Environment-Aware Learning of Smooth GNSS Covariance Dynamics for Autonomous Racing

TL;DR

This paper introduces LACE, a learning-based framework that models GNSS covariance dynamics as a stable, smooth system using neural networks, improving localization in autonomous racing environments.

Contribution

The paper presents a novel deep learning approach with stability guarantees for modeling GNSS measurement covariance dynamics in high-speed autonomous racing.

Findings

Enhanced localization accuracy in GNSS-degraded environments

Smoother covariance estimates leading to improved control stability

Demonstrated effectiveness on an autonomous racecar

Abstract

Ensuring accurate and stable state estimation is a challenging task crucial to safety-critical domains such as high-speed autonomous racing, where measurement uncertainty must be both adaptive to the environment and temporally smooth for control. In this work, we develop a learning-based framework, LACE, capable of directly modeling the temporal dynamics of GNSS measurement covariance. We model the covariance evolution as an exponentially stable dynamical system where a deep neural network (DNN) learns to predict the system's process noise from environmental features through an attention mechanism. By using contraction-based stability and systematically imposing spectral constraints, we formally provide guarantees of exponential stability and smoothness for the resulting covariance dynamics. We validate our approach on an AV-24 autonomous racecar, demonstrating improved localization performance and smoother covariance estimates in challenging, GNSS-degraded environments. Our results highlight the promise of dynamically modeling the perceived uncertainty in state estimation problems that are tightly coupled with control sensitivity.