Abstract
dc:descriptionIn the modern era, the development of autonomous robotic systems is becoming increasingly important, as the demand for more efficient and accurate task execution continues to grow across a wide range of applications. Particularly in the field of smart agriculture, the use of unmanned ground vehicles (UGVs) creates new opportunities for the automation of agricultural tasks and the improvement of agricultural production processes. However, their operation in complex and unstructured environments makes reliable and safe navigation one of the most significant challenges in agricultural robotics. Path planning is a fundamental component of autonomous navigation, as it determines how a vehicle moves from a starting point to a desired target while avoiding obstacles. This process must combine low computational cost, short path length, and smooth motion, ensuring that the resulting trajectory is compatible with the kinematic capabilities of a robotic system. The A* algorithm is one of the most widely used path-planning methods; however, it presents limitations in terms of computational efficiency and the geometric smoothness of the paths it generates. These challenges become more pronounced in agricultural environments, where UGVs are required to navigate among crops and obstacles, often with limited computational resources. The variability of the environment and the particular motion conditions increase the requirements for reliable and efficient path planning. At the same time, the abrupt changes in direction that characterize paths in grid-based environments pose an additional challenge to the smooth motion of the vehicles. Addressing these limitations is essential for the development of autonomous navigation systems capable of meeting the specific requirements of agricultural applications.In this dissertation, the primary objective is the development and evaluation of optimization techniques for the A* algorithm for the navigation of agricultural UGVs. To this end, different strategies for designing the heuristic function are investigated, aiming to reduce computation time and the extent of the search space without significantly increasing path length. In parallel, smoothing techniques are incorporated to transform grid-based paths into smoother trajectories suitable for autonomous vehicle motion. The proposed techniques are experimentally evaluated across different scales and obstacle configurations in order to investigate their effectiveness and the consistency of their performance.The contributions presented in this dissertation introduce new directions toward more efficient and adaptive solutions for the autonomous navigation of agricultural UGVs. Particular emphasis is placed on the development and evaluation of improved heuristic functions for the A* algorithm, as well as on the application of smoothing techniques aimed at reducing computational cost and improving the quality of the generated paths. The dissertation also highlights the importance of continued research and development in this field, as the demand for faster, more efficient, and reliable autonomous navigation in complex agricultural environments continues to grow.
Degree
thesis:*- Grantor dc:publisher
- University of Western Macedonia
- Year dc:date
- 2026
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Chatzisavvas, Antonios
- Χατζησάββας, Αντώνιος
Subjects
dc:subject × 17- Αυτόνομη πλοήγηση
- UGV
- Αλγόριθµος A*
- Έξυπνη γεωργία
- Σχεδιασµός διαδροµής
- Ευρετικές συναρτήσεις
- Autonomous navigation
- A* Algorithm
- Smart agriculture
- Path planning
- Heuristic functions
- Επιστήμη Ηλεκτρολόγου Μηχανικού, Ηλεκτρονικού Μηχανικού, Μηχανικού Η/Υ
- Επιστήμες Μηχανικού και Τεχνολογία
- Ρομποτική
- Electrical Engineering, Electronic Engineering, Information Engineering
- Engineering and Technology
- Robotics
Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/63104