Teaching an AI agent how to drive on complex racetracks using reinforcement learning

A tétel rövid nézete
dc.contributor.advisorBogacsovics, Gergő
dc.contributor.authorQin, Yuchen
dc.contributor.departmentDE--Informatikai Kar
dc.date.accessioned2023-04-11T10:58:55Z
dc.date.available2023-04-11T10:58:55Z
dc.date.created2023-04-05
dc.description.abstractThe main objective of this thesis is to educate a reinforcement learning (RL) agent on acquiring the skill of driving a car in the Unity environment. The central focus of this thesis is to train a reinforcement learning (RL) agent utilizing a deep neural network to effectively control and optimize its behavior in order to achieve a specific task or environment. This will be achieved through the application of various reward functions, utilizing the Proximal Policy Optimization (PPO) algorithm. Throughout the training process, I conducted extensive experimentation with various combinations of hyperparameters and network settings. Additionally, I explored different approaches for calculating rewards in order to identify the most effective method for achieving optimal RL agent performance.
dc.description.correctorN.I.
dc.description.courseComputer Science Engineering
dc.description.degreeBSc/BA
dc.format.extent46
dc.identifier.urihttps://hdl.handle.net/2437/349087
dc.language.isoen
dc.rights.accessHozzáférhető a 2022 decemberi felsőoktatási törvénymódosítás értelmében.
dc.subjectReinforcement Learning
dc.subjectUnity
dc.subjectAI
dc.subjectTrack
dc.subjectCar
dc.subject.dspaceDEENK Témalista::Informatika::Számítógéptudomány
dc.titleTeaching an AI agent how to drive on complex racetracks using reinforcement learning
Fájlok
Gyűjtemények
Theses (IK)