In the last few years, Aerial robots have stepped up in the commercial market and are being used in various commercial applications. These robots areused in agriculture, natural disaster, search and rescue and parcel delivery application domains. They have become more powerful tools to work independently with the help of artificial intelligence (AI) techniques and methods. This chapter covers the deep learning applications of AI for aerial robotics. For this purpose, different components of aerial robots are discussed and analysed. The discussion of deep learning algorithms is based on these components that use image and sensor data captured by the flying robots. These deep learning applications can be used in different application domains such as agriculture, indoor and outdoor autonomous robot navigation, autonomous search and rescue operation, image registration and localization of aerial robots. With the help of these deep learning applications, aerial robots become a more powerful tool to perform their specific mission in different application domains. This chapter summarizes the deep learning application of aerial robots, which are categorized into supervised, unsupervised and reinforcement learning.