Invited talk by Dr. Andriy Sarabakha

Abstract:

In recent years, many research activities have focused on unmanned aerial vehicles due to their usefulness in providing cost-effective solutions to dangerous, dirty and dull tasks. In many applications, these aerial robots must be able to fly autonomously in uncertain environments under variable operating conditions. In such circumstances, data-driven flight controllers propose alternative solutions to the model-based controllers without requiring a precise system model, which is often unavailable. This talk will introduce several artificial intelligence-based control techniques that enable drones to improve their performance under various operating conditions. The examination will go from the simplest static reinforcement learning controllers through fuzzy mapping-based controllers to online deep learning controllers.

Bio:

Dr. Andriy Sarabakha received his B.Sc. degree in Computer Engineering and his M.Sc. degree in Artificial Intelligence and Robotics, both with honours from "Sapienza" – University of Rome, Italy. In 2020, he received a Ph.D. degree in Mechanical and Aerospace Engineering from Nanyang Technological University, Singapore. He was a visiting scholar at the University of Toronto, Aarhus University and the Singapore University of Technology and Design. Between 2020 and 2022, he worked as a Senior Scientist at the Munich School of Robotics and Machine Intelligence at the Technical University of Munich, where he led the research at Flying Systems Arena and Underwater Systems Lab. Currently, he is a Presidential Postdoctoral Fellow at Nanyang Technological University, working on the continual learning and evolution for aerial robots.