Yazar "Altun, Yusuf" seçeneğine göre listele
Listeleniyor 1 - 14 / 14
Sayfa Başına Sonuç
Sıralama seçenekleri
Öğe Çeyrek Taşıt Aktif Süspansiyon Sistemi için LQR ve LQI DenetleyicilerininKarşılaştırılması(2017) Altun, YusufBu çalışma kara araçları için yol tutuşu ve yolcu konforu bakımından çok önemli bir yere sahip olan aktif süspansiyon sisteminin kontrolünü sunar. Kontrol sistemi için literatürde çeyrek taşıt modeli olarak bilinen ve aracın dörtte bir kütlesiyle tek teker sisteminden oluşan model kullanılmıştır. Öncelikle pasif ve aktif süspansiyon sistemlerinin matematiksel modelleri ortaya konulmuş ve aktif süspansiyon sistemi için kontrolörler tasarlanmıştır. Kontrolör tasarımları lineer matris eşitsizlikleri (LMI) ile optimizasyon yapılarak elde edilmiştir. Çalışmada lineer quadratik regülatör (LQR) kontrol ile lineer quadratik integral (LQI) kontrol tasarımları yapılmış ve yol bozucularına karşı performansları kıyaslanmıştır.Öğe Classification of Vehicles in Traffic and Detection Faulty Vehicles by Using ANN Techniques(Ieee, 2017) Başer, Ekrem; Altun, YusufNumber of vehicles in use worldwide per person and depending on this traffic accidents increase day by day. Until today, various studies have been made on detection of the faulty vehicles in traffic. This study; in accordance with this purpose, is realized in order to recognize, classify vehicles on a photo or video that captured on a autoban or road which has two lanes and detect faulty ones between them by using ANN(Artificial Neural Network) and image processing techniques. In this system that developed by using Convolutional Neural Network, the vehicles can be recognized, classified according to their vehicle types and faulty vehicles can be detected with drawing a frame around them.Öğe An Efficient Approach for Automatic Fault Classification Based on Data Balance and One-Dimensional Deep Learning(Mdpi, 2024) Ileri, Ugur; Altun, Yusuf; Narin, AliPredictive maintenance (PdM) is implemented to efficiently manage maintenance schedules of machinery and equipment in manufacturing by predicting potential faults with advanced technologies such as sensors, data analysis, and machine learning algorithms. This paper introduces a study of different methodologies for automatically classifying the failures in PdM data. We first present the performance evaluation of fault classification performed by shallow machine learning (SML) methods such as Decision Trees, Support Vector Machines, k-Nearest Neighbors, and one-dimensional deep learning (DL) techniques like 1D-LeNet, 1D-AlexNet, and 1D-VGG16. Then, we apply normalization, which is a scaling technique in which features are shifted and rescaled in the dataset. We reapply classification algorithms to the normalized dataset and present the performance tables in comparison with the first results we obtained. Moreover, in contrast to existing studies in the literature, we generate balanced dataset groups by randomly selecting normal data and all faulty data for all fault types from the original dataset. The dataset groups are generated with 100 different repetitions, recording performance scores for each one and presenting the maximum scores. All methods utilized in the study are similarly employed on these groups. From these scores, the use of 1D-LeNet deep learning classifiers and feature normalization resulted in achieving the highest overall accuracy and F1-score performance of 98.50% and 98.32%, respectively. As a result, the goal of this study was to develop an efficient approach for automatic fault classification, leveraging data balance, and additionally, to provide an analysis of one-dimensional deep learning and shallow machine learning-based classification methods. In light of the experimentation and comparative analysis, this study successfully achieves its stated goal by demonstrating that one-dimensional deep learning and data balance collectively emerge as the optimal approach, offering good prediction accuracy.Öğe Gain scheduling LQI controller design for LPV descriptor systems and motion control of two-link flexible joint robot manipulator(Balikesir University, 2018) Altun, YusufThis paper proposes a gain scheduling linear quadratic integral (LQI) servo controller design, which is derived from linear quadratic regulator (LQR) optimal control, for non-singular linear parameter varying (LPV) descriptor systems. It is assumed that state space matrices are non-singular since many mechanical systems do not have any non-singular matrices such as the natural state space forms of robotic manipulator, pendulum and suspension systems. A controller design is difficult for the systems due to rational LPV case. Therefore, the proposed gain scheduling controller is designed without the difficulty. Accordingly, the motion control design is implemented for two-link flexible joint robotic manipulator. Finally, the control system simulation is performed to prove the applicability and performance. © 2018 Balikesir University. All rights reserved.Öğe Improving YOLO Detection Performance of Autonomous Vehicles in Adverse Weather Conditions Using Metaheuristic Algorithms(Mdpi, 2024) Ozcan, Ibrahim; Altun, Yusuf; Parlak, CevahirDespite the rapid advances in deep learning (DL) for object detection, existing techniques still face several challenges. In particular, object detection in adverse weather conditions (AWCs) requires complex and computationally costly models to achieve high accuracy rates. Furthermore, the generalization capabilities of these methods struggle to show consistent performance under different conditions. This work focuses on improving object detection using You Only Look Once (YOLO) versions 5, 7, and 9 in AWCs for autonomous vehicles. Although the default values of the hyperparameters are successful for images without AWCs, there is a need to find the optimum values of the hyperparameters in AWCs. Given the many numbers and wide range of hyperparameters, determining them through trial and error is particularly challenging. In this study, the Gray Wolf Optimizer (GWO), Artificial Rabbit Optimizer (ARO), and Chimpanzee Leader Selection Optimization (CLEO) are independently applied to optimize the hyperparameters of YOLOv5, YOLOv7, and YOLOv9. The results show that the preferred method significantly improves the algorithms' performances for object detection. The overall performance of the YOLO models on the object detection for AWC task increased by 6.146%, by 6.277% for YOLOv7 + CLEO, and by 6.764% for YOLOv9 + GWO.Öğe Linear parameter varying feedforward control synthesis using parameter-dependent Lyapunov function(Springer, 2014) Altun, Yusuf; Gülez, KayhanThis paper presents the dynamic feedforward control synthesis for linear parameter varying (LPV) systems. It is assumed that all system matrices are dependent on varying parameters, which are measurable with sensor or observable. The parameters have bounded variation rates. Parameter-dependent Lyapunov function is used for the feedforward control synthesis such that the robust stability is assured for all varying parameters at the time of the operation. The method is formulated in terms of linear matrix inequalities for LPV feedforward controller that guarantees the stability of the transfer matrix having -gain. This compensator is designed by adding on the feedback controller in two degrees of freedom control configuration. This controller can be used for the disturbance attenuation or decreasing the tracking error. The numerical examples and simulations are given to provide the applicability of the proposed solution.Öğe A New Extended LMI-based Robust Gain Scheduled State Feedback H-2 Controller Design(Inst Control Robotics & Systems, Korean Inst Electrical Engineers, 2017) Altun, YusufThis paper proposes an improved robust H-2 state feedback control synthesis for the Linear Parameter Varying (LPV) systems by attaining the affine quadratic stability. In place of standard H-2 computation in the literature, a new H-2 computation based on extended Linear Matrix Inequality (LMI) is improved by means of the slack variable, where it is obtained by separation Lyapunov matrix from system matrix. State feedback H-2 synthesis is improved for the systems, and is more effective and less conservative than the common ones in the literature. Therefore, the less conservative results are obtained for gain scheduling controller design for LPV systems. The numerical examples are presented to show the superiority of the proposed controller design.Öğe Non-algorithmic Robust Static Output Feedback Control Designs for Parametric Uncertain Systems(Springer International Publishing Ag, 2018) Altun, YusufThis paper proposes novel static output feedback (SOF) control designs for the continuous-time systems under parametric uncertainty and linear time-invariant systems. The convex solutions based on the finite-dimensional linear matrix inequality are improved without any algorithmic approaches, thanks to the slack variable such that the solutions are achieved by isolation Lyapunov matrix from the system and controller matrix. In addition, the method is extended to H-2 SOF control design for the systems. Moreover, less conservative results are provided via the isolation in terms of the robust stability against the uncertainties. Finally, several numerical examples are presented to demonstrate the applicability of the proposed controller design.Öğe Otonom Araçla Genetik Algoritma Kullanılarak Haritalama ve Lokasyon(2020) Demir, Merve Nur; Altun, YusufTeknolojik gelişmeler ve bu zamana kadar biriken bilgilerin ışığında otonom sistemlerde muazzam bir ilerlemekaydedilmiştir. Bu sayede otonom sistemler çarpışmadan kaçınma, trafik işareti tespiti, haritalama vb. sayısız akıllıişlevleri gerçekleştirebilmektedir. Gerçek zamanlı otonom araçların en zorlu problemi aracın kendi kendineharitalandırma ve lokasyon işlemlerini yapabilmesidir. Genetik Algoritma (GA) kullanarak optimize edilmişlokasyon uygulaması ile otonom araçlar için sürüş güvenliğinin artması beklenmektedir. Bu çalışma da lazertabanlıbir lokalizasyon ve haritalama tekniğinin üzerine odaklanılmıştır. Gerçekleştirilen sistemde sanal bir test ortamıkurulmuş ve bir otonom araç üzerinde denemeler yapılmıştır. Çalışma kapsamında sanal makineler oluşturularaküzerlerine Linux işletim sistemi kurulmuştur. Sonra bu sanal makinelere ROS ortamında TurtleBot3 kurulmuş veiç mekân lokalizasyonu yapılarak bir harita elde edilmiştir. Bu harita genetik algoritma ile en kısa mesafelerinbulunmasını sağlamak için kullanılmaktadır. Gözlemler neticesinde simülasyon ortamındaki robot yüksekbaşarımla istenilen konuma gidebildiği sonucuna ulaşılmıştır.Öğe Parazitik Elemanların DC-DC Buck Dönüştürücünün PID Kontrolüne Etkisinin İncelenmesi(Duzce University, 2024) İnci, Mustafa; Altun, YusufDC-DC dönüştürücü devreleri Elektrikli Araçlar gibi birçok elektronik cihazda gerilimi belirli düzeye ayarlamak için kullanılmaktadır. DC-DC dönüştürücülerin yaygın kullanılan türü olan Buck DC-DC dönüştürücüler giriş gerilimini düşürerek çıkış gerilimini istenen bir referans gerilim değerinde sabit tutar. Bu çalışmada, Buck DC-DC dönüştürücülerdeki parazitik elemanların PID kontrolcüler üzerindeki etkisi incelenmiştir. Parazitik elemanlar Buck dönüştürücü sistem modeli üzerinde lineer olmayan etkiye neden olmaktadır. Literatürde yapılan çalışmalar genellikle parazitik elemanlar olmaksızın Buck dönüştürücü modeli üzerinde kontrolcüler tasarlanarak çıkış gerilimi kontrol edilmektedir. Alternatif olarak, parazitik elemanların etkisi de dikkate alınan model doğrusallaştırılarak lineer kontrolcüler tasarlanmaktadır. Bunlara ek olarak, parazitik elemanların etkisiyle birlikte tam model üzerinde lineer olmayan kontrolcüler tasarlanmaktadır. Bu çalışmada ise parazitik elemanların özellikle PID gibi lineer kontrolcü üzerindeki etkisi incelenmiştir.Öğe A Quest for Formant-Based Compact Nonuniform Trapezoidal Filter Banks for Speech Processing with VGG16(Springer Birkhauser, 2024) Parlak, Cevahir; Altun, YusufIn this text, we discuss the filter banks used for speech analysis and propose a novel filter bank for speech processing applications. Filter banks are building blocks of speech processing applications. Multiple filter strategies have been proposed, including Mel, PLP, Seneff, Lyon, and Gammatone filters. MFCC is a transformed version of Mel filters and is still a state-of-the-art method for speech recognition applications. However, 40 years after their debut, time is running out to launch new structures as novel speech features. The proposed acoustic filter banks (AFB) are innovative alternatives to dethrone Mel filters, PLP filters, and MFCC features. Foundations of AFB filters are based on the formant regions of vowels and consonants. In this study, we pioneer an acoustic filter bank comprising 11 frequency regions and conduct experiments using the VGG16 model on the TIMIT and Speech Command V2 datasets. The outcomes of the study concretely indicate that MFCC, Mel, and PLP filters can effectively be replaced with novel AFB filter bank features.Öğe The road disturbance attenuation for quarter car active suspension system via a new static two-degree-of-freedom design(2017) Altun, YusufThe main aim of this paper is to attenuate the effects of the road disturbance on the quarter-car active suspension system (ASS) for the passenger comfort by using design. Therefore, a new static disturbance compensator is proposed by using linear matrix inequality method such that the disturbance compensator and feedback controller are simultaneously designed for the disturbances in the linear time-invariant systems, which are measurable or predictable. They have static structure, and the disturbance compensator is designed on the feedforward path. The design is applied against the road disturbance affecting the quarter car ASS. The effectiveness of the design is demonstrated with the simulations.Öğe Spectro-Temporal Energy Ratio Features for Single-Corpus and Cross-Corpus Experiments in Speech Emotion Recognition(Springer Heidelberg, 2024) Parlak, Cevahir; Diri, Banu; Altun, YusufIn this study, novel Spectro-Temporal Energy Ratio features based on the formants of vowels, linearly spaced low-frequency, and logarithmically spaced high-frequency parts of the human auditory system are introduced to implement single- and cross-corpus speech emotion recognition experiments. Since the underlying dynamics and characteristics of speech recognition and speech emotion recognition differ too much, designing an emotion-recognition-specific filter bank is mandatory. The proposed features will formulate a novel filter bank strategy to construct 7 trapezoidal filter banks. These novel filter banks differ from Mel and Bark scales in shape and frequency regions and are targeted to generalize the feature space. Cross-corpus experimentation is a step forward in speech emotion recognition, but the researchers are usually chagrined at its results. Our goal is to create a feature set that is robust and resistant to cross-corporal variations using various feature selection algorithms. We will prove this by shrinking the dimension of the feature space from 6984 down to 128 while boosting the accuracy using SVM, RBM, and sVGG (small-VGG) classifiers. Although RBMs are considered no longer fashionable, we will show that they can achieve outstanding jobs when tuned properly. This paper discloses a striking 90.65% accuracy rate harnessing STER features on EmoDB.Öğe Tekil Olmayan Tanımlayıcı Sistemler İçin Kontrol Tasarımı(2017) Altun, YusufBu çalışmada, tekil olmayan descriptor (tanımlayıcı) belirsiz sistemler için kestirim yapılabilir yâda sensörle ölçülebilir bozucularıbastırmak için aynı anda statik ileri beslemeli ve geri beslemeli kontrol tasarımı önerilmektedir. Geri beslemeli ve ileri beslemelikontrol, doğrusal zamanla değişmeyen belirsiz sistemler için bozucuları bastırmak ve etkisini ortadan kaldırmak için geliştirilmiştir.Kontrol dizaynı, doğrusal matris eşitsizlikleri yardımıyla H kontrol teorisine dayanılarak gerçekleştirilmiştir. Robot kolu gibi çoğumekanik sistemler tekil olmayan belirsiz descriptor sistemler şeklinde modellenir. Bu yüzden bu çalışmada tekil olmayan belirsizsistemler ele alınmıştır. Önerilen tasarımın başarımı simülasyon örnekleriyle desteklenmektedir.
