Non-Connected Vehicle Detection Using Connected Vehicles

Non-Connected Vehicle Detection Using Connected Vehicles

Headshot of Srinivas Peeta. The link directs to their bio page.
Srinivas Peeta
Purdue University Logo. The link directs to the funded research led by this institution.

Principal Investigator(s):

Srinivas Peeta, Hockema Professor of Civil Engineering – Purdue University

Project Abstract:
Connected vehicle (CV) technologies are entering the realm of deployment. They have the potential to help drivers and vehicles make safe, reliable and informed decisions, and thereby to enhance network capacity and reduce congestion. However, during the transition to CV technologies, there will be mixed traffic streams of CVs (with vehicle-to-vehicle communication capabilities) and non-CVs. To improve the efficiency and reliability of traffic operations under mixed CV environments, there is the need not only for observable CV location data, but also unobservable non-CV location/trajectory to realize efficient and reliable CV-based applications. This study proposes a hidden Markov model, which is a probabilistic inference approach, to identify non-CV locations/trajectories. This methodology will be integrated with a cooperative-situation awareness framework. The proposed model will be analyzed using real-world vehicle trajectory data to aid the situational awareness of CVs under low market penetration rates.

Institution(s): Purdue University

Award Year2017

Research Thrust(s): Control & Operations, Enabling TechnologyModeling & Implementation

Project Form(s):