報告人: Prof. Fumin Zhang

講座日期：2020-12-10

講座時間：10:00

報告地點：騰訊會議（會議ID：537 442 673）

主辦單位：數(shù)學(xué)與系統(tǒng)科學(xué)研究院

講座人簡介：

Dr. Fumin Zhang is Professor in the School of Electrical and Computer Engineering at the Georgia Institute of Technology. He received a PhD degree in 2004 from the University of Maryland (College Park) in Electrical Engineering, and held a postdoctoral position in Princeton University from 2004 to 2007. His research interests include mobile sensor networks, maritime robotics, control systems, and theoretical foundations for cyber-physical systems. He received the NSF CAREER Award in September 2009 and the ONR Young Investigator Program Award in April 2010. He is currently serving as the co-chair for the IEEE RAS Technical Committee on Marine Robotics, associate editors for IEEE Journal of Oceanic Engineering, Robotics and Automation Letters, IEEE Transactions on Automatic Control, and IEEE Transactions on Control of Networked Systems.

講座簡介：

There is a perceivable trend for robots to serve as networked mobile sensing platforms that are able to collect data in challenging environments with difficulty for localization and communication.  The need for undisturbed operation of search and monitoring posts higher goals for sustainable autonomy. We propose a layered approach to achieve signal propagation over large swarms through active perception. Biological evidence from fish swarms has shown that active perception is used by animals to allow fast response to stimulations when only a few members are stimulated. Active perception has advantage over averaging consensus, such as reduced communication and faster signal propagation. After transferring this knowledge to the design of robotic swarms, we found that multiple perception layers can be overlaid on top of the feedback control layer to achieve complex swarm behaviors. The findings also lead to effective distributed optimization algorithms that are quite different from the known consensus-based algorithms. One key feature is the capability to handle vanishing and exploding gradients that often arise in machine learning. Our algorithms are rigorously analyzed and verified by experimental effort on mobile and flying robot platforms. The promising results demonstrates that bio-inspired autonomy might be preferred in aquatic environment that features severe limitation in localization and communication.