CN-121667049-B - Sowing method for improving sowing uniformity of light pasture seeds

CN121667049BCN 121667049 BCN121667049 BCN 121667049BCN-121667049-B

Abstract

The invention discloses a sowing method for improving sowing uniformity of light pasture seeds, which relates to the technical field of agricultural intelligent sowing and comprises the following steps of S1, acquiring soil humidity weight and temperature influence factor data of a target land from a preset database, constructing an initial distribution map by integrating soil humidity boundary and temperature distribution difference, carrying out data calibration on regional crop characteristics, determining basic influence parameters of seed drift in the land, S2, carrying out regional division treatment on the land according to the basic influence parameters of seed drift and combining topography fluctuation constraint and drift influence gradient, and regulating the region overlapping proportion through a boundary dynamic correction technology to obtain a preliminary partition layout.

Inventors

XUE KANG
JIANG ZIZHEN
TANG HAO
WANG CHUAN
WANG LIWEI
ZHANG JIN
LU BIYUN
CHEN WEI
WANG YAO
FAN BINBIN
AN MINHUI

Assignees

安徽省农业科学院农业机械装备与工程研究所

Dates

Publication Date: 20260505
Application Date: 20260211

Claims (8)

1. A sowing method for improving sowing uniformity of light pasture seeds, comprising: s1, acquiring soil humidity weight and temperature influence factor data of a target land block from a preset database, constructing an initial distribution map by integrating soil humidity boundary and temperature distribution difference, performing data calibration on regional crop characteristics, and determining basic influence parameters of seed drift in the land block; S2, according to basic influence parameters of seed drift, combining topography fluctuation constraint and drift influence gradient, carrying out regional division processing on the land, and adjusting the region overlapping proportion through a boundary dynamic correction technology to obtain a primary partition layout; the step S2 comprises the following steps: According to the basic influence parameters of seed drift and the topography fluctuation constraint, drift influence gradient data are obtained from a preset database, and parameter values are integrated in a weighted average mode by fusing the basic influence parameters and the topography fluctuation constraint, so that a gradient constraint matrix is obtained; Performing regional division processing on the land parcels through a gradient constraint matrix, integrating drift influence gradients, determining a range by a boundary expansion method, and determining division boundary lines; Aiming at the dividing boundary line, extracting region overlapping proportion data from the dividing boundary line by adopting a boundary dynamic correction technology, adjusting the overlapping proportion to correct the boundary in a scaling mode, and obtaining a corrected boundary set; According to the corrected boundary set, fusing partition parameter calibration data from a preset database, judging that if the overlapping proportion is lower than a preset threshold value, integrating layout boundary optimization through a boundary smoothing method, and determining an optimized partition map; Processing data in a gradient superposition mode by optimizing the partition map and combining constraint gradient fusion to obtain preliminary partition layout data, and obtaining a preliminary partition layout of the land block; S3, aiming at the primary partition layout, extracting soil humidity weight and temperature influence factors of each partition, calculating an expected yield fluctuation range of each partition by combining historical data comparison and yield prediction standard, judging that if the fluctuation range exceeds a preset threshold value, triggering boundary smoothing treatment, and determining an optimized partition structure; S4, analyzing the environmental interference variable and the growth cycle parameter of each subarea through the optimized subarea structure, and evaluating local seed accumulation risk points by combining the characteristics of the area crops to obtain the primary density distribution of the seed distribution; s5, aiming at the preliminary density distribution, detecting the coordinates of the vacant area in the balanced state of the area of the partition, superposing drift influence gradient data to carry out density correction, and if the corrected density distribution is still uneven, dynamically correcting and adjusting the overlapping proportion of the area through the boundary to obtain the final density distribution; S6, aiming at final density distribution, acquiring soil humidity boundary and temperature distribution difference data, and integrating the data in a superposition mode to obtain a path planning basis; and generating a seed sowing path of the unmanned aerial vehicle according to the path planning basis, judging a coverage missing area if the path does not cover all the subareas, and determining an adjustment requirement.
2. The method for improving uniformity of sowing light pasture seeds according to claim 1, wherein S1 comprises: Acquiring soil humidity weight and temperature influence factor data of a target land block from a preset database, and constructing an initial distribution map by integrating soil humidity boundary and temperature distribution difference; data calibration of the initial distribution map is carried out aiming at the characteristics of regional crops, and land slope factor data are obtained from a preset database and fused into the initial distribution map; judging a seed drift path according to the fused distribution map, and if the path deviation exceeds a preset threshold value, adjusting humidity weight and temperature influence factors; and determining basic influence parameters of seed drift in the land parcels through the adjusted maps.
3. The method for improving uniformity of sowing light pasture seeds according to claim 1, wherein S3 comprises: for the primary partition layout, acquiring soil humidity weight and temperature influence factors of each partition from a preset database, and calculating a comprehensive environment index in a weighted manner by fusing the soil humidity weight and the temperature influence factors to obtain a partition environment matrix; According to the partition environment matrix, extracting yield prediction benchmarks from a preset database by combining historical data comparison, calculating an expected yield fluctuation range by superposing the comprehensive environment index and the yield prediction benchmarks, and determining a fluctuation range set; Aiming at the fluctuation range set, judging that if the fluctuation range exceeds a preset threshold, triggering crop adaptability evaluation, acquiring adaptability parameters from the fluctuation range set, and processing data in a mode of matching crop types by integrating the adaptability parameters to obtain adaptability evaluation data; According to the adaptability evaluation data, adopting boundary smoothing processing to integrate irrigation adjustment planning, correcting the boundary in a smooth mode by expanding the adaptability evaluation data, and fusing soil humidity weight adjustment boundary curves to determine a correction boundary group; And fusing partition parameter calibration data from a preset database aiming at the correction boundary group, and determining an optimized partition structure by superposing the correction boundary group and the partition parameter calibration data processing layout.
4. The method for improving uniformity of sowing light pasture seeds according to claim 1, wherein S4 comprises: the method comprises the steps of obtaining environmental interference variables and growth cycle parameters of each partition through an optimized partition structure, calculating interference influence indexes by combining the crop characteristics of the regions in a weighted average mode, and summing the environmental interference variables multiplied by the crop characteristic weights to obtain an interference parameter set; For the interference parameter set, integrating the seed accumulation risk points by adopting a local risk assessment method in a grid division mode, and if the risk points exceed a preset threshold, triggering a density adjustment mechanism to process according to a uniform dispersion principle, so as to determine a risk point distribution map.
5. The method for improving uniformity of sowing light pasture seeds according to claim 4, wherein S4 further comprises: Extracting crop adaptation integration data from a preset database according to the risk point distribution map, and processing the distribution in a point-by-point addition mode by superposing the interference parameter set and the crop adaptation integration data to obtain primary density distribution; And aiming at the preliminary density distribution, fusing density distribution mapping and risk point positioning, performing uniformity calibration through curve fitting, and determining seed distribution uniformity indexes.
6. The method for improving uniformity of sowing light pasture seeds according to claim 1, wherein S5 comprises: For the preliminary density distribution, acquiring the coordinates of a vacant area in a zone area balanced state, integrating drift influence gradient data by superposing soil permeability data extracted from a preset database, wherein the soil permeability data represents the regional moisture flow rate, the drift influence gradient data reflects the seed position offset degree, and processing the two data in a weighted summation mode to obtain corrected density distribution; if the corrected density distribution is uneven, the deviation of the overlapping proportion of the region is judged, and the deviation is adjusted by adopting boundary dynamic correction, wherein the boundary dynamic correction moves the edge of the partition according to the deviation amplitude, and the equilibrium distribution parameter is determined.
7. The method for improving uniformity of sowing light pasture seeds according to claim 6, wherein S5 further comprises: Extracting partition coordinate positioning information from the balanced distribution parameters, and carrying out density calibration by combining influence data integration, wherein the influence data integration is derived from the superposition result of drift influence gradient data and soil permeability data, and calibrating in a point-by-point comparison mode to obtain an optimized density map; And according to the optimized density map fusion area equilibrium state, triggering a dynamic mechanism to acquire final density distribution, wherein the dynamic mechanism iterates and evenly disperses seed positions through equilibrium distribution parameters.
8. A sowing method for improving uniformity of sowing light pasture seeds as in claim 1, wherein S6 further comprises: Dynamically correcting a path track by adopting an adjustment requirement, and processing the track by fusing wind speed influence data and regional gradient information to obtain an optimized path; And extracting a working instruction from the optimized path, integrating the partition coordinate information and the seeding density threshold value, and outputting a complete sequence.

Description

Sowing method for improving sowing uniformity of light pasture seeds Technical Field The invention relates to the technical field of agricultural intelligent sowing, in particular to a sowing method for improving sowing uniformity of light pasture seeds. Background In the current grassland ecological restoration and animal husbandry development, pasture planting plays a vital role as a basic link for improving grassland productivity and ecological bearing capacity. Especially in the large-area restoration operation of the degraded grassland, whether the target of the expected forage grass yield can be accurately realized becomes a key index for measuring the sowing quality and the project effect. In the conventional grass seed sowing process, a land grass planter is generally used for sowing. The seeding equipment has certain efficiency advantages in a flat and accessible area, but has the problems of limited operation, difficult planning of paths, repeated seeding or missing seeding and the like when operating in mountain lands, sloping lands or large-area irregular plots. In order to improve the sowing efficiency, unmanned aerial vehicles are introduced in recent years to carry out aerial sowing operation, and the unmanned aerial vehicle has the advantages of rapid seed sowing, adaptation to complex terrains, labor saving and the like, and becomes an emerging means for large-area sowing. However, the characteristics of 'light weight, small particles and easy drifting' of pasture seeds are limited, and secondary diffusion, deviation and even accumulation phenomena in the seed sowing process are extremely easy to be caused due to the overlapping of the air washing flow under the flying rotor wing and the high altitude wind power disturbance in the unmanned aerial vehicle sowing process, so that uneven distribution of 'local dense and large-piece sparse' is formed on the ground, and the problems of uneven germination, large growth difference and severe yield fluctuation of the pasture in the later period are caused. The phenomenon of uneven seeding is particularly serious in large-area seeding tasks, and has become a main technical bottleneck for restricting popularization and application of unmanned aerial vehicle seeding technology. Even if the same setting of the sowing density is used, the germination rate and the survival rate of seeds of different areas are significantly different due to the micro-environmental differences such as soil humidity, temperature, topography fluctuation and the like. The problem that the seeding operation of pasture seeds is unstable is faced, the challenge of uncertain seedling effect is overcome, and the seeding accuracy and uniformity are difficult to guarantee under a large scale by the traditional empirically-set seeding parameters. Disclosure of Invention The invention aims to provide a sowing method for improving sowing uniformity of light pasture seeds, which solves the problems in the prior art. In order to achieve the above purpose, the invention provides a sowing method for improving the sowing uniformity of light pasture seeds, which comprises the following steps: s1, acquiring soil humidity weight and temperature influence factor data of a target land block from a preset database, constructing an initial distribution map by integrating soil humidity boundary and temperature distribution difference, performing data calibration on regional crop characteristics, and determining basic influence parameters of seed drift in the land block; S2, according to basic influence parameters of seed drift, combining topography fluctuation constraint and drift influence gradient, carrying out regional division processing on the land, and adjusting the region overlapping proportion through a boundary dynamic correction technology to obtain a primary partition layout; S3, aiming at the primary partition layout, extracting soil humidity weight and temperature influence factors of each partition, calculating an expected yield fluctuation range of each partition by combining historical data comparison and yield prediction standard, judging that if the fluctuation range exceeds a preset threshold value, triggering boundary smoothing treatment, and determining an optimized partition structure; S4, analyzing the environmental interference variable and the growth cycle parameter of each subarea through the optimized subarea structure, and evaluating local seed accumulation risk points by combining the characteristics of the area crops to obtain the primary density distribution of the seed distribution; s5, aiming at the preliminary density distribution, detecting the coordinates of the vacant area in the balanced state of the area of the partition, superposing drift influence gradient data to carry out density correction, and if the corrected density distribution is still uneven, dynamically correcting and adjusting the overlapping proportion of the area through the boundary to obtain t