Resilient Vehicle Navigation Utilizing Decentralized Integration of Solid-State LiDARs and 3-D Digital Maps

View at publisher Summarise with VIRAT
Authors: Mohamed A. Elsayed, Eslam Mounier, Abolemagd Noureldin
Cited by: 0
References: 33

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Solid-State LiDAR (SSL) technologies offer notable advantages in terms of cost, size, and durability. However, their limited field of view (FOV) and vulnerability to occlusions can hinder navigation performance when deployed as standalone sensors. Moreover, while map-based LiDAR navigation methods enable high-accuracy positioning, they remain inherently susceptible to issues such as map gaps and outdated data. This paper presents a decentralized fusion system for integrating multiple SSLs with high-accuracy 3-D digital maps to achieve robust vehicle navigation in complex urban environments. The proposed system employs a LiDAR-to-map registration (LMR) module, fused using a confidence-weighted strategy that dynamically adapts to registration quality and spatial consistency. An extended Kalman filter further integrates inertial and odometer data to refine the navigation solution. The system’s performance is validated through extensive experiments in both typical and challenging urban scenarios, including tests simulating sensor failures. Results demonstrate the effectiveness of the approach in delivering more resilient and accurate positioning to standalone SSL-based positioning under diverse challenging scenarios.

Fields of Study & Subjects

Robotics and Sensor-Based Localization Inertial Sensor and Navigation Autonomous Vehicle Technology and Safety

Publication details

Journal / source
IEEE Sensors Journal
Volume
26
Issue
9
Pages
13889–13904
Publication date
25 March 2026
Language
EN