Sequential Monte Carlo simulation of rare event probability in stochastic hybrid systems

Authors:	Krystul Jaroslav, University of Twente, Netherlands Blom Henk A.P., National Aerospace Laboratory NLR, Netherlands
Topic:	1.4 Stochastic Systems
Session:	Hybrid Stochastic Systems
Keywords:	Monte Carlo, Markov models, Probabilistic risk assessment,Stochastic jump processes, Uncertain dynamic systems

Abstract

Recently (Cerou et al., 2002) developed an elegant factorization of rareevent probabilities appearing in diffusion processes and otherstrong Markov processes, and a sequential Monte Carlo simulationapproach to estimate the factorized rare event probability. Thepaper extends this approach towards rarely switching diffusions,and demonstrates the effectiveness for a simple example.