TrackDiffuser: Nearly Model-Free Bayesian Filtering with Diffusion Model

TL;DR

TrackDiffuser introduces a nearly model-free Bayesian filtering method using diffusion models, effectively learning system dynamics from data and handling inaccuracies without explicit noise priors, outperforming traditional approaches.

Contribution

It reformulates Bayesian filtering as a conditional diffusion model, enabling robust state estimation without relying on accurate models or noise priors, a significant advancement in the field.

Findings

Outperforms classical and hybrid methods in non-linear, non-Gaussian scenarios.

Exhibits robustness to inaccuracies in system models.

Maintains interpretability similar to traditional filters.

Abstract

State estimation remains a fundamental challenge across numerous domains, from autonomous driving, aircraft tracking to quantum system control. Although Bayesian filtering has been the cornerstone solution, its classical model-based paradigm faces two major limitations: it struggles with inaccurate state space model (SSM) and requires extensive prior knowledge of noise characteristics. We present TrackDiffuser, a generative framework addressing both challenges by reformulating Bayesian filtering as a conditional diffusion model. Our approach implicitly learns system dynamics from data to mitigate the effects of inaccurate SSM, while simultaneously circumventing the need for explicit measurement models and noise priors by establishing a direct relationship between measurements and states. Through an implicit predict-and-update mechanism, TrackDiffuser preserves the interpretability advantage of traditional model-based filtering methods. Extensive experiments demonstrate that our framework substantially outperforms both classical and contemporary hybrid methods, especially in challenging non-linear scenarios involving non-Gaussian noises. Notably, TrackDiffuser exhibits remarkable robustness to SSM inaccuracies, offering a practical solution for real-world state estimation problems where perfect models and prior knowledge are unavailable.