Microscopic Estimation of Freeway Vehicle Positions from the Behaviors of Connected VehiclesPoster
The modern passenger vehicle is an incredibly sophisticated machine, with computer monitoring and control of most of its functions, including speed, acceleration, heading, and GPS-enabled position. Vehicles are beginning to communicate these data wirelessly to other vehicles and to roadside equipment. These communicating “probe” vehicles will drastically impact the way traffic is managed, allowing for more responsive traffic signals, more comprehensive traffic information, and more accurate travel time prediction. Research suggests that to begin experiencing these benefits, at least 20% of vehicles must act as probes, with benefits increasing with higher participation rates. Because of bandwidth limitations of and slow rollout of the technology, only a portion of vehicles on the roadway will be able to act as probes. Fortunately, the behavior of these probe vehicles may suggest the locations of nearby non-probe vehicles, thereby artificially augmenting the penetration rate and generating greater benefits. We propose an algorithm to predict the location of non-communicating vehicles based on the behaviors of nearby probe vehicles. By employing driver behavior models and rolling estimation techniques, the algorithm is able to predict the locations of 30% of vehicles with 9-meter accuracy when only 10% of vehicles are communicating, theoretically leading to immediate improvements in many probe vehicle applications.
University of Virginia Engineering Research Symposium
University of Virginia
Virginia Department of Transportation