| Autonomous driving technology is very important for developing intelligent agriculture and accelerating the upgrade of agricultural machinery.In order to make the agricultural machinery drive along the crop row,the satellite and vision integrated navigation control technology of crawler tractor was studied.In this paper,a method of obtaining crop rows and crop roots rows in visual image is presented,and a control strategy of satellite visual integrated navigation is researched,and a set of macro-micro navigation control system is constructed based on the steering characteristics of crawler tractor.In addition,the tractor navigation control system is tested and analyzed.The main research contents and results are as follows:1.The characteristics of crop color and crop row distribution in the image were analyzed.The image was segmented using the Super-Green algorithm(2G-R-B)and the mask method.The crop rows were detected using the multiline clustering method based on point-to-line distance and least squares.When initializing clustering,the horizontal bar method is used to obtain the crop row number information in the image.In image sequence processing,the crop row parameters detected from the previous image are used directly as the initial clustering line parameters of the next image.The experimental results show that the accuracy of crop root row detection was 96.79%,91.84%,and 86.34% respectively for the 6th,20 th,and 35 th days after transplantation.2.A crop root row detection algorithm based on the vanishing point method and the grow stage of crop rows was studied.Realize the regression from crop row parameters to crop root row parameters.On the 6th,20 th and 35 th day after rice transplantation,the success rates of root detection were 97.33%,90.82% and 84.15%,respectively.3.Based on the characteristics of satellite and visual navigation information,a preview point tracking strategy for crawler tractor navigation control is studied,and the integrated navigation control strategy is researched.The results show that the maximum lateral offset is 0.22 m when the tractor is running at a speed of about 0.1～0.2 m/s on the field with two rows of artificial seedlings.4.The integrated navigation control system is developed.The hardware of the system is mainly composed of electric steering wheel,industrial camera,satellite receiver,dual satellite antenna and macro-micro navigation controller,the software mainly consists of main program module,electric steering wheel module,camera module and satellite module.After the parameter identification of steering system,differential steering motion analysis and steering median calibration of electric steering wheel,the whole navigation control system can work stably and reliably. |
