The Vehicular Ad Hoc Network (VANET) holds promises for on-road security applications. In this paper, we utilize the VANET for surveillance purpose, tracking a noncooperative mobile target. We explore the possibilities of engaging Onboard Units (OBUs) and Roadside Units (RSUs) in a metropolitan VANET for tracking a vehicle that is on the run. The uncertainty associated with the unplanned locomotion of a vehicle in the metropolitan road network, that exhibits dynamic characteristics, such as different speed limits and time varying traffic congestion, makes vehicle tracking challenging. We present a tracking system composed of three operational modules: localization, tracking data collection and prediction of future locations of a target. Tracking messages are communicated among the OBUs and RSUs and are triggered on in probable areas where the target may be present. Therefore, another imperative element of the addressed problem is to scope the search to limit the number of OBUs and RSUs involved in the tracking operation. Our proposal does not presume any motion model for the target. A novel movement modeling technique utilizes OBU observations to classify the target's movement pattern. We propose a Dirichlet-multinomial model under the Bayesian estimation framework. The movement estimation is then exploited for predicting future locations of the target. The proposed method is analogous to chasing an on the run vehicle using police squad cars. We believe this approach holds potentials as an alternative to high-speed pursuits.

Additional Metadata
Keywords Bayesian estimation, broadcast control, intelligent transportation systems, intelligent vehicles, probabilistic path prediction, road network, target tracking, VANET
Persistent URL
Conference 2013 IEEE Intelligent Vehicles Symposium, IEEE IV 2013
Reza, T.A. (Tahsin Arafat), Barbeau, M, & Alsubaihi, B. (Badr). (2013). Tracking an on the run vehicle in a metropolitan VANET. Presented at the 2013 IEEE Intelligent Vehicles Symposium, IEEE IV 2013. doi:10.1109/IVS.2013.6629474