A method for path coordination and scheduling optimization of corn straw return to the field and fertilization operations

Abstract

This invention relates to the field of agricultural fertilization, and more particularly to a method for optimizing the path coordination scheduling of corn straw return to the field and fertilization operations. The method includes: obtaining standardized feature vectors for field units by standardizing multi-source raw data of farmland; obtaining a feature contribution balancing factor by analyzing the imbalance of distance components in the feature dimensions; obtaining a spatial membership smoothing factor by evaluating the activity of the neighborhood membership distribution of field units; obtaining an adaptive distance metric by combining the feature contribution balancing factor and the spatial membership smoothing factor; and obtaining a set of agricultural machinery path coordination scheduling instructions by hardening the final membership matrix. This method solves the problems of inaccurate partitioning results, fragmented management boundaries, and unreasonable agricultural machinery path scheduling in existing standard FCM clustering algorithms based on abnormal straw accumulation areas.

Description

Technical Field

[0001] This invention relates to the field of agricultural fertilization technology, and in particular to a method for path coordination and scheduling optimization of corn straw return to the field and fertilization operations. Background Technology

[0002] In modern precision agriculture, to achieve differentiated and refined management of large areas of farmland, it is typically necessary to divide the farmland into several management zones. This ensures that plots within the same area share similar characteristics in soil properties and crop growth potential, enabling the application of uniform field management practices. Currently, fuzzy C-means (FCM) clustering is a commonly used method for farmland management zoning. This method collects multi-source agronomic data from multiple sampling points within the farmland area, such as historical yield data, soil apparent electrical conductivity (ECa) data, and digital elevation model (DEM) data. Each sampling point is abstracted into a multi-dimensional feature vector, and the similarity between different plots is calculated using Euclidean distance. Based on this, the FCM algorithm iteratively optimizes and assigns similar plots to corresponding cluster categories, thus forming a management zoning map that reflects the spatial heterogeneity of farmland. This zoning method provides crucial data support and decision-making basis for precision farming activities such as variable fertilization and variable sowing. With the widespread adoption of straw return technology in agricultural production, farmland zoning methods increasingly require the incorporation of novel characteristic data, such as straw coverage or straw density, into cluster analysis to reflect the impact of straw distribution on soil nutrient cycling and crop growth. However, existing zoning methods based on the standard FCM algorithm have certain shortcomings when integrating straw density characteristics with traditional agronomic features. Because the Euclidean distance calculation applies equal weights to each feature dimension, when straw density values ​​in abnormally accumulated areas are significantly higher than surrounding areas due to overlapping operations or field deviance, this single-dimensional difference dominates the distance calculation, disrupting the balanced contribution of the original multi-dimensional features. This often results in these abnormal areas being incorrectly separated, forming fragmented boundaries or enclaves that do not match actual agronomic needs, thus distorting the management zoning results. More importantly, this distorted zoning result leads to inefficiency and resource waste in subsequent agricultural machinery path planning and operation scheduling. Therefore, how to effectively suppress the distance contamination effect caused by a single operational feature during the farmland zoning process, and obtain geometrically regular and executable zoning results while ensuring the authenticity of agronomic features, has become a technical problem that urgently needs to be solved in this field. Summary of the Invention

[0003] In view of this, the present invention aims to propose a path collaborative scheduling optimization method for corn straw returning to the field and fertilization operations, in order to solve the problems of inaccurate partitioning results, fragmented management boundaries, and unreasonable agricultural machinery path scheduling in existing standard FCM clustering algorithms.

[0004] To achieve the above objectives, the technical solution of the present invention is implemented as follows:

[0005] A method for path coordination and scheduling optimization of corn straw returning to the field and fertilization operations includes:

[0006] Step S1: Obtain standardized feature vectors for field units by standardizing the original data from multiple sources of farmland.

[0007] Step S2: Obtain the feature contribution balance factor by performing imbalance analysis on the distance components of the feature dimension;

[0008] Step S3: Obtain the spatial membership smoothing factor by evaluating the activity of the neighborhood membership distribution of field units;

[0009] Step S4: Obtain an adaptive distance metric by combining the joint feature contribution balancing factor and the spatial membership smoothing factor;

[0010] Step S5: Obtain the agricultural machinery path collaborative scheduling instruction set by hardening the final membership matrix.

[0011] Furthermore, the step of obtaining standardized feature vectors for field units by standardizing the multi-source raw data of farmland includes:

[0012] The target farmland area is sampled in a grid pattern to obtain all field units. Historical yield data of the field units is obtained using a combine harvester equipped with a yield monitor. Soil apparent electrical conductivity data of the field units is obtained using a soil electrical conductivity scanning device. Digital elevation model data of the field units is obtained using airborne or spaceborne remote sensing equipment. Straw density data of the field units is obtained using a UAV equipped with a multispectral or high-resolution visible light camera and combined with image analysis technology. The historical yield data, soil apparent electrical conductivity data, digital elevation model data, and straw density data are used to construct a multidimensional raw data matrix. Z-score standardization is performed on each feature dimension of the matrix to obtain the standardized feature vector corresponding to each field unit.

[0013] Furthermore, the step of obtaining the feature contribution balance factor by performing imbalance analysis on the distance components of the feature dimension includes:

[0014] Cluster centers are obtained during the clustering process; energy distribution analysis is performed on the difference data of feature dimensions between field units and cluster centers to obtain the distance contribution imbalance index; and nonlinear response processing is performed on the distance contribution imbalance index to obtain the feature contribution balance factor.

[0015] Furthermore, the step of performing energy distribution analysis on the feature dimension difference data between field units and cluster centers to obtain a distance contribution imbalance index includes:

[0016] For any target field unit among all field units and any target cluster center among all cluster centers, the square of the difference between the target field unit and the target cluster center in any target feature dimension is taken as the distance energy component of the target feature dimension; the distance energy components of all feature dimensions are accumulated to obtain the total energy distribution assessment; the fourth power of the difference between the target field unit and the target cluster center in all target feature dimensions is accumulated to obtain the energy concentration assessment.

[0017] The result of multiplying the number of feature dimensions by the energy concentration assessment is used as the numerator; the square of the total energy distribution assessment is used as the denominator, and the result of subtracting the constant 1 from the corresponding fraction is used as the distance contribution imbalance index.

[0018] Furthermore, the step of obtaining the feature contribution balance factor by performing nonlinear response processing on the distance contribution imbalance index includes:

[0019] Set the gain coefficient of the imbalance response; amplify the distance contribution imbalance index by applying the gain coefficient to obtain the amplified imbalance response value; perform arctangent nonlinear mapping on the amplified imbalance response value and normalize it by the ratio of constant 2 to pi to obtain the smooth correction increment between zero and constant 1; use the calculation result of adding the smooth correction increment to constant 1 as the characteristic contribution balance factor.

[0020] Furthermore, the step of obtaining a spatial membership smoothing factor by evaluating the activity of the neighborhood membership distribution of field units includes:

[0021] By statistically processing the neighborhood membership data of field units, the mean neighborhood membership data is obtained.

[0022] By performing variance analysis on the neighborhood membership data of field units, the neighborhood membership activity index was obtained.

[0023] By adaptively processing the deviation data between the membership degree of field units and the mean of their neighborhood, a spatial membership degree smoothing factor is obtained.

[0024] Furthermore, the step of statistically processing the neighborhood membership data of field units to obtain the mean neighborhood membership data includes:

[0025] Define the neighborhood range of the field unit. For any target field unit among all field units, obtain the set of neighborhood membership degrees of the target field unit through the neighborhood range of the field unit.

[0026] During the clustering process, the membership values ​​of all field units in the neighborhood membership set relative to the same target cluster center are accumulated. The accumulated value is then divided by the number of field units in the neighborhood to obtain the mean neighborhood membership data of the target field unit at the target cluster center.

[0027] Furthermore, the step of performing variance analysis on the neighborhood membership data of field units to obtain a neighborhood membership activity index includes:

[0028] Within the neighborhood centered on the target field unit, the membership degree values ​​of each field unit on all cluster centers are obtained; the membership degree values ​​of each field unit in the neighborhood on all cluster centers are used as the overall sample, and the variance value of the sample is calculated. The variance value is used as the neighborhood membership activity index.

[0029] Furthermore, the step of obtaining a spatial membership smoothing factor by adaptively processing the deviation data between the membership degree of field units and the mean of their neighborhood includes:

[0030] Obtain the membership degree value of the target field unit at the target cluster center, and calculate the difference between it and the mean membership degree data of the neighborhood; square the difference to obtain the deviation energy of the membership degree of the target field unit; add the deviation energy to the neighborhood membership degree activity index to obtain the neighborhood-corrected membership degree fluctuation value; obtain the spatial membership degree smoothing factor by calculating the ratio of the neighborhood-corrected membership degree fluctuation value to the neighborhood membership degree activity index.

[0031] Furthermore, the step of obtaining the adaptive distance metric by combining the joint feature contribution balancing factor and the spatial membership smoothing factor includes:

[0032] Obtain the original Euclidean distance between the target field unit and the target cluster center; obtain the distance value based on the feature dimension distance component correction by multiplying the original Euclidean distance by the feature contribution balance factor; obtain the adaptive distance metric between the target field unit and the target cluster center by multiplying the distance value based on the feature dimension distance component correction by the spatial membership smoothing factor.

[0033] Compared with the prior art, the present invention has the following advantages:

[0034] This invention presents a path-coordinated scheduling optimization method for corn straw return and fertilization operations. By introducing a feature contribution balancing factor and a spatial membership smoothing factor, it effectively suppresses the excessive dominance of single operational features such as straw density on the zoning results, while maintaining a reasonable weight distribution of agronomic background features in the zoning process. This not only accurately identifies high straw load areas with special management value but also avoids zoning result deviations caused by anomalies in single-dimensional features, ensuring that the generated management zoning is more consistent with the actual heterogeneous distribution of farmland and improving the scientific rigor and relevance of variable fertilization decisions. Furthermore, by introducing a neighborhood space smoothing mechanism in the membership calculation, this invention makes the management zoning boundaries more continuous and regular in geometric shape, avoiding the appearance of unreasonable enclaves and jagged boundaries on the zoning map. Combined with agricultural machinery operation scenarios, the optimized zoning results can be directly converted into path scheduling instructions, enabling agricultural machinery to adaptively adjust its operating speed and fertilization method when entering high straw load areas, achieving synchronous optimization of fertilization and straw return operations, thereby significantly improving the accuracy and efficiency of field operations. Attached Figure Description

[0035] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an undue limitation of the invention. In the drawings:

[0036] Figure 1 This is a flowchart illustrating a method for path coordination scheduling optimization of corn straw return to the field and fertilization operations, as described in an embodiment of the present invention. Detailed Implementation

[0037] The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0038] See Figure 1 This is a flowchart of a method for path collaborative scheduling optimization of corn straw returning to the field and fertilization operations provided in Embodiment 1 of the present invention. Figure 1 As shown, a path coordination scheduling optimization method for corn straw returning to the field and fertilization operations may include:

[0039] Step S1: Obtain standardized feature vectors for field units by standardizing the original data from multiple sources of farmland.

[0040] This step aims to establish a unified and comparable data foundation for subsequent management zoning analysis, ensuring that farmland characteristic data from different sources can be measured and calculated within the same scale space.

[0041] First, the target farmland area is divided into grids, with the number of plots determined based on the actual size of the farmland in the scenario. A corresponding spatial location is assigned to each plot. For each plot, the following multi-source raw data are acquired: historical yield data collected by a combine harvester equipped with a yield monitor, reflecting the crop's yield level during historical planting; soil apparent conductivity data collected by a soil conductivity scanning device, reflecting differences in soil texture and fertility; digital elevation model data acquired by airborne or spaceborne remote sensing equipment, reflecting the topographic relief and hydrological characteristics of the farmland area; and straw density data acquired by a drone equipped with a multispectral or high-resolution visible light camera combined with image processing technology, reflecting the spatial distribution of straw cover and accumulation after agricultural machinery operations.

[0042] After completing the data collection, the multi-source data corresponding to each field unit are used to construct a multi-dimensional feature vector, and a multi-dimensional original data matrix is ​​formed using all field units as samples. Since the various data sources differ, their physical dimensions and numerical ranges vary significantly. Directly using these data for clustering calculations would lead to unbalanced feature contributions; therefore, standardization is necessary. This invention employs the Z-score standardization method, which independently normalizes the mean and standard deviation of each feature dimension of the original data matrix: subtracting the arithmetic mean of that feature dimension from each value and dividing by the standard deviation of that feature dimension. After this processing, each field unit corresponds to a standardized feature vector, where the values ​​of each feature dimension are transformed into a dimensionless standardized result with a mean of zero and a variance of one.

[0043] This completes the process of obtaining standardized feature vectors for field units by standardizing multi-source raw data of farmland.

[0044] Step S2: Obtain the feature contribution balance factor by performing imbalance analysis on the distance components of the feature dimension.

[0045] This step aims to address the problem of single-feature distance contamination in existing technologies. The root of this problem lies in the fact that standard Euclidean distance, when quantifying the multidimensional feature similarity between a field unit and a cluster center, fails to identify and balance the contribution of each dimension to the total distance by summing the squares of differences across all feature dimensions with equal weights. In actual straw return operations, the data feature vectors of field units typically consist of two types of information: one type, composed of multiple dimensions such as yield and soil conductivity, reflects the inherent agronomic attributes of the field; the other type, introduced by factors such as overlapping harvester paths, presents a narrow band with a high amplitude on the single dimension of straw density. When a field unit located in an abnormal straw accumulation zone is used to calculate the distance to a cluster center located in a normal area, the huge difference in straw density alone will overwhelmingly account for a significant portion of the sum of Euclidean distances. At this point, the distance calculation result is no longer a true reflection of overall similarity but is contaminated by this single operational feature. Therefore, the logic of this step is that a truly reliable similarity measure must have a balanced composition of its final values. To this end, this invention constructs a feature contribution balancing factor. This factor analyzes the inherent quantitative relationships between the various single-dimensional distance components that constitute a specific distance calculation. It can quantify in real time the extent to which the result of that distance calculation is dominated by a single or a few feature dimensions. When a severe imbalance in contribution is identified, i.e., distance contamination, this factor generates a correction value significantly greater than 1 and applies it to the original distance calculation result. This is equivalent to adding a feature dominance analysis to each distance calculation. By penalizing and amplifying distance values ​​dominated by a single operational feature, the difference between these special areas (such as tall straw belts) and all other conventional agronomic areas is significantly increased in the feature space. This enhanced difference allows the FCM clustering algorithm to identify and separate these areas with significant characteristics in operational load, forming clusters with specific management significance, rather than incorrectly confusing them with areas with similar agronomic backgrounds.

[0046] In summary, the following steps are taken: First, cluster centers are obtained during the clustering process. Then, energy distribution analysis is performed on the difference data of feature dimensions between field units and cluster centers to obtain a distance contribution imbalance index. Specifically, for any target field unit among all field units and any target cluster center among all cluster centers, the square of the difference between the target field unit and the target cluster center on any target feature dimension is taken as the distance energy component of the target feature dimension. The distance energy components of all feature dimensions are summed to obtain the total energy distribution assessment. The fourth power of the difference between the target field unit and the target cluster center on all target feature dimensions is summed to obtain the energy concentration assessment. The result of multiplying the number of feature dimensions by the energy concentration assessment is taken as the numerator. The square of the total energy distribution assessment is taken as the denominator. The result of subtracting the constant 1 from the obtained fraction is taken as the distance contribution imbalance index.

[0047] After obtaining the distance contribution imbalance index, the nonlinear response processing is performed on the distance contribution imbalance index to obtain the feature contribution balance factor. Specifically, a gain coefficient for the imbalance response is set. In this embodiment of the invention, the gain coefficient for the imbalance response is set to 1.5. This coefficient is used to adjust the sensitivity of the arctangent function to the imbalance measurement. The larger the value of the gain coefficient, the more sensitive the feature contribution balance factor is to the imbalance of distance contribution. This gain coefficient can be adjusted according to the actual scenario and is not required. The distance contribution imbalance index is amplified by the gain coefficient to obtain the amplified imbalance response value. The amplified imbalance response value is then processed by arctangent nonlinear mapping and normalized by the ratio of constant 2 to pi to obtain a smooth correction increment between zero and constant 1. The result of adding the smooth correction increment to constant 1 is used as the feature contribution balance factor.

[0048] In one implementation, assume the first The first field unit in the The eigenvalues ​​in each feature dimension are: ;No. The cluster centers at the in The feature values ​​in each dimension are The number of feature dimensions is The gain coefficient of the imbalance response is Then the first The first plot unit and the first The expression for calculating the feature contribution balancing factor among the cluster centers is:

[0049]

[0050] in, Indicates the first The first plot unit and the first A factor for balancing the feature contributions among cluster centers; Represents pi; Represents the arctangent function; The gain coefficient representing the unbalanced response; Indicates the number of feature dimensions; Indicates the first The first field unit in the Feature values ​​in each feature dimension; Indicates the first The cluster centers at the in Feature values ​​in each dimension.

[0051] It should be noted that this invention constructs a solution to the problem of single-feature distance contamination. This section describes the construction of a distance contribution imbalance index. The aim is to establish a quantitative index that reflects the distribution of distance components contributed by each feature dimension within the overall distance composition. In the calculation of multidimensional distance, the square of each single-dimensional distance can be... This is considered as the energy contributed by that dimension, and This represents the total energy. This structure determines whether there is excessive concentration in the energy distribution by comparing the fourth power of the energy in each dimension with the square of the total energy. Its working principle is that the numerical value of this core structure directly reflects the degree of concentration of energy distribution in a single dimension. In a typical distance calculation, such as calculating the distance between two field units both located in the central region of a field and with similar agronomical attributes, although each single-dimensional distance... The values ​​of may differ, but their magnitudes are comparable after data standardization, and the energy distribution is relatively dispersed. In this case, the calculation result of the core structure will be a small positive value close to zero. This represents a distance composition state with relatively balanced contributions. At this time, after mapping the distance contribution imbalance index through the arctangent function, the value of the feature contribution balance factor will be close to 1, applying only a negligible correction to the original distance, thus ensuring the stability of the similarity measurement in most normal cases. However, when single-feature distance contamination occurs, such as when calculating the distance between a field unit located on a straw accumulation zone formed by overlapping harvester operations and a cluster center located in a normal area, the energy of only the straw density dimension becomes problematic. This will exceed all other agronomic feature dimensions by orders of magnitude, resulting in an extremely concentrated peak in the energy distribution. Since fourth-power operations amplify large values ​​far more strongly than quadratic operations, this will lead to... The value is dominated by this peak, resulting in a significantly positive number greater than zero for the core construction. The magnitude of this positive value precisely quantifies the degree of energy distribution imbalance in this distance calculation. Finally, by using the arctangent function to perform a nonlinear mapping on the distance contribution imbalance index, for distance calculations with a relatively balanced index, the distance contribution imbalance index of the core construction is close to zero. The correction effect after the arctangent function mapping is weak, ensuring that the feature contribution balance factor is close to... This protects the stability of conventional distance calculations. When a significant concentration of energy distribution indicates contamination of the distance calculation, the value of the distance contribution imbalance index increases, and the characteristic contribution balance factor increases smoothly accordingly, applying a correction to the distance corresponding to the degree of imbalance. When the distance contribution imbalance index reaches an extremely high level, the saturation characteristic of the arctangent function will cause the growth of the characteristic contribution balance factor to level off and saturate at a preset upper limit (in this embodiment, ). The bounded penalty mechanism ensures that sufficient but not excessive corrections are applied to the contaminated distance, fundamentally avoiding algorithm instability issues that may result from excessive penalties. This allows the feature contribution balancing factor to adaptively amplify and correct the contaminated distance, effectively suppressing distortion introduced by single operation traces during the FCM clustering iteration process, and guiding the clustering results back to the correct partitioning that is jointly dominated by multi-dimensional features and better reflects the intrinsic differences of the fields.

[0052] This completes the process of obtaining the feature contribution balance factor by performing imbalance analysis on the distance components of the feature dimension.

[0053] Step S3: Obtain the spatial membership smoothing factor by evaluating the activity of the neighborhood membership distribution of field units.

[0054] After completing step S2, by introducing a feature contribution balancing factor, this invention has been able to identify and separate areas with significant operational characteristics (high straw load) into independent management zones. However, the inherent defects of the FCM algorithm itself still exist. This defect manifests in the fact that when the algorithm calculates the membership degree of each field unit to each cluster center, its decision-making process occurs entirely within the multidimensional feature space, without analyzing the location of the field unit in the real geographic space and its neighborhood relationships. This spatial blindness leads to an isolated field unit that is similar in agronomical characteristics to a distant cluster center (e.g., a small saline-alkali patch surrounded by a large area of ​​ordinary soil) being assigned a membership degree drastically different from its geographic neighborhood. When such isolated points or elongated strips formed by them appear on the zoning map, they create geometrically discontinuous, irregular enclaves or jagged boundaries. These fragmented boundaries not only fail to meet the actual needs of large-scale farmland management but also introduce unnecessary complex turns and inefficient movements for subsequent automated agricultural machinery path planning based on the zoning map, reducing overall operational efficiency. Therefore, the final affiliation of a field unit should not be determined solely by its own characteristics, but also by its surrounding geographical environment. A reasonable management zone should exhibit a smooth and continuous spatial transition in the membership degrees of its members. To this end, this invention further constructs a spatial membership smoothing factor, which quantifies the degree of harmony between the membership state of a field unit and the macro-level membership trend of its neighborhood to perform a secondary correction on distance calculations. When the membership state of a field unit is inconsistent with its neighborhood environment, indicating a potential irregular boundary shape, this factor generates a correction value to increase the cost of such isolated allocation, thereby guiding the membership degree distribution towards a smoother spatial and more geometrically regular form during clustering iterations.

[0055] In summary, firstly, the mean value of neighborhood membership is obtained by statistically processing the neighborhood membership data of field units. Specifically, the neighborhood range of field units is defined, and for any target field unit among all field units, the neighborhood membership set of the target field unit is obtained through the neighborhood range of the field unit. During the clustering process, the membership values ​​of all field units in the neighborhood membership set relative to the same target cluster center are accumulated, and the corresponding accumulated value is divided by the number of field units in the neighborhood to obtain the mean value of neighborhood membership of the target field unit at the target cluster center.

[0056] After obtaining the mean neighborhood membership data of the target field unit at the target cluster center, the variance analysis of the neighborhood membership data of the field unit is further performed to obtain the neighborhood membership activity index. Specifically, within the neighborhood of the target field unit, the membership values ​​of each field unit at all cluster centers are obtained; the membership values ​​of each field unit in the neighborhood at all cluster centers are taken as the overall sample, and the variance value of the sample is calculated. The variance value is used as the neighborhood membership activity index.

[0057] Finally, by adaptively processing the deviation data between the membership degree of the field unit and the mean of the neighborhood, a spatial membership degree smoothing factor is obtained. Specifically, the membership degree value of the target field unit at the target cluster center is obtained, and the difference between it and the mean membership degree data of the neighborhood is calculated. The difference is squared to obtain the deviation energy of the membership degree of the target field unit. The result of adding the deviation energy to the neighborhood membership degree activity index is used as the neighborhood-corrected membership degree fluctuation value. The spatial membership degree smoothing factor is obtained by calculating the ratio between the neighborhood-corrected membership degree fluctuation value and the neighborhood membership degree activity index.

[0058] In one implementation, assume the first The neighborhood membership activity index of each field unit is ;No. The first field unit for the first The membership degree of each cluster center is ;No. The first field unit in the The mean membership data of the neighborhoods at the cluster centers are ;No. The neighborhood set of each field unit is Then the first The first field unit for the first The expression for calculating the spatial membership smoothing factor of each cluster center is:

[0059]

[0060] in, Indicates the first The first field unit for the first Spatial membership smoothing factor of cluster centers; Indicates the first The activity index of neighborhood membership of each field unit; Indicates the first The first field unit for the first The membership degree of each cluster center; Indicates the first The first field unit in the Mean membership data of neighborhoods at cluster centers.

[0061] It should be noted that the above formula does not penalize points whose membership degree differs from the mean of their neighborhood, but rather introduces the activity of the neighborhood membership degree. As an adaptive adjustment benchmark, it can intelligently distinguish membership degree fluctuations. Its working mechanism lies in the denominator of the formula... It represents a point The basic membership level of the neighborhood is chaotic, while the molecule This means that the field unit has been taken into consideration. After determining its own membership status, the overall membership disorder level of the neighborhood is determined. Therefore, the smoothing factor of the entire spatial membership quantifies the field unit. The relative increment contributed by the degree of disorder in the membership of its neighborhood. Within a cluster with highly consistent membership, the membership activity of its neighborhood... It will be very small. In such a stable environment, if a... Its neighborhood mean For significantly different field units (i.e., an enclave), the deviation term This results in the numerator being much larger than the denominator. In this case, the spatial membership smoothing factor will have a value significantly greater than 1, effectively correcting this allocation behavior that disrupts regional stability and thus suppressing the formation of isolated points. Conversely, in a fuzzy boundary region that is itself located at the intersection of multiple clusters, the membership degrees of points within its neighborhood to different clusters already vary considerably, leading to a high degree of membership activity. The denominator itself is quite large. In this environment, it's normal for the membership degree of a field unit to deviate somewhat from the mean of its neighborhood. A large denominator. This will cause the deviation term The impact on the final ratio is correspondingly weakened, and the value of the spatial membership smoothing factor remains close to 1. This makes the algorithm tolerant of such normal boundary transitions, avoiding rigidity of the administrative partition boundaries caused by over-smoothing.

[0062] In summary, the spatial membership smoothing factor achieves intelligent smoothing of the spatial distribution of membership degrees through this adaptive adjustment mechanism. It can accurately identify and correct membership assignments that may lead to irregular partition shapes, while protecting the fuzzy boundaries of normal gradations. Ultimately, it guides the algorithm to generate optimized partitioning results with more regular and continuous geometric shapes, which are more in line with the actual needs of farmland management and agricultural machinery operations.

[0063] Thus, the spatial membership smoothing factor was obtained by evaluating the activity of the neighborhood membership distribution of field units.

[0064] Step S4: Obtain an adaptive distance metric by combining the joint feature contribution balancing factor and the spatial membership smoothing factor.

[0065] After obtaining the feature contribution balancing factor and the spatial membership smoothing factor, the adaptive distance metric is obtained by combining the feature contribution balancing factor and the spatial membership smoothing factor. Specifically, the original Euclidean distance between the target field unit and the target cluster center is obtained; the distance value based on the feature dimension distance component is obtained by multiplying the original Euclidean distance by the feature contribution balancing factor; and the adaptive distance metric between the target field unit and the target cluster center is obtained by multiplying the distance value based on the feature dimension distance component by the spatial membership smoothing factor.

[0066] After obtaining the adaptive distance metric between the target field unit and the target cluster center, the target number of clusters in the FCM clustering process is set. In this embodiment of the invention, the target number of clusters is set to 8. The target number of clusters can be adjusted according to the fuzzy partitioning coefficient and partitioning entropy in the actual scenario. No requirement is made. The adaptive distance metric is used in the FCM clustering process to complete the clustering iteration and obtain the corresponding FCM clustering results.

[0067] Step S5: Obtain the agricultural machinery path collaborative scheduling instruction set by hardening the final membership matrix.

[0068] This step aims to transform the FCM clustering results obtained in step S4 into path-coordinated scheduling instructions that can guide agricultural machinery operations. First, based on the membership matrix output after the FCM clustering iteration convergence, the cluster membership of each field unit is hardened, that is, the field unit is assigned to the target management partition with the highest membership degree. After hardening, a geometrically continuous and well-defined farmland management partition map is obtained. This partition map not only reflects the differences in agronomic attributes such as yield, soil conductivity, and topography, but also identifies high-load areas formed by straw accumulation, providing spatial guidance for subsequent differentiated operations.

[0069] Subsequently, differentiated operational strategies were formulated based on the characteristic attributes of each management zone. For identified high straw load areas, a composite operational prescription was generated, incorporating additional fertilization measures and operational speed constraints. For example, an appropriate amount of quick-acting nitrogen fertilizer was added during fertilization, and a recommended low-speed operating value for the agricultural machinery in this area was set to ensure smooth operation in fields with thick straw cover. For conventional management areas, corresponding conventional fertilization prescriptions were generated based on their agronomic characteristics, maintaining normal operational speed and fertilization intensity. It should be noted that in the embodiments of this invention, the operating speed of the agricultural machinery in the high straw load area was set to 3-4 kilometers per hour, significantly lower than the 6-7 kilometers per hour operating speed in the conventional management area. This speed setting can maintain operational continuity while avoiding machinery vibration or blockage caused by straw accumulation, thereby ensuring the smooth execution of fertilization and straw return operations.

[0070] Finally, the aforementioned zoning maps and differentiated operation prescriptions are compiled into a path coordination scheduling instruction set that can be recognized by the agricultural machinery automatic control system. As the agricultural machinery travels along the preset operation path, its positioning system and control unit can determine in real time the management zone to which the current operation location belongs and invoke the corresponding operation prescription according to the instruction set. When the agricultural machinery enters a high straw load area, it automatically triggers fertilizer application adjustment and speed control, realizing the linkage optimization of straw returning and fertilization operations, thereby improving the accuracy and efficiency of field operations and reducing resource waste caused by unreasonable paths or mismatched operation strategies.

[0071] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for path coordination and scheduling optimization of corn straw returning to the field and fertilization operations, characterized in that, The method includes: Step S1: Obtain standardized feature vectors for field units by standardizing the original data from multiple sources of farmland. Step S2: Obtain the feature contribution balance factor by performing imbalance analysis on the distance components of the feature dimension; Step S3: Obtain the spatial membership smoothing factor by evaluating the activity of the neighborhood membership distribution of field units; Step S4: Obtain an adaptive distance metric by combining the joint feature contribution balancing factor and the spatial membership smoothing factor; Step S5: Obtain the agricultural machinery path collaborative scheduling instruction set by hardening the final membership matrix; The step of obtaining the feature contribution balance factor by performing imbalance analysis on the distance components of the feature dimension includes: obtaining cluster centers during the clustering process; obtaining a distance contribution imbalance index by performing energy distribution analysis on the feature dimension difference data between field units and cluster centers; and obtaining the feature contribution balance factor by performing nonlinear response processing on the distance contribution imbalance index. The method of obtaining an adaptive distance metric by combining a joint feature contribution balancing factor and a spatial membership smoothing factor includes: obtaining the original Euclidean distance between the target field unit and the target cluster center; obtaining a distance value corrected based on the feature dimension distance component by multiplying the original Euclidean distance by the feature contribution balancing factor; and obtaining an adaptive distance metric between the target field unit and the target cluster center by multiplying the distance value corrected based on the feature dimension distance component by the spatial membership smoothing factor.

2. The method for path coordination scheduling optimization of corn straw returning to the field and fertilization operations according to claim 1, characterized in that, The process of obtaining standardized feature vectors for field units by standardizing multi-source raw data of farmland includes: The target farmland area is sampled in a grid pattern to obtain all field units. Historical yield data of the field units is obtained using a combine harvester equipped with a yield monitor. Soil apparent electrical conductivity data of the field units is obtained using a soil electrical conductivity scanning device. Digital elevation model data of the field units is obtained using airborne or spaceborne remote sensing equipment. Straw density data of the field units is obtained using a UAV equipped with a multispectral or high-resolution visible light camera and combined with image analysis technology. The historical yield data, soil apparent electrical conductivity data, digital elevation model data, and straw density data are used to construct a multidimensional raw data matrix. Z-score standardization is performed on each feature dimension of the matrix to obtain the standardized feature vector corresponding to each field unit.

3. The method for path coordination scheduling optimization of corn straw returning to the field and fertilization operations according to claim 1, characterized in that, The method involves performing energy distribution analysis on the feature dimension differences between field units and cluster centers to obtain a distance contribution imbalance index, including: For any target field unit among all field units and any target cluster center among all cluster centers, the square of the difference between the target field unit and the target cluster center in any target feature dimension is taken as the distance energy component of the target feature dimension; the distance energy components of all feature dimensions are accumulated to obtain the total energy distribution assessment; the fourth power of the difference between the target field unit and the target cluster center in all target feature dimensions is accumulated to obtain the energy concentration assessment. The result of multiplying the number of feature dimensions by the energy concentration assessment is used as the numerator; the square of the total energy distribution assessment is used as the denominator, and the result of subtracting the constant 1 from the corresponding fraction is used as the distance contribution imbalance index.

4. The method for path coordination scheduling optimization of corn straw returning to the field and fertilization operations according to claim 1, characterized in that, The step of obtaining the feature contribution balance factor by performing nonlinear response processing on the distance contribution imbalance index includes: Set the gain coefficient of the imbalance response; amplify the distance contribution imbalance index by applying the gain coefficient to obtain the amplified imbalance response value; perform arctangent nonlinear mapping on the amplified imbalance response value and normalize it by the ratio of constant 2 to pi to obtain the smooth correction increment between zero and constant 1; use the calculation result of adding the smooth correction increment to constant 1 as the characteristic contribution balance factor.

5. The method for path coordination scheduling optimization of corn straw returning to the field and fertilization operations according to claim 1, characterized in that, The method of obtaining a spatial membership smoothing factor by evaluating the activity of the neighborhood membership distribution of field units includes: By statistically processing the neighborhood membership data of field units, the mean neighborhood membership data is obtained. By performing variance analysis on the neighborhood membership data of field units, the neighborhood membership activity index was obtained. By adaptively processing the deviation data between the membership degree of field units and the mean of their neighborhood, a spatial membership degree smoothing factor is obtained.

6. The method for path coordination scheduling optimization of corn straw returning to the field and fertilization operations according to claim 5, characterized in that, The step of statistically processing the neighborhood membership data of field units to obtain the mean neighborhood membership data includes: Define the neighborhood range of the field unit. For any target field unit among all field units, obtain the set of neighborhood membership degrees of the target field unit through the neighborhood range of the field unit. During the clustering process, the membership values ​​of all field units in the neighborhood membership set relative to the same target cluster center are accumulated. The accumulated value is then divided by the number of field units in the neighborhood to obtain the mean neighborhood membership data of the target field unit at the target cluster center.

7. The method for path coordination scheduling optimization of corn straw returning to the field and fertilization operations according to claim 5, characterized in that, The process involves performing variance analysis on the neighborhood membership data of field units to obtain a neighborhood membership activity index, including: Within the neighborhood centered on the target field unit, the membership degree values ​​of each field unit on all cluster centers are obtained; the membership degree values ​​of each field unit in the neighborhood on all cluster centers are used as the overall sample, and the variance value of the sample is calculated. The variance value is used as the neighborhood membership activity index.

8. The method for path coordination scheduling optimization of corn straw returning to the field and fertilization operations according to claim 5, characterized in that, The step of obtaining a spatial membership smoothing factor by adaptively processing the deviation data between the membership degree of field units and the mean of their neighborhood includes: Obtain the membership degree value of the target field unit at the target cluster center, and calculate the difference between it and the mean membership degree data of the neighborhood; square the difference to obtain the deviation energy of the membership degree of the target field unit; add the deviation energy to the neighborhood membership degree activity index to obtain the neighborhood-corrected membership degree fluctuation value; obtain the spatial membership degree smoothing factor by calculating the ratio of the neighborhood-corrected membership degree fluctuation value to the neighborhood membership degree activity index.

Images