Ahmet Mehmet Karadeniz

57204127635

Publications - 7

Mobile Robot Environment Representation Through Fuzzy Signatures-Integrated Quadtrees

Csaba Hajdu Ahmet Mehmet Karadeniz L. T. Koczy

Publication Name: Romanian Journal of Information Science and Technology

Publication Date: 2025-01-01

Volume: 28

Issue: 1

Page Range: 103-116

Description:

This paper presents an innovative environment representation technique for mobile robots, incorporating obstacle detection within their operational space. Leveraging the fuzzy signature method, this approach uses quadtrees for efficient data organization. A set of fuzzy rules evaluates feature points to ascertain the relevance of identified obstacles. These points and their fuzzy associations are systematically arranged using a quadtree structure. The environmental model is reconstructed by traversing this tree and applying the established fuzzy rules. This paper has achieved a high-resolution grid representation of 0.1m within a 20m×20m area. Notably, the inference operation completes in just 0.5 ms, underscoring the method’s efficiency. Additionally, the technique is optimized for low memory consumption, demonstrating effective resource management even on older PCs, such as an Intel Core Duo 2 with 16 GB RAM. This representation is designed to support advanced robotic functions, such as obstacle navigation in a distributed computing environment.

Open Access: Yes

DOI: 10.59277/ROMJIST.2025.1.09

Synergies between Fuzzy Signatures and Hypergraphs

Csaba Hajdu Ahmet Mehmet Karadeniz L. T. Koczy

Publication Name: Proceedings of the International Symposium on Applied Machine Intelligence and Informatics Sami

Publication Date: 2025-01-01

Volume: Unknown

Issue: Unknown

Page Range: 495-500

Description:

Fuzzy signatures have demonstrated effectiveness in various knowledge-representation domains, including medical decision-making systems and complex decision-making tasks across numerous fields. Even though the advancements in the field of fuzzy signatures have been substantial, the complete potential for developing a comprehensive graph-theoretical description format for this domain still needs to be fully realized. This paper introduces a novel hypergraph-based method for modeling fuzzy signatures, which offers a structured approach to their representation but also showcases the potential synergy between fuzzy signatures and hypergraphs. The proposed method is designed to improve fuzzy information representation and streamline the aggregation-based decision-making process. Future research is anticipated to extend the applicability of this method to control systems and robotics. Furthermore, the hypergraph-based model opens new avenues for the algebraic analysis of fuzzy signatures through tensor-based representations.

Open Access: Yes

DOI: 10.1109/SAMI63904.2025.10883273

Evaluating Deep Learning Algorithms for Freeway Mainstream Traffic Control

Ahmet Mehmet Karadeniz Mehran Amini L. T. Koczy

Publication Name: Lecture Notes in Networks and Systems

Publication Date: 2025-01-01

Volume: 1258 LNNS

Issue: Unknown

Page Range: 289-299

Description:

Traffic congestion is a universal problem that significantly impacts urban mobility and economic productivity. Accurate traffic flow prediction is crucial for efficient traffic management and congestion mitigation. Traditional methods often struggle to capture the complex temporal dependencies in traffic data. This study explores the effectiveness of Temporal Convolutional Network (TCN) models compared to Long Short-Term Memory (LSTM) models for predicting traffic volumes on freeway networks. Previous research has largely focused on LSTM models, leaving a gap in understanding the potential advantages of TCN models in this context. We address this gap by conducting a comprehensive comparison of LSTM and TCN models, training them on a dataset representing approximate traffic flow, and evaluating their performance using metrics such as Mean Squared Error (MSE), Mean Absolute Error (MAE), and coefficient of determination (R2). Our findings indicate that the TCN model outperforms the LSTM model, achieving lower MSE and MAE values and a higher R2 score. These results suggest that TCN models can more accurately predict traffic volumes under conditions with the least captured traffic data, offering a promising tool for real-time approximate traffic management and congestion prevention with reasonable prediction performance.

Open Access: Yes

DOI: 10.1007/978-3-031-81799-1_26

Fuzzy Systems Based Voltage Control of Buck Converter for Vehicles with 48V E/E Architecture

Ahmet Mehmet Karadeniz Nuri Furkan Kocak

Publication Name: Journal of Electrical and Electronics Engineering

Publication Date: 2024-10-01

Volume: 17

Issue: 2

Page Range: 17-22

Description:

This paper investigates the efficiency of PID and Fuzzy Logic Control approaches as a voltage control method in vehicles with 48V E/E architecture. Then compares the mentioned methods output on Buck converter for 48V to 12V. Buck converter aims to convert 48V to 12V to be able to supply power to low voltage systems such as electric power steering systems within vehicles. 48V E/E architecture is becoming a common architecture among electric vehicles and power distribution through 48V batteries, hence it is chosen for this study. Our study presents that Fuzzy Logic Controller has a better performance in terms of response time and settling time on controlling output voltage of Buck converter which converts 48V to 12V.

Open Access: Yes

DOI: DOI not available

Transfer Learning-Based Steering Angle Prediction and Control with Fuzzy Signatures-Enhanced Fuzzy Systems for Autonomous Vehicles

Á Ballagi Ahmet Mehmet Karadeniz L. T. Koczy

Publication Name: Symmetry

Publication Date: 2024-09-01

Volume: 16

Issue: 9

Page Range: Unknown

Description:

This research introduces an innovative approach for End-to-End steering angle prediction and its control in electric power steering (EPS) systems. The methodology integrates transfer learning-based computer vision techniques for prediction and control with fuzzy signatures-enhanced fuzzy systems. Fuzzy signatures are unique multidimensional data structures that represent data symbolically. This enhancement enables the fuzzy systems to effectively manage the inherent imprecision and uncertainty in various driving scenarios. The ultimate goal of this work is to assess the efficiency and performance of this combined approach by highlighting the pivotal role of steering angle prediction and control in the field of autonomous driving systems. Specifically, within EPS systems, the control of the motor directly influences the vehicle’s path and maneuverability. A significant breakthrough of this study is the successful application of transfer learning-based computer vision techniques to extract respective visual data without the need for large datasets. This represents an advancement in reducing the extensive data collection and computational load typically required. The findings of this research reveal the potential of this approach within EPS systems, with an MSE score of 0.0386 against 0.0476, by outperforming the existing NVIDIA model. This result provides a 22.63% better Mean Squared Error (MSE) score than NVIDIA’s model. The proposed model also showed better performance compared with all other three references found in the literature. Furthermore, we identify potential areas for refinement, such as decreasing model loss and simplifying the complex decision model of fuzzy systems, which can represent the symmetry and asymmetry of human decision-making systems. This study, therefore, contributes significantly to the ongoing evolution of autonomous driving systems.

Open Access: Yes

DOI: 10.3390/sym16091180

Robot environment representation based on Quadtree organization of Fuzzy Signatures

Á Ballagi Csaba Hajdu Ahmet Mehmet Karadeniz L. T. Koczy

Publication Name: Saci 2021 IEEE 15th International Symposium on Applied Computational Intelligence and Informatics Proceedings

Publication Date: 2021-05-19

Volume: Unknown

Issue: Unknown

Page Range: 509-514

Description:

This paper presents a novel approach to mobile robot environment representation to hold information on detected obstacles. The method is inspired by fuzzy signature-based formalism and is based on classical quadtrees as a data indexing structure. Each detected feature point is evaluated by a fuzzy-ruleset defining the presumed significance of each detected object. Feature points and their fuzzy-mapping are indexed in a classical quadtree-based fashion. During the reconstruction of the environment representation, inference is done by the traversal on the constructed tree using accumulated fuzzy-ruleset. Our goal is to use this representation format for further robotic tasks such as obstacle avoidance in a distributed computational environment.

Open Access: Yes

DOI: 10.1109/SACI51354.2021.9465566

Exploring Fuzzy Signatures in Sensor Fusion: A Comparative Study with the Complementary Filter

Á Ballagi Ahmet Mehmet Karadeniz L. T. Koczy

Publication Name: Cinti 2024 IEEE 24th International Symposium on Computational Intelligence and Informatics Proceedings

Publication Date: 2024-01-01

Volume: Unknown

Issue: Unknown

Page Range: 69-74

Description:

Sensing has become a pivotal element in the development of autonomous systems with the advancement of the technology. These systems operate on a sense-think-act cycle to execute tasks, necessitating the integration of multiple sensors. The challenge of synthesizing meaningful information from diverse data sources escalates with the complexity of the data. This study tackles the issue of sensor data complexity by investigating the potential of Fuzzy Signatures, which are promising in handling complex data due to their hierarchical structures. The main goal is to present a concept for sensor fusion based on Fuzzy Signatures, which may facilitate their use in autonomous system tasks. To demonstrate this concept, accelerometer and gyroscope data are utilized, with results compared to those from a Complementary Filter providing insight into the sensor fusion capabilities of Fuzzy Signatures. The study also underscores the importance of aggregation operators in Fuzzy Signatures, focusing on the Max and WRAO (weighted relevance aggregation operator) aggregation operators. The potential to employ various aggregation operators or to develop new ones for specific applications is highlighted. The findings indicate that Fuzzy Signatures could be an effective solution for sensor fusion challenges, offering prospects for enhancement and broader application in autonomous systems.

Open Access: Yes

DOI: 10.1109/CINTI63048.2024.10830837