Yazar "Tutsoy, Önder" seçeneğine göre listele
Listeleniyor 1 - 20 / 36
Sayfa Başına Sonuç
Sıralama seçenekleri
Öğe A Machine Learning-Based 10 Years Ahead Prediction of Departing Foreign Visitors by Reasons: A Case on Turkiye(Mdpi, 2022) Tutsoy, Önder; Tanrikulu, CeydaThe most important underlying reasons for marketing failures are incomplete understanding of customer wants and needs and the inability to accurately predict their future behaviors. This study develops a machine learning model to estimate the number of departing foreign visitors from Turkiye by reasons for the next 10 years to gain a deeper understanding of their future behaviors. The data between 2003 and 2021 are extensively analyzed, and a multi-dimensional model having a higher-order fractional-order polynomial structure is constructed. The resulting model can predict the 10 reasons of departing foreign visitors for the next 10 years and can update the predictions every year as new data becomes available as it has stable polynomial parameters. In addition, a batch-type genetic algorithm is modified to learn the unknown model parameters by considering the disruptions, such as the coup attempt in 2016 and the COVID-19 pandemic outbreak in 2019, termed as uncertainties. Thus, the model can estimate the overall behavior of the departing foreign visitors in the presence of uncertainties, which is the dominant character of the foreign visitors by their reasons. Furthermore, the developed model is utterly data-driven, meaning it can be trained with the data collected from different cities, regions, and countries. It is predicted that the departing foreign visitors for all reasons will increase at various rates between 2022 and 2031, while the increase in transit visitors is predicted to be higher than the others. The results are discussed, and suggestions are given considering the marketing science. This study can be helpful for global and local firms in tourism, governmental agencies, and civil society organizations.Öğe A novel adaptive PD-type iterative learning control of the PMSM servo system with the friction uncertainty in low speeds(Public Library Science, 2023) Riaz, Saleem; Qi, Rong; Tutsoy, Önder; Iqbal, JamshedHigh precision demands in a large number of emerging robotic applications strengthened the role of the modern control laws in the position control of the Permanent Magnet Synchronous Motor (PMSM) servo system. This paper proposes a learning-based adaptive control approach to improve the PMSM position tracking in the presence of the friction uncertainty. In contrast to most of the reported works considering the servos operating at high speeds, this paper focuses on low speeds in which the friction stemmed deteriorations become more obvious. In this paper firstly, a servo model involving the Stribeck friction dynamics is formulated, and the unknown friction parameters are identified by a genetic algorithm from the offline data. Then, a feedforward controller is designed to inject the friction information into the loop and eliminate it before causing performance degradations. Since the friction is a kind of disturbance and leads to uncertainties having time-varying characters, an Adaptive Proportional Derivative (APD) type Iterative Learning Controller (ILC) named as the APD-ILC is designed to mitigate the friction effects. Finally, the proposed control approach is simulated in MATLAB/Simulink environment and it is compared with the conventional Proportional Integral Derivative (PID) controller, Proportional ILC (P-ILC), and Proportional Derivative ILC (PD-ILC) algorithms. The results confirm that the proposed APD-ILC significantly lessens the effects of the friction and thus noticeably improves the control performance in the low speeds of the PMSM.Öğe A novel deep machine learning algorithm with dimensionality and size reduction approaches for feature elimination: thyroid cancer diagnoses with randomly missing data(Oxford Univ Press, 2024) Tutsoy, Önder; Sumbul, Hilmi ErdemThyroid cancer incidences endure to increase even though a large number of inspection tools have been developed recently. Since there is no standard and certain procedure to follow for the thyroid cancer diagnoses, clinicians require conducting various tests. This scrutiny process yields multi-dimensional big data and lack of a common approach leads to randomly distributed missing (sparse) data, which are both formidable challenges for the machine learning algorithms. This paper aims to develop an accurate and computationally efficient deep learning algorithm to diagnose the thyroid cancer. In this respect, randomly distributed missing data stemmed singularity in learning problems is treated and dimensionality reduction with inner and target similarity approaches are developed to select the most informative input datasets. In addition, size reduction with the hierarchical clustering algorithm is performed to eliminate the considerably similar data samples. Four machine learning algorithms are trained and also tested with the unseen data to validate their generalization and robustness abilities. The results yield 100% training and 83% testing preciseness for the unseen data. Computational time efficiencies of the algorithms are also examined under the equal conditions.Öğe A Novel Exploration-Exploitation-Based Adaptive Law for Intelligent Model-Free Control Approaches(IEEE-Inst Electrical Electronics Engineers Inc, 2023) Tutsoy, Önder; Barkana, Duygun Erol; Balikci, KemalModel-free control approaches require advanced exploration-exploitation policies to achieve practical tasks such as learning to bipedal robot walk in unstructured environments. In this article, we first construct a comprehensive exploration-exploitation policy that carries quality knowledge about the long-term predictor and the control policy, and the control signal of the model-free algorithms. Therefore, the developed model-free algorithm continues exploration by adjusting its unknown parameters until the desired learning and control are accomplished. Second, we provide an utterly model-free adaptive law enriched with the exploration-exploitation policy and derived step-by-step using the exact analogy of the model-based solution. The obtained adaptive control law considers the control signal saturation and the control signal (input) delay. Performed Lyapunov stability analysis ensures the convergence of the adaptive law that can also be used for intelligent control approaches. Third, we implement the adaptive algorithm in real time on a challenging benchmark system: a fourth-order, coupled dynamics, input saturated, and time-delayed underactuated manipulator. The results show that the proposed adaptive algorithm explores larger state-action spaces and treats the vanishing gradient problem in both learning and control. Also, we notice from the results that the learning and control properties of the adaptive algorithm are optimized as required.Öğe A Novel Parametric Model for the Prediction and Analysis of the COVID-19 Casualties(IEEE-Inst Electrical Electronics Engineers Inc, 2020) Tutsoy, Önder; Colak, Sule; Polat, Adem; Balikci, KemalCoronavirus disease (COVID-19) outbreak has affected billions of people, where millions of them have been infected and thousands of them have lost their lives. In addition, to constraint the spread of the virus, economies have been shut down, curfews and restrictions have interrupted the social lives. Currently, the key question in minds is the future impacts of the virus on the people. It is a fact that the parametric modelling and analyses of the pandemic viruses are able to provide crucial information about the character and also future behaviour of the viruses. This paper initially reviews and analyses the Susceptible-Infected-Recovered (SIR) model, which is extensively considered for the estimation of the COVID-19 casualties. Then, this paper introduces a novel comprehensive higher-order, multi-dimensional, strongly coupled, and parametric Suspicious-Infected-Death (SpID) model. The mathematical analysis results performed by using the casualties in Turkey show that the COVID-19 dynamics are inside the slightly oscillatory, stable (bounded) region, although some of the dynamics are close to the instability region (unbounded). However, analysis with the data just after lifting the restrictions reveals that the dynamics of the COVID-19 are moderately unstable, which would blow up if no actions are taken. The developed model estimates that the number of the infected and death individuals will converge zero around 300 days whereas the number of the suspicious individuals will require about a thousand days to be minimized under the current conditions. Even though the developed model is used to estimate the casualties in Turkey, it can be easily trained with the data from the other countries and used for the estimation of the corresponding COVID-19 casualties.Öğe Active fault-tolerant control of quadrotor UAVs with nonlinear observer-based sliding mode control validated through hardware in the loop experiments(Pergamon-Elsevier Science Ltd, 2023) Ahmadi, Karim; Asadi, Davood; Merheb, Abdelrazzak; Yaser Nabavi-Chashmi, Seyed-; Tutsoy, ÖnderMultirotor unmanned aerial vehicles (UAV) are highly prone to motor faults, which can arise from defective motors or damaged propellers. Motor faults severely change the multirotor UAV's dynamics and therefore endanger flight safety and reliability since the controller loses its efficiency. To cope with such crucial problems, a fault-tolerant controller is proposed in this paper for full control of a quadrotor UAV with motor faults. The proposed fault-tolerant approach consists of the nonlinear observer technique and the Sliding Mode Control (SMC). The designed novel nonlinear observer predicts the effects of motor faults on quadrotor dynamics and it is augmented with an SMC to create a fault-tolerant controller. In addition, the nonlinear observer enhances the robustness of the SMC against the uncertainties and disturbances acting on the quadrotor during flight. Any actuator fault will be treated as a disturbance detected by the nonlinear observer and will be attenuated directly by the proposed SMC. The disturbance attenuation capability achieved by the nonlinear observer decreases the amount of control action expected from the SMC, which results in advanced robustness without sacrificing the nominal control performance. The performance of the proposed nonlinear observer SMC (NOSMC) is demonstrated through simulations and testbed experiments. The results show that the proposed fault-tolerant controller effectively recovers the full control of the quadrotor with motor faults up to 40% while tracking the predefined trajectory and rotation angles as desired.Öğe Adaptive estimator design for unstable output error systems: A test problem and traditional system identification based analysis(Sage Publications Ltd, 2015) Tutsoy, Önder; Colak, SuleA key open question in adaptive estimator design is how to assure that the parameters of the proposed algorithms are converging to their almost correct solutions; hence, the learning algorithm is unbiased. Moreover, determining the speed of parameter convergence is important as it provides insight about the performance of the learning algorithms. The main contributions of the article are fourfold: the first one is that the article, initially, introduces an adaptive estimator to learn the discounted Q-function and approximate optimal control policy without requiring linear, discrete time, unstable output error system dynamics, but using only the noisy system measurements. The simulation results show that the adaptive estimator minimizes the stochastic cost function and temporal difference error and also learns the approximate Q-function together with the control policy. The second one is consideration of a different approach by taking a simple test problem to investigate issues associated with the Q-function's representation and parametric convergence. In particular, the terminal convergence problem is analyzed with a known optimal control policy where the aim is to accurately learn only the Q-function. It is parameterized by terms which are functions of the unknown plant's parameters and the Q-function's discount factor, and their convergence properties are analyzed and compared with the adaptive estimator. The third one is to show that even though the adaptive estimator with a large Q-function discount factor yields larger control feedback gains, so that faster state converges upright, the learning problem is badly conditioned; hence, the parameter convergence is sluggish, as the Q-function discount factor approaches the inverse of the dominant pole of the unstable system. Finally, the fourth one is comparison of the state output learned by the adaptive estimator with the ones obtained from traditional system identification algorithms. Simulation result for a higher order unstable output error system shows that the adaptive estimator closely follows the real system output whereas the system identification algorithms do not.Öğe An analysis of value function learning with piecewise linear control(Taylor & Francis Ltd, 2016) Tutsoy, Önder; Brown, MartinReinforcement learning (RL) algorithms attempt to learn optimal control actions by iteratively estimating a long-term measure of system performance, the so-called value function. For example, RL algorithms have been applied to walking robots to examine the connection between robot motion and the brain, which is known as embodied cognition. In this paper, RL algorithms are analysed using an exemplar test problem. A closed form solution for the value function is calculated and this is represented in terms of a set of basis functions and parameters, which is used to investigate parameter convergence. The value function expression is shown to have a polynomial form where the polynomial terms depend on the plant's parameters and the value function's discount factor. It is shown that the temporal difference error introduces a null space for the differenced higher order basis associated with the effects of controller switching (saturated to linear control or terminating an experiment) apart from the time of the switch. This leads to slow convergence in the relevant subspace. It is also shown that badly conditioned learning problems can occur, and this is a function of the value function discount factor and the controller switching points. Finally, a comparison is performed between the residual gradient and TD(0) learning algorithms, and it is shown that the former has a faster rate of convergence for this test problem.Öğe Bozulmuş İnsansız Hava Araçları İçin Minimum Mesafe Ve Minimum Zaman Optimal Yol Planlamalarının Çok Boyutlu Makine Öğrenmesi Yaklaşımları Ile Başarımı(2023) Tutsoy, Önder; Polat, Adem; Hendoustanı, Davood Asadıİnsansız Hava Araçları (İHA) bilinmeyen ortamlarda, dinamik çevre koşullarında görev yapabilmekte ve beklenen ya da beklenmeyen birçok arızalarla karşılaşabilmektedirler. Bu sebeplerden dolayı otonom bir İHA, acil durumlarda minimum mesafe veya minimum sürede en uygun konuma inebilecek özellikler ile donatılmalıdırlar. Hasarlar ve bozulmalar, kararsız (unstable) ve belirsiz (uncertain) İHA dinamiklerini (dynamics) değiştirdiğinden, yol planlama (path planning) algoritmaları uyarlanabilir (adaptive) ve modelden bağımsız (model free) olmalıdır. Bunların yanında, İHA için tasarlanan yol planlama optimizasyon problemleri, gerçek zamanlı uygulamaların başarımı için hayati olan, aktüatör doygunluklarını (actuator saturations), kinematik ve dinamik kısıtlamaları (kinematic and dynamic constraints) dikkate almalıdır. Bu nedenle bu projede, bir İHA?nın bozulması sonucu ortaya çıkan parametrik belirsizlikleri (parametric uncertainties) ve çeşitli kısıtlamaları dikkate alan üç boyutlu yol planlama algoritmaları quadrotorlar için geliştirmiştir. Bu projede, öteleme (translation), dönme (rotation), Euler açıları (Euler angle), ilgili minimum zaman ve minimum mesafe kontrol sinyalleri çok boyutlu parçacık sürü optimizasyonu (multi-dimensional particle swarm optimization) ve çok boyutlu genetik algoritması (multi-dimensional genetic algorithm) meta-sezgisel makine öğrenmesi yaklaşımları ile elde edilmiştir. Algoritmalar hem simülasyon ortamında hem de deneysel ortamlarda değerlendirilmiş ve performansları karşılaştırılmıştır. Bu projenin bütçesi ile sadece lisans, yüksek lisans ve doktora öğrencilerine burs sağlanmış ve bu alanda yeni projelerimizin alt yapısı oluşturularak yeni projeler sunulmuştur. Projenin sonuçlarında, çok boyutlu genetik algoritmanın kısıtlamalar altında daha kısa minimum mesafe ve minimum zaman yolları üretebildiği gösterilip doğrulanmıştır. Gerçek zamanlı deneyler, quadrotorun mevcut maksimum rotor hızlarını kullanarak üretilen hedef yolu tam olarak izleyebildiği ayrıca kanıtlanmıştır.Öğe Chaotic dynamics and convergence analysis of temporal difference algorithms with bang-bang control(Wiley, 2016) Tutsoy, Önder; Brown, MartinReinforcement learning is a powerful tool used to obtain optimal control solutions for complex and difficult sequential decision making problems where only a minimal amount of a priori knowledge exists about the system dynamics. As such, it has also been used as a model of cognitive learning in humans and applied to systems, such as humanoid robots, to study embodied cognition. In this paper, a different approach is taken where a simple test problem is used to investigate issues associated with the value function's representation and parametric convergence. In particular, the terminal convergence problem is analyzed with a known optimal control policy where the aim is to accurately learn the value function. For certain initial conditions, the value function is explicitly calculated and it is shown to have a polynomial form. It is parameterized by terms that are functions of the unknown plant's parameters and the value function's discount factor, and their convergence properties are analyzed. It is shown that the temporal difference error introduces a null space associated with the finite horizon basis function during the experiment. The learning problem is only non-singular when the experiment termination is handled correctly and a number of (equivalent) solutions are described. Finally, it is demonstrated that, in general, the test problem's dynamics are chaotic for random initial states and this causes digital offset in the value function learning. The offset is calculated, and a dead zone is defined to switch off learning in the chaotic region. Copyright (C) 2015 John Wiley & Sons, Ltd.Öğe COVID-19 Epidemic and Opening of the Schools: Artificial Intelligence-Based Long-Term Adaptive Policy Making to Control the Pandemic Diseases(IEEE-Inst Electrical Electronics Engineers Inc, 2021) Tutsoy, ÖnderEven though the COVID-19 pandemic has endured to be a serious threat for the societies, the state authorities have been seeking policies to re-open the schools and universities. It is clear that opening of the schools will cause more COVID-19 casualties, but the key question is how many students should attend the schools daily while keeping the casualties under control. In this paper, an artificial intelligence based long-term policy making algorithm has been developed to generate time varying policies for opening of the schools part-by-part. The key aim of the algorithm is to produce policies which maximize the number of the students attending the schools while minimizing the pandemic casualties under the worst-case uncertainties. The proposed algorithm consists of a multi-input-multi-output, uncertain, and adaptive background parametric model which is externally manipulated by the produced adaptive policy. Its long-term predictor assesses the possible future casualties under the current policy and its policy maker generates alternative solutions that minimize the future casualties. The results confirm that the proposed algorithm is able to generate effective policies which minimize the COVID-19 casualties while maximize the number of the students attending the schools under the worst-case uncertainties.Öğe CPG BASED RL ALGORITHM LEARNS TO CONTROL OF A HUMANOID ROBOT LEG(Acta Press, 2015) Tutsoy, ÖnderAutonomous humanoid robots equipped with learning capabilities are able to learn tasks such as sitting down, standing up, balancing, walking and running. In this paper, central pattern generator (CPG) based reinforcement learning (RL) algorithm is applied to a robot leg with 3-links to balance it at upright by reducing dimensionality of the learning problem from 6 to 2. MapleSim is used for the leg modelling and this model is combined with the CPG based RL algorithm by utilizing Modelica and Maple software properties. Maple multi-body analysis template and Modelica custom component template allow symbolic inverse kinematics solution for the leg to be obtained. Thus, time and information lost in case of using a numerical solution are eliminated. The learning results show that the value function is maximized, temporal difference error is significantly reduced to zero and the leg is balanced at upright.Öğe Deep self-supervised machine learning algorithms with a novel feature elimination and selection approaches for blood test-based multi-dimensional health risks classification(Bmc, 2024) Tutsoy, Önder; Koc, Gizem GulBackgroundBlood test is extensively performed for screening, diagnoses and surveillance purposes. Although it is possible to automatically evaluate the raw blood test data with the advanced deep self-supervised machine learning approaches, it has not been profoundly investigated and implemented yet.ResultsThis paper proposes deep machine learning algorithms with multi-dimensional adaptive feature elimination, self-feature weighting and novel feature selection approaches. To classify the health risks based on the processed data with the deep layers, four machine learning algorithms having various properties from being utterly model free to gradient driven are modified.ConclusionsThe results show that the proposed deep machine learning algorithms can remove the unnecessary features, assign self-importance weights, selects their most informative ones and classify the health risks automatically from the worst-case low to worst-case high values.Öğe Design and Comparison Base Analysis of Adaptive Estimator for Completely Unknown Linear Systems in the Presence of OE Noise and Constant Input Time Delay(Wiley, 2016) Tutsoy, ÖnderIn this paper, an adaptive estimator (AE) is introduced to learn the approximate Q-function and control policy by only using the noisy states and control signals of the unknown linear, discrete time systems having constant input time delay. The system measurements are uncertain owing to output error (OE)-type noise acting randomly on the system measurements. Therefore, this research differs from the designed AE in the literature since previous research ignores the role of the external random disturbances on AE-based learning. In order to compare the AE-based learning results with traditional system identification (SI) approaches, a modified version of the OE model structure for unstable systems is reviewed and parameters of a second-order unstable system with constant input time delay are identified. The simulation results demonstrate that the designed AE efficiently minimizes the stochastic cost function and the temporal difference error by learning the approximate solution for the Hamilton-Jacobi-Bellman (HJB) equation. It is noted that the error in the Q-function obtained with the AE is slightly larger than the Q-function attained with the identified OE parameters. However, as the noise standard deviation increases, the error in the AE-based learning results reduces whereas the error in the OE-based learning increases. This indicates that even though the added random noise deteriorates the performance of the OE predictor, it improves the learning efficiency of the AE since it acts like exploration noise.Öğe Design and Implementation of a Facial Character Analysis Algorithm for Humanoid Robots(Cambridge Univ Press, 2019) Gongor, Fatma; Tutsoy, ÖnderHumanoid robots (HR) equipped with a sophisticated facial character analysis (FCA) algorithm can able to initiate crucial improvements in human-robot interactions. This paper, for the first time in the literature, proposes a three-stage FCA algorithm for the HR. At the initial stage of this algorithm, the HR detects the face with the Viola-Jones algorithm, and then important facial distance measurements are obtained with the geometric-based facial distance measurement technique. Finally, the measured facial distances are evaluated with the physiognomy science to reveal the characteristic properties of a person. Even though the proposed algorithm can be implemented to all HR, in this paper, it has been specifically applied to NAO HR. The reliability of the developed FCA algorithm is verified by analyzing each terminal decision about the character and its connection with the measured facial distances in the anatomy science.Öğe Design of a completely model free adaptive control in the presence of parametric, non-parametric uncertainties and random control signal delay(Elsevier Science Inc, 2018) Tutsoy, Önder; Barkana, Duygun Erol; Tugal, HarunIn this paper, an adaptive controller is developed for discrete time linear systems that takes into account parametric uncertainty, internal-external non-parametric random uncertainties, and time varying control signal delay. Additionally, the proposed adaptive control is designed in such a way that it is utterly model free. Even though these properties are studied separately in the literature, they are not taken into account all together in adaptive control literature. The Q-function is used to estimate long-term performance of the proposed adaptive controller. Control policy is generated based on the long-term predicted value, and this policy searches an optimal stabilizing control signal for uncertain and unstable systems. The derived control law does not require an initial stabilizing control assumption as in the ones in the recent literature. Learning error, control signal convergence, minimized Q-function, and instantaneous reward are analyzed to demonstrate the stability and effectiveness of the proposed adaptive controller in a simulation environment. Finally, key insights on parameters convergence of the learning and control signals are provided. (C) 2018 ISA. Published by Elsevier Ltd. All rights reserved.Öğe Development of a Multi-Dimensional Parametric Model With Non-Pharmacological Policies for Predicting the COVID-19 Pandemic Casualties(IEEE-Inst Electrical Electronics Engineers Inc, 2020) Tutsoy, Önder; Polat, Adem; Colak, Sule; Balikci, KemalCoronavirus Disease 2019 (COVID-19) has spread the world resulting in detrimental effects on human health, lives, societies, and economies. The state authorities mostly take non-pharmacological actions against the outbreak since there are no confirmed vaccines or treatments yet. In this paper, we developed Suspicious-Infected-Death with Non-Pharmacological policies (SpID-N) model to analyze the properties of the COVID-19 casualties and also estimate the future behavior of the outbreak. We can state the key contributions of the paper with three folds. Firstly, we propose the SpID-N model covering the higher-order internal dynamics which cause the peaks in the casualties. Secondly, we parametrize the non-pharmacological policies such as the curfews on people with chronic disease, people age over 65, people age under 20, restrictions on the weekends and holidays, and closure of the schools and universities. Thirdly, we explicitly incorporate the internal and coupled dynamics of the model with these multi-dimensional non-pharmacological policies. The corresponding higher-order and strongly coupled model has utterly unknown parameters and we construct a batch type Least Square (LS) based optimization algorithm to learn these unknown parameters from the available data. The parametric model and the predicted future casualties are analyzed extensively.Öğe Doctor Robots: Design and Implementation of a Heart Rate Estimation Algorithm(Springer, 2022) Gongor, Fatma; Tutsoy, ÖnderPopulations are ageing and the healthcare costs are increasing accordingly. Humanoid Robots (HRs) perform basic but crucial health checkups such as the heart rate can efficiently meet the healthcare demands. This paper develops a 9-stage heart rate estimation algorithm and implements it to an HR. The 9-stages cover the recognition of the face with the Viola-Jones algorithm, determination of the facial regions with the geometric-based facial distance measurement technique, extraction of the forehead and cheek regions, tracking of these facial regions with the Hierarchical Multi Resolution algorithm, decomposition of the facial regions in the Red-Green-Blue (RGB) color channels, averaging and normalization of the RGB color data, elimination of the artifacts with the Adaptive Independent Component Analysis (ICA) technique, calculating the power spectrum of the data with the Fast Fourier Transform (FFT) technique, and finally determining the peaks inside the threshold reflecting the human heart rate boundaries. One of the key contributions of this paper is building and incorporating the Hierarchical Multi Resolution technique in the heart rate estimation algorithm to eliminate the deteriorating effects of the human and camera motions. A further contribution of this paper is generating a rule-based approach to discard the effects of the sudden movements. These two contributions have noticeably improved the accuracy of the heart rate estimation algorithm in the dynamic environments. The algorithm has been assessed extensively with 5 different experimental scenarios consisting of 33 conditions.Öğe Estimation of DC Motor Parameters Using Least Square-based Optimization Algorithm(Institute of Electrical and Electronics Engineers Inc., 2023) Tolun, Omer Can; Tutsoy, ÖnderIn daily life, Direct Current (DC) motors are employed in virtually all applications due to their ease of operation, simple construction, and affordability. Therefore, a proper mathematical model for DC motors is essential for developing model-based controllers and predicting system responses. In this paper, the Least Square-based Autoregressive-eXogenous (LS-ARX) optimization algorithm has been developed to estimate unknown parameters of the DC motor model. Furthermore, the unknown parameters of the motor have been estimated by using the Nonlinear Least Square (NLS) and Pattern Search (PS) methods from the MATLAB optimization toolbox. For the purpose of estimating the DC motor parameters, an experimental setup has been constructed to measure the angular velocity of the motor utilizing an Arduino and two different sensors (photon interrupter and hall-effect sensors). To transfer data from the Arduino program to the MATLAB environment, a serial connection has been established between the Arduino and the Python program. Utilizing the proposed algorithm and the MATLAB optimization toolbox, the accuracy of the estimation process has been evaluated and compared through coefficient of determination. As a result of the comparison, it can be observed that the LS-ARX optimization algorithm is extremely robust, and the unknown parameters of the DC motor have been estimated with a high degree of accuracy. © 2023 IEEE.Öğe Graph Theory Based Large-Scale Machine Learning With Multi-Dimensional Constrained Optimization Approaches for Exact Epidemiological Modeling of Pandemic Diseases(IEEE Computer Soc, 2023) Tutsoy, ÖnderMulti-dimensional prediction models of the pandemic diseases should be constructed in a way to reflect their peculiar epidemiological characters. In this paper, a graph theory-based constrained multi-dimensional (CM) mathematical and meta-heuristic algorithms (MA) are formed to learn the unknown parameters of a large-scale epidemiological model. The specified parameter signs and the coupling parameters of the sub-models constitute the constraints of the optimization problem. In addition, magnitude constraints on the unknown parameters are imposed to proportionally weight the input-output data importance. To learn these parameters, a gradient-based CM recursive least square (CM-RLS) algorithm, and three search-based MAs; namely, the CM particle swarm optimization (CM-PSO), the CM success history-based adaptive differential evolution (CM-SHADE), and the CM-SHADEWO enriched with the whale optimization (WO) algorithms are constructed. The traditional SHADE algorithm was the winner of the 2018 IEEE congress on evolutionary computation (CEC) and its versions in this paper are modified to create more certain parameter search spaces. The results obtained under the equal conditions show that the mathematical optimization algorithm CM-RLS outperforms the MA algorithms, which is expected since it uses the available gradient information. However, the search-based CM-SHADEWO algorithm is able to capture the dominant character of the CM optimization solution and produce satisfactory estimates in the presence of the hard constraints, uncertainties and lack of gradient information.
