Out-of-Distribution Robustness with Deep Recursive Filters

TL;DR

This paper introduces a method combining deep neural networks with recursive filters to improve out-of-distribution noise robustness in state estimation for robotics, demonstrating enhanced accuracy and efficiency in simulations and real-world data.

Contribution

It presents a novel approach integrating deep learning with recursive filtering for robust state and uncertainty estimation under out-of-distribution noise.

Findings

Improved state and uncertainty estimation over baselines.

Approximately 3x computational efficiency gain.

Successful application to both simulated and real-world scenarios.

Abstract

Accurate state and uncertainty estimation is imperative for mobile robots and self driving vehicles to achieve safe navigation in pedestrian rich environments. A critical component of state and uncertainty estimation for robot navigation is to perform robustly under out-of-distribution noise. Traditional methods of state estimation decouple perception and state estimation making it difficult to operate on noisy, high dimensional data. Here, we describe an approach that combines the expressiveness of deep neural networks with principled approaches to uncertainty estimation found in recursive filters. We particularly focus on techniques that provide better robustness to out-of-distribution noise and demonstrate applicability of our approach on two scenarios: a simple noisy pendulum state estimation problem and real world pedestrian localization using the nuScenes dataset. We show that our approach improves state and uncertainty estimation compared to baselines while achieving approximately 3x improvement in computational efficiency.