农机轮式机器人自动驾驶控制研究及试验  

Research and Experiment on Autonomous Driving Control ofAgricultural Machinery Wheeled Robot

作　　者：余科松 武志明[1] 马玮 Yu Kesong;Wu Zhiming;Ma Wei(School of Agricultural Engineering,Shanxi Agricultural University,Jinzhong 030801,China)

机构地区：[1]山西农业大学农业工程学院,山西晋中030801

出　　处：《农机化研究》2025年第7期271-278,共8页Journal of Agricultural Mechanization Research

基　　金：山西省重点研发项目(201903D221029)。

摘　　要：农机人工驾驶需要驾驶员依靠自身主观经验进行操作控制,在农忙时期,高强度的工作压力会增加操作难度,降低作业精度,甚至引发安全问题。为此,对农机自动驾驶控制方案展开了研究,旨在实现一种基于STM32F107单片机的低成本农机轮式机器人控制系统。系统由ARM控制核心、定位采集和记录模块、转向控制模块和接口电路等多个组件组成,能够实时获取农机的位置信息和规划路径,可根据定位偏差信息自动发布转向控制命令,从而达到自动导航作业的目的。主要研究内容分两点:一是基于二维平面坐标系,通过边界坐标点进行了作业路径规划,生成了作业基准线;二是进行了农机移动特性分析,建立了农机动力学和运动学数学模型,以此为基础研究了转向控制与路径追踪的数学关系,并提出了基于农机航偏角控制的追踪方案。同时,为减少ARM运算量、提升追踪效果,采用模糊控制思想,建立了模糊控制查询表,构建了基于路径偏差和航向误差的转向角模糊控制规则,并分别设置了模糊输入输出的隶属度函数,将复杂运算简化为简单的查表操作。试验结果表明:模型车系统初始化时间为4.4 s,在平均车速为0.618 m/s的情况下,路径追踪的横向偏差不超过5.4 cm。Manual driving of agricultural machinery requires operators to rely on their subjective experience for operational control.During busy farming periods,high intensity work pressure can increase operational difficulty,reduce operational accuracy,and even cause safety issues.In response to this,this article conducted research on the automatic driving control scheme of agricultural machinery,aiming to implement a low-cost agricultural machinery wheeled robot control system based on the STM32F107 microcontroller.The system was composed of multiple components such as ARM control core,positioning acquisition and recording module,steering control module,and interface circuit.It could obtain real-time position information and plan paths of agricultural machinery,and automatically issue steering control commands based on positioning deviation information,thus achieving the purpose of automatic navigation homework.The main research content was divided into two points:Firstly,based on a two-dimensional plane coordinate system,the task path was planned through boundary coordinate points,and the task baseline was generated;Secondly,the movement characteristics of agricultural machinery were analyzed,and the mathematical models of dynamics and kinematics of agricultural machinery were established.On this basis,the mathematical relationship between steering control and path tracking was studied,and the tracking scheme based on the yaw angle control of agricultural machinery was proposed.At the same time,in order to reduce the amount of ARM calculation and improve the tracking effect,the fuzzy control idea was used to establish a fuzzy control query table,constructed fuzzy control rules for steering angle based on path deviation and heading error,and set membership functions for fuzzy input and output,simplifying complex operations into simple table lookup operations.The test showed that the initialization time of the model car system was 4.4 s,at an average speed of 0.618 m/s,the lateral deviation of path tracking did not exceed

关 键 词：农机自动驾驶 单片机 路径规划 路径追踪 模糊控制 

分 类 号：S24[农业科学—农业电气化与自动化]