基于动力学建模的农机路径跟踪鲁棒滑模控制算法研究  

Robust Sliding Mode Control Method for Path Tracking of Unmanned Agricultural Vehicles Based on Dynamic Modeling

作　　者：葛志康 白晓平[1,3] 王卓[1,3] 王枭雄 Ge Zhikang;Bai Xiaoping;Wang Zhuo;Wang Xiaoxiong(Shenyang Institute of Automation,Chinese Academy of Sciences,Shenyang 110016,China;University of Chinese Academy of Sciences,Beijing 100049,China;Institutes for Robotics and Intelligent Manufacturing,University of Chinese Academy of Sciences,Shenyang 110169,China)

机构地区：[1]中国科学院沈阳自动化研究所,沈阳110016 [2]中国科学院大学,北京100049 [3]中国科学院机器人与智能制造创新研究院,沈阳110169

出　　处：《农机化研究》2025年第8期1-9,共9页Journal of Agricultural Mechanization Research

基　　金：国家重点研发计划项目(2021YFD2000305)。

摘　　要：为了解决无人农业车辆在复杂路面条件下作业时存在参数摄动、未知扰动而导致控制精度下降的问题,首先进行了农业轮式车辆侧向动力学分析,建立了考虑前视距离的车辆动力学模型;然后设计了一种基于系统未知参数边界条件的鲁棒滑模控制方法,对系统参数大幅摄动、外部干扰和作业过程中的路面环境变化具有强鲁棒性,并且在实际应用时无需获取未知或难以测量的车辆动力学参数的精确值。田间试验结果表明:在路面条件发生突然变化的情况下,滑模控制方法的跟踪平均误差绝对值为0.1609 m,相对于状态反馈控制(State Feedback Control,SFC)方法提高了43.8%;跟踪误差绝对值极值为0.3650 m,相对于SFC控制方法提高了13.5%。在拖拉机以2.5 m/s的速度进行连续作业时,滑模控制方法跟踪平均误差绝对值为0.0980 m,最大偏差为0.2682 cm,标准差为0.1163,相比于SFC控制方法分别提升了34.80%、29.23%、40.27%,满足无人农业车辆在复杂环境下导航作业的实际需求。In order to solve the decrease in control accuracy problemscaused by parameter perturbation and unknown disturbances of unmanned agricultural vehicles in complex road condition,a lateral dynamics analysis of agricultural wheeled vehicles was conducted and vehicle dynamics model considering forward looking distance was established primarily.Then,a robust sliding mode control method based on boundary conditions of unknown system parameters was designed,whichhad strong robustness against significant perturbation of system parameters,external disturbances,and changes in road environment during operation.In practical applications,it was unnecessarytoobtain accurate values of unknown or difficult to measure vehicle dynamic parameters.The field test results revealed that in the case of sudden changes in road conditions,thetracking average error absolute value ofslidingmode control method was 0.1609 m,which was 43.8%higher than the path tracking method of State Feedback Control(SFC).The extreme absolute value of tracking error was 0.3650 m,which was 13.5%higher than the path tracking method of SFC.When the tractor was operating continuously at the speed of 2.5 m/s,theslidingmode control methodhad a tracking average error absolute value of0.0980 m,with a maximum deviation of 0.2682 cm and a standard deviation of 0.1163.Compared with the SFC control method,it was improved by 34.80%,29.23%,and 40.27%,respectively,which could meet the actual needs of navigation and operation in complex environments for unmanned agricultural vehicles.

关 键 词：无人农机 路径跟踪 滑模控制 鲁棒性 动力学建模 

分 类 号：S219.032.1[农业科学—农业机械化工程]