Unmanned Aerial Vehicle Path Planning for Traffic Estimation and Detection of Non-Recurrent Congestion

07/31/2018

∙

by   Cesar N. Yahia, et al.

∙

0

∙

share

Unmanned aerial vehicles (drones) can be used in traffic and road monitoring applications. We investigate the benefit of using drones for simultaneous traffic state estimation and incident detection. Specifically, we propose a coupled planning and estimation framework where we adaptively navigate a drone to minimize the uncertainty on parameter and traffic state estimates. We show that the use of a drone provides significant improvement in incident detection under congested conditions. Without a drone, the estimation procedure in congested conditions is not able to distinguish between observations due to congestion under normal operating conditions and similar observations due to a reduction in capacity.