摘要
为提高结构化工况下无人驾驶汽车轨迹跟踪的精确性,提出一种融合纵向车速跟踪控制器的预瞄轨迹跟踪控制算法。针对四轮转向的特性,基于传统纯跟踪算法设计了四轮转向纯跟踪控制模型;针对纯跟踪算法跟踪效果受车速影响较大的缺陷,设计了纵向车速跟踪控制器并采用n维插值表方式在不同曲率处设置不同车速;将四轮转向纯跟踪算法融合纵向车速跟踪控制器,以达到更好的跟踪效果。仿真与实车试验表明,试验平台可较精确地跟踪规划轨迹并平稳抵达目标终点,实车轨迹跟踪过程中最大横向误差控制在0.497m。
To improve the tracking accuracy of unmanned vehicles under structured working conditions,a preview trajectory tracking control algorithm with longitudinal vehicle speed tracking controller is proposed.Firstly,according to the characteristics of four-wheel steering,a pure pursuit control model of four-wheel steering was designed based on the traditional pure pursuit algorithm.Secondly,aiming at the defect that the tracking effect of pure pursuit algorithm was greatly affected by the vehicle speed,a longitudinal vehicle speed tracking controller was designed,and different vehicle speeds were set at different curvatures by N-dimensional interpolation table.Finally,four-wheel steering pure pursuit algorithm was combined with longitudinal speed tracking controller to achieve better tracking effect.The simulation and real vehicle test show that the test platform can accurately track the planned trajectory and smoothly reach the target destination,and the maximum lateral error in the track tracking process of real vehicle is controlled at 0.497m.
作者
徐豪迪
李强
梁洪波
杨爱喜
XU Hao-di;LI Qiang;LIANG Hong-bo;YANG Ai-xi(School of Mechanical and Energy Engineering,Zhejiang University of Science and Technology,Hangzhou Zhejiang 310023,China;Department of Automobile and Engineering,Anhui Vocational and Technical College of Transportation,Hefei Anhui 230051,China;College of Engineers,Zhejiang University,Hangzhou Zhejiang 310015,China)
出处
《计算机仿真》
2024年第4期145-150,共6页
Computer Simulation
基金
浙江省自然科学基金项目(LY21E050001)
汽车新技术安徽省工程技术研究中心开放基金(QCKJ202105)。
关键词
无人驾驶
纯跟踪算法
四轮转向
纵向车速跟踪控制器
实车试验
Unmanned driving
Purepursuit algorithm
Four-wheel steering
Longitudinal vehicle speed tracking controller
Real vehicle verification