关键词: 小麦小区育种, 北斗卫星定位, 相对坐标系, 算数平均值滤波, 卡尔曼滤波

Abstract: Breeding experts need to accurately divide the sowing areas in wheat plot breeding. However, in the sowing process of the traditional wheat plot planter, the lines in each plot cannot be aligned precisely, leading to confusion in the breeding area and affecting the breeding effect. In order to solve the problem of precise alignment of the planting path in wheat plots, Beidou satellite positioning was combined with wheat plot breeding on the basis of long-term experiments. Firstly, the longitude and latitude of the planter was obtained, and the plot path model of the planter was built through the establishment of relative coordinate system to obtain the planned path, which replaced the traditional manual line drawing process. Using the method proposed in this paper, the sowing accuracy of wheat plot was increased by 32% within the error range of 5 cm. At the same time, considering the complexity of farmland operation environment and limited ability of manual driving, Beidou satellite positioning itself also had centimeter-level error. In order to further improve the accuracy of path planning, this paper put forward a kind of arithmetic average value filtering and the algorithm of Kalman filtering fusion. The fused algorithm performed iterative recursive processing on the slope value of the reference path in the relative coordinate system. Through experiments, the arithmetic mean filtering and Kalman filtering combined algorithm could effectively correct the slope value of the actual calculation, with the minimum correction range of 20.1% and the maximum correction range of 4 871.4%. The research results in this paper could provide reference for the accurate planning of the Beidou satellite positioning in the planting path.

Key words: wheat plot breeding, Beidou satellite positioning, relative coordinate system, arithmetic average filtering, Kalman filtering