Abstract
This article aims to introduce an innovative framework and associated construction content for the modern smart highway, and to succinctly provide viable recommendations for its rapid development. This study examines the construction frameworks and guidelines across 13 provinces and cities, identifying their commonalities and variations. The new framework is categorized into five layers: Perception, Support, Data, Computing, and Application. The construction key points in these layers are concisely explained and the Application layer is further segmented into six facets: Safety, Efficiency, Green, Convenience, and Toll Collection. This study analyzes the critical application scenarios in various provinces and cities. It aids in rectifying the disparity in understanding during actual construction across these provinces, providing a classification criteria for smart highway construction content. These guidelines will significantly influence future project construction improvements.
What Problem Does This Paper Address?
As of early 2023, 13 provinces and cities in China—including major tech hubs like Beijing, Zhejiang, and Guangdong—had released independent "Smart Expressway Construction Guides." While this surge in local innovation is positive, it has led to a "fragmentation problem." Different regions are developing smart roads based on divergent architectural philosophies, ranging from functional layer models to hardware-centric "cloud-edge-end" hierarchies.
This lack of a unified framework creates a "bulk version" of smart highway development, making large-scale integration and cross-province interoperability nearly impossible. For the end-user, this results in an inconsistent driving experience. This research seeks to bridge these gaps by establishing a standardized classification criterion.
Methods
The study involved a meta-analysis of the 13 official provincial and municipal guides released between 2020 and 2023. The researchers synthesized these disparate perspectives into a unified five-layer framework:
- Perception Layer: The "eyes" of the system, including sensors, traffic monitoring, and weather stations.
- Support Layer: The "nervous system," encompassing 5G/LTE-V/WiFi, power supply, and information security.
- Data Layer: The "memory," consisting of high-precision maps and big data centers.
- Computing Layer: The "brain," featuring cloud control platforms and edge computing.
- Application Layer: The "value" delivered through safety, efficiency, green energy, and tolling solutions.
Additionally, the team analyzed 17 specific application scenarios to distinguish between "mandatory" requirements (widely adopted) and "personalized" ones (region-specific).
Key Insights and Ideas
A critical finding of the study is the varying focus on "all-weather traffic." While some provinces aim for total 24/7 environmental immunity, others focus on automated warnings for hazards like fog and ice. There is also a significant shift toward decentralized intelligence, with "Edge Computing" processing data at the roadside to minimize latency for safety-critical applications.
By "heat mapping" these scenarios, the authors identified that Information Services and Service Areas are the most universally prioritized, whereas Active Management and Vehicle Platooning remain experimental or localized to specific high-tech corridors.
Key Results
The paper successfully standardizes technical requirements across the five layers:
- Perception: Categorized into subject monitoring (infrastructure health), traffic status (flow/incidents), and meteorology.
- Computing: Proposed a multi-level storage mechanism where data is processed in edge clouds and selectively uploaded to central clouds to balance bandwidth.
- Application Facets: Segmented into six clear categories: Safety, Efficiency, Green, Convenience, Toll Collection, and Construction/Maintenance.
- Scenario Consensus: "Vehicle-Road Coordination" and "Emergency Management" are emerging as high-consensus priorities across the national landscape.
Limitations and Future Work
Despite the progress, several challenges remain. There is a risk of "passive, involuted releases" where provinces continue to produce slightly different guides without achieving true interoperability. Future efforts must move beyond individual "smart points" (e.g., a single tunnel) toward "line" (entire highways) and "surface" (national network) integration. Furthermore, the framework must remain flexible enough to accommodate future advancements like Level 5 autonomous driving.
Summary updated