RNN-Based Control of Robotic Systems with Various Constraints
Shuai Li
Faculty of Information Technology and Electrical Engineering, University of Oulu, Finland
The utilization of robotic arms for an array of tasks is burgeoning across various sectors, ranging from industrial applications to our daily lives. With the ever-evolving landscape of machine learning, we now have the opportunity to revolutionize autonomous robot control. While existing neural network-based robot control methods excel in terms of statistical average performance, they often fall short in guaranteeing worst-case performance. This talk will present our research on the development of recurrent neural networks (RNNs) based models with neural weights derived from models for robot control. Our approach harnesses the adaptability of neural networks while ensuring certified stability, making it a viable solution for addressing safety-critical tasks with a focus on guaranteed performance.
Biosketch
Shuai (Steven) Li
received the bachelor's degree in precision mechanical engineering from the Hefei University of Technology, China, master degree in automatic control engineering from the University of Science and Technology of China, and the Ph.D. degree in electrical and computer engineering from the Stevens Institute of Technology, USA. He is presently a full professor with Faculty of Information Technology and Electrical Engineering, University of Oulu, Finland. His current research interests include dynamic neural networks, robotic networks, machine learning. Steven is a fellow of IET (Institute of Engineering and Technology), BCS (British Computer Society), IMA (Institute of Mathematics and its Applications) and a member of EurAsc (European Academy of Sciences).