Towards Intention-Aware and Resilient Systems Using a Hidden Mode Hybrid System
Sze Zheng Yong, PhD, MIT
The Polytechnic School
Ira A. Fulton Schools of Engineering
Monday, March 7, 2016
10 a.m.
Goldwater Center (GWC) 487, Tempe Campus [map]
Free to attend
Seminar is free to watch via Abode Connect
Abstract
Most autonomous robots must operate in the presence of uncertainties in the intentions and decisions of robots or humans in the absence of communication. However, assuming the worst-case scenario can result in an undesirably conservative solution. Moreover, there has been a growing concern and an urgent need for providing resilience to critical infrastructures, such as the national power grid, to ensure the integrity and availability of these safety-critical systems despite malicious attacks or unforeseen faults. We will consider these problems through the lens of a highly expressive modeling framework known as the hidden mode hybrid system with unknown inputs (i.e., a partially observed system with mixed continuous and discrete dynamics). The literature on feedback control and estimation approaches for such systems is relatively sparse. In this talk, we will discuss the development of the very first inference algorithms for simultaneously estimating states, unknown inputs and hidden modes of stochastic switched linear systems, along with a rigorous analysis of their properties. These algorithms provide the initial steps towards the design of intention-aware and resilient systems.
Biosketch
Sze Zheng Yong is currently a postdoctoral associate with the Laboratory for Information and Decision Systems (LIDS) at Massachusetts Institute of Technology
(MIT) under the supervision of Prof. Emilio Frazzoli. He has obtained a Dipl.-Ing.(FH) degree in automotive engineering with a specialization in mechatronics and control systems from the Esslingen University of Applied Sciences, Germany in 2008 and his S.M. and Ph.D. degrees in mechanical engineering from MIT in 2010 and 2016, respectively. His research interests lie in the broad area of control and estimation of hidden mode hybrid systems, with applications to intention-aware autonomous systems and resilient cyber-physical systems.
