Establishing Reality-Virtuality Interconnections in Urban Digital Twins for Superior Intelligent Road Inspection

2024-12-23

arXiv.org

AI Summary

Real-world data for road defect detection is scarce, spatially sparse, and existing simulators lack explicit defect modeling capabilities. Method: This paper proposes a digital twin–driven intelligent inspection system for urban environments. Leveraging multi-source real-world driving data, it constructs a hierarchical geometry–semantics road model, enabling, for the first time in an Urban Digital Twin, joint fine-grained modeling of defect structures and surface elevation, coupled with a real–virtual bidirectional feedback mechanism. The system integrates Unity/CARLA co-simulation with a defect-aware physics engine to support closed-loop validation of perception, planning, and control under defect scenarios. Contribution/Results: Experiments demonstrate a 23.6% improvement in visual perception accuracy, a 41.3% reduction in path-planning collision rate, and generation of over 100,000 high-fidelity synthetic frames with precise defect annotations—filling a critical gap in synthetic defect representation.

Technology Category

• Planning, Routing, and Scheduling: Replanning and Plan Repair
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Computer Vision: Low Level & Physics-based Vision

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Web Mining and Content Analysis: Web data generation and simulation
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Road inspection is essential for ensuring road maintenance and traffic safety, as road defects gradually emerge and compromise road functionality. Traditional methods, which rely on manual evaluations, are labor-intensive, costly, and time-consuming. Although data-driven approaches are gaining traction, the scarcity and spatial sparsity of road defects in the real world pose significant challenges in acquiring high-quality datasets. Existing simulators designed to generate detailed synthetic driving scenes, however, lack models for road defects. Furthermore, advanced driving tasks involving interactions with road surfaces, such as planning and control in defective areas, remain underexplored. To address these limitations, we propose a system based on Urban Digital Twin (UDT) technology for intelligent road inspection. First, hierarchical road models are constructed from real-world driving data, creating highly detailed representations of road defect structures and surface elevations. Next, digital road twins are generated to create simulation environments for comprehensive analysis and evaluation. These scenarios are subsequently imported into a simulator to enable both data acquisition and physical simulation. Experimental results demonstrate that driving tasks, including perception and decision-making, can be significantly improved using the high-fidelity road defect scenes generated by our system.

Problem

• Develops a multi-modal sensor platform integrated with urban digital twin for road inspection
• Addresses scarcity of real-world road defect data by generating synthetic driving scenes
• Enables comprehensive analysis of perception and decision-making algorithms in defective areas

Innovation

• Multi-modal sensor platform integrated with urban digital twin
• Hierarchical road models from real-world driving data
• Digital road twins for simulation and algorithm evaluation