Machine learning (ML)/artificial intelligence (AI) is making advances faster that the society is able to absorb, understand and assimilate them in areas such as image recognition, natural language processing, and data analytics; at the same time feedback control that employ AI and ML are becoming more pervasive and critical.Today, application of learning controllers can be found in areas as diverse as robotics and autonomous systems, process control, energy or smart grid, civil infrastructure, healthcare, manufacturing, automotive, transportation, entertainment, and consumer appliances. Moreover, controllers designed in discrete-time have the important advantage that they can be directly implemented in digital form using modern-day embedded hardware. Unfortunately, discrete-time design using Lyapunov stability analysis is far more complex than the continuous-time counterpart since the first difference in Lyapunov function is quadratic in the states and not linear as in the case of continuous-time.Further, optimal control of uncertain linear or nonlinear dynamic systems is a major challenge. By incorporating learning feature, optimal adaptive control of such uncertain dynamical systems in discrete-time can be designed.

In this talk, an overview of online learning-based feedback control framework of robotics/autonomous system in discrete-time will be discussed. Subsequently, learning-based optimal adaptive control of uncertain nonlinear dynamic systems will be presented in a systematic manner based on reinforcement learning (RL)/approximate dynamic programming (ADP). Challenges in developing the three generation of learning controllers will be addressed using robotics and autonomous systems.The talk will conclude with a short discussion of open research problems in the area of deep learning/AI based control of such systems.

Dr. Jagannathan Sarangapani (or S. Jagannathan) received his doctoral degree from University of Texas at Arlington in 1994. He is presently at the Missouri University of Science and Technology (former University of Missouri-Rolla) where he is a Professor and Rutledge-Emerson Endowed Chair in the Dept of Electrical and Computer Engineering with a courtesy appointment in Dept of Computer Science. He is also served as Interim Director of Intelligent Systems Center, and the Site Director for the graduated NSF Industry/University Cooperative Research Center on Intelligent Maintenance Systems for over 13 years.  

His research interests include learning and adaptation, deep learning based adaptive and neural network control, networked control systems/cyber physical systems, sensor networks, prognostics, and autonomous systems/robotics. He has coauthored 189 peer reviewed journal articles, 289 refereed IEEE conference articles, several book chapters and six books. He holds 21 patents. He is a Fellow of IEEE, US National Academy of Inventors, IET (UK), Institute of Measurement and Control, UK and Asia Association of Artificial Intelligence.  He received NSF Career Award in 2000, Caterpillar Research Excellence Award in 2001, Boeing Pride Achievement Award in 2007, IEEE Control System Society’s Transition to Practice Award in 2018, and several Faculty Excellence and Teaching Excellence Awards at Missouri University of Science and Technology.