| Obtaining the information of farmland topography,soil moisture,fertility and others is the premise of precision agriculture.For high-precision farmland information,field measurement must be carried out in the field.At present,manual collection is the main way in actual agricultural production and scientific research activities.There are many challenges such as high labor intensity,low efficiency and insufficient data in the process of colletcting.A point-to-point tracking control system of mobile robot based on RTK-GNSS was designed,which was used to drive the farmland information collecting robot traversing a series of planned points in the field to achieve high-precision tracking operation.Firstly,the hardware of the robot was designed,and the control performance is tested.Secondly,the kinematics model of the robot is established,and the grid method is used to generate the field information collection points.After the global path planned,speed and steering controllers is further completed.Based on the robot motion control model and the designed control algorithm,the software is designed.Finally,the feasibility of the designed point tracking control algorithm is verified by tests.The results show that on cement pavement,the average deviation of point tracking is 0.027 m and the maximum deviation is 0.043 m,on the lawn pavement,average deviation 0.035 m and maximum deviation 0.074 m.The designed high-precision point-to-point control system and experimental research can provide technical support for the automation of farmland information collection.It has certain theoretical innovation and application value.The specific research contents are as follows:(1)Hardware designThe overall scheme of the hardware system of farmland information collection robot is proposed,and the main hardware components such as the chassis of the wheeled mobile robot,RTK-GNSS,on-board computer,microcontroller,steering module,and drive module are designed respectively.(2)Design of point tracking control algorithmPoint tracking kinematics model of the mobile robot is established.The steering control law and speed control law of the robot are designed according to different tracking scenarios in the information collection process.A polygon field segmentation method to generate information collection points is designed.The turning connection path is planned according to the collection points generated by the calculation example to complete the path planning.(3)Software designIn the MATLAB environment,the host computer software is designed by modular programming method,and the main program,serial communication subroutine,RTK-GNSS message parsing subroutine and path tracking subprogram are completed.DC motor driver and steering motor driver are written in the client computer controller.MATLABGUI tools are used to optimize the human-computer interface.(4)Control system verification testsControl parameters are determined by repeated tests with variable parameters.Continuous point tracking tests are carried out and tracking data and tracking accuracy are analyzed.To realize the initial application of the system,elevation data is processed to draw a topographic map. |