Research On Autonomous Variable Spray Control System In Orchard Based On Lidar

Chemical pesticides are the main methods for pest control.The current common way of plant protection in orchard is continuous spray operation with artificial spray or driving air sprayer,which is low in efficiency and environment,and at the same time endangers the health of operators.In order to improve the efficiency and autonomy of orchard sprayer,reduce the labor intensity of operators and promote the efficiency of pesticide reduction,this paper studied the autonomous variable spray control system in orchard.A self steering algorithm for orchard was proposed based on 3D LIDAR.A variable spray model based on vertical structure was established,and a variable spray control system was designed to match the plant canopy characteristics in real time.The main research contents are as follows:(1)In this paper,the structure of orchard variable sprayer is designed by Solid Works 3D software.A variable spray system composed of spray support,sprinkler head,water pump,variable spray solenoid valve and air delivery system is built;Ackerman mobile chassis can carry out walking control through remote control or autonomous navigation to realize man-machine separation and reduce direct contact between personnel and liquid medicine;An orchards autonomous variable spray system was built,which was based on 3D lidar perception,2-D lidar detection of plant canopy characteristics,microcomputer processing algorithm,independent control of sprinkler flow and STM32F1 based controller.(2)In order to improve the efficiency and autonomy of orchard variable spray machine,an autonomous navigation algorithm for orchard was proposed based on 3D LIDAR.Firstly,in order to improve the accuracy and real-time of fruit tree detection,a fruit tree detection method based on 3D lidar and optimized DBSCAN algorithm is proposed;Then,the RANSAC algorithm is used to fit the straight line at the center of the left and right fruit trees,and the tree row center is taken as the target navigation line;Finally,the pure tracking algorithm is used to control the orchards variable sprayer autonomously driving along the guided path.The experimental results show that the average positive detection rate of fruit tree detection of the optimized DBSCAN algorithm is increased from 89.42% to 95.61%,and the algorithm time is significantly reduced.When the spray speed is 0.3m/s,0.5m/s and 0.8m/s,the mean lateral deviation of tree tracking is 4.6cm,5.4cm and5.9cm.This method can meet the requirements of autonomous spray operation between orchard sprayers.(3)In order to improve the utilization rate of liquid medicine and reduce environmental pollution,a variable spray control system was designed to match the plant canopy characteristics in real time.In order to calculate the output flow of each nozzle independently,a variable application model based on vertical structure was proposed.The real-time spray characteristics of the two dimensional lidar were tested by combining the real-time scanning of the canopy characteristics,the spray speed and the delay spray control.The experimental results show that the accuracy of laser based canopy volume measurement is high.Compared with traditional constant spray,variable spray can effectively reduce pesticide consumption on the basis of pest control.