Abstract:In order to meet the needs of precision agriculture for crop location information, achieve accurate plant position of crops, and solve the problem of low resource utilization rate and high operation cost caused by repeated identification and positioning of plants in each part of field management, a seed location method at planting process was proposed. Seed position, which was regarded as plant location, was mapped and saved during planting process, and it could be shared and used in the follow-up field management links. This method was based on a seeder equipped with a satellite receiver and four seed quality detection sensors. Seeder was mounted with an automatic navigation tractor. Tractor and seeder were hinge joint, relative positions changing during the operation, thus, satellite receiver was installed on seeder frame girder instead of tractor cab roof. Real-time kinematic (RTK) receiver collected satellite observation data from satellite antenna and fixed base station data from radio antenna at a frequency of 10Hz, and formed differential data in system for real-time analytical processing, to achieve accurate acquisition of information such as the position of seeding unit and coordinated universal time (UTC). Seed quality detection sensors were triggered when seed passed through seeds tube, and signal acquisition circuit recorded trigger time and row number at the same time, then transmitted the information to main controller. Seed touched ditch time was determined in combination with UTC time, trigger time and time difference. Position of seeding unit at seed touched ditch time was determined by using equal velocity interpolation method. A mathematic model of seeding unit and planter monomer relative position was established based on spatial location of the seeder, planter monomer position could be extrapolated by this model. Combining with E-discrete element method (EDEM) numerical simulation and dynamic simulation, time and distance difference compensation model was established, seeds localization could be 〖JP3〗predicted. The operation quality of the seed location method was evaluated for the operation speed(3km/h, 5km/h and 7km/h), sowing spacing(0.2m,0.3m and 0.4m) and GPS frequency (1Hz,5Hz and 10Hz) by deviation of implantation position (DIP ) and accuracy of implantation position (AIP ). Field trails results showed that operation speed and GPS frequency significantly affected DIP (p<0.01), sowing spacing affected DIP (p<0.05); operation speed affected AIP (p<0.05), while sowing spacing and GPS frequency had no significant effect on AIP (p>0.05). With the decrease of operation speed, the increase of sowing spacing and GPS frequency, the DIP was decreased. With the decrease of operation speed, the AIP was increased. At the operation speed of 3km/h, sowing spacing of 0.4m and GPS frequency of 10Hz, the system gave the best DIP and AIP of 24.3mm and 88.9%, respectively. This method can provide theoretical reference for synchronous generation of seed map in seeding process, provide position information support for each link of field management in crop growth process, such as precise weeding, top dressing and application, and provide technical and data support for digital and intelligent precision operations.