引用本文:张家旭,周时莹,施正堂,赵健,朱冰.采用滑模条件积分的无人驾驶汽车弯道超车路径规划与跟踪控制[J].控制理论与应用,2021,38(2):197~205.[点击复制]
ZHANG Jia-xu,ZHOU Shi-ying,SHI Zheng-tang,ZHAO Jian,ZHU Bing.Path planning and tracking control for corner overtaking of driverless vehicle using sliding mode technique with conditional integrators[J].Control Theory and Technology,2021,38(2):197~205.[点击复制]
采用滑模条件积分的无人驾驶汽车弯道超车路径规划与跟踪控制
Path planning and tracking control for corner overtaking of driverless vehicle using sliding mode technique with conditional integrators
摘要点击 1951  全文点击 685  投稿时间：2020-03-01  修订日期：2020-09-10
查看全文  查看/发表评论  下载PDF阅读器
DOI编号  10.7641/CTA.2020.00110
  2021,38(2):197-205
中文关键词  弯道超车  路径规划与跟踪控制  五次多项式曲线  滑模条件积分控制
英文关键词  corner overtaking  path planning and tracking control  quintic polynomial curve  sliding mode technique with conditional integrators
基金项目  国家自然科学基金项目(51775235), 国家重点研发计划课题(2018YFB0105103)资助.
作者单位E-mail
张家旭* 吉林大学 汽车仿真与控制国家重点实验室 zhjx_686@163.com 
周时莹 中国第一汽车集团有限公司 智能网联研发院  
施正堂 浙江亚太机电股份有限公司 智能汽车控制系统研究院  
赵健 吉林大学 汽车仿真与控制国家重点实验室  
朱冰 吉林大学 汽车仿真与控制国家重点实验室  
中文摘要
      针对无人驾驶汽车弯道超车场景, 基于五次多项式曲线提出一种路径规划算法和基于滑模条件积分控制 方法提出一种路径跟踪控制策略. 首先, 为了简化无人驾驶汽车弯道超车路径规划过程, 将其分解成驶入超车道路 径、保持超车道内行驶路径和驶回原车道路径3部分, 并基于五次多项式曲线分别规划出满足汽车侧向加速度约束 和动态障碍物避障约束的驶入超车道路径和驶回原车道路径, 3部分路径平滑连接构成无人驾驶汽车弯道超车路 径. 随后, 为了高精度的跟踪规划的路径, 采用滑模条件积分控制方法设计具有强鲁棒性优点且无抖振的无人驾驶 汽车路径跟踪控制策略, 并采用李雅普诺夫稳定性理论证明其收敛性. 最后, 进行了算法的仿真验证, 结果表明: 所 提出无人驾驶汽车路径规划算法和路径跟踪控制策略可以引导无人驾驶汽车安全、舒适的完成弯道超车操作, 并 且相对于斯坦利方法, 具有更快的响应速度和更高的跟踪精度.
英文摘要
      Aiming at the corner overtaking scene of driverless vehicle, a path planning algorithm is proposed based on quintic polynomial curve and a path tracking control strategy is proposed based on sliding mode technique with conditional integrators. Firstly, in order to simply the process of the corner overtaking path planning, it is divided into the first segment from the original lane to the overtaking lane, the second segment on the overtaking lane, and the third segment from the overtaking lane to the original lane, and then the first segment and the third segment meeting the constraints of the vehicle lateral acceleration and the dynamic obstacle avoidance are planned based on quintic polynomial curve. All the three segments are connected smoothly to construct the corner overtaking path. Then, in order to track the planned path with high precision, the path tracking control strategy with strong robustness and without chattering is designed based on sliding mode technique with conditional integrators, and its convergence is proved by Lyapunov stability theory. Finally, the algorithm is verified by simulation, and the results show that the proposed path planning algorithm and path tracking control strategy can guide the driverless vehicle to achieve the corner overtaking operation safely and comfortably, and have faster response speed and higher tracking accuracy than Stanley method.