- 著者
-
織田 利彦
- 雑誌
- 研究報告高度交通システムとスマートコミュニティ(ITS) (ISSN:21888965)
- 巻号頁・発行日
- vol.2019-ITS-78, no.8, pp.1-6, 2019-08-30
Recently owing to fiscal constraints, installation of traffic control systems and equipment has been reduced. In addition, the renewal and repair of the infirastruct has been delayed remarkably. These situations have led to decrepit infrastructure. Ahove all, the operations of vehicle detectors monitoring traffic flows have degraded and failed to function property. VICS (Vehicle Information & Communication System) Center has provided drivers with useful traffic information, and travel time service has offered excellent assistance in route choice decision. However, the degraded detectors has resulted in missing traffic flow data and lowered the quality of the service. Moreover, unusual traffic flows caused by large scale and special events such as annual functions have often occurred, and heavy congestion can be expected on the srurounding roads of events site. This paper describes a study on travel time estimation for urban roadways with missing detector data in extraordinary traffic flows by using an ensemble learning algorithm. In order to verify the effectiveness of the approach, we conducted an experimental study on arterial roads around Nakayama Racecourse on the days of the horse racing, and found that our approach has the potential to estimate the missing travel time data and enable the prediction of traffic flow to enhance the information service.