Velibor Trujic · Hochschule Campus Wien
Analyse von Reinforcement Learning für Wegfindung
SubmittedJune 2023
Extent59 pages
LanguageGerman
Research focus
The thesis compares deep reinforcement learning with classical pathfinding algorithms such as A* and Dijkstra. A three-dimensional Unity environment is used to examine computational efficiency, adaptability and overall performance, with implications for game development, robotics and autonomous navigation.
Research questionWhat are the advantages and disadvantages of deep reinforcement learning compared with classical pathfinding algorithms in video games?
Artificial neural networks · Deep learning · Machine learning · Reinforcement learning · Pathfinding
View official record