车辆跟随控制策略的状态可达集建模及验证方法
作者:
作者单位:

山东理工大学交通与车辆工程学院

作者简介:

通讯作者:

中图分类号:

TP273

基金项目:

国家自然科学基金项目(面上项目,61573009);山东省自然科学基金(ZR2018LF009,ZR2018PEE016)


A Modeling and Verification Method of State Reachable Set for Vehicle Following Control Strategy
Author:
Affiliation:

School of Transportation and Vehicle Engineering, Shandong University of Technology

Fund Project:

The National Natural Science Foundation of China (General Program,Grant No. 61573009);Natural Science Foundation of Shandong province, China (Grant No. ZR2018LF009,ZR2018PEE016)

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    由于车辆动态行为的不确定性,造成跟随车辆控制的不确定性。然而传统车辆跟随控制方法只能针对车辆的单一行为动态的不确定性进行建模,无法遍历车辆所有可能的控制输入,因而既无法一次性提供跟随策略下完整的可行控制方案,也不足以在理论上保证对策略安全检测的可信性.为此提出车辆跟随控制策略的状态可达集建模及验证方法.该方法将控制策略转换为可用可达集计算和表征的多级安全判定事件,利用随机可达集的状态遍历特征描述车辆控制输入的不确定性,通过对可达集交集的判断,辨识所有初始条件对应的危险与安全控制行为,为驾驶员提供完整的可选择控制方案,然后利用马尔科夫链逼近可达集,近似表达车辆行为的不确定性,依据驾驶员行为习惯,统计验证策略的安全性,从而实现对控制策略的有效建模分析.实验结果表明,提出的建模及验证方法不仅完备地表征了车辆不确定行为,提供了交通情形中跟随策略相应的完整控制方案,也实现了对策略安全性的精确验证.

    Abstract:

    The uncertainty of the following vehicle control is caused by the uncertainty of vehicle dynamic behavior. However, traditional vehicle following control methods can only formulate the uncertainty of specific dynamic of vehicle and be impossible to traverse all the possible system inputs with finite operations. These methods are insufficient to provide all the viable control decision according to the following strategy at once, and unable to ensure the reliability of safety detection of strategy in theory. Therefore, a modeling and verifiction method of state reachable set for vehicle following control strategy is proposed in this work. Based on the transformation of control strategy into the corresponding multi-level safety judgment events that can be calculated and represented by reachable set, the stochastic reachable set is applied to characterize the uncertain control input of vehicle given its ergodicity in this method. According to the judgment of the intersection of state reachable sets for related vehicles, both dangerous behaviors and safety behaviors can be detected refering to the control strategy. The whole alternative control schemes are provided to the drivers of following vehicles. The stochastic reachable sets of vehicles are abstracted to Markov chains that can be used to express the states change of system, and verify the safety of the following control strategy statistically based on the drivers behavior habits. Therefore, an effective modeling and analysis of the control strategy can be realized. The experiment results show that the proposed modeling and verfication method not only characterizes the uncertain behavior process of vehicle completely, provides the integrated control decision schemes according to the follwing strategy under the traffic situation, but also verifies the safety of control strategy accurately.

    参考文献
    相似文献
    引证文献
引用本文
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2019-11-11
  • 最后修改日期:2021-01-25
  • 录用日期:2020-03-18
  • 在线发布日期:
  • 出版日期: