Geo-ORBIT: A Federated Digital Twin Framework for Scene-Adaptive Lane Geometry Detection

2025-07-11

AI Summary

Dynamic road geometry awareness remains a critical bottleneck in digital twin traffic systems; existing approaches rely on static maps or expensive sensors, exhibiting limited generalizability, scalability, and privacy protection. This paper proposes FedMeta-GeoLane—a novel framework integrating a lightweight lane detection model (GeoLane), meta-learning (Meta-GeoLane), and federated meta-learning to enable dynamic geometric modeling solely from roadside camera trajectory data. Leveraging CARLA-SUMO co-simulation, it establishes a high-fidelity digital twin supporting cross-regional personalized model customization and privacy-preserving collaborative training. Experiments across multi-city scenarios demonstrate a 18.7% reduction in geometric error, a 23.4% improvement in zero-shot generalization to unseen scenes, and a 62.5% decrease in communication overhead—significantly enhancing scalability, adaptability, and deployment efficiency.

Technology Category

• Natural Language Processing: Language Grounding & Multi-modal NLP
• Planning, Routing, and Scheduling: Planning with Language Models
• Computer Vision: Large Vision Models

Application Category

• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Digital Twins (DT) have the potential to transform traffic management and operations by creating dynamic, virtual representations of transportation systems that sense conditions, analyze operations, and support decision-making. A key component for DT of the transportation system is dynamic roadway geometry sensing. However, existing approaches often rely on static maps or costly sensors, limiting scalability and adaptability. Additionally, large-scale DTs that collect and analyze data from multiple sources face challenges in privacy, communication, and computational efficiency. To address these challenges, we introduce Geo-ORBIT (Geometrical Operational Roadway Blueprint with Integrated Twin), a unified framework that combines real-time lane detection, DT synchronization, and federated meta-learning. At the core of Geo-ORBIT is GeoLane, a lightweight lane detection model that learns lane geometries from vehicle trajectory data using roadside cameras. We extend this model through Meta-GeoLane, which learns to personalize detection parameters for local entities, and FedMeta-GeoLane, a federated learning strategy that ensures scalable and privacy-preserving adaptation across roadside deployments. Our system is integrated with CARLA and SUMO to create a high-fidelity DT that renders highway scenarios and captures traffic flows in real-time. Extensive experiments across diverse urban scenes show that FedMeta-GeoLane consistently outperforms baseline and meta-learning approaches, achieving lower geometric error and stronger generalization to unseen locations while drastically reducing communication overhead. This work lays the foundation for flexible, context-aware infrastructure modeling in DTs. The framework is publicly available at https://github.com/raynbowy23/FedMeta-GeoLane.git.

Problem

• Dynamic roadway geometry sensing lacks scalability and adaptability
• Large-scale Digital Twins face privacy and efficiency challenges
• Real-time lane detection needs context-aware personalization

Innovation

• Federated meta-learning for privacy-preserving adaptation
• Lightweight lane detection from trajectory data
• Real-time DT synchronization with traffic simulators