Establishing Reality-Virtuality Interconnections in Urban Digital Twins for Superior Intelligent Road Inspection and Simulation
Yikang Zhang, Chuangwei Liu, Jiahang Li, Yingbing Chen, Jie CHENG, Rui Fan
AI summary
Problem
Traditional road inspection is labor-intensive, and data-driven methods suffer from scarce, sparse real-world defect data, while existing simulators lack realistic 3D road defect models needed for accurate perception and control testing.
Approach
The authors develop a portable multi-sensor platform to capture real-world road geometry and defects, then use a hierarchical model creator and digital twin generator to seamlessly integrate these into a simulator for synthetic data generation and physical simulation.
Key results
- Portable multi-modal sensor platform for high-fidelity real-world road data collection
- Autonomous pipeline for reconstructing hierarchical 3D road surface and defect models
- Digital twin generator enabling seamless integration of defects into simulator environments
- Comprehensive benchmark demonstrating improved perception transfer and flexible defect-avoidance planning
Why it matters
Enables scalable, high-fidelity simulation and data synthesis for autonomous driving, reducing reliance on scarce real-world defect data and improving road safety assessment.
Abstract
Road inspection is crucial for maintaining road ser- viceability and ensuring traffic safety, as road defects gradually develop and compromise functionality. Traditional inspection methods, which rely on manual evaluations, are labor-intensive, costly, and time-consuming. While data-driven approaches are gaining traction, the scarcity and spatial sparsity of real-world road defects present significant challenges in acquiring high- quality datasets. Existing simulators designed to generate detailed synthetic driving scenes, however, lack models for road defects. Moreover, advanced driving tasks that involve interactions with road surfaces, such as planning and control in defective areas, remain underexplored. To address these limitations, we propose a multi-modal sensor platform integrated with an urban digital twin (UDT) system for intelligent road inspection. First, hier- archical road models are constructed from real-world driving data collected using vehicle-mounted sensors, resulting in highly detailed representations of road defect structures and surface elevations. Next, digital road twins are generated to create sim- ulation environments for comprehensive analysis and evaluation of algorithm performance. These scenarios are then imported into a simulator to facilitate both data acquisition and physical simulation. Experimental results demonstrate that driving tasks, including perception and decision-making, benefit significantly from the high-fidelity road defect scenes generated by our system.