| The harvesting of field vegetables is an important link in the agricultural production of field planting,and the harvesting efficiency will directly affect the final agricultural planting income and production efficiency.For vegetables with inconsistent maturity stages,there are problems such as heavy labor,difficult handling,and low efficiency in manual harvesting.The overall low level of mechanization is an important reason for restricting harvesting efficiency.In view of the above problems,taking broccoli and cauliflower as an example,the solution in this paper is: the operators carry out selective picking,and the harvesting and transportation platform independently follows the operators,picking and releasing as they go,so as to realize the loading and transportation tasks of cauliflower vegetables.For this reason,this paper designs a self-following selective harvesting and transportation platform for cauliflower vegetables that is suitable for field environments and has good passability and intelligence.The platform greatly reduces the labor intensity of the staff and improves the harvesting efficiency.The main research content of this paper includes(1)By comparing the advantages and disadvantages of various following target positioning technologies,combined with the characteristics of the actual field environment,UWB positioning technology is selected as the following positioning technology for the selective harvesting platform.On this basis,three typical UWB positioning algorithms are compared,and the PDOA algorithm is selected as the positioning algorithm for UWB positioning according to the calculation complexity,installation difficulty,precision and other indicators;the measured two-axis average error sum does not exceed 11 cm,but the data with large variance is generally unstable;the Kalman filter is introduced,and the positioning data is processed through the Kalman filter,and the overall accuracy is improved by about 51% compared with the original.(2)In order to solve the problem that natural illumination affects the image quality of the camera,the graying effect of different components under the three different color models of RGB,YUV and HSI is compared and analyzed,and the method of using the U component in the YUV color model as the optimal component in the graying step is determined;in order to reduce the adverse effects of uneven gray distribution on image segmentation,an improved OTSU method is used for image segmentation;an improved boundary feature point extraction algorithm is proposed,which effectively filters out the abnormal feature points in the process of extracting crop row boundary feature points,and improves the accuracy of straight line fitting.(3)On the basis of obtaining a stable and reliable navigation reference line,multiple selective harvesting platform coordinate systems are constructed,and the pose parameters of the harvesting platform are calculated through the geometric relationship between these coordinate systems,and the calculated platform pose deviation is used as The input of PID control,the rotation speed of steering wheel is taken as output,and the platform linear following method based on PID control has been developed;the key parameters K＿p,K＿i,and K＿d of the PID controller designed in this paper are finally determined to be 2.53,0.037,and 0.21 through the method of multi-scenario experiments.(4)According to the actual operation requirements in the field environment,the crawler chassis and telescopic stage mechanism of the selective harvesting platform are designed.Select the hardware equipment required for the harvesting platform,such as high-power motors,motor drivers,electromagnetic relays,and electric push rods.Build a control system with automatic follow and remote control mode and complete the design and wiring of related circuits.(5)Carry out relevant tests on the selective harvesting transport platform in the actual environment.The load transport capacity and field transport test results show that the harvest transport platform has a load capacity of more than 300 kg and a maximum driving speed of more than 4km/h;the climbing performance test results show that the platform can pass through a 15° slope at different speeds;the minimum turning radius test results It shows that the deviation of the maximum turning radius of the platform does not exceed 8% of the ideal minimum turning radius;the platform pose error test shows that the error standard between the calculated value of the lateral deviation and the heading angle deviation obtained by the geometric calculation of the platform compared with the actual measured value The differences are 1.17 cm and 0.6°respectively;the straight-line following effect test shows that the maximum vertical error is 0.32 m,the average standard deviation is 0.12 m,the maximum lateral error is4.91 cm,and the average standard deviation is 2.04 cm.The test results show that the platform can meet the needs of actual field operations and complete the task of autonomously following and loading and transporting cauliflower vegetables during harvesting.This paper designs a self-following transportation platform for selective harvesting of field cauliflower vegetables.By combining target personnel positioning and visual navigation technology,the platform motion control algorithm is developed to realize the platform’s straight-line automatic following function between crop rows,which effectively reduces the workload.The labor intensity of personnel is improved,and the harvesting efficiency of crops is improved. |
PDF Full Text Request