Scholarlink

<p>Low-speed merge scenarios bring challenges to automated driving techniques since there is strong interaction and lack of coordination between vehicles. To address this problem, a novel game-theoretic decision-making framework for autonomous vehicles in low-speed merge scenarios is proposed. The trajectory planning of vehicles is carried out based on fifth-order polynomial curves, which in turn constructs the action space for vehicles’ decision-making. Cost functions considering safety and efficiency are designed for both the merging vehicle and the target vehicle to quantify the cost or payoffs for traffic participants. Next, a hierarchical decision-making framework for autonomous vehicles is proposed, consisting of up-level dynamic game decision-making and low-level security protection. Finally, through the simulations in Matlab, the effectiveness of the proposed algorithm for low-speed merge scenarios is verified.</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 novel game-theoretic decision-making framework for autonomous vehicles in low-speed merge scenarios

  • 期刊论文

  • 发表日期：

    2026年1月10日

查看全文
  • Qian He,
  • Lei Zhong,
  • Xinchang Liu,
  • Jianbin Qiu,
  • Jiaqiu Zheng,
  • Qiyu Wang

摘要

Low-speed merge scenarios bring challenges to automated driving techniques since there is strong interaction and lack of coordination between vehicles. To address this problem, a novel game-theoretic decision-making framework for autonomous vehicles in low-speed merge scenarios is proposed. The trajectory planning of vehicles is carried out based on fifth-order polynomial curves, which in turn constructs the action space for vehicles’ decision-making. Cost functions considering safety and efficiency are designed for both the merging vehicle and the target vehicle to quantify the cost or payoffs for traffic participants. Next, a hierarchical decision-making framework for autonomous vehicles is proposed, consisting of up-level dynamic game decision-making and low-level security protection. Finally, through the simulations in Matlab, the effectiveness of the proposed algorithm for low-speed merge scenarios is verified.