.png){kind=icon}
.png){kind=icon}
Return
Hot Spot Identification for the Deployment of Roadside Units: Install Only Where Lives Are Saved
.png){kind=icon}
PRE
AI
DOI：10.1061/JUPDDM.UPENG-6165.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Roadside units (RSUs) are crucial for improving the safety of connected vehicles, yet their effectiveness heavily depends on strategic deployment. Traditional methods prioritize RSU installation in high-collision-risk areas (hot spots), but these locations may not yield the greatest safety improvements. This paper redefines hot spots as road areas where RSU deployment achieves the greatest collision risk reduction and proposes a cost-effective method for identifying these hot spots. The proposed method not only helps save more lives but also minimizes the identification cost. It reduces both data consumption and identification time compared with traditional approaches. It requires only small-sample real-world data, which are augmented to generate the additional data needed for identification. Additionally, an accelerated sampling strategy helps minimize the time required for the identification process. The simulation results show that this approach achieves a 46.7% greater risk reduction, a 43% reduction in data costs, and a 40% faster identification of hot spots compared with the state-of-the-art method.
Keywords:
Roadside units
Hot spot identification
Risk exposure reduction
Accelerated sampling
Cost-effective deployment
Journal
J
IF:
1.9
Papers:
87
Citations:
2.6K