Scholarlink

With the popularization of scenarios where intelligent connected vehicles (ICVs) and human-driven vehicles (HDVs) coexist, the current high-priority vehicle traffic strategy is difficult to effectively play a role in mixed traffic flow. Therefore, a reinforcement learning based method is proposed. Firstly, use SUMO to build the model. Secondly, the Proximal Policy Optimization (PPO) algorithm is adopted to adjust the longitudinal speed of ICVs, the longitudinal spacing between basic units of sparse heterogeneous mixed traffic flow, and collaborate with high-priority vehicles to change lanes and overtake. Finally, validate the model in different scenarios. The results indicate that this method is suitable for scenarios with heterogeneous mixed traffic flow and full ICVs; Compared to the lane pre-clearance strategy, this strategy reduces the passing time of high-priority vehicles by 17.39% and 5.09% in distance, respectively; The overall number of lane changes has decreased by 75%, significantly reducing the impact on weekly traffic.

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

Decision-Making for Priority Vehicle Transit Based on Multi-agent Reinforcement Learning

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Yi Yang,
  • Zhongguo Huang,
  • Qing Gu,
  • Yu Meng,
  • Huazhen Fang

摘要

With the popularization of scenarios where intelligent connected vehicles (ICVs) and human-driven vehicles (HDVs) coexist, the current high-priority vehicle traffic strategy is difficult to effectively play a role in mixed traffic flow. Therefore, a reinforcement learning based method is proposed. Firstly, use SUMO to build the model. Secondly, the Proximal Policy Optimization (PPO) algorithm is adopted to adjust the longitudinal speed of ICVs, the longitudinal spacing between basic units of sparse heterogeneous mixed traffic flow, and collaborate with high-priority vehicles to change lanes and overtake. Finally, validate the model in different scenarios. The results indicate that this method is suitable for scenarios with heterogeneous mixed traffic flow and full ICVs; Compared to the lane pre-clearance strategy, this strategy reduces the passing time of high-priority vehicles by 17.39% and 5.09% in distance, respectively; The overall number of lane changes has decreased by 75%, significantly reducing the impact on weekly traffic.