We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs.NI

Change to browse by:

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Networking and Internet Architecture

Title: Framework for Highway Traffic Profiling using Connected Vehicle Data

Abstract: The connected vehicle (CV) data could potentially revolutionize the traffic monitoring landscape as a new source of CV data that are collected exclusively from original equipment manufactures (OEMs) have emerged in the commercial market in recent years. Compared to existing CV data that are used by agencies, the new-generation of CV data have certain advantages including nearly ubiquitous coverage, high temporal resolution, high spatial accuracy, and enriched vehicle telematics data (e.g., hard braking events). This paper proposed a traffic profiling framework that target vehicle-level performance indexes across mobility, safety, riding comfort, traffic flow stability, and fuel consumption. The proof-of-concept study of a major interstate highway (i.e., I-280 NJ), using the CV data, illustrates the feasibility of going beyond traditional aggregated traffic metrics. Lastly, potential applications for either historical analysis and even near real-time monitoring are discussed. The proposed framework can be easily scaled and is particularly valuable for agencies that wish to systemically monitoring regional or statewide roadways without substantial investment on infrastructure-based sensing (and the associated on-going maintenance costs)
Comments: 7 pages, 8 figures
Subjects: Networking and Internet Architecture (cs.NI); Systems and Control (eess.SY)
Cite as: arXiv:2209.14245 [cs.NI]
  (or arXiv:2209.14245v1 [cs.NI] for this version)

Submission history

From: Zijia Zhong [view email]
[v1] Tue, 6 Sep 2022 13:18:23 GMT (4490kb,D)

Link back to: arXiv, form interface, contact.