LF-GNSS: Towards More Robust Satellite Positioning with a Hard Example Mining Enhanced Learning-Filtering Deep Fusion Framework

TL;DR

This paper introduces LF-GNSS, a deep learning-based framework that enhances satellite positioning accuracy in urban environments by intelligently analyzing signals, dynamically weighting measurements, and prioritizing challenging signals during training.

Contribution

The paper proposes a novel learning-filtering deep fusion framework with hard example mining and DOP-based feature representation for robust satellite positioning in urban scenarios.

Findings

Demonstrates superior accuracy over traditional methods.

Validates effectiveness on public and private datasets.

Enhances robustness through hard example mining.

Abstract

Global Navigation Satellite System (GNSS) is essential for autonomous driving systems, unmanned vehicles, and various location-based technologies, as it provides the precise geospatial information necessary for navigation and situational awareness. However, its performance is often degraded by Non-Line-Of-Sight (NLOS) and multipath effects, especially in urban environments. Recently, Artificial Intelligence (AI) has been driving innovation across numerous industries, introducing novel solutions to mitigate the challenges in satellite positioning. This paper presents a learning-filtering deep fusion framework for satellite positioning, termed LF-GNSS. The framework utilizes deep learning networks to intelligently analyze the signal characteristics of satellite observations, enabling the adaptive construction of observation noise covariance matrices and compensated innovation vectors for Kalman filter input. A dynamic hard example mining technique is incorporated to enhance model robustness by prioritizing challenging satellite signals during training. Additionally, we introduce a novel feature representation based on Dilution of Precision (DOP) contributions, which helps to more effectively characterize the signal quality of individual satellites and improve measurement weighting. LF-GNSS has been validated on both public and private datasets, demonstrating superior positioning accuracy compared to traditional methods and other learning-based solutions. To encourage further integration of AI and GNSS research, we will open-source the code at https://github.com/GarlanLou/LF-GNSS, and release a collection of satellite positioning datasets for urban scenarios at https://github.com/GarlanLou/LF-GNSS-Dataset.