| Rapeseed is one of the four major oilseed crops in the world,with many uses such as oil extraction,feed,vegetables,honey source,and tourism,and has an important role in global agriculture.In China,rapeseed is the largest oil crop,and the main production areas are divided into winter-sowing and spring-sowing regions.The winter-sowing areas are mainly concentrated in the Yangtze River basin,where the previous crop is mostly rice.However,the aging rural labor force and the decrease in labor force efficiency have posed many challenges to rapeseed production.To address these issues,mechanical and intelligent operations are being developed to replace traditional human labor.In order to meet the requirements of unmanned sowing operations in the region,this study conducted research on the construction of track-type and wheeled unmanned sowing equipment,as well as key technologies such as navigation,path planning,tracking,and control strategies.The main study components and findings are as follows:(1)In order to meet the current lack of field rape unmanned seeding equipment in the middle and lower reaches of the Yangtze River,a crawler and wheeled rape unmanned seeding equipment has been developed based on the based on Beidou positioning system according to different land sizes.The track-type rape unmanned seeding equipment is designed for irregular and scattered plots,while the wheeled rape unmanned seeding equipment is designed for relatively large and regular plots.Furthermore,the software system of the rape unmanned seeding equipment is designed and developed based on the Qt platform.(2)In order to address the issues of the current A and B point translation method,which is time-consuming,labor-intensive,and has poor transfer operation ability,this paper proposes a path planning technology of "UAV Reconstruction Map + Unmanned Operation Path Planning on the Map".The test verified that the plane error of the high-precision map reconstructed by the UAV was 4.58 cm at the maximum,and the average deviation was3.23 cm.A method of map reconstruction by UAV is proposed,and the plane precision of the high-precision map reconstructed by UAV can reach centimeter-level precision.A global path planning algorithm of "Bow Turning Mode + Shuttle Operation" is proposed for track-type rape unmanned seeding equipment,and a global path planning algorithm of "Two-Back Three-Cut Fishtail Turning Mode + Shuttle Operation" is proposed for wheeled rape unmanned seeding equipment.The simulation results show that the turning mode of tracked and wheeled unmanned operation equipment can reduce the area occupied by turning and improve the land utilization rate.The proposed map-based path planning method enables automatic global job path generation on the map interface.(3)A fuzzy adaptive pure pursuit controller is proposed for the navigation adaptability of unmanned seeding equipment in the process of unmanned oilseed rape seeding operation at different operating speeds.Kinematic models of both tracked and wheeled vehicles were constructed,and the lateral and heading deviations were calculated based on the geometric relationship.Fuzzy rules were then constructed for both tracked and wheeled rape unmanned seeding equipment.Simulation analysis shows that the fuzzy adaptive pure pursuit controller was able to track straight paths with smooth lines and small steady-state errors,and the results of tracked and wheeled equipment were consistent.The field test results show that the maximum tracking deviation of the fuzzy adaptive pure pursuit controller is no more than 0.082,0.086,0.092 m when the speed of crawler-type rape unmanned seeding equipment is 0.5,0.8,1.2m/s,and the average absolute deviation is no more than 0.012,0.010,0.013 m,respectively.The maximum tracking deviation of the fuzzy adaptive pure pursuit controller is no more than 0.047,0.051,0.063 m,and the average absolute deviation is no more than 0.015,0.051,0.019 m when the speed of wheeled rape unmanned seeding equipment is 0.53,1.02,1.49m/s.The maximum tracking deviation of the fuzzy adaptive pure pursuit controller is no more than 0.047,0.051,0.063 m,and the average absolute deviation is no more than 0.015,0.051,0.019 m.(4)In order to solve the switching control and machine control problems of rape unmanned seeding equipment in linear operation path and non-operation path,a finite state machine(FSM)-based control strategy was proposed.After analyzing the manual driving habits,the state set and switching conditions were divided according to the operation path.The Bow U-turn model control strategy applicable to tracked rape seeding equipment,and the two-back three-cut fishtail U-turn model control strategy applicable to wheeled rape unmanned seeding equipment were designed respectively.The results of road and field tests demonstrated the effectiveness of the unmanned seeding equipment proposed navigation control strategy and U-turn model,which met the requirements and accuracy of the turn of rape unmanned seeding equipment.(5)In this study,unmanned seeding experiments were conducted on track-type and wheel-type rape equipment.The results verified the feasibility of "UAV Reconstruction Map + Unmanned Operation Path Planning on the Map",which allows users to draw a field in the software map,set the operation width,and achieve automatic planning of the unmanned operation path.The seeding machine can be operated with just one key,and the operation parameters such as the lifting and lowering of the seeder,the rotation speed of the rotary tillage,and the operation speed and machine width can be automatically adjusted according to the agronomic requirements of the rape operation.The average absolute deviation of the path tracking was found to be less than 3.5 cm,and the precision of the row was less than 2.5 cm,thus meeting the requirements of the rape seeding operation.The unmanned seeding equipment and system for rape researched in this project have been applied in several areas in Hubei Province,with a core demonstration area of more than100 mu. |
PDF Full Text Request