions. We conclude that knife-edge-like wavefront sensors, such as the PWFS, are not recommended for use with LGS for a 40 m telescope, as they will operate in a low-performance regime, given the size of the extended object.

Estimation of optimal process parameters for minimum wear in the application of SiC/RHA reinforced Al7075 hybrid composites using ANN, ANFIS, and GA

Increasing demand for high-performance materials has led to the exploration of composite materials for enhanced mechanical properties. In this study, a composite of silicon carbide particulate and rice husk ash (RHA) in varying proportions was utilized to reinforce an aluminum alloy (Al7075) hybrid composite fabricated through the stir casting technique. Microstructure examination via an optical microscope ensured the homogeneous distribution of reinforced particles. Wear was evaluated using a pin-on-disc apparatus, considering material factors (% of SiC and % of RHA) and mechanical wear factors (load applied, speed of rotation, and sliding distance). Experimental data were used to develop artificial neural network and adaptive neural fuzzy inference system models, which demonstrated high predictive accuracy. An objective function, formulated to minimize wear via regression analysis, guided the application of a genetic algorithm to determine optimal process parameters. The optimal combination, resulting in a minimum wear of 34.5 µm, comprised 12% SiC, 7% RHA, a sliding speed of 1.9 m/s, an applied load of 11.5 N, and a sliding distance of 715 mm. This study concludes with recommendations for further research and implications for composite material design and optimization.