Traffic microscopic simulation models are able to represent traffic conditions and their evolution over time. They take into account the geometrical aspects of transport infrastructures, driver behavior and, step by step, the vehicles’ movements on the road network. The most widespread applications of these models are referred to accident management, route guidance problems (possibility of evaluating control strategies in real time), adaptive control of traffic-light and ramp metering (management and control of the ramps and flow).To accomplish this goal observational data of vehicles’ kinematic characteristics have to be acquired. These elements are not always easily available. In scientific literature there are several techniques for the detection of traffic flow characteristics that are generally classified in: 'infrastructure-based' techniques with fixed devices (intrusive or non-intrusive), and 'non infrastructure-based' techniques, that employ mobile devices such as probe vehicles or detection systems with cameras.The aim of this paper is to compare speed profiles of probe vehicles acquired at the same time with two different 'non infrastructure-based' survey methods. The first one, commonly used in several scientific research, is based on positioning data recorded by a GPS unit. In this work, a differential GPS installed on board is used to improve data position accuracy. The second one is based on the assumption that it is possible to acquire kinematic characteristics of vehicles by a video camera mounted on an Unmanned Aerial Vehicle (UAV) that resumes study area from the nadiral point of view. Some Ground Control Points (GCP) are positioned inside the investigated area to minimize the errors of video processing analysis. Frames extracted from the video, and consequently georeferred, are useful to obtain speed profiles of the analyzed vehicles. These last steps are essential to minimize position errors due both to wind and to rotors vibrations.Further aim of this study is to determine the minimum number of GCP that minimizes the difference between speed profiles acquired by UAV and differential GPS.
|Numero di pagine||1|
|Stato di pubblicazione||Published - 2014|