車両追従モデル構築に利用する簡易模擬データの提案  [in Japanese] Proposal of Simple Hypothesis Data Utilized in Process of Constructing Car-following Model  [in Japanese]

Access this Article

Search this Article

Author(s)

Abstract

我々はファジィニューラルネットワークを用いて,実測データを学習するだけで,車両追従が再現できるモデルを提案した.しかし,その車両追従モデルは実測データに依存するため,交通シミュレーションへ応用するには,どのぐらい,そしてどのような実測データを用意する必要があるのかが問題となる.実際の運転においては,車両の最大速度や加減速範囲が道路状況や交通信号などの交通状況に制限される. 一方,先行車速度の追従車への伝播遅れ時間は運転個性によって違う.さらに,実交通では交通信号などの交通状況の影響で,「停止・加速・減速・停止」のような速度変化が多い.本研究では,以上のような特性に注目し,速度変動がサイン波状である簡易模擬データを提案した.そして,実測データの代わりに簡易模擬データによって,車両追従モデルを構築する可能性を検討する.By using the Fuzzy Neural Network (FNN), a car-following model can be built to represent the actual driving. However we have a problem how many, and what kind of actual data we should prepare for simulating practical traffic because the car-following model rely on the actual data. In the actual driving, the maximum speed or the range of acceleration for cars is limited by road conditions and traffic signals, while the time lag of speed between the preceding and the following cars is different from the character of drivers. Moreover, there is variety of the speed just as stop, acceleration, deceleration, and stop in the actual driving. In this research, we propose simple hypothesis data in which the variation of speed of the car is like sinusoidal wave because we pay attention to the above characteristics. And we discuss the possibility for constructing the car-following model by the simple hypothesis data instead of the actual data.

By using the Fuzzy Neural Network (FNN), a car-following model can be built to represent the actual driving. However we have a problem how many, and what kind of actual data we should prepare for simulating practical traffic because the car-following model rely on the actual data. In the actual driving, the maximum speed or the range of acceleration for cars is limited by road conditions and traffic signals, while the time lag of speed between the preceding and the following cars is different from the character of drivers. Moreover, there is variety of the speed just as stop, acceleration, deceleration, and stop in the actual driving. In this research, we propose simple hypothesis data in which the variation of speed of the car is like sinusoidal wave because we pay attention to the above characteristics. And we discuss the possibility for constructing the car-following model by the simple hypothesis data instead of the actual data.

Journal

  • 情報処理学会論文誌数理モデル化と応用(TOM)

    情報処理学会論文誌数理モデル化と応用(TOM) 48(SIG15(TOM18)), 47-54, 2007-10-15

    Information Processing Society of Japan (IPSJ)

References:  15

Codes

  • NII Article ID (NAID)
    110006419281
  • NII NACSIS-CAT ID (NCID)
    AA11464803
  • Text Lang
    JPN
  • Article Type
    Article
  • ISSN
    1882-7780
  • NDL Article ID
    8951795
  • NDL Call No.
    Z74-C192
  • Data Source
    CJP  NDL  NII-ELS  IPSJ 
Page Top