汽车工程 ›› 2024, Vol. 46 ›› Issue (2): 201-210.doi: 10.19562/j.chinasae.qcgc.2024.02.002

• • 上一篇    下一篇

面向强制换道场景的智能网联汽车协同换道策略

关书睿1,李克强1,周俊宇1,石佳1,孔伟伟2,罗禹贡1()   

  1. 1.清华大学车辆与运载学院,汽车安全与节能国家重点实验室,北京 100084
    2.中国农业大学工学院,北京 100083
  • 收稿日期:2023-06-29 修回日期:2023-07-28 出版日期:2024-02-25 发布日期:2024-02-23
  • 通讯作者: 罗禹贡 E-mail:lyg@tsinghua.edu.cn
  • 基金资助:
    国家重点研发计划项目(2022YFB2503200)

A Cooperative Lane Change Strategy for Intelligent Connected Vehicles Oriented to Mandatory Lane Change Scenarios

Shurui Guan1,Keqiang Li1,Junyu Zhou1,Jia Shi1,Weiwei Kong2,Yugong Luo1()   

  1. 1.School of Vehicle and Mobility,Tsinghua University,State Key Laboratory of Automotive Safety and Energy,Beijing  100084
    2.Engineering College,China Agricultural University,Beijing  100083
  • Received:2023-06-29 Revised:2023-07-28 Online:2024-02-25 Published:2024-02-23
  • Contact: Yugong Luo E-mail:lyg@tsinghua.edu.cn

摘要:

智能网联汽车的协同换道技术已得到广泛研究,但现有协同换道策略难以解决强制换道场景下的车辆协同问题,或对上游交通造成的负面影响较大。本文旨在面向强制换道场景需求,提出一种考虑理论最小安全距离的两阶段协同换道策略。首先,提出了两车协同换道系统控制架构,并面向车辆强制换道场景制定了协同换道方案。随后,设计了间距调整、协同换道两阶段的滚动时域轨迹规划方法,并求解理论最小安全距离作为间距调整阶段末态约束,解决了现有研究间距策略保守的问题。最后,通过仿真与硬件在环试验验证了所提策略的有效性、优越性与计算实时性。结果表明,所提策略可有效提升换道成功率,在保障换道安全性的条件下降低了换道行为对交通的影响,且在实际边缘云平台的实时计算、通信环境下具备可用性。

关键词: 智能网联汽车, 协同换道, 强制换道, 轨迹规划

Abstract:

Collaborative lane change technology for intelligent connected vehicles has been widely studied, but existing strategies can hardly solve the problem of vehicle collaboration in mandatory lane change scenarios or may cause notable impact on upstream traffic. For mandatory lane change scenarios demand, a two-stage cooperative lane change strategy considering theoretical minimal safety space is proposed in this paper. Firstly, the control architecture for a two-vehicle cooperative lane change system is proposed and a collaborative lane change scheme is developed for mandatory lane change scenarios. Then, a two-stage receding-horizon trajectory planning strategy of spacing adjustment and collaborative lane change is designed, where the theoretical minimum safe distance is embedded as a constraint of spacing adjustment stage, to solve the problem of conservative spacing strategies in existing research. Finally, numerical simulation and hardware in-loop experiments are performed to verify the effectiveness, advantages and computational real-time performance of the proposed strategy. The results show that the proposed strategy can effectively improve the success rate of lane change, reduce the negative traffic impact while ensuring lane change safety, and is also applicable in real time computing and communication environment of actual edge cloud platform.

Key words: intelligent and connected vehicles, cooperative lane change, mandatory lane change, trajectory planning