DiffPF: Differentiable Particle Filtering with Generative Sampling via Conditional Diffusion Models

TL;DR

DiffPF introduces a novel differentiable particle filtering method using conditional diffusion models, enabling accurate, high-dimensional, and multimodal state estimation with significant performance improvements over existing methods.

Contribution

It is the first to integrate conditional diffusion models into particle filtering, enhancing posterior sampling and state estimation accuracy.

Findings

Achieves 82.8% improvement on a multimodal localization benchmark.

Attains 26% better accuracy on the KITTI visual odometry task.

Outperforms existing differentiable filtering baselines across various scenarios.

Abstract

This paper proposes DiffPF, a differentiable particle filter that leverages diffusion models for state estimation in dynamic systems. Unlike conventional differentiable particle filters, which require importance weighting and typically rely on predefined or low-capacity proposal distributions. DiffPF learns a flexible posterior sampler by conditioning a diffusion model on predicted particles and the current observation. This enables accurate, equally-weighted sampling from complex, high-dimensional, and multimodal filtering distributions. We evaluate DiffPF across a range of scenarios, including both unimodal and highly multimodal distributions, and test it on simulated as well as real-world tasks, where it consistently outperforms existing filtering baselines. In particular, DiffPF achieves an 82.8% improvement in estimation accuracy on a highly multimodal global localization benchmark, and a 26% improvement on the real-world KITTI visual odometry benchmark, compared to state-of-the-art differentiable filters. To the best of our knowledge, DiffPF is the first method to integrate conditional diffusion models into particle filtering, enabling high-quality posterior sampling that produces more informative particles and significantly improves state estimation.