Agricultural planting decision-making method and system based on multi-source information fusion

Abstract

The application discloses a kind of agricultural planting decision-making method and system based on multi-source information fusion, it is related to agricultural planting technical field, the method includes: based on multi-source data, field is divided into multiple microblocks, and the state image of each microblock is generated;Acquire agricultural operation constraint data;Determine the earliest sowing date of each microblock, suitable sowing time window and agricultural operation ability parameter;With the preset mechanical harvesting window as constraint condition, determine multiple candidate varieties for each microblock;Determine the sowing priority of each microblock, and generate the sowing order and variety configuration scheme of each microblock;In the sowing execution process, real-time field operation state data is acquired, according to the actual sowing information of the sown microblock in each microblock, the sowing order and variety configuration scheme of the unsown microblock in each microblock are corrected.The present application provides reliable decision support for unified mechanical harvesting.

Description

Technical Field

[0001] This application relates to the field of agricultural planting technology, and more specifically, to an agricultural planting decision-making method and system based on multi-source information fusion. Background Technology

[0002] Agricultural planting decisions, as a core aspect of agriculture, are crucial for improving land productivity, reducing production risks, and optimizing resource allocation. In recent years, with the continuous development of technologies such as IoT sensing, remote sensing monitoring, and weather forecasting, data-driven planting decision-making methods have gradually become a research hotspot, providing decision support for agricultural production, including sowing date recommendations, variety selection, and field management. Existing planting decision-making schemes typically use individual plots as the basic decision-making unit, providing unified sowing date recommendations and variety configuration schemes based on input information such as weather forecasts, soil moisture monitoring, and variety growth period models. These schemes have achieved good application results in large-scale, plain-area agricultural production.

[0003] However, in small-scale rice, corn, or soybean planting scenarios in hilly areas, factors such as shared agricultural machinery queuing and scheduling, reduced field road accessibility after rain, and micro-topographical undulations within the field often prevent the entire sowing operation from being completed on the same day. Instead, different micro-plots are forced to be sown in batches on different dates. Existing methods treat the field as a whole unit with the same sowing date and variety, ignoring the differences in low-lying areas, drainage speed, and road access conditions among different micro-plots within the field. This makes it impossible to determine the earliest sowing date, suitable sowing time window, and matching variety for each micro-plot. Furthermore, the generated decision-making scheme is a static plan. If the actual sowing date deviates or the field conditions change during the sowing process, the generated sowing sequence and variety configuration cannot be dynamically adjusted according to the actual operation information of the sown micro-plots. This results in unsown micro-plots still using the initial plan, further exacerbating the misalignment of heading and maturity periods among different micro-plots within the same field, ultimately making it difficult to complete a unified harvest within the preset mechanical harvesting window.

[0004] There is currently no effective solution to the above problems. Summary of the Invention

[0005] This application provides an agricultural planting decision-making method and system based on multi-source information fusion to solve the above-mentioned technical problems.

[0006] This application provides an agricultural planting decision-making method based on multi-source information fusion, including: S1, acquiring multi-source data, dividing the field into multiple micro-blocks based on the multi-source data, and determining the low-lying degree, drainage speed, and road access conditions of each micro-block to generate a state profile of each micro-block; S2, acquiring agricultural machinery operation constraint data; determining the earliest sowing date, suitable sowing time window, and agricultural machinery operation capability parameters of each micro-block based on the agricultural machinery operation constraint data and the state profile; S3, using a preset mechanical harvesting window as a constraint, and considering the crop variety's growth period, accumulated temperature requirements, and tolerance to moisture and late sowing, combined with the... S4. Based on the status profile, the earliest sowing date, the suitable sowing time window, the agricultural machinery operation capability parameters, the candidate varieties, and the theoretical sowing date, the sowing priority of each micro-block is determined, and the sowing sequence and variety configuration scheme of each micro-block are generated; S5. During the sowing process, field operation status data is acquired in real time, and the sowing sequence and variety configuration scheme of the unsown micro-blocks in each micro-block are corrected based on the actual sowing information of the sown micro-blocks in each micro-block.

[0007] This application provides an agricultural planting decision-making system based on multi-source information fusion, comprising: a state profile generation module, used to acquire multi-source data, divide the field into multiple micro-blocks based on the multi-source data, and determine the low-lying degree, drainage speed, and road access conditions of each micro-block to generate a state profile of each micro-block; a sowing parameter prediction module, used to acquire agricultural machinery operation constraint data; and determine the earliest sowing date, suitable sowing time window, and agricultural machinery operation capability parameters of each micro-block based on the agricultural machinery operation constraint data and the state profile; and a variety matching module, used to match varieties based on a preset mechanical harvesting window as a constraint, and according to the crop variety's growth period, accumulated temperature requirement, and tolerance to moisture and late sowing. The system, in conjunction with the suitable sowing time window, identifies multiple candidate varieties for each micro-block and determines the theoretical sowing date corresponding to each candidate variety. The scheme generation module determines the sowing priority of each micro-block based on the status profile, the earliest sowing date, the suitable sowing time window, the agricultural machinery operation capability parameters, the candidate varieties, and the theoretical sowing date, and generates the sowing sequence and variety configuration scheme for each micro-block. The correction module acquires real-time field operation status data during the sowing process and corrects the sowing sequence and variety configuration scheme for the unsown micro-blocks in each micro-block based on the actual sowing information of the sown micro-blocks.

[0008] Based on the embodiments provided in this application, by acquiring multi-source data, the field is subdivided into multiple micro-blocks, and the depression level, drainage rate, and road accessibility of each micro-block are determined to form a state profile of each micro-block. On this basis, agricultural machinery operation constraint data is acquired, and combined with the state profile, the earliest sowing date, suitable sowing time window, and agricultural machinery operation capability parameters are determined for each micro-block. This allows subsequent sowing decisions to no longer rely on the assumption of homogeneity within the field, but rather accurately reflects the spatial differences in the drainage rate of post-rainwater accumulation and the actual accessibility of agricultural machinery among different micro-blocks. This solves the problem of existing methods treating the field as a unified unit with the same sowing date and variety, ignoring the differences in depression level, drainage rate, and road accessibility among different micro-blocks within the field, and avoiding misjudgments of the unified sowing date due to differences in micro-topography and road conditions.

[0009] By using a preset mechanical harvesting window as a constraint, and based on the crop variety's growth period, accumulated temperature requirements, and tolerance to moisture and late sowing, combined with the suitable sowing time window determined for each micro-block, multiple candidate varieties and their corresponding theoretical sowing dates are matched for each micro-block. Then, by combining the comprehensive status profile, the earliest sowing date, the suitable sowing time window, agricultural machinery operation capability parameters, candidate varieties and their theoretical sowing dates, the sowing sequence and variety configuration scheme for each micro-block are generated. This allows variety selection and sowing date planning to no longer be conducted in isolation, but rather the global mechanical harvesting window is incorporated as a hard constraint into the decision-making process. Even if there is a difference of several days in the actual sowing date between micro-blocks due to batch sowing, the maturity combination of the selected varieties can still bring the maturity dates of each micro-block into the same mechanical harvesting window. At the same time, by comprehensively considering spatial heterogeneity, time window, and variety characteristics in the ranking decision, the overall planning is ensured to serve the goal of unified harvesting from the initial scheme level. This solves the problem of the initial planning level in static decision-making, which lacks the ability to adjust unsown blocks based on the actual operation information of sown micro-blocks. That is, before the intervention of dynamic correction, the initial scheme itself already has the ability to converge the maturity dates of micro-blocks with different sowing dates into the same mechanical harvesting window.

[0010] By acquiring real-time field operation status data during sowing and making rolling corrections to the sowing sequence and variety configuration schemes of unsown micro-plots based on the actual sowing information of already sown micro-plots, the initial plan can continuously respond to uncertainties such as deviations from the plan in actual sowing progress and changes in field moisture due to sudden weather changes, rather than remaining unchanged after sowing begins. This solves the problem that existing decision-making schemes are static plans, and if there are deviations in the actual sowing date or changes in field conditions during sowing, the generated sowing sequence and variety configuration cannot be dynamically adjusted according to the actual operation information of already sown micro-plots, resulting in unsown micro-plots still using the initial plan. This ensures that the maturity period of each micro-plot always converges to the same mechanical harvesting window throughout the entire sowing cycle, providing decision support from initial planning to execution correction for unified mechanical harvesting in batch sowing scenarios in hilly areas. Attached Figure Description

[0011] The accompanying drawings, which are included to provide a further understanding of embodiments of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings: Figure 1 A flowchart illustrating an optional agricultural planting decision-making method based on multi-source information fusion according to an embodiment of this application; Figure 2 A flowchart illustrating an optional variety configuration for compensating for sowing time differences according to an embodiment of this application; Figure 3 This is a flowchart illustrating the deviation propagation and dynamic correction during an optional seeding process according to an embodiment of this application; Figure 4 This is a structural diagram of an optional agricultural planting decision-making system based on multi-source information fusion according to an embodiment of this application; Figure 5 This is a schematic diagram of the structure of an optional electronic device according to an embodiment of this application.

[0017] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0019] According to one aspect of the embodiments of this application, such as Figure 1 As shown, this application provides an agricultural planting decision-making method based on multi-source information fusion, including:

[0020] S1. Acquire multi-source data, divide the field into multiple micro-blocks based on the multi-source data, and determine the low-lying degree, drainage speed and road access conditions of each micro-block to generate a status profile of each micro-block.

[0021] S2, acquire agricultural machinery operation constraint data; based on the agricultural machinery operation constraint data and status profile, determine the earliest sowing date, suitable sowing time window and agricultural machinery operation capability parameters for each micro-block;

[0022] S3, with the preset mechanical harvesting window as a constraint, determines multiple candidate varieties for each micro-block based on the crop variety's growth period, accumulated temperature requirements, and tolerance to moisture and late sowing, combined with the suitable sowing time window, and determines the theoretical sowing date corresponding to each candidate variety.

[0023] S4. Based on the status profile, earliest sowing date, suitable sowing time window, agricultural machinery operation capacity parameters, candidate varieties and theoretical sowing date, determine the sowing priority of each micro-block, and generate the sowing sequence and variety configuration scheme of each micro-block.

[0024] S5 acquires real-time field operation status data during the sowing process and modifies the sowing sequence and variety configuration scheme of the unsown micro-blocks in each micro-block based on the actual sowing information of the sown micro-blocks in each micro-block.

[0025] Furthermore, the multi-source data includes: field boundary data, UAV micro-topographic data, historical drainage records, soil moisture data, and rainfall forecast data; S1, acquire the multi-source data, divide the field into multiple micro-blocks based on the multi-source data, and determine the depression level, drainage rate, and road accessibility of each micro-block, generating a status profile of each micro-block, including:

[0026] Determine the geographical extent of the fields based on the field boundary data;

[0027] Based on the micro-topographical data from the drone, the elevation information within the field was extracted, multiple low-lying areas were identified, and the degree of low-lying in each area was determined.

[0028] Based on historical receding water records and elevation information, the receding rate of each low-lying area was determined;

[0029] Based on soil moisture data and rainfall forecast data, assess the impact of muddy conditions on traffic on various road sections within the field, and determine the road access conditions for each micro-block.

[0030] Based on the degree of depression, the rate of water receding, and the accessibility of roads, the field is divided into multiple micro-blocks, and a status profile of each micro-block is generated.

[0031] Taking a 45-mu (approximately 3 hectares) paddy field in a hilly area, planned for rice cultivation, as an example, the field has a gentle east-west slope, naturally formed low-lying valleys, and only one dirt road for mechanized farming around its perimeter. The following steps are used to generate a state profile of each micro-block within this field:

[0032] Obtain multi-source data. Specifically, retrieve the field boundary data from the local agricultural and rural affairs department's land registration database, and determine the geographical range of the field based on the field boundary data. This range includes the entire 45 mu area of ​​the field, which is distributed on a gentle east-west slope, as well as the connecting sections of the surrounding farm roads.

[0033] UAV micro-topographic data of the field was acquired. A UAV equipped with LiDAR was used to perform a cruise scan at an altitude of 80 meters above the field, acquiring surface elevation point cloud data. After denoising and extracting ground points from the point cloud data, a digital elevation model of the field was constructed. Using the arithmetic mean of all ground elevation points within the field as the baseline elevation, the subsidence depth of each location relative to the baseline elevation was calculated grid-by-grid. Based on the UAV micro-topographic data, elevation information within the field was extracted. Low-lying areas with a subsidence depth greater than 5 cm and a continuous contiguous area exceeding 200 square meters were identified as multiple low-lying areas. Three low-lying areas were identified, located in the northwest corner, the east-central part, and the southeast corner of the field, respectively. Based on the distribution range of subsidence depth, the degree of low-lying areas was determined and classified into three levels: areas with a subsidence depth of 5 cm to 15 cm were classified as mild low-lying areas, areas with a subsidence depth of 15 cm to 30 cm as moderate low-lying areas, and areas with a subsidence depth exceeding 30 cm as severe low-lying areas. Therefore, the low-lying area in the northwest corner is classified as moderately low-lying, the low-lying area in the central-eastern part is classified as slightly low-lying, and the low-lying area in the southeast corner is classified as severely low-lying.

[0034] Obtain historical drainage records for this field. Retrieve observation records of water receding after each rainfall event from the field management files for the past five years. These records include the number of calendar days required for the field to clear of significant water accumulation after the rainfall ceases. Based on these historical drainage records and the aforementioned elevation information, determine the drainage rate for each low-lying area. Specifically, categorize and statistically analyze the records according to the degree of low-lying area. For areas of the same low-lying level, calculate the distribution range of historical drainage days. If the historical drainage days for a certain area are mainly concentrated in 1 to 2 days, its drainage rate is determined to be rapid; if concentrated in 3 to 5 days, it is determined to be medium-speed; and if it exceeds 5 days, it is determined to be slow. For this field, the historical receding water duration in the moderately low-lying northwest corner is 4 to 6 days, corresponding to a medium-speed receding rate; the historical receding water duration in the slightly low-lying east-central area is 2 to 3 days, corresponding to a medium-to-fast receding rate; and the historical receding water duration in the severely low-lying southeast corner is 7 to 10 days, corresponding to a slow receding rate. For the high and dry areas that are not low-lying, the historical receding water duration is all less than 1 day, corresponding to a rapid receding rate.

[0035] Soil moisture data and rainfall forecast data for the field were acquired. Soil moisture sensors were installed on the farm roads surrounding the field to acquire real-time soil volumetric water content data for each road section, and rainfall forecast data for the area for the next 72 hours was also acquired. Based on the soil moisture data and rainfall forecast data, the impact of the muddy conditions on traffic on each road section within the field was assessed, and the road access conditions for each micro-block were determined. Specifically, based on current soil moisture monitoring values ​​and the cumulative rainfall forecast, the muddy passability of farm roads under current and post-rain conditions is assessed: when the current soil moisture content is below 20% and the forecast cumulative rainfall is less than 10 mm, the road accessibility is considered good, and wheeled agricultural machinery can pass directly; when the current soil moisture content is between 20% and 35% or the forecast cumulative rainfall is between 10 mm and 30 mm, it is considered average, and wheeled agricultural machinery faces the risk of slipping, requiring reduced load or the use of tracked agricultural machinery; when the current soil moisture content is above 35% or the forecast cumulative rainfall exceeds 30 mm, it is considered poor, and wheeled agricultural machinery cannot pass, requiring waiting for the road to dry or laying temporary roadbed materials. Simultaneously, aerial imagery from drones is used to identify water accumulation and rut depth on the farm road surface to further verify the road accessibility. The assessment showed that the soil moisture in the southwestern and east-central areas of the field adjacent to the farm roads was good and the forecast rainfall was low, so the road access conditions were good. The soil moisture content in the northwestern and northern areas adjacent to the farm roads was high, so the road access conditions were average. The farm road in the southeastern corner was low-lying and prone to water accumulation, and the muddy conditions would be significantly aggravated after the forecast rainfall, so the road access conditions were poor.

[0036] Based on the aforementioned factors of low-lying area, drainage rate, and road accessibility, the field was divided into multiple micro-blocks. Specifically, the spatial distribution data of the three dimensions—low-lying area, drainage rate, and road accessibility—were overlaid, and the division was based on the spatial continuity of these three factors. When all three factors were of the same level within a continuous area, that area was classified as a micro-block; when any dimension showed a change in level in space, the boundary between micro-blocks was formed. The 45-mu (approximately 3.3 hectares) plot of land was ultimately divided into six micro-plots: The first micro-plot is the southwestern, dry area with no depression, rapid drainage, and good road accessibility; the second micro-plot is the northwestern, moderately depressed area with moderate drainage, moderate drainage, and fair road accessibility; the third micro-plot is the eastern, centrally depressed area with slight depression, moderate to fast drainage, and good road accessibility; the fourth micro-plot is the southeastern corner, severely depressed with slow drainage and poor road accessibility; the fifth micro-plot is the northeastern, gently sloping area with slight depression, rapid drainage, and good road accessibility; and the sixth micro-plot is the northern area near the farm road with no depression, rapid drainage, and fair road accessibility.

[0037] For each micro-block, a status profile is generated. This profile records, in structured data format, the micro-block's unique identifier, area, boundary coordinates, low-lying level, drainage rate level, road accessibility level, and the elevation deviation range, historical drainage days range, and road mud assessment criteria used to generate these levels. This status profile serves as the foundational input data for subsequent steps S2 to S5, determining the earliest sowing date, suitable sowing time window, and variety selection plan for each micro-block.

[0038] Furthermore, the agricultural machinery operation constraint data includes shared agricultural machinery reservation queuing data, historical agricultural machinery arrival delay data, agricultural machinery operation speed data, and future rainfall forecast data; S2, acquire agricultural machinery operation constraint data; based on the agricultural machinery operation constraint data and status profile, determine the earliest sowing date, suitable sowing time window, and agricultural machinery operation capacity parameters for each micro-block, including:

[0039] Based on the shared agricultural machinery reservation queue data, determine the reservation time slot for agricultural machinery;

[0040] For example, shared agricultural machinery reservation queue data is obtained from the dispatch platform of a local agricultural machinery cooperative. The reserved field is for a tracked rotary tiller and seeder. The reservation information shows that after the agricultural machinery has completed its work in the previous field, its reserved work period is from the morning to the afternoon of April 8th. At the same time, the reservation queue data also includes the agricultural machinery's subsequent reservation tasks in adjacent fields. Based on this, the upper limit of the agricultural machinery's available work time in this field can be determined.

[0041] Based on historical arrival delay data of agricultural machinery, the scheduled operation time is adjusted to obtain the estimated arrival time of agricultural machinery in each micro-block;

[0042] Obtain the historical arrival delay data for this agricultural machinery. The historical arrival records for the previous three fields show an average arrival delay of approximately 0.7 days. Based on this data, the scheduled work time is adjusted, changing the estimated arrival time of the agricultural machinery from the morning of April 8th to the evening of April 8th, thus obtaining the adjusted estimated arrival time of the agricultural machinery in each micro-plot.

[0043] The duration of agricultural machinery operation is calculated based on the agricultural machinery operation speed data and the area of ​​each micro-block;

[0044] For example, the agricultural machinery's operating speed data is obtained. The machinery's operating speed is 3 mu per hour. Based on the area of ​​each micro-block, the agricultural machinery's operating time for each micro-block can be calculated separately. For example, the third micro-block has an area of ​​approximately 9 mu and an operating time of approximately 3 hours; the sixth micro-block has an area of ​​approximately 5 mu and an operating time of approximately 1.7 hours.

[0045] Based on future rainfall forecast data, determine the duration of the impact of rainfall on field operations;

[0046] For example, future rainfall forecast data for the region is obtained. The forecast shows moderate rain from the evening of April 7th to the early morning of April 8th, expected to stop on the morning of April 8th. Combining the receding rate in the status profiles of each micro-block, the duration of the rainfall's impact on field operations is determined: the impact duration for the first micro-block (rapid receding rate) is 0.5 days, for the second micro-block (medium receding rate) it is 1 day, and for the fourth micro-block (slow receding rate) it is 2 days.

[0047] Based on the expected arrival time, agricultural machinery operation time, and duration of impact, determine the earliest sowing date and suitable sowing time window for each micro-block;

[0048] Based on agricultural machinery operation speed data and scheduled operation time periods, determine the parameters of agricultural machinery operation capacity.

[0049] For example, the earliest sowing date for each micro-block is determined based on the expected arrival time, the duration of agricultural machinery operation in each micro-block, and the duration of rainfall impact. Taking the third micro-block as an example, the rainfall impact duration in this micro-block is 0.5 days, and it reaches a workable state by noon on April 8th. Given an operation duration of 3 hours, its earliest sowing date is April 8th. The rainfall impact duration in the fourth micro-block is 2 days, and its earliest sowing date is April 10th. The suitable sowing time window for each micro-block begins with the earliest sowing date, and the end date is determined based on a combination of the water receding rate and subsequent weather forecasts. Micro-blocks with faster water receding rates have relatively later end dates, while micro-blocks with slower water receding rates have shorter windows, to avoid further delays caused by subsequent rainfall.

[0050] Based on the maximum continuous operating time of the agricultural machinery per day (approximately 10 hours) and its operating speed (3 mu / hour), the agricultural machinery operation capacity parameter is determined to be a maximum daily operating area of ​​approximately 30 mu. This parameter is used in subsequent steps to allocate each micro-block to each operation day according to the sowing priority. The specific values ​​of the above parameters are only embodiments of the present invention and are not unique, and their calculation methods do not affect the implementation of the present invention.

[0051] Furthermore, in S3, using a preset mechanical harvesting window as a constraint, and based on the crop variety's growth period, accumulated temperature requirements, and tolerance to moisture and late sowing, combined with a suitable sowing time window, multiple candidate varieties are determined for each micro-block, and the theoretical sowing date corresponding to each candidate variety is determined, including:

[0052] Determine the target maturity date based on the preset mechanical harvesting window, where the target maturity date is located within the preset mechanical harvesting window;

[0053] In this embodiment, the preset mechanical harvesting window is a time interval pre-set based on local crop types, historical agricultural calendars, combine harvester cross-regional operation scheduling plans, and climate patterns. It includes a clearly defined start and end date. For example, under a rice-wheat rotation system in a certain region, the mechanical harvesting window is typically set between May 25th and June 10th each year. Within this window, the target maturity date is determined as follows: the middle date of the window is preferentially selected as the baseline maturity target for variety configuration in each micro-block, in order to reserve a buffer for earlier or later maturity; simultaneously, it can also be reversed based on the sowing time of the next crop, selecting the date most closely connected to the next crop's agricultural activities as the target maturity date.

[0054] For each microblock,

[0055] Based on the depression level and drainage rate in the state profile of the micro-block, combined with historical drainage records and soil moisture data from multiple sources, the deviation coefficient of the effective accumulated temperature accumulation rate of the micro-block relative to the reference field is determined.

[0056] Based on the growth period and accumulated temperature requirements of each crop variety, and combined with the deviation coefficient, the accumulated temperature requirements of each crop variety are converted into the estimated calendar growth period of the micro-block.

[0057] Specifically, in micro-plots with higher depression levels or slower water receding rates, the effective accumulated temperature rate is significantly lower than that of well-drained reference plots due to prolonged waterlogging and slower soil temperature recovery. This deviation coefficient is defined as the ratio of the actual accumulated temperature rate of each micro-plot to that of the reference plot. Based on this deviation coefficient, according to the growth period and accumulated temperature requirements of each crop variety, the standard accumulated temperature requirement recorded in the variety approval data is divided by this deviation coefficient, and then converted into an estimated calendar growth period in days according to the local daily average temperature. This achieves the conversion of accumulated temperature requirement into an estimated calendar growth period for the micro-plot.

[0058] Based on the tolerance to late sowing characteristics among the moisture tolerance and late sowing tolerance characteristics of each crop variety, establish the correspondence between the sowing delay and the growth period compression of each crop variety in this micro-block;

[0059] Based on the target maturity date, the estimated calendar growth period, and the correspondence between the sowing delay and the growth period reduction, the theoretical sowing date of each crop variety in the micro-block is deduced from the target maturity date, and the theoretical maturity date of the corresponding crop variety when sowing on the theoretical sowing date is determined.

[0060] Determine whether the theoretical maturity date falls within the preset mechanical harvesting window, and determine whether the theoretical sowing date falls within the suitable sowing time window for the micro-block;

[0061] Crop varieties whose theoretical maturity date falls within the preset mechanical harvesting window and whose theoretical sowing date falls within the suitable sowing time window of the micro-block are added to the candidate variety set of the micro-block, and the corresponding theoretical sowing date is used as the theoretical sowing date corresponding to each candidate variety in the candidate variety set.

[0062] Regarding the correspondence between sowing delay and growth period compression, late-sown varieties will accelerate their growth process by actively shortening the vegetative growth stage after sowing is delayed. Establishing this correspondence requires pre-calibrating the compression response parameters of each crop variety based on its late-sown tolerance characteristics within the wet-to-dry-tolerance and late-sown-tolerance traits, using the results of staggered sowing trials in variety approval data or specialized field trials. When working backward from the target maturity date to calculate the theoretical sowing date for each crop variety in a micro-block, the estimated calendar growth period of the variety in that micro-block is subtracted from the target maturity date. A reverse iterative correction is then performed based on the aforementioned correspondence between sowing delay and growth period compression. That is, if the backward-calculated sowing date is delayed compared to the variety's normal sowing date, the potential compression is predicted based on the number of days of delay, and the starting point for backward calculation is adjusted. The iteration terminates when the sum of the backward-calculated theoretical sowing date and the estimated calendar growth period calculated from that date equals the target maturity date, thus achieving self-consistency. This backward calculation method is only one embodiment of the present invention and is not the only one; other equivalent backward calculation methods are also within the scope of protection of this invention.

[0063] After completing the above reverse calculation, a dual judgment is performed: determining whether the theoretical maturity date of the corresponding crop variety falls within the preset mechanical harvesting window when sown according to the theoretical sowing date, and determining whether the theoretical sowing date falls within the suitable sowing time window of the micro-block. Only crop varieties whose theoretical maturity date falls within the preset mechanical harvesting window and whose theoretical sowing date falls within the suitable sowing time window of the micro-block are added to the candidate variety set of the micro-block, and the corresponding theoretical sowing date is used as the theoretical sowing date corresponding to each candidate variety in the candidate variety set. The variety approval data is a well-known technology and will not be described in detail in this embodiment.

[0064] Furthermore, based on the late-sowing tolerance characteristic among the moisture tolerance and late-sowing tolerance characteristics of each crop variety, the corresponding relationship between the sowing delay and the growth period reduction of each crop variety in this micro-block is established, including:

[0065] When the number of mature records under different sowing periods in the historical receding water record data is lower than the preset number threshold, mature period data of the micro-block under different sowing period scenarios are generated to expand the sample size.

[0066] When establishing the correlation between sowing delay and growth period compression, if the number of maturity records under different sowing dates in the historical flooding data is lower than a preset threshold (e.g., less than 3 or 5 records), the sample size is expanded numerically. Specifically, a crop growth model based on environmental factors such as temperature and light is used to simulate and generate maturity data for the micro-block under different sowing date scenarios. This supplements the data foundation for the variety's tolerance to late sowing and ensures that the calibration of the compression response curve is statistically significant.

[0067] Based on the late-sowing tolerance characteristics of each crop variety, a segmented correspondence between the amount of sowing delay and the amount of growth period compression was established. The segmented correspondence divides the amount of sowing delay into the initial delay range, the range exceeding the critical nodes of the variety, and the range close to the physiological lower limit determined by the early maturity of the variety.

[0068] Based on the late-sowing tolerance characteristics of various crop varieties, a segmented correspondence between sowing delay and growth period compression was established. This segmented correspondence has the following crop physiological significance: the initial delay range refers to when the sowing delay is small (e.g., within 3 days), the compression of the variety's growth period is not significant or the compression rate is low; the range exceeding the critical node of the variety refers to when the sowing delay exceeds a certain critical value for the variety (this node varies depending on the variety and needs to be determined by staggered sowing trials in the variety approval data, for example, 5 days for some varieties and 7 days for others), the variety enters an emergency response stage mainly focused on shortening the growth period, and the growth period compression rate increases significantly; the range approaching the physiological lower limit determined by the variety's early maturity refers to when the sowing delay is so large that it approaches the physiological limit determined by the variety's early maturity, the compressibility space of the variety is very limited, and the compression rate decreases again.

[0069] The specific critical days and compression rates for each segment vary by variety and need to be determined through variety approval trials or field staggered sowing trials. To facilitate understanding of the segmentation logic of this scheme, an example is given below: Taking a certain medium-maturing indica rice variety as an example, according to its staggered sowing trial data, when the sowing date is delayed by less than 3 days, it is within the initial delay range, and the growth period compression rate is about 0.1 to 0.2 days per day, with a weak compression effect; when the sowing date is delayed by more than 5 days, it enters the range beyond the critical node of the variety, and the compression rate jumps significantly to 0.6 to 0.8 days per day, with the variety responding to the delayed sowing date by significantly shortening the vegetative growth period; when the sowing date is delayed by more than 15 days, it approaches the physiological lower limit determined by the early maturity of the variety, and the compression rate drops back to less than 0.1 days per day, with the compressible space basically exhausted.

[0070] It is important to emphasize that the above values ​​(such as 3 days, 5 days, 15 days, and compression rates of 0.1 to 0.2, 0.6 to 0.8, and below 0.1) are merely examples to illustrate the segmentation logic and calculation principles of this scheme. The actual calibration values ​​for different varieties and different ecological zones must be based on the measured data of the varieties. This scheme does not limit the specific values ​​mentioned above, and this does not affect the implementation of this scheme. The method for determining the segmentation critical values ​​is as follows: Based on the staggered sowing and maturity records in the variety approval announcement, a response curve is plotted with the sowing delay as the horizontal axis and the actual number of days of growth as the vertical axis. The inflection point where the slope of the curve changes significantly is the critical node of the variety, and the inflection point where the slope slows down again is the point of approaching the physiological lower limit.

[0071] Among them, the rate of compression of the reproductive period after exceeding the critical node of the variety is greater than the rate of compression of the reproductive period within the initial delay range, while the rate of compression of the reproductive period when approaching the physiological lower limit is less than the rate of compression of the reproductive period after exceeding the critical node of the variety.

[0072] Based on the moisture tolerance characteristic among the moisture tolerance and late sowing tolerance characteristics, the candidate variety set was screened, including:

[0073] Based on the degree of depression in the status profile, determine the depression level of each micro-block;

[0074] For micro-blocks with a depression level higher than a preset threshold, varieties with a moisture resistance level lower than the preset moisture resistance level are removed from the candidate variety set.

[0075] Each candidate variety in the selected candidate variety set of each micro-block is sorted from earliest to latest according to the theoretical sowing date to form the sowing date-maturity correspondence sequence of that micro-block;

[0076] Calculate the rate of change of maturity date between adjacent candidate varieties in the sowing-maturity correspondence sequence. The rate of change of maturity date is the magnitude of the change in maturity date caused by the number of days the sowing date is delayed.

[0077] Based on the moisture tolerance characteristic within the moisture tolerance and late-sowing tolerance characteristics, the candidate variety set is screened. First, the degree of depression in each micro-block is determined according to the depression level in the condition profile. A preset threshold, for example, is set as moderate depression, as the dividing line. For micro-blocks with a depression level higher than the preset threshold, varieties with a moisture tolerance characteristic level lower than the preset moisture tolerance level are removed from the candidate variety set. The moisture tolerance characteristic of the varieties is quantitatively graded, divided into three levels: high moisture tolerance, medium moisture tolerance, and low moisture tolerance, based on the waterlogging resistance assessment results in the variety approval data. The preset moisture tolerance level is, for example, setting varieties with a moisture tolerance level of "medium" or higher to be retained.

[0078] After screening, the candidate varieties in the selected candidate variety set for each micro-block are sorted from earliest to latest according to their theoretical sowing date, forming the sowing date-maturity correspondence sequence for that micro-block. Based on this, the rate of change of maturity date between adjacent candidate varieties in this sowing date-maturity correspondence sequence is calculated. This rate of change of maturity date is the magnitude of the change in maturity date caused by a unit number of days the sowing date is delayed. This rate of change provides a sensitivity parameter for the compensation calculation of maturity date offset by subsequent variety replacement. The quantitative grading of the moisture tolerance characteristics of each variety is a well-known technique and will not be elaborated upon in this embodiment.

[0079] Furthermore, S4, based on the status profile, earliest sowing date, suitable sowing time window, agricultural machinery operation capacity parameters, candidate varieties, and theoretical sowing date, determines the sowing priority of each micro-block and generates the sowing sequence and variety configuration scheme for each micro-block, including:

[0080] Based on the earliest sowing date and road access conditions of each micro-block, the sowing priority of each micro-block is determined, and the initial operation sequence of agricultural machinery is generated.

[0081] Before determining the planting priority, the preset allowable deviation of maturity date must be defined. This deviation is used to determine the maximum tolerable difference in maturity date between adjacent micro-plots. Its setting needs to be determined comprehensively based on actual operational factors such as the daily operating capacity of the combine harvester and the ratio of total field area to harvest efficiency. For example, it can be set to 2 days or 3 days after comprehensive calculation.

[0082] Based on the earliest sowing date and road access conditions of each micro-block, the sowing priority of each micro-block is determined, and an initial operation sequence for agricultural machinery is generated. Specifically, the earlier the earliest sowing date, the higher the priority of the micro-block; when the earliest sowing dates are the same, the micro-blocks with better road access conditions (e.g., accessible roads and close proximity to the field entrance / exit) are given priority for operation.

[0083] Based on the agricultural machinery operation capacity parameters, the micro-blocks in the initial operation sequence of agricultural machinery are allocated to each operation day according to the sowing priority, and the actual sowing date of each micro-block is determined.

[0084] Based on the agricultural machinery operation capacity parameters, the micro-blocks in the initial agricultural machinery operation sequence are allocated to each operation day according to sowing priority, thus determining the actual sowing date for each micro-block. The agricultural machinery operation capacity parameters are jointly determined by the maximum continuous operation time and operating speed of the agricultural machinery each day, such as the maximum daily operating area. When the operation time of a micro-block exceeds the remaining operation capacity for that day, the micro-block is either cut off and continued to the next day, or the entire micro-block is postponed to the next day as the first operation target. This processing method is only one embodiment of the present invention and is not the only one; other equivalent operation allocation methods are also within the protection scope of the present invention.

[0085] For every two consecutive micro-blocks in the initial operation sequence of agricultural machinery, determine the actual sowing date of the earlier micro-block and the actual sowing date of the later micro-block in order to calculate the actual sowing date difference;

[0086] Based on the theoretical sowing date and theoretical maturity date of each candidate variety in the candidate variety set of each microblock, and the actual sowing date of each microblock, determine the actual maturity date of each candidate variety under the actual sowing date of each microblock;

[0087] The difference between the actual maturity dates is calculated based on the actual maturity dates of the earlier microblocks and the actual maturity dates of the later microblocks based on the initial candidate varieties.

[0088] Determine whether the difference between the actual maturity date and the actual maturity date exceeds the preset allowable deviation of the maturity date; the initial candidate variety is the candidate variety with the smallest difference between the theoretical sowing date and the actual sowing date.

[0089] For every two consecutive micro-blocks in the initial operation sequence of agricultural machinery, the actual sowing date of the earlier micro-block and the actual sowing date of the later micro-block are determined in order to calculate the actual sowing date difference, that is, the number of days between the actual sowing date of the later micro-block and the actual sowing date of the earlier micro-block.

[0090] Based on the theoretical sowing and maturity dates of each candidate variety in the candidate variety set of each micro-block, and the actual sowing date of each micro-block, the actual maturity date of each candidate variety under the actual sowing date of each micro-block is determined. When the actual sowing date of a micro-block is inconsistent with the theoretical sowing date corresponding to its candidate variety, the theoretical maturity date of the variety is corrected to the actual maturity date under the actual sowing date by using the aforementioned correspondence between sowing delay and growth period compression, combined with the effective accumulated temperature accumulation rate deviation coefficient of the micro-block.

[0091] Specifically, the deviation of the actual sowing date of the micro-block from the theoretical sowing date of the variety (i.e., the number of days the actual sowing date is delayed or advanced) is determined. If the actual sowing date is later than the theoretical sowing date, a positive delay occurs; if it is earlier, a negative delay occurs (in the case of early sowing). This deviation is substituted into the correspondence between sowing delay and growth period compression to obtain the corresponding growth period compression or extension. Specifically, when the actual sowing date is later than the theoretical sowing date, the variety will experience varying degrees of growth period compression due to the late sowing, and the compression amount changes with the deviation according to the aforementioned segmented rule; when the actual sowing date is earlier than the theoretical sowing date, there is usually no compression effect, but the impact of low-temperature stress that may be faced due to early sowing on the growth process needs to be considered, and this impact is reflected by the deviation coefficient.

[0092] Because the actual daily temperature sequence experienced by the micro-block after the actual sowing date is determined may differ from the historical average temperature used in the theoretical calculation, and the overall growth stage shifts later after late sowing, often corresponding to a lower temperature environment. Therefore, it is necessary to correct the accumulated temperature accumulation efficiency during the remaining growth period after the actual sowing date using this deviation coefficient. The specific method includes dividing the remaining accumulated temperature requirement of the variety (i.e., the total standard accumulated temperature minus the accumulated temperature before the actual sowing date) by the corrected effective accumulated temperature accumulation rate of the micro-block under the current sowing date (i.e., the reference rate multiplied by the deviation coefficient), to obtain the re-estimated length of the remaining growth period in calendar days.

[0093] Starting from the theoretical maturity date, the calendar shift between the actual sowing date and the theoretical sowing date is first added, and then the number of days of growth period compression caused by late sowing (or the number of days of extension caused by early sowing) is subtracted. At the same time, the number of days of the remaining growth stage is recalculated according to the deviation coefficient. After daily iterative convergence, the actual maturity date of the variety under the actual sowing date conditions in the micro-block is finally obtained.

[0094] It should be noted that the above-mentioned daily iterative calculation process can be implemented by a computer program through cyclically accumulating the effective accumulated temperature: starting from the actual sowing date, the effective accumulated temperature, after correction by the deviation coefficient, is accumulated daily until the total accumulated temperature requirement of the variety is reached, and the stopping date is the actual maturity date. This daily iterative method is a conventional technique in the field of agricultural model simulation, and the specific programming implementation is not unique; this embodiment will not elaborate on it here.

[0095] The difference between the actual maturity dates of earlier micro-plots and those of later micro-plots (based on the initial candidate varieties) is calculated. The initial candidate varieties are those with the smallest difference between their theoretical and actual sowing dates, aiming to closely approximate the predetermined agricultural machinery operation sequence and minimize subsequent variety adjustments. The difference in actual maturity dates is then assessed to determine if it exceeds a preset allowable deviation, thus necessitating subsequent variety replacement.

[0096] Furthermore, after determining whether the difference in actual maturity dates exceeds the preset allowable deviation in maturity dates, S4 also includes:

[0097] If the difference between the actual maturity dates exceeds the preset allowable deviation of maturity dates, then from the candidate variety set of the subsequent micro-blocks, select the candidate variety that makes the difference between the actual maturity dates fall within the preset allowable deviation of maturity dates and the actual maturity date of the subsequent micro-block falls within the preset mechanical harvesting window, and use it as the current configuration variety for the subsequent micro-block; select a variety to compensate for the maturity date offset caused by the difference in actual sowing dates.

[0098] If the difference between the actual maturity dates does not exceed the preset allowable deviation of maturity dates, the initial candidate variety will be used as the current configuration variety for the subsequent microblocks.

[0099] Arrange the micro-blocks according to their actual sowing dates from earliest to latest to form a sowing sequence; summarize the current varieties configured in each micro-block to form a variety configuration scheme; the optimization objective of the sowing sequence and variety configuration scheme is to ensure that the actual maturity date of each micro-block falls within the preset mechanical harvesting window, and the difference between the actual maturity dates of adjacent micro-blocks falls within the preset allowable deviation range of maturity dates.

[0100] When the difference between the actual maturity dates of two adjacent microblocks exceeds a preset allowable deviation for maturity dates, a new variety needs to be selected for the subsequent microblock. This involves: traversing each variety in the candidate variety set for the subsequent microblock; for each variety, comparing its actual maturity date at the actual planting date of the microblock with the actual maturity date of the preceding microblock, and calculating the difference; from all varieties whose differences fall within the preset allowable deviation range for maturity dates, prioritizing the variety that minimizes the difference (i.e., has the closest maturity dates); if multiple varieties can make the difference fall within the allowable range, prioritizing the variety that makes the actual maturity date of the subsequent microblock fall within a preset mechanical harvesting window, and using this as the current configured variety for that microblock. If no variety in the candidate variety set of the subsequent microblock can simultaneously meet both of the above conditions, the initial candidate varieties for that microblock remain unchanged, and the difference in actual maturity dates between the adjacent microblock pairs is recorded as a pending conflict, entering the subsequent rollback mechanism for processing. If the difference between the actual maturity dates does not exceed the preset allowable deviation of maturity dates, the initial candidate variety will be used as the current configuration variety for the subsequent microblock.

[0101] Once all consecutive microblock pairs have been evaluated, the microblocks are arranged from earliest to latest according to their actual sowing dates to form a sowing sequence; the current varieties configured in each microblock are summarized to form a variety configuration plan.

[0102] When the current configuration of a micro-block changes, the process is propagated forward along the job sequence, including:

[0103] The actual maturity date of the microblock is re-determined based on the changed current configuration variety;

[0104] For the next microblock following the current microblock in the initial operation sequence of agricultural machinery, determine the actual maturity date of the current microblock, determine the actual maturity date of the next microblock according to the currently configured variety, and calculate the difference between the actual maturity dates;

[0105] Determine whether the difference between the actual maturity dates exceeds the preset allowable deviation of the maturity dates;

[0106] If the difference is exceeded, then from the candidate variety set of the next microblock, select the candidate variety that makes the difference between the actual maturity dates fall within the allowable deviation range and the actual maturity date of the next microblock fall within the preset mechanical harvesting window, and use it as the current configuration variety for the next microblock.

[0107] Continue propagating along the job sequence until a microblock is encountered where the difference between the actual maturity date and the actual maturity date does not exceed the allowable deviation, or until the end of the job sequence is reached.

[0108] The backward propagation is triggered when a microblock in the work sequence changes its current configured variety due to variety replacement. This change alters the actual maturity date of the microblock, potentially affecting the maturity date difference between it and the next microblock. The specific implementation of backward propagation is as follows: First, determine the position of the microblock undergoing variety replacement in the initial work sequence of the agricultural machinery. Starting from this position, traverse all microblocks following that microblock sequentially. For a pair of adjacent microblocks currently being processed, the earlier microblock has already determined its final current configured variety, and its actual maturity date under its respective actual sowing date has been recalculated. For the later microblock, first calculate its actual maturity date under its current configured variety (or initial candidate variety if it has not yet been replaced), then calculate the difference between its actual maturity date and that of the earlier microblock, and determine whether this difference exceeds a preset allowable deviation in maturity date. If the deviation exceeds the limit, then from the candidate variety set of the subsequent micro-block, following the same traversal and selection rules as described above, a candidate variety is selected that ensures the difference in actual maturity dates falls within the allowable deviation range, and the actual maturity date of the subsequent micro-block falls within the preset mechanical harvesting window. This candidate variety is then selected as the current configuration variety for the subsequent micro-block. If the deviation does not exceed the limit, it indicates that the adjacent micro-block pair has met the allowable deviation requirements. In this case, the propagation along the operation sequence terminates because the actual maturity date of the subsequent micro-block has not changed, and the difference between it and the previous micro-block remains unchanged, so there is no need to continue propagation. If there is no candidate variety in the subsequent micro-block that meets the conditions, then the conflict position is recorded and a rollback mechanism is triggered.

[0109] Each time the current configuration variety is changed, a two-way verification is performed synchronously, including:

[0110] The actual maturity date of the microblock is re-determined based on the changed current configuration variety;

[0111] For the microblock preceding this microblock in the initial operation sequence of agricultural machinery, determine the actual maturity date of the preceding microblock, determine the actual maturity date of this microblock according to the currently configured variety, and calculate the difference between the actual maturity dates;

[0112] Determine whether the difference in actual maturity dates exceeds the allowable deviation;

[0113] If the difference exceeds the limit, then from the candidate variety set of the previous micro-block, a new candidate variety that makes the difference between the actual maturity dates fall within the allowable deviation range will be selected as a local backtracking.

[0114] If the partial backtracking is successful, the current configuration of the previous microblock is updated, and the process continues to propagate forward along the operation sequence.

[0115] If the local backtracking fails, the rollback mechanism is triggered.

[0116] Each time the current variety configuration is changed, a bidirectional verification is performed synchronously. The purpose of bidirectional verification is to prevent unidirectional backward propagation from disrupting the adjacent relationships in the processed segments that already meet the allowable deviation. Specifically, this includes: whenever a variety replacement in a microblock causes a change in its actual maturity date, backward propagation is first triggered to check the compatibility of the microblock with its successor microblock; simultaneously, forward backtracking is triggered to check the compatibility of the microblock with its predecessor microblock (if it exists). The forward backtracking process is as follows: for the microblock preceding the current microblock in the job sequence, the current variety configuration of the predecessor microblock and its actual maturity date under the actual sowing date are determined. The actual maturity date of the microblock with the replaced current variety configuration is determined, and the difference between the two is calculated. It is then determined whether the difference exceeds the allowable deviation. If it does, it indicates that the variety replacement has disrupted the compatibility with the predecessor microblock. In this case, all candidate varieties in the candidate variety set of the predecessor microblock are traversed, and a new candidate variety that makes the difference fall within the allowable deviation range is selected as a local backtracking step. If the local backtracking is successful (i.e., the previous microblock can be replaced with another variety to meet the allowable deviation), the current configured variety of the previous microblock is updated, and backward propagation is triggered again from the previous microblock as the new starting point to check whether the variety change further affects the compatibility of the previous or subsequent microblocks. If the local backtracking cannot find a suitable variety in the candidate variety set of the previous microblock, it means that the current conflict cannot be resolved by local variety adjustment, and the rollback mechanism is triggered.

[0117] The rollback mechanism includes:

[0118] Relax the allowable deviation threshold for the maturity date between the current conflicting microblock and its predecessor;

[0119] Using the relaxed allowable deviation threshold as a constraint, re-execute the propagation along the job sequence until the propagation terminates;

[0120] After the spread is terminated, fix the variety configuration and operation sequence of the currently processed segment, and use the actual operation time and agricultural machinery status of the fixed segment as the new initial conditions to continue processing the segments in the operation sequence that are after this segment.

[0121] When a segment in the operation sequence that is after a fixed segment still has adjacent microblock pairs whose actual maturity dates differ from the allowable deviation after backpropagation without moving the fixed segment, a global rollback is triggered: starting from the last fixed segment, the segments are unlocked level by level and backpropagation is performed again until the current configuration varieties of all microblocks in the entire field are compatible with the operation sequence.

[0122] When local backtracking fails, it is confirmed that the current conflict occurs between a pair of adjacent microblocks. To resolve this conflict, the allowable deviation threshold for maturity dates between these adjacent microblocks is relaxed, for example, by adjusting the original allowable deviation threshold upwards. Each relaxation step is one calendar day, but the maximum allowable deviation after relaxation should not exceed the maximum dispersion allowed by the mechanical harvesting window. That is, the extreme relaxation value of this deviation should not cause the difference in maturity dates between earlier-maturing and later-maturing microblocks to exceed the total length of the mechanical harvesting window. With the relaxed allowable deviation threshold as a constraint, starting from the later microblock in the adjacent microblock pair where the conflict occurred, propagation is re-executed along the operation sequence until propagation terminates (i.e., encountering a pair of microblocks whose actual maturity date difference does not exceed the relaxed allowable deviation, or propagation reaches the end of the operation sequence). After propagation terminates, the variety configuration and operation sequence of the currently processed segment (all microblocks from the start of the operation sequence to the end of the propagation) are fixed and will not participate in subsequent adjustments. Using the actual operation time and agricultural machinery status (including the calendar date at the end of the fixed segment and the consumed agricultural machinery operation capacity) as new initial conditions, the remaining segments in the operation sequence following that segment are processed. If the conflict still cannot be eliminated after relaxing to a preset upper limit threshold, the actual maturity date difference between the conflicting micro-block pairs is fixed as the maximum acceptable deviation to ensure the feasibility of the plan; during the field management stage of the micro-block, the maturity date can be fine-tuned through agronomic measures such as irrigation or chemical control.

[0123] The trigger condition for global rollback is: when, in the remaining segments after the fixed segments, without modifying the fixed segments, there are still adjacent micro-block pairs whose actual maturity dates differ by more than the allowable deviation after backpropagation, it indicates that the current fixing strategy is causing subsequent segments to fail to converge. In this case, a global rollback is executed. Specifically, starting from the last fixed segment, fixed segments are unlocked sequentially forward. For each unlocked segment, the previously fixed portions are used as new initial conditions, and backpropagation is performed again on that segment and all subsequent segments. If convergence still fails after propagation, the next fixed segment is unlocked, and so on, until the current configuration of varieties and operational sequences in all micro-blocks of the entire field are compatible. The convergence criterion for compatibility is: the difference in actual maturity dates of all adjacent micro-blocks in the entire field falls within the preset allowable deviation range for maturity dates, and the actual maturity dates of all micro-blocks fall within the preset mechanical harvest window, including both dates no later than the end date of the window and dates no earlier than the start date of the window. Early maturity is also detrimental to unified harvesting. The specific parameters and termination conditions for the relaxed threshold, fallback mechanism, and global fallback described above are only one embodiment of the present invention and are not unique. Other equivalent conflict resolution methods are all within the protection scope of the present invention.

[0124] In one specific implementation, such as Figure 2 As shown, this embodiment uses six micro-blocks of a 45-mu (approximately 3 hectares) paddy field as an example to illustrate that the actual sowing dates of each micro-block differ due to variations in low-lying areas, drainage speed, and road access conditions. Taking two adjacent micro-blocks in the sowing sequence as an example, the actual sowing date for the first micro-block A (the third micro-block, with mild low-lying areas and a moderate to fast drainage speed) was April 8th, while the actual sowing date for the second micro-block B (the fourth micro-block, with severe low-lying areas and a slow drainage speed) was April 10th due to the longer duration of rainfall. This results in a two-day difference in the actual sowing date between the two.

[0125] Because micro-block A and micro-block B were planted on different dates, even if they were initially planted with the same variety, their maturity dates would shift due to the different planting times. Micro-block B, being planted two days later, had a correspondingly delayed maturity date. To ensure that the difference in actual maturity dates between adjacent micro-blocks falls within a preset allowable deviation range for maturity dates (set to two days in this embodiment), it is necessary to determine whether this shift exceeds the preset allowable deviation for maturity dates.

[0126] For example, calculations show that the actual maturity date of microblock A under its initial candidate varieties is June 1st, and the actual maturity date of microblock B under its initial candidate varieties is June 5th, with a difference of 4 days, exceeding the allowable deviation by 2 days. In this case, variety replacement is performed in the candidate variety set of microblock B: Each variety in the candidate variety set of microblock B is traversed, and the actual maturity date of each variety under the actual sowing date of April 10th is calculated. The variety that makes the difference between the actual maturity dates of microblock B and microblock A fall within the allowable deviation range, and the actual maturity date of microblock B falls within the preset mechanical harvesting window (May 25th to June 10th), is selected as the current configured variety for microblock B. In this embodiment, after replacing the original medium-maturing variety with an early-maturing variety in microblock B, its actual maturity date is advanced to June 2nd, differing from June 1st of microblock A by 1 day, falling within the allowable deviation range.

[0127] When a variety change causes a shift in the actual maturity date of a microblock, this change propagates downstream in the work sequence. For example, if microblock B's actual maturity date changes from June 5th to June 2nd due to a variety change, it is necessary to further verify whether the difference between the actual maturity date of microblock B and its next microblock C (the fifth microblock, with an actual sowing date of April 11th) still meets the allowable deviation requirement. If not, the variety change is then performed on microblock C, forming a chain of adjustments that propagate downstream in the work sequence.

[0128] Each time the current configuration variety is changed, a bidirectional verification is performed synchronously. At each node in the backward propagation, not only is the compatibility between the microblock and its successor microblock verified, but also its compatibility with its predecessor microblock is backtracked. For example, after microblock B changes its variety, in addition to verifying the difference with microblock C, it is also necessary to re-verify whether the difference with microblock A has changed due to the change. If the difference between microblock B and microblock A exceeds the allowable deviation after the variety change, a partial backtracking is required for microblock A, reselecting a variety from its candidate variety set so that the difference falls back within the allowable deviation range. If the partial backtracking is successful, the current configuration variety of microblock A is updated, and the backward propagation restarts from microblock A. If the partial backtracking fails, it indicates that the current conflict cannot be resolved by adjusting the local variety.

[0129] When local backtracking fails to resolve the conflict, a rollback mechanism is triggered. Specifically, the allowable deviation threshold for the maturity date between the current conflicting microblock and its predecessor is relaxed, for example, from 2 days to 3 days. The backward propagation is then re-executed with the relaxed threshold as a constraint. After the propagation terminates, the variety configuration and operation sequence of the currently processed segment are fixed, and the agricultural machinery status at the end of the fixed segment is used as the new initial condition to continue processing subsequent segments.

[0130] If, after fixing a segment, there are still irreconcilable differences in maturity dates between adjacent micro-plots after backward propagation, without altering the fixed segments, a global rollback is triggered: starting from the last fixed segment, the rollback is initiated level by level, proceeding backwards again until the actual maturity date difference between all adjacent micro-plots in the entire field falls within the preset allowable deviation range for maturity dates, and the actual maturity dates of all micro-plots fall within the preset mechanical harvesting window (including both dates no later than the end date and no earlier than the start date). Early maturity is also detrimental to uniform harvesting. At this point, the planting sequence and variety configuration scheme for each micro-plot are determined.

[0131] Furthermore, S5, during the sowing process, acquires field operation status data in real time, and based on the actual sowing information of the sown micro-blocks in each micro-block, modifies the sowing sequence and variety configuration scheme of the unsown micro-blocks in each micro-block, including:

[0132] Actual sowing information includes the sowing date and the actual variety used; field operation status data includes actual rainfall data, field water level observation data, road traffic status data, and agricultural machinery arrival status data.

[0133] The sowing date and actual variety configuration of the sown micro-blocks will be treated as fixed conditions that will not be changed.

[0134] Based on the sowing date, actual variety configuration, and field operation status data of the sown micro-plots, determine the expected maturity date of the sown micro-plots;

[0135] Based on the current varietal configuration and actual sowing date of each unsown microblock, the expected maturity date of each unsown microblock is determined.

[0136] Following the sowing order of each microblock, starting from the last sown microblock and the first unsown microblock thereafter, check the difference in expected maturity date between adjacent microblocks one by one;

[0137] If the difference between the expected maturity dates exceeds the preset allowable deviation of the maturity date, the excess portion will be propagated backward as a deviation to be absorbed.

[0138] For each unsown microblock, the range of maturity date adjustment for that unsown microblock is determined based on the correspondence between the sowing delay and the growth period compression established in S3, and the absorption deviation is shared within the maturity date adjustment range through variety replacement.

[0139] The sowing sequence and variety configuration scheme of the unsown microplots are modified so that the difference between the expected maturity date of the unsown microplots and the expected maturity date of the sown microplots falls within the preset allowable deviation range of maturity date.

[0140] In the S5 correction phase, the preset allowable deviation for maturity date follows the same value set in S4 (e.g., 2 days or 3 days), which is used to establish a unified maturity date tolerance judgment standard between planted microblocks and unplanted microblocks.

[0141] During execution, the actual sowing information for the sown micro-blocks includes the actual sowing date and the actual variety selected. These two data points are fixed after sowing, forming a fixed condition for subsequent corrections. This means they cannot be changed due to subsequent optimizations, ensuring that completed field operations and agricultural records remain undisturbed. Field operation status data includes real-time rainfall data (obtained through field weather stations or regional meteorological data interfaces), field water level observation data (reported through water level sensors or manual inspections), road traffic status data (obtained through agricultural machinery vehicle-mounted terminal positioning and manual feedback), and agricultural machinery arrival status data (obtained through the shared agricultural machinery platform's real-time interface). The methods for acquiring the above real-time data are all conventional in the fields of agricultural IoT and agricultural machinery informatization, and will not be elaborated upon in this embodiment.

[0142] The expected maturity date for sown micro-plots is determined as follows: Starting from the actual sowing date of the micro-plot, and considering the varietal characteristics of the varieties actually configured in the plot, the maturity date calculation logic established in S4 is invoked (i.e., utilizing the effective accumulated temperature accumulation rate deviation coefficient and the correspondence between sowing delay and growth period compression of the micro-plot). Simultaneously, the previously used historical average data is replaced with daily measured meteorological data that has occurred since the actual sowing date, thereby calculating the most likely maturity date under the current conditions. The expected maturity date for unsown micro-plots is calculated based on their current varietal configuration and expected sowing date, temporarily extrapolated from historical average meteorological data, serving as a benchmark for subsequent adjustments.

[0143] During the pairwise inspection, starting from the last sown micro-block and the first unsown micro-block following it, the difference in expected maturity dates between adjacent micro-blocks is checked pairwise along the sowing sequence. If the difference falls within the allowable deviation range, the inspection continues to the next pair of adjacent micro-blocks; if the difference exceeds the allowable deviation, the excess is defined as the absorbable deviation, the value of which is equal to the difference in expected maturity dates between the adjacent micro-block pair minus the preset allowable maturity date deviation, and the excess is carried forward to subsequent unsown micro-blocks for variety replacement compensation. The starting point setting method for the pairwise inspection is only one embodiment of the present invention and is not the only one; other starting point setting methods do not affect the implementation of this solution.

[0144] Furthermore, by using variety substitution to share the absorption bias within the range of maturity date adjustments, including:

[0145] For unseeded micro-blocks that receive a deviation to be absorbed, determine whether the deviation to be absorbed falls within the range of maturity date adjustment.

[0146] If the target is completely met, then from the candidate variety set of the unsown microblock, the variety that matches the maturity date adjustment amount to the absorption deviation is selected as the current configuration variety for the unsown microblock.

[0147] If a portion falls into the range, then from the candidate variety set of the unsown microblock, the variety that makes the maturity date adjustment amount reach the boundary value of the maturity date adjustment range is selected as the current configuration variety of the unsown microblock. The unsown microblock shares part of the deviation to be absorbed and the remaining deviation to be absorbed is passed on to the next unsown microblock.

[0148] If the error exceeds the limit, the variety that makes the maturity date adjustment reach the maximum value of the maturity date adjustment range will be selected from the candidate variety set of the unsown microblock and used as the current configuration variety for the unsown microblock. The unsown microblock will share the deviation corresponding to its maximum capacity and continue to transmit all remaining deviations to be absorbed to the next unsown microblock.

[0149] The adjustment range for the maturity date of each unsown microblock refers to the range of maturity date changes that the microblock can achieve by changing varieties in its candidate variety set. Specifically, the expected maturity date of the earliest-matured variety in the candidate variety set for the microblock at the current actual sowing date is taken as the lower limit, and the expected maturity date of the latest-matured variety is taken as the upper limit. The difference between the two is the adjustment range for the microblock. This adjustment range directly reflects the microblock's ability to absorb maturity date deviations: the larger the range, the stronger the absorption capacity; the smaller the range, the weaker the absorption capacity.

[0150] When the deviation to be absorbed propagates along the work sequence, it shall be handled in the following three ways:

[0151] In scenario one, when the absorption deviation falls completely within the adjustment range of the maturity date of the unsown micro-block (the absolute value of the absorption deviation does not exceed the adjustment range), a variety is selected from the candidate variety set of the micro-block, and its actual maturity date is adjusted in the opposite direction of the deviation relative to the original configured variety, thereby completely eliminating the deviation.

[0152] In scenario two, when the portion of the deviation to be absorbed falls within the maturity date adjustment range of the unsown microblock (the absolute value of the deviation to be absorbed is greater than the adjustment range but does not exceed twice the range), the microblock selects a variety that allows the maturity date adjustment to reach the boundary value of its adjustment range, completes the distribution of the deviation within its own capacity, and continues to transmit the remaining unabsorbed portion to the next unsown microblock.

[0153] Scenario 3: When the deviation to be absorbed completely exceeds the maturity date adjustment range of the unsown microblock (the absolute value of the deviation to be absorbed is greater than twice the adjustment range), the microblock selects the variety that makes the maturity date adjustment reach the maximum value of its adjustment range. That is, the microblock shares the deviation with the maximum capacity, but all the deviation to be absorbed continues to be transmitted to the next microblock for processing.

[0154] The above three scenarios constitute a complete mechanism for the step-by-step distribution of deviations along the operational sequence, ensuring that each micro-block reasonably shares the correction pressure based on its own variety structure adjustment capabilities. The calculation method for the maturity date adjustment range in this embodiment is only a preferred option and is not the only one; other definition methods do not affect the implementation of this scheme.

[0155] Furthermore, during the process of checking and transmitting data backward pair by pair, the reservation status of agricultural machinery in each unsown micro-block is checked:

[0156] For unsown micro-blocks where agricultural machinery reservations have not been confirmed, the sowing sequence is allowed to be adjusted within the suitable sowing time window, and the absorption deviation is allowed to be shared within the maturity date adjustment range through variety replacement; the suitable sowing time window includes the suitable sowing time window updated with real-time data;

[0157] For unsown micro-blocks where agricultural machinery reservations have been confirmed and the work period cannot be changed, it is forbidden to adjust the sowing order. Only variety replacement is allowed to share the absorption deviation within the range of maturity date adjustment. The remaining absorption deviation will be forcibly transmitted to the next unsown micro-block.

[0158] If there are no unplanted microblocks after the unplanted microblock that cannot be changed during the operation period, or if the adjustment range of the maturity date of all subsequent unplanted microblocks cannot absorb the remaining deviation to be absorbed, then the allowable deviation threshold of the maturity date between the unplanted microblock and its predecessor is relaxed, so that the threshold after the relaxation of the remaining deviation to be absorbed is compatible.

[0159] During the backward propagation of deviations, the agricultural machinery reservation status of each unsown micro-block has a binding effect on the corrective behavior and should be treated differently.

[0160] The status of agricultural machinery reservations is obtained by querying the reservation confirmation status and work period lock-in status of each micro-block in real time through the shared agricultural machinery platform interface. For micro-blocks where agricultural machinery reservations are not confirmed, the sowing sequence can be adjusted within the suitable sowing time window. Variety substitution can also be used to mitigate absorption deviations within the maturity date adjustment range. The suitable sowing time window needs to be dynamically updated based on real-time field operation status data. For example, if actual rainfall causes a decrease in the rate of field water receding, the end date of the suitable sowing time window will be postponed or shortened accordingly. The suitable sowing time window includes the suitable sowing time window updated with real-time data.

[0161] For micro-blocks where agricultural machinery reservations have been confirmed and the work period cannot be changed, the sowing order is prohibited from being adjusted. Only variety replacement is allowed to share the deviation to be absorbed within the adjustment range of its maturity date. If there is still a remaining deviation that cannot be absorbed after variety replacement in the micro-block, it will be forcibly transferred to the next unsown micro-block.

[0162] When there are no unplanted micro-blocks following the aforementioned micro-block whose operating period cannot be changed, or when the adjustment range for the maturity date of all subsequent unplanted micro-blocks has been exhausted and the remaining deviation cannot be absorbed, as a last resort, the allowable deviation threshold for the maturity date between the micro-block and its predecessor is relaxed. The relaxation method is to increase the allowable deviation from the original value (e.g., 2 or 3 days) by one step (e.g., to 5 days). If this is still insufficient, it can be further increased to 7 days or a larger value. However, the upper limit of the relaxation should not exceed the maximum time span allowed by the preset mechanical harvesting window to avoid making whole-field harvesting infeasible, ultimately ensuring that the threshold after the relaxation of the remaining deviation to be absorbed is compatible. The specific step size and upper limit of this relaxation threshold are only one implementation method in this embodiment and are not unique; other setting methods do not affect the implementation of this solution.

[0163] Furthermore, when there are no varieties in the candidate variety set of unsown microblocks that can achieve the required maturity date adjustment:

[0164] Based on real-time field operation status data, update the water receding speed and road access conditions in the status profile of the unsown micro-block;

[0165] Based on the updated receding rate and road access conditions, the deviation coefficient of the effective accumulated temperature accumulation rate of the unseeded microplot relative to the reference plot is re-determined, and the suitable planting time window for the unseeded microplot is re-determined.

[0166] Based on the newly determined suitable sowing time window, and combined with the growth period and accumulated temperature requirements of each crop variety, the candidate variety set for the unsown micro-block was re-screened.

[0167] The scope of re-screening is limited to the currently unsown microblock and the preset number of adjacent unsown microblocks. The candidate variety set of sown microblocks and unsown microblocks in the sowing sequence that precede the currently unsown microblock remains unchanged.

[0168] When there is no variety in the candidate variety set of an unsown microblock that can achieve the required adjustment amount of maturity date, a real-time re-screening mechanism is triggered.

[0169] The trigger condition for re-screening is: in the candidate variety set of this micro-block, the adjustment range of the maturity date of all varieties cannot cover the current value of the deviation to be absorbed.

[0170] The input data for re-screening needs to be updated based on real-time field operation status data: First, based on the latest acquired rainfall data, field water level observation data, and road access status data, update the receding rate and road access conditions in the micro-block status profile; Second, based on the updated receding rate, recalculate the deviation coefficient of the effective accumulated temperature accumulation rate of the micro-block relative to the reference field according to the same logic in S1 and S4 (the slower the receding, the smaller the deviation coefficient); At the same time, based on the updated road access conditions and receding rate, redetermine the suitable sowing time window of the micro-block according to the same logic in S2 (delayed receding may cause the window to shift later or shorten); Finally, based on the redetermined suitable sowing time window, combined with the growth period and accumulated temperature requirements of each crop variety, re-screen the candidate variety set of the micro-block according to the screening logic completely consistent with S4.

[0171] To avoid the calculation scope from expanding indefinitely due to re-screening, this scheme limits the scope of impact: the preset number refers to the maximum number of unsown micro-blocks affected by re-screening, which can be set to 3 or 5 depending on the actual job sequence length and computing resources. Specifically, the scope of re-screening is limited to the currently unsown micro-block and the preset number of adjacent unsown micro-blocks; the candidate variety set of already sown micro-blocks and unsown micro-blocks preceding the currently unsown micro-block in the sowing sequence remains unchanged. The specific value of this preset number is only one implementation method and is not unique; other values ​​do not affect the implementation of this scheme.

[0172] Furthermore, the sowing sequence and variety configuration scheme for unsown micro-plots were revised, including:

[0173] Arrange the unsown micro-blocks according to the revised sowing date from earliest to latest to form the revised sowing order;

[0174] The current variety configuration scheme is compiled by summarizing the varieties determined after the unsown micro-blocks are assigned and re-screened;

[0175] Verify whether the expected maturity dates of all micro-plots in the entire field fall within the preset mechanical harvesting window, and whether the difference in expected maturity dates between adjacent micro-plots along the planting sequence falls within the preset allowable deviation range for maturity dates.

[0176] It should be noted that the search space of the backpropagation, local backtracking, and global rollback mechanisms in this application is limited by the number of micro-blocks within the same field. In small-scale planting scenarios in hilly areas, the number of micro-blocks within the same field is typically 3 to 15. The candidate variety set is pre-screened and limited by S3, and the propagation direction is unidirectional along the operation sequence. Each propagation step fixes the current node, and there is no loop oscillation. Therefore, the above convergence process can be completed in a finite number of steps without the need to set additional truncation conditions.

[0177] After the correction is completed, the specific implementation method for correcting the sowing order and variety configuration scheme of the unsown micro-blocks is as follows: arrange each unsown micro-block in order of the corrected sowing date from earliest to latest, and splice it with the fixed sowing order of the sown micro-blocks to form the corrected sowing order; at the same time, summarize the current configuration varieties finally determined after deviation sharing and re-screening of each unsown micro-block with the actual configuration varieties of the sown micro-blocks to form the corrected variety configuration scheme.

[0178] After the corrections are complete, a final verification is performed. The verification includes two necessary conditions:

[0179] Condition 1: Verify whether the expected maturity date of all micro-plots (including those that have been sown and those that have not) in the entire field falls within the preset mechanical harvesting window, i.e., no earlier than the start date of the window and no later than the end date of the window;

[0180] Condition 2: Verify whether the expected maturity date difference between adjacent microblocks arranged along the final sowing sequence all fall within the preset allowable deviation range of maturity date.

[0181] If both of the above conditions are met simultaneously, the current solution is output as the final solution and executed. If the verification fails, the deviation value is treated as a new deviation to be absorbed, and the correction process is re-executed, forming an iterative loop. To avoid the endless looping of the correction process leading to the exhaustion of computing resources, this embodiment sets a preset maximum number of iterations as a termination condition, such as 10 or 20 times. When the number of iterations reaches this preset upper limit, the correction is forcibly terminated, the current solution is output as the final solution, and a deviation that has not fully converged is marked during output for manual review and intervention by job schedulers. The specific value of the maximum number of iterations is only one implementation method in this embodiment and is not unique; other values ​​do not affect the implementation of this solution. Setting the maximum number of iterations is a conventional method in the field of computer program flow control, and will not be elaborated upon in this embodiment.

[0182] In one specific implementation, such as Figure 3 As shown, during the sowing process, the actual sowing information of the sown micro-blocks, namely the sowing date and the actual variety configured, is fixed after sowing is completed, constituting a fixed condition for subsequent corrections and cannot be changed due to subsequent optimizations. In this embodiment, micro-blocks 1 to 3 have been sown, and their sowing dates and actual variety configured are already determined, becoming unchangeable fixed conditions.

[0183] Starting with the last planted microplot (microplot 3) and the first unplanted microplot (microplot 4) after it, the difference in expected maturity dates between adjacent microplots is checked pair by pair along the planting sequence. The calculated expected maturity date for microplot 3 is June 1st, and for microplot 4, based on the currently configured variety, it is June 5th. The difference is 4 days, while the preset allowable deviation for maturity dates is 2 days. This difference exceeds the allowable deviation, and the excess is defined as the deviation to be absorbed and propagated to the next unplanted microplot (microplot 4).

[0184] Each unsown micro-plot adjusts its maturity date range according to its own capabilities, sharing the absorption deviation accordingly. The maturity date adjustment range is the range of maturity date changes achievable by changing varieties from the candidate variety set for that micro-plot. For micro-plot 4, the expected maturity date of the earliest maturing variety among its candidate varieties is May 31, and the expected maturity date of the latest maturing variety is June 8, with an adjustment range of 8 days. Since the absorption deviation is 2 days, which falls entirely within this adjustment range, a variety is selected from the candidate variety set of micro-plot 4, and its actual maturity date is advanced by 2 days compared to the original selected variety, thus completely eliminating the deviation.

[0185] If the expected absorption deviation is 5 days, and the adjustment range of micro-block 4 is only 3 days, then it is a partial fall-in case: micro-block 4 selects varieties to make the maturity date adjustment amount reach the boundary value of the adjustment range (3 days), sharing part of the deviation, and the remaining 2 days continue to be transmitted to the next unplanted micro-block (micro-block 5). If the expected absorption deviation is 8 days, and the adjustment range of micro-block 4 is 3 days, then it is a complete exceedance case: micro-block 4 selects varieties to make the maturity date adjustment amount reach the maximum value of the adjustment range (3 days), and all the expected absorption deviation (8 days) continues to be transmitted to the next micro-block for processing.

[0186] During the deviation propagation process, the agricultural machinery reservation status of each unsown micro-block has a binding effect on the corrective action. For micro-blocks with unconfirmed agricultural machinery reservations, the sowing sequence is allowed to be adjusted within the suitable sowing time window, and the deviation to be absorbed can be shared through variety substitution. For micro-blocks with confirmed agricultural machinery reservations and unchangeable operation periods, the sowing sequence is prohibited from being adjusted, and the deviation to be absorbed can only be shared through variety substitution; if there is still a residual deviation after variety substitution, it is forcibly propagated to the next unsown micro-block. When there are no unsown micro-blocks after the current micro-block or the subsequent micro-blocks are unable to absorb the residual deviation, the allowable deviation threshold for maturity dates between the current micro-block and the previous micro-block is relaxed to make the relaxed thresholds for residual deviations compatible, for example, the allowable deviation is relaxed from 2 days to 3 days or 5 days, but the upper limit of the relaxation does not exceed the maximum span allowed by the mechanical harvesting window. The deviation is propagated step by step along the operation sequence, and each micro-block shares the burden step by step according to its own capacity, until all the deviations to be absorbed are completely eliminated or the correction termination condition is met. Finally, the expected maturity dates of all micro-plots in the entire field were verified to fall within the preset mechanical harvesting window, and the difference in expected maturity dates between adjacent micro-plots along the planting sequence was within the preset allowable deviation range for maturity dates. The correction was then completed.

[0187] According to another aspect of the embodiments of this application, an agricultural planting decision-making system based on multi-source information fusion is also provided. For example... Figure 4 As shown, the system includes:

[0188] The status profile generation module 41 is used to acquire multi-source data, divide the field into multiple micro-blocks based on the multi-source data, determine the low-lying degree, drainage speed and road access conditions of each micro-block, and generate a status profile of each micro-block.

[0189] The sowing parameter prediction module 42 is used to acquire agricultural machinery operation constraint data; based on the agricultural machinery operation constraint data and status profile, it determines the earliest sowing date, suitable sowing time window and agricultural machinery operation capability parameters for each micro-block;

[0190] The variety matching module 43 is used to determine multiple candidate varieties for each micro-block based on the crop variety’s growth period, accumulated temperature requirements, and moisture tolerance and late sowing characteristics, combined with the appropriate sowing time window, with the preset mechanical harvesting window as a constraint, and to determine the theoretical sowing date corresponding to each candidate variety.

[0191] The scheme generation module 44 is used to determine the sowing priority of each micro-block based on the status profile, the earliest sowing date, the suitable sowing time window, the agricultural machinery operation capacity parameters, candidate varieties and theoretical sowing dates, and to generate the sowing sequence and variety configuration scheme of each micro-block.

[0192] The correction module 45 is used to acquire field operation status data in real time during the sowing process, and correct the sowing sequence and variety configuration scheme of the unsown micro-blocks in each micro-block based on the actual sowing information of the sown micro-blocks in each micro-block.

[0193] It should be noted that the embodiments implemented on the system side in this application can be referenced to the embodiments implemented on the method side, and will not be described in detail here.

[0194] According to another aspect of the embodiments of this application, an electronic device for implementing the above-described agricultural planting decision-making method based on multi-source information fusion is also provided. This electronic device may be... Figure 5 The terminal device or server shown. This embodiment uses this electronic device as an example of a server. Figure 5 As shown, the electronic device includes a memory 402, a processor 404, and a transmission device 406. The memory 402 stores a computer program, and the processor 404 is configured to execute the steps of any of the above method embodiments through the computer program.

[0195] Optionally, in this embodiment, the aforementioned electronic device may be located in at least one of a plurality of network devices in a computer network.

[0196] Optionally, the transmission device 406 is used to receive or send data via a network. Specific examples of the network described above may include wired and wireless networks. In one example, the transmission device 406 includes a Network Interface Controller (NIC), which can be connected to other network devices and a router via a network cable to communicate with the Internet or a local area network. In another example, the transmission device 406 is a Radio Frequency (RF) module used to communicate with the Internet wirelessly. Furthermore, the electronic device also includes a display 408 and a connection bus 410, which connects the various module components within the electronic device.

[0197] The above are merely preferred embodiments of the present invention and do not limit the scope of the patent. Any equivalent structural or procedural transformations made based on the description and drawings of the present invention, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of the present invention.

Claims

1. An agricultural planting decision-making method based on multi-source information fusion, characterized in that, include: S1. Acquire multi-source data, divide the field into multiple micro-blocks based on the multi-source data, and determine the low-lying degree, drainage speed and road access conditions of each micro-block to generate a status profile of each micro-block. S2, acquire agricultural machinery operation constraint data; based on the agricultural machinery operation constraint data and the status profile, determine the earliest sowing date, suitable sowing time window and agricultural machinery operation capability parameters for each micro-block; S3, using the preset mechanical harvesting window as a constraint, and based on the crop variety's growth period, accumulated temperature requirements, and tolerance to moisture and late sowing, combined with the suitable sowing time window, determine multiple candidate varieties for each micro-block, and determine the theoretical sowing date corresponding to each candidate variety. S4. Based on the status profile, the earliest sowing date, the suitable sowing time window, the agricultural machinery operation capability parameters, the candidate varieties, and the theoretical sowing date, determine the sowing priority of each micro-block, and generate the sowing sequence and variety configuration scheme for each micro-block. S5 acquires real-time field operation status data during the sowing process and modifies the sowing sequence and variety configuration scheme of the unsown micro-blocks in each micro-block based on the actual sowing information of the sown micro-blocks in each micro-block.

2. The agricultural planting decision-making method based on multi-source information fusion according to claim 1, characterized in that, The multi-source data includes: field boundary data, UAV micro-topographic data, historical drainage records, soil moisture data, and rainfall forecast data; S1 includes: Based on the field boundary data, the geographical range of the field is determined; Based on the micro-topography data from the UAV, the elevation information within the field is extracted, multiple low-lying areas are identified, and the degree of low-lying in each area is determined. Based on the historical receding water record data and the elevation information, the receding water rate of each low-lying area is determined; Based on the soil moisture data and the rainfall forecast data, assess the impact of the muddy conditions of each road section within the field on traffic, and determine the road access conditions for each micro-block; Based on the degree of depression, the rate of water receding, and the road accessibility, the field is divided into multiple micro-blocks, and a status profile of each micro-block is generated.

3. The agricultural planting decision-making method based on multi-source information fusion according to claim 1, characterized in that, The agricultural machinery operation constraint data includes shared agricultural machinery reservation queuing data, historical agricultural machinery arrival delay data, agricultural machinery operation speed data, and future rainfall forecast data; S2 includes: Based on the shared agricultural machinery reservation queuing data, the reservation time slot for agricultural machinery is determined; Based on the historical arrival delay data of the agricultural machinery, the scheduled operation time is corrected to obtain the estimated arrival time of the agricultural machinery in each micro-block; The operation time of the agricultural machinery is calculated based on the agricultural machinery operation speed data and the area of ​​each micro-block; Based on the aforementioned future rainfall forecast data, determine the duration of the impact of rainfall on field operations; Based on the estimated arrival time, the duration of agricultural machinery operation, and the duration of impact, determine the earliest sowing date and suitable sowing time window for each micro-block; The agricultural machinery operation capacity parameters are determined based on the agricultural machinery operation speed data and the scheduled operation time period.

4. The agricultural planting decision-making method based on multi-source information fusion according to claim 1, characterized in that, S3 includes: Based on the preset mechanical harvesting window, a target maturity date is determined, wherein the target maturity date is located within the preset mechanical harvesting window; For each microblock, Based on the depression level and drainage rate in the state profile of the micro-block, combined with the historical drainage records and soil moisture data in the multi-source data, the deviation coefficient of the effective accumulated temperature accumulation rate of the micro-block relative to the reference field is determined. Based on the growth period and accumulated temperature requirements of each crop variety, and combined with the aforementioned deviation coefficient, the accumulated temperature requirements of each crop variety are converted into an estimated value of the calendar growth period of the micro-block. Based on the tolerance to late sowing characteristics among the moisture tolerance and late sowing tolerance characteristics of each crop variety, establish the correspondence between the sowing delay and the growth period compression of each crop variety in this micro-block; Based on the target maturity date, the estimated calendar growth period, and the correspondence between the sowing delay and the growth period compression, the theoretical sowing date of each crop variety in the micro-block is deduced from the target maturity date, and the theoretical maturity date of the corresponding crop variety when sowing on the theoretical sowing date is determined. Determine whether the theoretical maturity date falls within the preset mechanical harvesting window, and determine whether the theoretical sowing date falls within the suitable sowing time window for the micro-block; Crop varieties whose theoretical maturity date falls within the preset mechanical harvesting window and whose theoretical sowing date falls within the suitable sowing time window of the micro-block are added to the candidate variety set of the micro-block, and the corresponding theoretical sowing date is used as the theoretical sowing date corresponding to each candidate variety in the candidate variety set.

5. The agricultural planting decision-making method based on multi-source information fusion according to claim 4, characterized in that, The step of establishing the correspondence between the sowing delay and the growth period compression of each crop variety within the micro-block, based on the late-sowing tolerance characteristic among the wet-tolerance and late-sowing tolerance characteristics of each crop variety, includes: When the number of mature records under different sowing periods in the historical water receding record data is lower than a preset threshold, mature period data of the micro-block under different sowing period scenarios are generated to expand the sample size. Based on the late-sowing tolerance characteristics of each crop variety, a segmented correspondence between the amount of sowing delay and the amount of growth period compression is established; the segmented correspondence divides the amount of sowing delay into an initial delay range, a range exceeding the critical nodes of the variety, and a range close to the physiological lower limit determined by the early maturity of the variety. Among them, the rate of compression of the reproductive period after exceeding the critical node of the variety is greater than the rate of compression of the reproductive period within the initial delay range, while the rate of compression of the reproductive period when approaching the physiological lower limit is less than the rate of compression of the reproductive period after exceeding the critical node of the variety.

6. The agricultural planting decision-making method based on multi-source information fusion according to claim 4, characterized in that, S4 include: Based on the earliest sowing date and road access conditions of each micro-block, the sowing priority of each micro-block is determined, and the initial operation sequence of agricultural machinery is generated. Based on the agricultural machinery operation capacity parameters, the micro-blocks in the initial operation sequence of the agricultural machinery are allocated to each operation day according to the sowing priority, and the actual sowing date of each micro-block is determined. For every two consecutive micro-blocks in the initial operation sequence of the agricultural machinery, the actual sowing date of the earlier micro-block is determined, and the actual sowing date of the later micro-block is determined, in order to calculate the actual sowing date difference; Based on the theoretical sowing date and theoretical maturity date of each candidate variety in the candidate variety set of each microblock, and the actual sowing date of each microblock, determine the actual maturity date of each candidate variety under the actual sowing date of each microblock; The difference between the actual maturity dates is calculated based on the actual maturity dates of the earlier microblocks and the actual maturity dates of the later microblocks based on the initial candidate varieties. Determine whether the difference between the actual maturity date and the actual maturity date exceeds the preset allowable deviation of the maturity date; the initial candidate variety is the candidate variety with the smallest difference between the theoretical sowing date and the actual sowing date.

7. The agricultural planting decision-making method based on multi-source information fusion according to claim 6, characterized in that, After determining whether the difference in actual maturity dates exceeds a preset allowable deviation in maturity dates, S4 further includes: If the difference between the actual maturity dates exceeds the preset allowable deviation of maturity dates, then from the candidate variety set of the subsequent micro-blocks, a candidate variety that makes the difference between the actual maturity dates fall within the preset allowable deviation of maturity dates and the actual maturity date of the subsequent micro-block falls within the preset mechanical harvesting window is selected as the current configuration variety for the subsequent micro-block; the selection is used to compensate for the maturity date offset caused by the difference in actual sowing dates. If the difference between the actual maturity dates does not exceed the preset allowable deviation of maturity dates, then the initial candidate variety will be used as the current configuration variety for the subsequent microblock. The micro-blocks are arranged from earliest to latest according to their actual sowing dates to form a sowing sequence; the current varieties configured in each micro-block are summarized to form a variety configuration scheme; wherein, the optimization objective of the sowing sequence and variety configuration scheme is to ensure that the actual maturity date of each micro-block falls within the preset mechanical harvesting window, and the difference between the actual maturity dates of adjacent micro-blocks falls within the preset allowable deviation range of maturity dates.

8. The agricultural planting decision-making method based on multi-source information fusion according to claim 4, characterized in that, S5 include: The actual sowing information includes the sowing date and the actual variety configured, and the field operation status data includes rainfall data, field water level observation data, road traffic status data, and agricultural machinery arrival status data; The sowing date and actual variety configuration of the sown micro-blocks will be treated as fixed conditions that will not be changed. Based on the sowing date, actual variety configuration, and field operation status data of the sown micro-plots, determine the expected maturity date of the sown micro-plots; and estimate the expected maturity date of each unsown micro-plot.

9. The agricultural planting decision-making method based on multi-source information fusion according to claim 8, characterized in that, After determining the expected maturity date for each microblock, S5 also includes: Following the sowing order of each microblock, starting from the last sown microblock and the first unsown microblock thereafter, check the difference in expected maturity date between adjacent microblocks one by one; If the difference between the expected maturity dates exceeds the preset allowable deviation of the maturity date, the excess portion will be carried forward as a deviation to be absorbed. For each unsown microblock, the maturity date adjustment range of the unsown microblock is determined based on the correspondence between the sowing delay and the growth period compression established in S3, and the deviation to be absorbed is shared within the maturity date adjustment range by variety replacement; The sowing sequence and variety configuration scheme of the unsown microplots are modified so that the difference between the expected maturity date of the unsown microplots and the expected maturity date of the sown microplots falls within the preset allowable deviation range of maturity date.

10. An agricultural planting decision-making system based on multi-source information fusion, characterized in that, include: The status profile generation module is used to acquire multi-source data, divide the field into multiple micro-blocks based on the multi-source data, determine the low-lying degree, drainage speed and road access conditions of each micro-block, and generate a status profile of each micro-block. The sowing parameter prediction module is used to acquire agricultural machinery operation constraint data; based on the agricultural machinery operation constraint data and the status profile, it determines the earliest sowing date, suitable sowing time window, and agricultural machinery operation capability parameters for each micro-block. The variety matching module is used to determine multiple candidate varieties for each micro-block based on the crop variety's growth period, accumulated temperature requirements, and tolerance to moisture and late sowing, combined with the suitable sowing time window, with a preset mechanical harvesting window as a constraint, and to determine the theoretical sowing date corresponding to each candidate variety. The scheme generation module is used to determine the sowing priority of each micro-block based on the status profile, the earliest sowing date, the suitable sowing time window, the agricultural machinery operation capability parameters, the candidate varieties and the theoretical sowing date, and to generate the sowing sequence and variety configuration scheme of each micro-block. The correction module is used to acquire field operation status data in real time during the sowing process, and correct the sowing sequence and variety configuration scheme of the unsown micro-blocks in each micro-block based on the actual sowing information of the sown micro-blocks in each micro-block.

Images