Development of Advanced Algorithms for Autonomous Vehicle Navigation
Session
Computer Science and Communication Engineering
Description
This paper addresses the development of advanced algorithms for autonomous vehicle navigation, focusing on the creation of an intelligent platform capable of safely planning and managing vehicle movement without human intervention. The main objective is to design a functional and reliable system by applying modern artificial intelligence techniques, optimization methods, and sensor data processing (sensor fusion). The study begins with a review of the existing literature on autonomous navigation, identifying key approaches such as Machine Learning, Simultaneous Localization and Mapping (SLAM), and Path Planning. It then presents the development methods for the autonomous navigation system using machine learning through simulation, implemented with Python, Jupyter Notebook, and GitHub. The paper also provides system testing and performance evaluation in simulated environments. The results show significant improvements in navigation accuracy, obstacle avoidance, and overall safety. Finally, recommendations for future work are presented, including real-world testing and the integration of path planning algorithms with machine learning to achieve a safer and more reliable autonomous system.
Keywords:
Autonomous Navigation, Machine Learning, Path Planning, End-to-End Learning, Convolutional Neural Networks (CNN), Data Augmentation, SLAM
Proceedings Editor
Edmond Hajrizi
ISBN
978-9951-982-41-2
Location
UBT Lipjan, Kosovo
Start Date
25-10-2025 9:00 AM
End Date
26-10-2025 6:00 PM
DOI
10.33107/ubt-ic.2025.88
Recommended Citation
Hoti, Liridon; Qehaja, Besnik; and Sllamniku, Jeton, "Development of Advanced Algorithms for Autonomous Vehicle Navigation" (2025). UBT International Conference. 20.
https://knowledgecenter.ubt-uni.net/conference/2025UBTIC/CS/20
Development of Advanced Algorithms for Autonomous Vehicle Navigation
UBT Lipjan, Kosovo
This paper addresses the development of advanced algorithms for autonomous vehicle navigation, focusing on the creation of an intelligent platform capable of safely planning and managing vehicle movement without human intervention. The main objective is to design a functional and reliable system by applying modern artificial intelligence techniques, optimization methods, and sensor data processing (sensor fusion). The study begins with a review of the existing literature on autonomous navigation, identifying key approaches such as Machine Learning, Simultaneous Localization and Mapping (SLAM), and Path Planning. It then presents the development methods for the autonomous navigation system using machine learning through simulation, implemented with Python, Jupyter Notebook, and GitHub. The paper also provides system testing and performance evaluation in simulated environments. The results show significant improvements in navigation accuracy, obstacle avoidance, and overall safety. Finally, recommendations for future work are presented, including real-world testing and the integration of path planning algorithms with machine learning to achieve a safer and more reliable autonomous system.