| In view of the need for artificial driving in the process of plant protection in the orchard,the pesticide is easy to cause the fruit poisoning.Unreasonable spraying of pesticides,it can pollute the groundwater,and even cause environmental pollution and other issues.The object of this paper is to research the operation planning,row recognition and motion control technology.It is helpful to improve the working conditions,the quality and precision of the operation,which can finally achieve the purpose of separation.Firstly,this paper analyzes the background and significance of the research,and analyzes the research progress of the global path planning,local path planning and autonomous navigation in three aspects.In view of the problems existing in the current research on the technology of plant protection robot,the overall design scheme and the research route of the system are developed.Using the method of high precision GPS module and digital map technology carry out the operation planning of plant protection robot.Firstly,the feature points of row are obtained by using the land mapping technology.Then,with the help of GPRS wireless communication technology,the information is loaded into the Baidu maps API,and the feature map is constructed to determine the scheduled route of the work area of the robot.At the same time,it can realize the real-time positioning of the row,the continuous spraying of the broken point,the track playback and other functions.Based on the laser radar scanning technology,the environmental information of the row is collected,and the least square method is used to extract the boundary line and the navigation center line.The position and orientation information of the relative navigation center line of the plant protection robot is obtained,and the position and attitude information is output by CAN communication.The motion control part is mainly to establish the kinematics model,the pure tracking model and the steering control model of the plant protection robot.At the same time,a fuzzy controller is designed to realize the accurate path following of the robot,which is based on the fuzzy controller.At the end of the paper,some experiments are carried out to validate the recognition andmotion control.The experiment mainly includes stability ranging distance measuring instrument based on laser radar,the average error is 2.2mm;error calibration experiments,the maximum error is 25mm;the maximum error angle calibration is 3.53%,which meets the basic requirements for recognition of the ridge.The CAN communication experiment between the USB host and the main controller is designed to realize the transmission process of the relative position and orientation information of the robot.In the end,the simulation and the experiment of the movement of the grape and the grape were carried out to further illustrate the effectiveness and feasibility of the algorithm. |