Controllers with Minimal Observation Power (Application to Timed Systems)

Title: Controllers with Minimal Observation Power (Application to Timed Systems)
Speaker:  Dr. Peter Bulychev (Aalborg University, Denmark)
Time: 14:30, Friday, May 11, 2012
Venue: Lecture room, 3rd Floor, Building 5#, State Key Laboratory of Computer Science, Institute of Software, Chinese Academy of Sciences
Abstract: We consider the problem of controller synthesis under imperfect information in a setting where there is a set of available observable predicates equipped with a cost function. The problem that we address is the computation of a subset of predicates sufficient for control and whose cost is minimal. Our solution avoids a full exploration of all possible subsets of predicates, uses an on-the-fly algorithm for solving the underlying games with imperfect information and reuses some information between different iterations. We apply our approach to timed systems. We have developed a prototype of tool and show the interest of our optimization algorithm on two case studies.

 Biography:

Peter Bulychev was born in Moscow, Russia, and obtained MsC (2007) and PhD (2010) degrees from the Computer Science department at Lomonosov Moscow State University. In his PhD thesis he studied two topics. The first is about the computation of various simulation relations between finite-state Kripke structures. As a part of this topic he developed a BDD-based tool for computing strong, weak, stuttering, etc. simulations between NuSMV models. The second topic is about detection of clones in software systems. The clone detection tool Clone Digger (http://clonedigger.sourceforge.net/) developed by Peter Bulychev has been supported by Google and participated in Google Summer of Code 2008.

 Starting from 2011 Peter Bulychev works as a postdoctoral researcher at Aalborg University, Denmark. He studies various applications of Timed Automata, including controller synthesis and statistical model checking. Peter Bulychev participates in the development of UPPAAL model checker (http://uppaal.org/) and the open-source library for the Timed Automata manipulation (http://people.cs.aau.dk/~pbulychev/python.html).