搭载改进滑模控制的自动驾驶汽车轨迹跟踪控制  

Trajectory tracking control of autonomous vehicles with optimized sliding mode control

作　　者：李磊 李军[1,2] 张世义[3] LI Lei;LI Jun;ZHANG Shiyi(School of Mechatronics and Vehicle Engineering,Chongqing Jiaotong University,Chongqing 400074,China;Chongqing Key Laboratory of Rail Vehicle System Integration and Control(Chongqing Jiaotong University),Chongqing 400074,China;School of Shipping and Naval Engineering,Chongqing Jiaotong University,Chongqing 400074,China)

机构地区：[1]重庆交通大学机电与车辆工程学院,重庆400074 [2]重庆市轨道交通车辆系统集成与控制重点实验室(重庆交通大学),重庆400074 [3]重庆交通大学航运与船舶工程学院,重庆400074

出　　处：《汽车安全与节能学报》2020年第4期503-510,共8页Journal of Automotive Safety and Energy

基　　金：国家自然科学基金资助项目(51305472)。

摘　　要：为了提高自动驾驶汽车在各种干扰因素下轨迹跟踪控制的鲁棒性,提出了一种改进滑模控制的鲁棒控制器。通过采用时变误差模型来设计滑模面,而不是恒定滑模面,进而缩短滑模控制器到达滑动阶段的时间。利用了MATLAB对改进后的滑模控制器,进行了滑模面仿真、变道工况仿真以及超车工况仿真,并与未改进的滑模控制器进行了对比。结果表明:改进后的滑模控制器到达滑动阶段的时间减少了33.10%;在变道仿真工况中鲁棒性提高了7.44%;在超车仿真工况中鲁棒性提高了10.45%。因而,在自动驾驶汽车轨迹跟踪控制研究中,该控制器可以提升控制鲁棒性。A robust controller was proposed with an improved sliding-mode-controller to improve the robustness of autonomous-vehicle trajectory-tracking under various disturbance factors.Designed a slidingmode-surface instead of a constant sliding-mode-surface by adopting the time-varying error model to shorten the time for the sliding-mode-controller to reach the sliding phase.Simulated an optimized slidingmode-controller for the driving condition with sliding-mode-surface,the lane change,and the overtaking by using a MATLAB software.The simulation results were compared with that by the unoptimized sliding-modecontroller.The results show that the time is reduced by 33.10%for the optimized sliding-mode-controller to reach the sliding phase;the robustness is increased by 7.44%in the lane change simulation conditions;and the robustness is increased by 10.45%in the overtaking simulation conditions.Therefore,the controller improves the control robustness in the trajectory-tracking control research.

关 键 词：自动驾驶汽车 轨迹跟踪 鲁棒性控制 滑模控制 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]