Research On Embedded Precision Fertilization Control System Based On Grid Prescription Map

Prescription fertilization technology is an efficient means of variable fertilization in precision agriculture.By creating a grid prescription map of the fertilizer requirements for the land to be fertilized,fertilizer can be accurately applied,which not only improves the utilization rate of fertilizers,but also reduces the environmental pollution caused by improper fertilization.This article takes the grid prescription map as the research object,and combines embedded Linux system,RTK differential positioning technology,INS inertial navigation system,and spatial interpolation algorithm to design an embedded precision fertilization control system based on the grid prescription map.The main research contents are as follows:(1)The overall structure of the precision fertilization control system was designed,including hardware selection and circuit design for the fertilization terminal and fertilization execution unit.Based on the hardware platform of the fertilization terminal,a Linux system was built,and the fertilization application program was developed using QT Creator.(2)The RTK positioning technology was optimized by using the IMU inertial device and GNSS/INS combination algorithm,which outputs position,velocity,and attitude information with an update frequency of over 100 Hz.A simulation was conducted on a set of moving vehicle data using MATLAB,and the results indicated that the proposed solution can significantly improve the output frequency of speed and position while ensuring positioning accuracy.This resolves the issue of traditional agricultural machinery being unable to adjust spraying rates in a timely manner due to low GPS positioning frequencies.(3)A single-point positioning model for straight and turning fertilization was established,and the spatial coordinates of each nozzle were calculated using the GPS antenna position.To further improve the fertilization accuracy,a feedforward control method was used to modify the established positioning model,and a fertilization lag compensation algorithm for traction agricultural machinery was proposed.The fertilization lag time was determined through bench tests.To verify the practical effect of the above model and algorithm,fertilization tests were conducted in a simulation experiment.The results showed that after adopting the lag compensation algorithm in the fertilization terminal,the maximum lag distance of 8.78 meters can be corrected to within1.24 meters.(4)The original grid prescription map was optimized to invert the continuous distribution of actual soil fertilizer requirements.Taking a local prescription as the research object,a "one in two" cross-validation method was designed to verify the processing effect of four interpolation algorithms.The results showed that the Kriging interpolation method with the Gaussian model as the variogram function had the best processing effect,with a root mean square error value of 0.143946 in cross-validation.To test the accuracy of this interpolation algorithm in fertilization,simulation experiments were conducted on three fertilization states: straight,right turn,and left turn.The results showed that the interpolation method can significantly optimize the distribution of original fertilizer requirements,and the total and average spraying volumes after optimization still maintain high accuracy.The average accuracies of 20 sets of nozzles under three different fertilization states were 97.62%,95.31%,and 94.42%,respectively.