Skip to content

LAA

Seminar

Solving games without determinization

Piterman, N (EPF Lausanne)
Thursday 18 May 2006, 11:00-12:00

Seminar Room 1, Newton Institute

Abstract

The synthesis of reactive systems requires the solution of two-player games on graphs with omega-regular objectives. When the objective is specified by a linear temporal logic formula or nondeterministic Buechi automaton, then previous algorithms for solving the game require the construction of an equivalent deterministic automaton. However, determinization for automata on infinite words is extremely complicated, and current implementations fail to produce deterministic automata even for relatively small inputs. We show how to construct, from a given nondeterministic Buechi automaton, an equivalent nondeterministic parity automaton N that is good for solving games with objective N. The main insight is that a nondeterministic automaton is good for solving games if it fairly simulates the equivalent deterministic automaton. In this way, we omit the determinization step in game solving and reactive synthesis. The fact that our automata are nondeterministic makes them surprisingly simple, amenable to symbolic implementation, and allows an incremental search for winning strategies.

Presentation

[ppt ]

Audio

MP3MP3 Real AudioReal Audio

Back to top ∧