Parallel subspace sampling for particle filtering in dynamic bayesian networks

Abstract

Monitoring the variables of real world dynamic systems is a difficult task due to their inherent complexity and uncertainty. Particle Filters (PF) perform that task, yielding probability distribution over the unobserved variables. However, they suffer from the Curse of dimensionality problem: the number of particles grows exponentially with the dimensionality of the hidden state space. The problem is aggravated when the initial distribution of the variables is not well known, as happens in global localization problems. We present a new parallel PF for systems whose variable dependencies can be factored into a Dynamic Bayesian Network. The new algorithms significantly reduce the number of particles, while independently exploring different subspaces of hidden variables to build particles consistent with past history and measurements. We demonstrate this new PF approach on some complex dynamical system estimation problems, showing that our method successfully localizes and tracks hidden states in cases where traditional PFs fail.