Unmanned aerial vehicles (UAVs) are gaining popularity for a range of purposes such as surveillance, monitoring, and mapping. One of the key challenges in UAVs is path planning, which involves finding a safe and efficient route through the environment while avoiding obstacles. In recent years, reinforcement learning algorithms, such as SARSA, have been applied to path planning in UAVs, with promising results. In this research paper, we propose a SARSA-based path planning algorithm for UAVs that uses a camera to capture images of the environment. The camera captures real-time images of the environment, which are processed to identify obstacles and plan a safe path. The SARSA algorithm is used to learn the optimal path based on the UAV's current state and the environment's feedback. To evaluate the proposed algorithm's performance, we conducted simulations in various environments with different levels of complexity. The results show that our algorithm can successfully navigate the UAV through the environment while avoiding obstacles and achieving optimal performance. We also compared our algorithm's performance to other path planning algorithms commonly used in UAVs, demonstrating that our algorithm outperforms them in terms of efficiency and safety.
Path Planning for UAV “Drones” Using SARSA: Enhancing Efficiency and Performance
01.11.2023
889156 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Central type dynamic path inducing method based on Sarsa learning
Europäisches Patentamt | 2015
|Cruise dynamic pricing based on SARSA algorithm
Taylor & Francis Verlag | 2021
|Enhancing Wildlife Detection Using Thermal Imaging Drones: Designing the Flight Path
DOAJ | 2025
|