Normalizing Flow-based Differentiable Particle Filters
This addresses the limitation of existing differentiable particle filters in complex real-world scenarios by allowing flexible density estimation, which is incremental as it builds on prior work but introduces a novel integration of normalizing flows.
The paper tackles the problem of performing joint sequential state estimation and model learning for non-linear non-Gaussian state-space models in complex environments by introducing a differentiable particle filtering framework that uses normalizing flows to build its dynamic model, proposal distribution, and measurement model, enabling valid probability densities and adaptive learning without predefined distribution families, with performance evaluated through numerical experiments.
Recently, there has been a surge of interest in incorporating neural networks into particle filters, e.g. differentiable particle filters, to perform joint sequential state estimation and model learning for non-linear non-Gaussian state-space models in complex environments. Existing differentiable particle filters are mostly constructed with vanilla neural networks that do not allow density estimation. As a result, they are either restricted to a bootstrap particle filtering framework or employ predefined distribution families (e.g. Gaussian distributions), limiting their performance in more complex real-world scenarios. In this paper we present a differentiable particle filtering framework that uses (conditional) normalizing flows to build its dynamic model, proposal distribution, and measurement model. This not only enables valid probability densities but also allows the proposed method to adaptively learn these modules in a flexible way, without being restricted to predefined distribution families. We derive the theoretical properties of the proposed filters and evaluate the proposed normalizing flow-based differentiable particle filters' performance through a series of numerical experiments.