Differentiable Particle Filters: End-to-End Learning with Algorithmic Priors

TL;DR

This paper introduces differentiable particle filters (DPFs), enabling end-to-end learning of state estimation models that incorporate algorithmic priors, leading to significant error reduction and improved generalization in localization tasks.

Contribution

The paper presents a novel differentiable implementation of particle filters that allows end-to-end training with learnable models, enhancing performance and interpretability.

Findings

Reduced error rates by approximately 80% in experiments.

Learned localization in a policy-agnostic way, improving generalization.

Demonstrated benefits on both simulated and real data.

Abstract

We present differentiable particle filters (DPFs): a differentiable implementation of the particle filter algorithm with learnable motion and measurement models. Since DPFs are end-to-end differentiable, we can efficiently train their models by optimizing end-to-end state estimation performance, rather than proxy objectives such as model accuracy. DPFs encode the structure of recursive state estimation with prediction and measurement update that operate on a probability distribution over states. This structure represents an algorithmic prior that improves learning performance in state estimation problems while enabling explainability of the learned model. Our experiments on simulated and real data show substantial benefits from end-to- end learning with algorithmic priors, e.g. reducing error rates by ~80%. Our experiments also show that, unlike long short-term memory networks, DPFs learn localization in a policy-agnostic way and thus greatly improve generalization. Source code is available at https://github.com/tu-rbo/differentiable-particle-filters .