Scholarlink

<p>In urban expressway traffic systems, the merging area, where multiple traffic streams converge, is a critical node that frequently experiences traffic conflicts. It has become a significant bottleneck that constrains both the operational efficiency and safety of the road network. Relying on the high-precision motion control and efficient information exchange capabilities of Connected Automated Vehicles (CAVs), the traffic efficiency and collaborative scheduling in the merging area can be significantly improved. This paper proposes a method for optimizing the merging sequence and planning the trajectory based on a two-layer architecture. The upper layer employs an improved Monte Carlo Tree Search (MCTS) algorithm to achieve global optimization of the vehicle merging sequence. The lower layer transforms the trajectory planning problem into an optimal control problem, where a fuzzy logic evaluation mechanism is applied to adjust the trajectory based on multiple criteria, integrating both safety and comfort considerations. The proposed MCTS-based strategy effectively enhances coordination in multi-vehicle merging scenarios and significantly reduces delay under medium-to-high traffic demand, while the improvement is relatively limited under low-demand conditions.</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

A Dual-Layer Optimization Framework for Merging Sequence and Trajectory Planning of Connected Automated Vehicles

  • 期刊论文

  • 发表日期：

    2026年4月21日

查看全文
  • Yanling Zhao,
  • Huan Zhang,
  • Guohui Liu,
  • Jianghong Li,
  • Xin Liu

摘要

In urban expressway traffic systems, the merging area, where multiple traffic streams converge, is a critical node that frequently experiences traffic conflicts. It has become a significant bottleneck that constrains both the operational efficiency and safety of the road network. Relying on the high-precision motion control and efficient information exchange capabilities of Connected Automated Vehicles (CAVs), the traffic efficiency and collaborative scheduling in the merging area can be significantly improved. This paper proposes a method for optimizing the merging sequence and planning the trajectory based on a two-layer architecture. The upper layer employs an improved Monte Carlo Tree Search (MCTS) algorithm to achieve global optimization of the vehicle merging sequence. The lower layer transforms the trajectory planning problem into an optimal control problem, where a fuzzy logic evaluation mechanism is applied to adjust the trajectory based on multiple criteria, integrating both safety and comfort considerations. The proposed MCTS-based strategy effectively enhances coordination in multi-vehicle merging scenarios and significantly reduces delay under medium-to-high traffic demand, while the improvement is relatively limited under low-demand conditions.