混合自动驾驶场景多换道需求下的主动间隙适配和换道序列规划

Proactive Gap Adaption and Sequence Planning for Multiple Lane-changing Requests under Mixed Autonomous Vehicle Flow

为解决城市道路拥挤状态下车辆可行换道间隙少、换道时间长,进而导致停车等待、堵塞后续车流等问题,本文基于智能网联环境下车辆之间的交互协同,提出了一种混合自动驾驶场景多换道需求下的主动间隙适配和换道序列规划模型。首先,采用多项式和三角函数分别描述了换道过程中的空间轨迹和速度曲线,得出了换道间隙可行性判别依据。在此基础上,构造了换道启动可行状态集合,并构建了单个换道请求下的间隙适配和协同换道的最优控制模型。然后,考虑多换道需求,构建了主动间隙适配和换道序列规划模型,对换道序列进行整体优化。最后,设计了数值仿真实验,验证了本文模型可对多换道需求进行时空优化。仿真实验结果表明,本文模型能够降低换道行为对城市道路通行能力的影响,且适用于不同的交通需求,模型可提升24%的道路通过量。

During rush hours,the feasible gaps for the lane-changing behavior of vehicles decreases,and lane-changing duration increases.Vehicles that fail to change lanes have to wait and hence,block the upstream traffic flow.As connected and autonomous vehicles(CAVs)communicate and cooperate,a proactive gap adaption and lane-changing sequence planning model that works in a mixed CAV scenario when multiple lane-changing requests coexist was established in this study.First,polynomial and trigonometric functions were employed to express the spatial trajectory and the speed profile during lane changing,based on which the lane-changing feasibility criterion was established.Next,the feasible state set of the lane-changing initiation moment was constructed.The cooperative lane changing for a single lane-changing request was formulated as an optimal control model.For multiple lane-changing requests,the lane-changing sequence was optimized as a whole.The numerical simulations show that the proposed model can optimize the spatial-temporal trajectory.The software simulations reveal that the model can decrease the influence of lane-changing behavior on throughput at different demand levels.The improvement was up to 24%.