| Straw rotary burying operation can fertilize the ground,improve the soil structure,while avoiding environmental pollution caused by straw burning.At present,the operation process of straw burying and returning to the field is only based on the driver’s feeling to judge the position of the machine,which leads to problems such as heavy plowing,missed plowing and unstable depth of burying.The intelligent operation of straw burying can effectively reduce the driver’s work intensity,reduce the operation time and improve the operation effect of straw burying.In order to improve the adaptability of the navigation and tillage depth control system to the straw burying and returning machine,this paper carried out the research on the key technology of navigation and tillage depth control of straw burying and returning operation for the environmental characteristics such as soft soil,high moisture content and residual straw bulge and surface puddle after harvesting machine operation.It mainly includes five parts:information collection,information processing,navigation control,tillage depth control and GUI operation interface,etc.Simulation analysis and experimental verification were also carried out.The main studies are as follows:（1）The research progress of domestic and foreign agricultural machinery navigation,tillage depth control,visual path extraction and operation path planning technologies were systematically compared and analyzed,and the straw rotary burial navigation and tillage depth control system was built on this basis.The electronic-control hydraulic lifting system was built based on the original mechanical hydraulic lifting system of the unit,and the high pressure hydraulic oil was directly delivered to the lifting oil pump through the electric control proportional hydraulic lifting valve to realize the lifting of the rear suspension of the operating unit.The results showed that this electronic-hydraulic steering system achieved fast and accurate control of front wheel steering,and this electro-hydraulic lifting system achieved accurate control of three-point suspension lifting arm.（2）In the straw burying and returning operation,the operating unit is affected by the sliding phenomenon of soft and wet soil,rice stubble and residual straw.Two influencing factors,front wheel slip angle and rear wheel slip angle,were introduced into the classical two-wheeled vehicle model to more accurately reflect the motion state of straw rotary burying and returning operation.The inversion of the three-dimensional chain system was used to obtain the non-linear control law of the turning angle,and the value of the control law coefficient can be adjusted to adjust the performance of the system.The slip prediction model accurately represented the nonlinear motion caused by the complex environment in the straw rotary burying and returning operation,and provided a suitable kinematic model for accurate tracking of the straw rotary burying and returning path.A variable-gain single-neuron PID path tracking control algorithm was proposed to further optimize the adaptability of the control algorithm to environmental changes through real-time adaptive nonlinear adjustment of the gain coefficients by input and output errors;a path tracking control strategy was constructed to control the tractor to perform steering operations and realize the path tracking operation of the operating unit,which improves the robustness and accuracy of the path tracking control system.Simulink simulation and road test of path tracking were carried out.The path tracking Simulink simulation results showed that the variable-gain single-neuron PID path tracking control algorithm has the fastest convergence speed,smaller overshoot and the best signal following performance compared with the conventional PID and single neuron PID control algorithms.The road test showed that the steady-state regulation time is shorter in linear tracking,and the maximum lateral deviation and the average lateral deviation of the variable-gain single-neuron PID were smaller;in curve tracking,the maximum lateral deviation and the average lateral deviation of the variable-gain single-neuron PID were the smallest.It could be seen that the tracking effect of variable-gain single-neuron PID was the best,which meets the requirements of straw rotary burying navigation accuracy.（3）Based on the surface environment and soil characteristics of the straw field,a four-link rear suspension lifting motion model was established,a variable-gain single-neuron PID tillage depth control algorithm was designed,a tillage depth control strategy was constructed,and a comparative analysis was conducted through simulation and performance tests to provide a theoretical basis for field operation.The relationship equation between the angle of the upper swing arm and the height of the working equipment was established by building a rear suspension lifting motion model.The robustness and accuracy of the rotary depth control system were improved.Simulink simulation and performance improvement tests were conducted.Simulation results showed that compared with conventional PID and single neuron PID control algorithm,the variable-gain single-neuron PID tillage depth control algorithm had the fastest convergence speed,smaller overshoot,and the best signal following performance;the tillage depth performance test showed that the variable-gain single-neuron PID tillage depth control algorithm had the shortest lifting time and lower overshoot,and the overall lifting process was smoother.The variable-gain single-neuron PID tillage depth control algorithm had the best lifting control performance and met the requirements of rice straw rotary burial plowing depth control accuracy.（4）Based on the motion performance parameters of the straw rotary burying operation unit and the requirements of rotary burying operation,the starting route of the boundary was obtained by machine vision,and the maximum inter-row set path planning method was designed based on the starting route with bow turn around,which provides a feasible operation path for the straw rotary burying navigation operation.For the optimal operation path of the regular field,one of the field boundaries was selected as the starting route.In order to extract the starting route for the straw area of the field,an image segmentation method based on H-component was proposed;a morphological image filter was designed;a starting route of straw boundary extraction based on Hough matrix and RANSAC algorithm was proposed;the algorithm used Hough matrix to extract the straw boundary feature points and remove the redundant points,and then the RANSAC algorithm reduced the noise caused by different straw shapes and improved the accuracy of the initial route extraction.The extraction experiments showed that the recognition accuracy of the path extraction method of Hough matrix combined with RANSAC algorithm was 96.21%～90.45%under different light intensities and backgrounds,respectively,and the algorithm took no more than 0.51 s.The method was better and more accurate in real time compared with the traditional Hough transform;it was more accurate compared with the least squares method and was suitable for the starting route of straw fields fast extraction.The set paths under different areas of regular field blocks were analyzed,and the maximum spanning method was selected for set path planning in order to ensure the smooth turnaround of the operation unit;the road test showed that the method met the characteristics of large turning radius and no reversing when the tractor operates,and satisfied the requirements of unmanned operation of straw rotary burying and return to the field.（5）Field experiments were carried out to analyze the field control effects of path tracking control system and tillage depth control system under different algorithms,to verify the feasibility of the navigation and tillage depth control system designed in this paper;and a comprehensive field of navigation and tillage depth control of straw rotary burying returning unit was carried out to provide reference for the field application of straw rotary burying returning navigation and tillage depth control system.The field test results of path tracking control showed that when the speed of the working machine was about1.15 m/s,the maximum lateral deviation of the linear tracking of the variable-gain single-neuron PID controller was not more than 0.071 m,the average absolute deviation was not more than 0.031 m,and the standard deviation was not more than 0.038 m.In the straight path tracking,compared with the conventional PID control algorithm,the maximum lateral deviation and the average absolute deviation control accuracy of the variable-gain single-neuron PID control algorithm respectively improved by 53.08%and 51.72%;the maximum lateral deviation and the average absolute deviation control accuracy respectively improved by 39.00%and 28.21%,compared with the single neuron PID control algorithm.In the curve path tracking,the maximum lateral deviation and average absolute deviation control accuracy of the variable-gain single-neuron PID respectively improved by 44.73%and40.39%,compared with the conventional PID control algorithm;and the maximum lateral deviation and average absolute deviation control accuracy respectively improved by 21.81%and 21.11%,compared with the single neuron PID control algorithm.Compared with the conventional PID and single neuron PID navigation control algorithms,the variable-gain single-neuron PID adaptive control algorithm could reduce the influence of the straw rotary burying returning operation environment on the tracking accuracy and enhance the adaptability and robustness of the navigation controller to the operation environment by adjusting the proportionality coefficient K1 in real time.The results of the tillage depth control field test showed that when the operating speed was about 0.61 m/s under the control of three different algorithms,the basic trend of the rotary burying tool follows the surface height change was the same,but the overshoot of the variable-gain single-neuron PID tillage depth control was the smallest.The average depth of straw burying and returning operation of the variable-gain single-neuron PID tillage depth control algorithm was 15.49 cm,and its stability coefficient of burying depth was 96.09%,which was 5.85%higher than that of conventional PID control algorithm,and4.37%higher than that of single neuron PID control algorithm;its straw burying rate was94.74%,which was 4.38%higher than that of conventional PID control algorithm,and 3.49%higher than that of single neuron PID control algorithm.Compared with conventional PID and single-neuron PID control algorithm,the variable-gain single-neuron PID control algorithm could make the straw rotary burying operation tool follow the complex surface environment in real time by adjusting the proportional coefficient K2 in real time,which enhanced the adaptability and robustness of the tillage depth control system to the complex operation environment.This algorithm enhanced the adaptability and robustness of the tillage depth control system to the complex operating environment,and finally achieved a more stable operating effect to meet the technological requirements of straw rotary burying operation.The results of a comprehensive field test on the navigation and burying depth control system of straw rotary burying to the field showed that the maximum deviation of path tracking navigation system was 0.133 m,the average absolute deviation was 0.047 m and the standard deviation of the deviation was 0.052 m when the rotary burying operation was carried out at an operating speed of 0.61 m/s.The average tillage depth was 13.3 cm,the stability coefficient of tillage depth was 88.35%,and the straw burying rate was 91.41%.Therefore,it could be seen that the comprehensive test of the navigation and tillage depth control system has achieved a relatively stable operation effect,and the measured items met the requirements of the evaluation index.This system meets the technical requirements of straw rotary burying and returning to the field,and is suitable for automatic straw rotary burying and returning to the field under unmanned driving. |
PDF Full Text Request