Leveraging V2X for Collaborative HD Maps Construction Using Scene Graph Generation

2025-02-14

AI Summary

High-definition (HD) map construction traditionally relies on expensive, specialized survey vehicles and suffers from high costs and poor real-time update capability. To address this, this paper proposes a lightweight, local geometric-layer mapping method leveraging V2X-enabled collaboration and scene graph modeling. Our approach models lane centerlines as compact, transmissible graph structures and integrates onboard vision-based detection, graph neural networks, and scene graph generation to perform localized geometric perception at the edge. Through V2X communication and cloud-edge coordination, it enables multi-vehicle joint optimization and dynamic geometric aggregation. Crucially, the method eliminates dependence on dedicated mapping hardware, enabling low-cost, real-time, and evolvable HD map co-construction. Evaluated on the nuScenes dataset, our method achieves significantly higher lane association prediction accuracy than existing state-of-the-art approaches.

Technology Category

• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Intelligent Robots: Localization, Mapping, and Navigation
• Machine Learning: Graph-based Machine Learning

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Systems and Infrastructure for Web, Mobile and WoT: Location- and context-aware Web and WoT applications and services

Abstract

High-Definition (HD) maps play a crucial role in autonomous vehicle navigation, complementing onboard perception sensors for improved accuracy and safety. Traditional HD map generation relies on dedicated mapping vehicles, which are costly and fail to capture real-time infrastructure changes. This paper presents HDMapLaneNet, a novel framework leveraging V2X communication and Scene Graph Generation to collaboratively construct a localized geometric layer of HD maps. The approach extracts lane centerlines from front-facing camera images, represents them as graphs, and transmits the data for global aggregation to the cloud via V2X. Preliminary results on the nuScenes dataset demonstrate superior association prediction performance compared to a state-of-the-art method.

Problem

• HD map construction
• V2X communication
• Scene Graph Generation

Innovation

• V2X communication
• Scene Graph Generation
• HDMapLaneNet framework