Differentiable Factor Graph Optimization for Learning Smoothers

TL;DR

This paper introduces a differentiable factor graph smoother that can be trained end-to-end for improved state estimation, demonstrating significant accuracy gains in object tracking and visual odometry.

Contribution

It presents a novel end-to-end learning approach for factor graph-based smoothers, unrolling the optimizer for joint learning of system models and state estimation.

Findings

Significant accuracy improvements over baselines in object tracking.

Enhanced visual odometry performance.

Open-source code and tools provided for reproducibility.

Abstract

A recent line of work has shown that end-to-end optimization of Bayesian filters can be used to learn state estimators for systems whose underlying models are difficult to hand-design or tune, while retaining the core advantages of probabilistic state estimation. As an alternative approach for state estimation in these settings, we present an end-to-end approach for learning state estimators modeled as factor graph-based smoothers. By unrolling the optimizer we use for maximum a posteriori inference in these probabilistic graphical models, we can learn probabilistic system models in the full context of an overall state estimator, while also taking advantage of the distinct accuracy and runtime advantages that smoothers offer over recursive filters. We study this approach using two fundamental state estimation problems, object tracking and visual odometry, where we demonstrate a significant improvement over existing baselines. Our work comes with an extensive code release, which includes training and evaluation scripts, as well as Python libraries for Lie theory and factor graph optimization: https://sites.google.com/view/diffsmoothing/