Scholarlink

<p>This study explores a deep reinforcement learning mechanism to obtain cooperative driving control of connected autonomous vehicles at a roundabout. A roundabout is one of the intersections without signal control, and we focus on it because high-level cooperation is required to merge and branch off at each junction. This study introduces three new mechanisms to the previous model that enable learning cooperative control: (<InlineEquation ID="IEq1"> <EquationSource Format="TEX">\(\textrm{i}\)</EquationSource> <EquationSource Format="MATHML"><math> <mtext>i</mtext> </math></EquationSource> </InlineEquation>) curriculum learning that decreases the new vehicle departure interval during the training so that the number of vehicles in a roundabout gradually increases, (<InlineEquation ID="IEq2"> <EquationSource Format="TEX">\(\textrm{ii}\)</EquationSource> <EquationSource Format="MATHML"><math> <mtext>ii</mtext> </math></EquationSource> </InlineEquation>) utilization of information about the vehicle’s destination, and (<InlineEquation ID="IEq3"> <EquationSource Format="TEX">\(\textrm{iii}\)</EquationSource> <EquationSource Format="MATHML"><math> <mtext>iii</mtext> </math></EquationSource> </InlineEquation>) an additional penalty for approaching walls and other vehicles for safe driving. We conducted simulation experiments to investigate the effectiveness of the proposed methods. The experimental results showed that the proposed methods enable the acquisition of cooperative vehicle control that can arrive at each destination while reducing the collision rate.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Exploration of deep reinforcement learning method for cooperative control of connected automated vehicles at a roundabout

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年8月28日

查看全文
  • Reo Nakaya,
  • Tomohiro Harada,
  • Yukiya Miura,
  • Kiyohiko Hattori,
  • Johei Matsuoka

摘要

This study explores a deep reinforcement learning mechanism to obtain cooperative driving control of connected autonomous vehicles at a roundabout. A roundabout is one of the intersections without signal control, and we focus on it because high-level cooperation is required to merge and branch off at each junction. This study introduces three new mechanisms to the previous model that enable learning cooperative control: ( \(\textrm{i}\) i ) curriculum learning that decreases the new vehicle departure interval during the training so that the number of vehicles in a roundabout gradually increases, ( \(\textrm{ii}\) ii ) utilization of information about the vehicle’s destination, and ( \(\textrm{iii}\) iii ) an additional penalty for approaching walls and other vehicles for safe driving. We conducted simulation experiments to investigate the effectiveness of the proposed methods. The experimental results showed that the proposed methods enable the acquisition of cooperative vehicle control that can arrive at each destination while reducing the collision rate.