Korsoveczki, GyulaAlTamimi, Jebril2023-12-202023-12-202023-12-01https://hdl.handle.net/2437/364139This article discusses how drones are transforming various industries, such as agriculture and surveillance, yet they still face difficulties in autonomously navigating complex environments. To address this, the article introduces a method that combines artificial intelligence (AI) with drone systems to enhance their ability to avoid obstacles. The approach involves the use of deep reinforcement learning (DRL) techniques to better the navigation skills of drones. The authors detail their integrated hardware and software strategy, emphasizing their preliminary findings and the obstacles they encountered. The primary aim of this research is to further the use of autonomous drones in real-life situations.86enDeep Reinforcement Learning (DRL)Virtual EnvironmentSimulation EnvironmentTwin-Delayed Deep Deterministic Policy Gradient(TD3)DEENK Témalista::Engineering Sciences::Electrical EngineeringDEENK Témalista::Engineering Sciences::EngineeringHozzáférhető a 2022 decemberi felsőoktatási törvénymódosítás értelmében.