Speaker: Dongkyu Choi
Date & Time: 2018년 7월 30일(월) 17:00
Where: 융대원 D-123호
Robotic platforms these days offer an impressive set of physical
capabilities. Drones are capable of long, reliable flights and ground
robots can negotiate very challenging terrains. But such robots still
lack high-level facilities that provide human-level, long-term
autonomy that enables them to work flawlessly with human counterparts.
This talk reviews a research effort to address this issue in the
context of a cognitive architecture, ICARUS, in three important
directions: 1) unifying continuous control with symbolic reasoning; 2)
goal reasoning; and 3) episodic capabilities for explainable autonomy.
After describing these previous and ongoing work, we will conclude by
providing a few discussion points for future research and their
implications to the integrated systems technology.
Dr. Dongkyu Choi is an Assistant Professor of Aerospace
Engineering at the University of Kansas. Dr. Choi’s research focuses
on using agent architectures to control physical platforms like
unmanned aerial systems and other robots. He has been intimately
involved in the development of the ICARUS cognitive architecture,
which he uses in many of his research to enable long-term, explainable
autonomy for robots. Dr. Choi received a B.S. in Mechanical and
Aerospace Engineering from Seoul National University in 2001 and
received his M.S. in 2003 and Ph.D. in 2010 from the Department of
Aeronautics and Astronautics at Stanford University.
초청자 : 융합과학부 지능형융합시스템전공 박재흥 교수 (email@example.com)