Design and Control of Multi-Agent Robotic Systems

Overview

In this talk, I will present my research on designing multi-agent robotic systems and developing theory and algorithms for system control. The first part of the talk will present my research on designing an animal-borne remote imaging system and developing theory and algorithm for distributed sensor fusion that enables estimation and detection of animal group movements using the system. I will explain data collection capability of the system and share system deployment experiences in monitoring a wild life of African animals In the second part, I will describe my current work on developing a fleet of autonomous robotic vessels. The research aims at building autonomous vessels that can transport people and goods and construct dynamic structures in the canal of Amsterdam. For this purpose, the vessels are designed to be latched and connected to each other and also to self-reconfigure their connections. I will explain how we design such self-reconfigurable robotic vessels and a coordination algorithm to maneuver multiple vessels and to build floating structures on water surface.

Bio

Shinkyu Park received the Ph.D. degree in electrical engineering from the University of Maryland in 2015. From January to June in 2016, he was a postdoctoral researcher at National Geographic Society and University of Maryland, where he investigated distributed algorithm and animal-borne sensing system design. Since June 2016, he is a postdoctoral associate at Massachusetts Institute of Technology, where his research focuses on building a fleet of autonomous robotic vessels and designing coordination strategy to enable self-reconfiguration of the vessels. His current research interests are in robotics, multi-robot control, optimization, and game theory.