题目： Local Control and Estimation in Multi-Agent Systems

时间：2011年12月2日（周五），15:30

地点：广知C楼511会议室

The talk will consist of two parts: one on complex Laplacians for formation control and the other on estimation of a common coordinate system.

It is known that real-valued graph Laplacians play an important role in collective coordination of multi-agent systems, particularly on consensus and consensus related problems. In the talk, I will introduce complex-valued Laplacians for graphs whose edges are attributed with complex weights and present some local control strategies for leader-follower networks via complex Laplacians. The main goal is to control the shape of a formation of point agents in the plane using a simple linear interaction rule related to complex Laplacians.
On the other hand, one of the main issues in motion coordination for autonomous agents like mobile robots is the absence of a common reference frame. In this talk, I will introduce an estimation scheme for a common reference frame based on local information measurement and exchange. Then I will show how local coordination control problems using relative state feedback can be converted to control problems using absolute state feedback. The approach makes some coordination problems of heterogeneous agents also solvable.

Biography:

Zhiyun Lin is a professor of Systems Control at the Department of Systems Science and Engineering, College of Electrical Engineering, Zhejiang University, China. He received his Ph.D. degree in Electrical and Computer Engineering from University of Toronto, Canada, in 2005. From 2005 to 2007, he was a postdoctoral researcher in the same department at University of Toronto. He then joined College of Electrical Engineering, Zhejiang University in 2007. During the years of 2009-2011, he also held a visiting professor position at the Australian National Unviersity and University of Cagliari, Italy. His current research includes multi-agent (multi-robot) systems, wireless sensor network, switched and hybrid systems, and feedback control of biped robots.