当前位置: X-MOL 学术Transp. Res. Part C Emerg. Technol. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
A platoon-based cooperative optimal control for connected autonomous vehicles at highway on-ramps under heavy traffic
Transportation Research Part C: Emerging Technologies ( IF 8.3 ) Pub Date : 2023-03-10 , DOI: 10.1016/j.trc.2023.104083
Yongjie Xue , Xiaokai Zhang , Zhiyong Cui , Bin Yu , Kun Gao

To improve traffic efficiency at highway on-ramps under heavy traffic, this study proposes a platoon-based cooperative optimal control algorithm for connected autonomous vehicles (CAVs). The proposed algorithm classifies CAVs on both mainline and on-ramp into multiple local platoons (LPs) according to their initial conditions (i.e., spacing and speed), which enables the algorithm to adapt to time-varying traffic volume. A distributed cooperative control for multiple LPs is designed which projects on-ramp LPs onto mainline to transform the complex 2-D multi-platoon cooperation problem into a 1-D platoon following control problem. An optimal control is applied to further consider the strict nonlinear safety spacing constraint and state limitations (e.g., maximum speed and acceleration), and an analytical solution to the optimal control is derived based on Pontryagin’s maximum principle. The consensus of intra-platoon and inter-platoon are analyzed, and sufficient conditions of the consensus are mathematically deducted based on Lyapunov stability theorem. Numerical simulations are conducted for different traffic demand levels and demand splits to verify the effectiveness of the proposed algorithm. The sensitivity analysis of maximum platoon sizes for mainline and on-ramp LPs is performed. A comparison with a baseline virtual platooning merging strategy is conducted, and results show that the proposed algorithm could significantly improve the average travel speed and traffic efficiency, and reduce total travel time.



中文翻译:

交通繁忙情况下高速公路匝道上联网自动驾驶车辆的基于队列的协同优化控制

为了提高交通繁忙情况下高速公路入口匝道的交通效率,本研究提出了一种基于队列的联网自动驾驶车辆 (CAV) 协作优化控制算法。所提出的算法根据初始条件(即间距和速度)将干线和匝道上的 CAV 分为多个本地排 (LP),这使该算法能够适应时变交通量。设计了一种用于多个 LP 的分布式协同控制,它将入口匝道 LP 投射到主线上,将复杂的二维多排合作问题转化为一维排跟随控制问题。应用最优控制以进一步考虑严格的非线性安全间距约束和状态限制(例如,最大速度和加速度),并基于Pontryagin极大值原理推导了最优控制的解析解。分析了排内和排间的共识,基于Lyapunov稳定性定理数学推导了共识的充分条件。针对不同的交通需求水平和需求拆分进行数值仿真,验证了所提算法的有效性。对干线和匝道 LP 的最大排规模进行了敏感性分析。与基线虚拟队列合并策略进行了比较,结果表明,所提出的算法可以显着提高平均行驶速度和交通效率,并减少总行驶时间。基于李亚普诺夫稳定性定理,从数学上推导了共识的充分条件。针对不同的交通需求水平和需求拆分进行数值仿真,验证了所提算法的有效性。对干线和匝道 LP 的最大排规模进行了敏感性分析。与基线虚拟队列合并策略进行了比较,结果表明,所提出的算法可以显着提高平均行驶速度和交通效率,并减少总行驶时间。基于李亚普诺夫稳定性定理,从数学上推导了共识的充分条件。针对不同的交通需求水平和需求拆分进行数值仿真,验证了所提算法的有效性。对干线和匝道 LP 的最大排规模进行了敏感性分析。与基线虚拟队列合并策略进行了比较,结果表明,所提出的算法可以显着提高平均行驶速度和交通效率,并减少总行驶时间。对干线和匝道 LP 的最大排规模进行了敏感性分析。与基线虚拟队列合并策略进行了比较,结果表明,所提出的算法可以显着提高平均行驶速度和交通效率,并减少总行驶时间。对干线和匝道 LP 的最大排规模进行了敏感性分析。与基线虚拟队列合并策略进行了比较,结果表明,所提出的算法可以显着提高平均行驶速度和交通效率,并减少总行驶时间。

更新日期:2023-03-10
down
wechat
bug