Mobile Phone-Based Traffic State Estimation ? The Penetration Rate Issue ?

概要

論文の詳細を見る
This paper studies the relationship between the penetration rate and the effectiveness of the mobile phone-based traffic state estimation. The study reveals that the higher the penetration rate, the better the traffic state estimation is. In the real world applications, however, there is no-way to assure the penetration rate to be always relevant, especially when the system has just been launched. This study proposes notable solutions to improve the effectiveness of the traffic state estimation. Two novel "velocity-density inference" models, namely the "adaptive" and the "adaptive feedback" velocity-density inference circuits, and a neural network-based prediction model are proposed to assure the effectiveness of the traffic state estimation in cases of low penetration rate. These innovations are practically meaningful since they help to guarantee a high accurate traffic state estimation, even with a very low penetration rate, namely 0%, for instance. The experimental evaluations reveal the effectiveness as well as the robustness of the proposed solutions.
2011-05-26