Holistic view of the road transportation system based on real-time data sharing mechanism

Tao Li, Dong Xiang, Hao Junfeng, Ping Yin, Xiaoxue Xu, Lai Maokai, Yuan Li, Ting PengarXiv 2024 · DOI · arXiv

Abstract

Traditional manual driving and single-vehicle-based intelligent driving have limitations in real-time and accurate acquisition of the current driving status and intentions of surrounding vehicles, leading to vehicles typically maintaining appropriate safe distances from each other. Yet, accidents still frequently occur, especially in merging areas; meanwhile, it is difficult to comprehensively obtain the conditions of road infrastructure. These limitations not only restrict the further improvement of road capacity but also result in irreparable losses of life and property. To overcome this bottleneck, this paper constructs a space-time global view of the road traffic system based on a real-time sharing mechanism, enabling both road users and managers to timely access the driving intentions of nearby vehicles and the real-time status of road infrastructure.

Introduction

Current autonomous driving systems often suffer from a form of "tunnel vision," where perception is limited to the immediate vicinity visible to on-board sensors like LiDAR and cameras. This creates two critical vulnerabilities: intent uncertainty and infrastructure blindness. While a vehicle can detect a lead car slowing down, it cannot inherently know the driver's underlying intention—such as a planned lane change—until the maneuver begins. Furthermore, individual vehicles struggle to detect macro-level road anomalies, such as rockfalls or bridge fractures, until they are within dangerous proximity.

This paper proposes a shift from isolated vehicle intelligence to a collaborative "spatio-temporal global view." By utilizing enhanced Road Section Management Units (RSMUs) and C-V2X communication, the system allows vehicles to share not just their current status, but their future driving intentions, effectively eliminating perception blind spots.

Proposed System Framework

The proposed architecture centers on the interaction between the Vehicle Intelligent Unit (VIU) and the Road Section Management Unit (RSMU).

Road Section Management Unit (RSMU)

The RSMU is an evolution of the traditional Road Side Unit (RSU). It serves as a local coordinator that aggregates high-fidelity data from all vehicles within its jurisdiction. Beyond simple data relay, the RSMU monitors infrastructure via high-definition sensors and broadcasts a unified "global view" back to road users.

Functional architecture of the Road Section Management Unit (RSMU)

Vehicle Intelligent Unit (VIU)

The VIU is an on-board system responsible for collecting internal vehicle data—such as speed, steering angle, and braking status—and broadcasting the vehicle's intended maneuvers to the RSMU. This allows the system to treat a driver's intention as a shareable data point, enabling proactive rather than reactive safety adjustments.

Communication Standards and Performance

To support this high-density data exchange, the paper advocates for the C-V2X (Cellular Vehicle-to-Everything) standard over the older DSRC (Dedicated Short-Range Communication). C-V2X provides the necessary range, latency, and reliability required for high-speed highway environments.

Comparison of DSRC and C-V2X performance indicators

As shown in the comparative analysis, C-V2X ensures zero packet loss in high-density scenarios where DSRC may fail, and it offers significantly wider distance coverage, which is essential for communication between vehicles and infrastructure.

Deployment Strategy and Handover Logic

The paper provides specific engineering constraints for highway deployment to ensure continuous coverage and accuracy:

  • Optimized Spacing: RSMUs should be placed approximately 1.2 km apart. This balances the theoretical communication range (up to 1 km) with a "maximum effective connection" of 600 m to ensure overlap.
  • Parallel Communication Handover: To prevent data loss during transitions between RSMU zones, the authors propose a parallel link system. A vehicle establishes a connection with the upcoming RSMU while still connected to the current one, ensuring a seamless handover of the global view.
  • High-Risk Zones: Specific placement is recommended at 100m before and 50m after exit and entrance ramps to manage the complexities of merging zones.

Conclusion

By moving the focus from individual vehicle autonomy to a holistic road segment view, this framework addresses the fundamental limitations of current intelligent transportation systems. While the current research focuses on highway scenarios, future work will aim to adapt this deployment logic to the more complex environments of urban roads and various highway grades to optimize global traffic flow.

Summary updated

Abstract

Traditional manual driving and single-vehicle-based intelligent driving have limitations in real-time and accurate acquisition of the current driving status and intentions of surrounding vehicles, leading to vehicles typically maintaining appropriate safe distances from each other. Yet, accidents still frequently occur, especially in merging areas; meanwhile, it is difficult to comprehensively obtain the conditions of road infrastructure. These limitations not only restrict the further improvement of road capacity but also result in irreparable losses of life and property. To overcome this bottleneck, this paper constructs a space-time global view of the road traffic system based on a real-time sharing mechanism, enabling both road users and managers to timely access the driving intentions of nearby vehicles and the real-time status of road infrastructure.
BibTeX
@misc{li2024holistic,
  author = {Tao Li and Dong Xiang and Hao Junfeng and Ping Yin and Xiaoxue Xu and Lai Maokai and Yuan Li and Ting Peng},
  title = {{Holistic view of the road transportation system based on real-time data sharing mechanism}},
  year = {2024},
  eprint = {2407.03187},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2407.03187},
}
Figure