GPR-Based Positioning System for Robust Autonomous Vehicle Navigation: A Low-Latency SLAM Framework with V2X Capabilities in Mixed-GPS Environments

Akshay Pramod Roy, Mubarik Mohamoud, Venkata Ashok Masimukku, and Vamshi Krishna Uppununthala

Peer Reviewed

Abstract:	Accurate localization is critical for autonomous vehicles (AVs), enabling safe navigation and interaction with their environment. Traditional localization methods, such as Global Navigation Satellite System (GNSS), often face limitations in urban canyons, forests, and remote areas where GNSS signals are unreliable. While other sensor modalities like light detection and ranging (LiDAR) sensors and cameras offer potential solutions, they still fall short in certain conditions. This paper presents a Ground Penetrating Radar (GPR) based vehicle positioning system that addresses these limitations. We propose a rasterized GPR data model that efficiently represents subsurface features by discretizing them into multiple data layers, reducing storage and transmission requirements. This model integrates with existing sensor modalities, such as Global Navigation Satellite System (GNSS), Inertial Measurement Units (IMUs), and odometry, to create maps using a low-latency Simultaneous Localization and Mapping (SLAM) framework optimized for mixed-GPS environments. The GPR-SLAM system, combined with a Vehicleto-Everything (V2X) data transmission architecture using the Message Queuing Telemetry Transport (MQTT) protocol, enables real-time map generation and sharing among vehicles, ensuring accurate localization even with partial data reception. Our experimental results demonstrate that GPR based maps have a small data footprint, remaining within the bandwidth requirements of standard networks. Furthermore, the system proves effective in mixed-GNSS environments, where GNSS signals are degraded or unavailable, showcasing its potential for off-road and urban applications with limited or disrupted network conditions. This research highlights GPR as a promising technology for robust and scalable localization. Index Terms—Autonomous vehicles, GPR, GNSS-denied, SLAM, V2X communication, MQTT.
Published in:	2025 IEEE/ION Position, Location and Navigation Symposium (PLANS) April 28 - 1, 2025 Salt Lake Marriott Downtown at City Creek Salt Lake City, UT
Pages:	907 - 916
Cite this article:	Roy, Akshay Pramod, Mohamoud, Mubarik, Masimukku, Venkata Ashok, Uppununthala, Vamshi Krishna, "GPR-Based Positioning System for Robust Autonomous Vehicle Navigation: A Low-Latency SLAM Framework with V2X Capabilities in Mixed-GPS Environments," 2025 IEEE/ION Position, Location and Navigation Symposium (PLANS), Salt Lake City, UT, April 2025, pp. 907-916.
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In