Deep Learning for Unmanned Aerial Vehicles: Advancements, Applications, Challenges, and Future Directions

Significant advancements have been achieved in the field of unmanned aerial vehicles (UAVs) in recent times, marking a notable stride forward, allowing them to autonomously perform a wide range of tasks. One crucial factor contributing to these advancements is deep learning, which involves teaching artificial neural networks to glean insights and make informed choices based on extensive and intricate datasets. Deep learning has played a pivotal role in enhancing the capabilities of UAVs across various operations. This review paper provides a comprehensive overview of deep learning techniques in the context of UAVs, with a focus on their adaption to UAV-specific scenarios. Moreover, it delves into the applications of deep learning for UAVs, including object discovery and shadowing, independent navigation, scene understanding, and mission planning. Additionally, it highlights emerging trends and future directions, such as lightweight deep learning architectures, edge computing, model interpretability, and the integration of deep learning with other emerging technologies. This paper also centers on the application of deep learning in UAVs, categorized according to distinct learning algorithms and varying assessment standards.