APPLICABILITY OF FEEDBACK-BASED TRAFFIC STATE ESTIMATOR FOR ANALYSIS OF TRAFFIC ACCIDENTS

DOI Web Site 5 References

Bibliographic Information

Other Title
  • フィードバック型交通状態推定手法の事故発生状況分析への適用可能性
  • フィードバックガタ コウツウ ジョウタイ スイテイ シュホウ ノ ジコ ハッセイ ジョウキョウ ブンセキ エ ノ テキヨウ カノウセイ

Search this article

Abstract

The focus of traffic control system of freeway has been slightly shifted from mitigating traffic congestion towards preventing traffic accident during the last decade as the upgrade of the freeway network connection. Due to the intensive effort to investigate the factors relating to the occurrence of traffic accident in the past studies, it has been revealed that the risk of traffic accident gets higher in the critical and congested traffic flow situation. However, the difficulty in collecting the precise and adequate data of traffic flow limits the further understanding of the question; what induces traffic accidents?<br>To overcome the limitation, this study investigates the applicability of a feedback-based traffic state estimator for the further understanding of the situation inducing traffic accidents. The feedback system is composed of Extended Kalman Filter (EKF) and Cell Transmission Model (CTM), which is extended to represent the influence of road alignment on traffic flow at sag sections. The parameters of a fundamental diagram, which are considered as an indicator of the efficiency of traffic flow, are set as the state variables in EKF. Then, this system is applied to 9 situations where traffic accidents occurred and the variation of the estimated state variables just before and after the traffic accidents is investigated. As a result, it is revealed: (i) the feedback system can estimate rational traffic state where traffic accidents occur, and (ii) in some cases the efficiency of traffic flow tends to deteriorate just before the traffic accidents.

Journal

References(5)*help

See more

Related Projects

See more

Keywords

Details 詳細情報について

Report a problem

Back to top