Speaker Professor Hyo-Sang Shin
Cranfield University
Time/Place Sep. 27 (Mon) 15:30
Building 133 Room 316-1
Abstract
The seminar talk will introduce computational flight control concepts and an corresponding control law that is developed by Autonomous and Intelligent Systems (AIS) Group at Cranfield University, based on Deep Reinforcement Learning (DRL). Benefits of implementing computational flight control or guidance algorithms, especially the ones based on the model-based optimal control concepts, have been successfully demonstrated. At the beginning, this talk will discuss potential benefits of implementation of emerging Reinforcement Learning techniques to the flight control problems, beyond model-based computational control algorithm. Then, the main challenges for developing such a control law and barriers preventing those flight control laws to be implemented in practice will be discussed. Next, the talk will focus on suggesting some concepts and philosophies that could help to remove those barriers and hence achieve the potential benefits, followed by introduction on our research development in achieving the proposed key concept. The talk will then introduce a few numerical examples of the proposed DRL-based flight control law. The talk will be concluded by discussing lessons learned, future research directions and potential collaborations.
Biography
Prof. Hyo-Sang Shin is Professor of Guidance, Control and Navigation Systems in Centre for Autonomous and Cyberphysical Systems at Cranfield University. He received his BSc from Pusan National University on Aerospace Engineering in 2004 and gained an MSc on flight dynamics, guidance and control in Aerospace Engineering from KAIST and a PhD on cooperative missile guidance from Cranfield University in 2006 and 2011, respectively. He has published over 180 journal and conference papers and has been invited for many lectures, invited talks and keynotes both in Universities and industries mainly on flight control and guidance, sensor, data and information fusion, cooperative control and aircraft ISTAR applications. He has been also involved in many research programmes and has been the coordinator of a few European projects. Prof. Shin has significant experiences and track records in flight control, guidance, and sensor/data/information fusion related research and development projects. He was leading Team A, who won the Swarm Challenge sponsored by BAE System in 2019, from Cranfield University as the academic supervisor. Prof. Shin is a member of various technical committees (IFAC Technical Committee on Aerospace Control, IEEE Technical Committee on Aerial Robotics and Unmanned Aerial Systems), and program and editorial boards. He was programme co-chair of the 21st IFAC Symposium on Aerospace Control and programme chair of the 2019 IEEE Conference on Research, Education and Development of Unmanned Aircraft Systems (RED-UAS). He is also an Associate Editor of IEEE Transactions on Aerospace and Electronic Systems and also Journal of Intelligent and Robotic Systems. Prof. Shin is Head of Autonomous and Intelligent Groups and leading relevant research areas in Cranfield University. The current research activities he is leading and being involved include, but not being limited to, data-centric guidance and control, decision making on multi-agent systems, information-driven sensing and fusion, multiple target tracking, and improvement of multiple vehicle cooperation.