Fertilizer applicator in-field path planning method for soil mechanical compaction and reduction

Abstract

The invention discloses a fertilizer applicator in-field path planning method for soil mechanical compaction and reduction, and relates to the field of agricultural machinery dispatching. The invention aims to solve the problem that a fertilizing machine can roll soil in a field operation link. The method comprises the following steps: constructing a path code set containing turning node coordinates, turning direction angles and fertilizer applicator operation speeds in a fertilizer applicator operation path; and carrying out genetic operation on the path coding set by utilizing a genetic algorithm to obtain an optimal path. On the premise of ensuring the working efficiency, the soil compaction area is reduced, the shortest path planning of the in-field operation of the fertilizer applicator is deployed, and the high-efficiency and low-damage precise operation is realized.

Description

Technical Field

[0001] This invention belongs to the field of agricultural machinery scheduling. Background Technology

[0002] The sustainable development of modern agriculture faces numerous challenges, among which soil compaction has become an increasingly serious global problem. With the continuous improvement of agricultural mechanization, the frequent operation of large and heavy agricultural machinery in farmland leads to the destruction of soil structure, reduced porosity, and increased bulk density. This compaction not only directly damages the physical, chemical, and biological properties of the soil, significantly reducing soil aeration, water conductivity, and nutrient availability, but also hinders crop root growth, ultimately leading to reduced crop yields. Studies have shown that the compaction effect caused by agricultural machinery operations can have long-term or even irreversible impacts on farmland productivity, with high costs and time-consuming remediation processes.

[0003] Among various field operations, fertilization, due to its time-sensitive nature, relatively high frequency (especially topdressing), and the need to carry heavy fertilizer loads, has become a key factor in inducing and exacerbating soil compaction. Traditional fertilizer applicator path planning often prioritizes operational efficiency (such as shortest path, fewest turns) and operational quality (such as fertilizer uniformity and coverage), while paying insufficient attention to the cumulative effect of soil compaction caused by repeated mechanical rolling. Although the conventional zigzag full-coverage path can ensure uniform fertilization, it leads to the machine rolling over almost all areas within the field, especially forming a high-intensity compaction zone in the center line area, exacerbating the compaction problem. Summary of the Invention

[0004] This invention aims to solve the problem of soil compaction caused by fertilizer applicators during field operations. It provides a method for planning the field path of fertilizer applicators to minimize soil compaction.

[0005] The first aspect of this application provides a method for field path planning of fertilizer applicators for reducing soil mechanical compaction, including:

[0006] Construct a path code set that includes the coordinates of turning nodes, turning angles, and operating speed of the fertilizer applicator in its operating path;

[0007] The optimal path is obtained by performing genetic operations on the path encoding set using a genetic algorithm.

[0008] In one possible design, the fitness function of the genetic algorithm includes:

[0009] The fertilizer applicator's operating speed is adaptively adjusted based on the field selection weights. The expression for the adaptive function of the fertilizer applicator's operating speed is:

[0010] ,

[0011] in, Choose weights for the fields. The first in the fertilizer applicator's operating path The operating speed of the fertilizer applicator at each turning point. This is the maximum design speed of the fertilizer applicator;

[0012] Penalty for turning At the minimum, the first Turning angle at each turning point The optimal adaptive function expression for the turning direction angle is:

[0013] ,

[0014] in, represent Total number of turns during the time period Additional penalty for turning. Indicates the current turning radius. Minimum turning constraint radius, For turning variables, Taking 1 indicates a turn. Setting it to 0 indicates going straight.

[0015] In one possible design, the field selection weights The expression is:

[0016] ,

[0017] in, For the first Compaction sensitivity of each turning node To achieve maximum compaction sensitivity across the entire path, For fields The probability of task priority.

[0018] In one possible design, the method for calculating the probability of operation priority for a field includes:

[0019] Calculate the unevenness of each field plot;

[0020] The probability of selecting each plot is calculated using the following formula based on the unevenness of the plot:

[0021] ,

[0022] in, For fields The probability of choosing, and Fields and The unevenness, The total number of fields;

[0023] The operation priority probability of the field is calculated using the selection probability of each field according to the following formula:

[0024] ,

[0025] in, Indicates field The probability of job priority. This represents the maximum probability of selecting a field.

[0026] In one possible design, the method for calculating the minimum turning constraint radius includes:

[0027] The area, centroid coordinates, compactness, and concavity / convexity of each field are calculated and normalized to obtain the feature vector.

[0028] The area feature vector of each field is updated based on the aforementioned feature vector;

[0029] The central convergence is calculated using the area feature vectors of each field before and after the update;

[0030] Calculate the operation speed based on the central convergence.

[0031] The minimum turning constraint radius is calculated using the operating speed.

[0032] In one possible design, the calculation of the area, centroid coordinates, compactness, and concavity / convexity of each plot, followed by normalization, yields a feature vector, including:

[0033] The feature vector includes the normalized total feature vector of the fields and the area feature vector of each field:

[0034] ,

[0035] ,

[0036] in, For normalized fields The total eigenvector, For normalized fields The area eigenvector, This indicates normalization processing. For fields The centroid coordinates before normalization For fields Compactness before normalization For fields Concavity and convexity before normalization For fields Area before normalization;

[0037] The step of updating the area feature vector of each field based on the feature vector includes:

[0038] ,

[0039] in, Indicates the updated field The area characteristic value, Represents the set of all fields. for eigenvalues;

[0040] The calculation of central convergence using the area feature vectors of each field before and after the update includes:

[0041] ,

[0042] in, For the central convergence, Describes the Euclidean norm. Indicates the updated field The area eigenvector.

[0043] In one possible design, calculating the job speed based on the central convergence includes:

[0044] ,

[0045] in, For fields The speed of operation The maximum design speed of the machine, For soil compaction sensitivity, This refers to the tire's contact patch. The coefficient of friction between the tire and the soil. For safety reasons, This refers to the load capacity of the fertilizer applicator.

[0046] In one possible design, calculating the minimum turning constraint radius using the operating speed includes:

[0047] ,

[0048] in, For fields Minimum turning constraint radius, It is the acceleration due to gravity. The lateral adhesion coefficient, This refers to the wheelbase of the fertilizer applicator. This is for a safety margin.

[0049] In one possible design, the constraint conditions for the minimum turning radius include:

[0050] ,

[0051] in, For the planned turning radius, This is the minimum turning radius of the machine.

[0052] The second aspect of this application provides a field path planning device for fertilizer applicators aimed at reducing soil mechanical compaction. The field path planning device for fertilizer applicators aimed at reducing soil mechanical compaction includes a processor and a memory. The memory stores at least one instruction, which is loaded and executed by the processor to implement the field path planning method for fertilizer applicators aimed at reducing soil mechanical compaction as described above.

[0053] A third aspect of this application provides a computer storage medium, characterized in that the computer storage medium stores at least one instruction, which is loaded and executed by a processor to implement the above-described method for field path planning of fertilizer applicators for reducing soil mechanical compaction.

[0054] The beneficial effects of this application are:

[0055] By using RTK-GPS differential positioning technology to obtain field boundary information for fertilizer applicators, the area of ​​soil compaction can be reduced while ensuring operational efficiency. The shortest path planning for fertilizer applicator operations within the field can be deployed to achieve high-efficiency, low-damage, and precise operations. Attached Figure Description

[0056] Figure 1 A schematic diagram of a field path planning method for fertilizer applicators aimed at reducing soil compaction;

[0057] Figure 2 This is a flowchart of the genetic algorithm. Detailed Implementation

[0058] 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 them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. It should be noted that, unless otherwise specified, the embodiments and features in the embodiments of the present invention can be combined with each other.

[0059] Specific implementation method one: Refer to Figure 1 and Figure 2This embodiment specifically describes the field path planning method for fertilizer applicators aimed at reducing soil mechanical compaction, which includes:

[0060] I. Use GPS differential positioning technology to collect field information and obtain the set of field boundary points. :

[0061] ,

[0062] in, Indicates reaching the boundary point The point in time, Boundary points in the field coordinates This represents the total number of boundary points in the field.

[0063] II. Field Clustering

[0064] (1) Calculate the area of ​​the field :

[0065] .

[0066] Calculate the centroid coordinates :

[0067] ,

[0068] .

[0069] Compactness The expression for dividing the fields into regular shapes is as follows:

[0070] ,

[0071] in, Perimeter of the field;

[0072] Convexity The expression representing the curvature of the field boundary is as follows:

[0073] ,

[0074] in, The area of ​​a field is calculated for the region of its convex boundary points.

[0075] (2) Normalize the centroid coordinates, area, compactness, and concavity of the fields to eliminate the influence of numerical range differences on the clustering algorithm.

[0076] Feature vector:

[0077] ,

[0078] in, For normalized fields The total eigenvector, For normalized fields The area eigenvector, This indicates normalization processing. For fields The centroid coordinates before normalization For fields Compactness before normalization For fields Concavity and convexity before normalization For fields Area before normalization.

[0079] The feature normalization expression is:

[0080] ,

[0081] in, Represents the normalized eigenvalues, used to eliminate dimensional differences; and These represent the maximum and minimum values ​​of the characteristic in the field, respectively; for eigenvalues.

[0082] (3) Center selection. This stage shows that the farther away the field is, the higher the probability of it being selected as the center. This ensures that new centers will appear in areas far away from existing centers, thus making the center points scattered and random.

[0083] Selection probability:

[0084] ,

[0085] in, For fields The probability of selection; the higher the probability, the more likely it is to be selected. and Fields and The unevenness, This represents the total number of fields.

[0086] (4) Update of field area feature vector:

[0087] ,

[0088] ,

[0089] in, Indicates the updated field The area characteristic value; Represents the set of all fields; Indicates field The original area before normalization; For fields Normalized eigenvalues; For fields The area eigenvector.

[0090] (5) Calculate the central convergence, i.e., the maximum difference between the area vectors of the fields before and after the update, when the central convergence... Less than the convergence threshold This indicates that the field clustering effect is good. The convergence criterion expression is:

[0091] ,

[0092] in, For the Euclidean norm, This represents the updated area feature vector.

[0093] 3. Fertilizer applicator operating parameters and path status matching

[0094] (1) Job priority generation:

[0095] After clustering, the tasks are ranked for each field. Probability selection is then performed. High-yielding fields, due to their high representativeness, should be prioritized for operation to reduce the idle running of cluster fertilizer applicators and the complexity of scheduling.

[0096] ,

[0097] in, Indicates field The probability of job priority, whose value represents the order of job priority from high to low within the same field; This represents the maximum probability of selecting a field.

[0098] (2) Dynamic adaptation of operation speed:

[0099] ,

[0100] in, For fields The speed of operation The maximum design speed of the machine, For soil compaction sensitivity, This refers to the tire's contact patch. The coefficient of friction between the tire and the soil. For safety reasons, This refers to the load capacity of the fertilizer applicator; Central convergence is a parameter used to reflect cluster stability. Clusters that approach zero are more stable. Increasing the cluster size makes it more unstable; stable clusters have higher field similarity and their velocities tend to be closer to a certain value. If clustering is unstable, reduce the operation speed and operate with conservative and safe parameters to achieve high efficiency.

[0101] (3) Modeling of turning radius constraints:

[0102] ,

[0103] in, For fields The minimum turning constraint radius; It is the acceleration due to gravity; The lateral adhesion coefficient; This refers to the wheelbase of the fertilizer applicator; This is for a safety margin.

[0104] Constraints:

[0105] ,

[0106] in, For the planned turning radius, This is the minimum turning radius of the machine.

[0107] Current turning radius of fertilizer applicator Downward turning angle Less than the maximum turning angle , .

[0108] 4. Genetic algorithm for decompression of real path planning.

[0109] (1) Genetic algorithm chromosome encoding based on fertilizer applicator operation parameters.

[0110] A work path passes through multiple fields. When the fertilizer applicator turns, it is turning to another field, and turning is also a choice of the next field to work on.

[0111] Design initial path coding: , ,

[0112] in, , This represents the total number of turning points in the fertilizer applicator's operating path. The first in the fertilizer applicator's operating path The coordinates of each turning node; The first in the fertilizer applicator's operating path The turning angle at each turning node The first in the fertilizer applicator's operating path The operating speed of the fertilizer applicator at each turning point.

[0113] and The initial values ​​are respectively and .

[0114] (2) Design of the fitness function.

[0115] Select weights :

[0116] ,

[0117] in, For the first Compaction sensitivity of each turning node (between 0 and 1, the higher the value, the easier it is to compact). Maximum compaction sensitivity across the entire path; Indicates field The probability of task priority.

[0118] Velocity adaptive function:

[0119] .

[0120] The starting point of the path is fixed at the entrance of the field, which is the first... Assign appropriate path speeds to each turning node. .

[0121] Penalty for turning At its minimum, the number of... Turning angle at each turning point The optimal value is:

[0122] ,

[0123] in, Additional penalty for turning; Indicates the current turning radius. ; represent Total number of turns during the time period; For turning variables, ,when When the value is 1, it indicates that a turn is made upon reaching the end of the field; otherwise, continue straight along the current path.

[0124] (3) Genetic operations based on fitness functions.

[0125] Selection operation: Random selection For each individual, select the one with the best fitness in the current population.

[0126] Cross operation:

[0127] The order of creating parent generation 1 samples is: ABCDEF; the order of creating parent generation 2 samples is: abcdef.

[0128] Randomly select sub-paths; the order of child 1 is: ABcdEF, and the order of child 2 is: abCDef.

[0129] Detecting turning penalties during cross-operations To determine if the minimum value is achieved, we can verify the degree of sharp turns in the path and optimize the smoothness of turns by improving the intersection sequence.

[0130] Mutation operations, adaptive mutation reduces compaction sensitivity:

[0131] ,

[0132] in, For the first The probability of variation under low-sensitivity compaction at each turning node; Indicates compaction cost The probability of variation at the minimum value; Indicates compaction cost The probability of mutation at the maximum value; For the first The compaction sensitivity of each turning node.

[0133] Compacted cost :

[0134] ,

[0135] in, For the first The number of times each turning node is run over; It is a non-linear coefficient used to intensify the penalty of repeated compaction.

[0136] (4) Elite retention strategy: After the operation, individuals are ranked according to their fitness, and individuals with the former elite proportion are retained. , The remaining individuals are selected randomly.

[0137] (5) Termination conditions.

[0138] Maximum number of iterations:

[0139] The maximum number of iterations is reached, satisfying the termination condition.

[0140] Fitness convergence threshold:

[0141] ,

[0142] This represents the optimal fitness threshold read at time t; This represents the optimal fitness threshold before 50 iterations at time t.

[0143] (6) Output the optimal path.

[0144] Discrete representation of the optimal path: Each waypoint contains four-dimensional information. .

[0145] While ensuring the basic needs of fertilization operations, such as coverage integrity, uniformity, and operational efficiency, the genetic algorithm is used to optimize the trajectory of the fertilizer applicator in the field, minimizing unnecessary mechanical compaction. This path optimization strategy to reduce compaction is an important manifestation of the concepts of precision agriculture and sustainable soil management at the level of agricultural machinery operation scheduling.

[0146] V. Path accuracy verification.

[0147] Path deviation control helps reduce the probability of compaction of field furrows during the operation of fertilizer applicators.

[0148] Lateral error control :

[0149] ,

[0150] in, To accommodate the width of the fertilizer applicator, The minimum overlap rate.

[0151] Fertilizer machine operation compaction overlap rate :

[0152] ,

[0153] in, This indicates the width of the furrow where fertilizer application overlaps.

[0154] Closure error test:

[0155] ,

[0156] in, and These are the horizontal and vertical coordinates of the initial point of the work path, respectively. The lateral error of the current path point m and ; This is the tolerance factor; Perimeter of the field; Closure error represents the deviation of the straight-line distance between the starting point and the ending point of a path.

[0157] While the invention has been described herein with reference to specific embodiments, it should be understood that these embodiments are merely examples of the principles and applications of the invention. Therefore, it should be understood that many modifications can be made to the exemplary embodiments, and other arrangements can be designed without departing from the spirit and scope of the invention as defined by the appended claims. It should be understood that different dependent claims and features described herein can be combined in ways different from those described in the original claims. It is also understood that features described in conjunction with individual embodiments can be used in other described embodiments.

Claims

1. A method for in-field path planning of a fertilizer applicator oriented to the reduction of mechanical compaction of the soil, characterized by, The application relates to a method for constructing a minimum turning radius constraint path for a fertilizer distributor. The method comprises the following steps: A path code set comprising coordinates of turning nodes in a fertilizer distributor operation path, a turning direction angle and an operation speed of the fertilizer distributor is constructed; 2. The method for in-field path planning of a fertilizer applicator oriented to the reduction of mechanical compaction of soil according to claim 1, characterized in that, A genetic algorithm is used to perform genetic operation on the path code set to obtain an optimal path. The fitness function of the genetic algorithm comprises: ， wherein, selecting weights for the field, selecting a fertilizer applicator operating speed for the fertilizer applicator at a turn node in the fertilizer applicator operating path, selecting a fertilizer applicator operating speed for the fertilizer applicator at a turn node in the fertilizer applicator operating path, is a maximum design speed for the fertilizer applicator; when turning penalty minimum, the turn direction angle at the turning node optimal, the turn direction angle adaptive function expression is: ， wherein, represents the total number of turns in the time period, is a penalty for turns, denotes the current turn radius, is the minimum turn constraint radius, is a turn variable, takes the value 1 for a turn, takes the value 0 for a straight.

3. The method for in-field path planning of a soil-conservation-oriented applicator according to claim 2, characterized in that, the field selection weight The expression is: ， in, For the first Compaction sensitivity of each turning node To achieve maximum compaction sensitivity across the entire path, For fields The probability of task priority.

4. The method for in-field path planning of a soil-conservation-oriented applicator according to claim 3, characterized in that, The operation speed of the fertilizer distributor is adaptively adjusted according to a field selection weight, and the expression of an adaptive function of the operation speed of the fertilizer distributor is as follows: A method for calculating an operation priority probability of a field comprises the following steps: The concave-convex degree of each field is calculated; ， wherein, is the selection probability of the field plot, and are the concave-convex degrees of the field plots, is the total number of field plots;​​​ The selection probability of each field is calculated according to the following formula: ， wherein, represents the priority probability of the operation of the field block represents the priority probability of the operation of the field block represents the maximum probability of selecting the field block.

5. The soil-conservation-mechanic-compaction-reduction-oriented in-field path planning method for a fertilizer machine according to claim 2, characterized in that, The operation priority probability of each field is calculated according to the selection probability of each field according to the following formula: A method for calculating the minimum turning radius constraint comprises the following steps: The area, the centroid coordinates, the compactness and the concave-convex degree of each field are respectively calculated, and a feature vector is obtained after normalization processing; The area feature vector of each field is updated according to the feature vector; The center convergence degree is calculated by using the area feature vectors of each field before and after updating; The operation speed is calculated according to the center convergence degree; 6. The in-field path planning method for soil-conservation-oriented mechanical compaction-reducing applicator according to claim 5, characterized in that, The minimum turning radius constraint is calculated by using the operation speed. The method for respectively calculating the area, the centroid coordinates, the compactness and the concave-convex degree of each field and performing normalization processing to obtain the feature vector comprises the following steps: ， ， in, For normalized fields The total eigenvector, For normalized fields The area eigenvector, This indicates normalization processing. For fields The centroid coordinates before normalization For fields Compactness before normalization For fields Concavity and convexity before normalization For fields Area before normalization.

7. The in-field path planning method for soil-conservation-oriented mechanical compaction-reducing applicator according to claim 6, characterized in that, The feature vector comprises a normalized total feature vector of the field and an area feature vector of each field: ， wherein, denotes the area characteristic value of the field plot after the update, denotes the set of all field plots, is the characteristic value of . The method for updating the area feature vector of each field according to the feature vector comprises the following steps: wherein, is the central convergence, denotes the Euclidean norm, denotes the updated field area feature vector.

8. The method for in-field path planning of a soil-conservation-oriented machine for reducing compaction of soil according to claim 7, characterized in that, The method for calculating the center convergence degree by using the area feature vectors of each field before and after updating comprises the following steps: ， wherein, is the field size, is the working speed, is the maximum design speed of the implement, is the soil compaction sensitivity, is the tire ground contact area, is the tire-soil friction coefficient, is the safety factor, is the fertilizer on-board weight.

9. The method of claim 8, wherein the method further comprises: The method for calculating the operation speed according to the center convergence degree comprises the following steps: ， wherein, is the minimum turning radius of the field, is the minimum turning radius of the field, is the acceleration due to gravity, is the lateral adhesion coefficient, is the wheel base of the fertilizer machine, is the safety margin.

10. The in-field path planning method for soil-conservation-oriented mechanical compaction-reducing applicator according to claim 9, characterized in that, The method for calculating the minimum turning radius constraint by using the operation speed comprises the following steps: The constraint condition of the minimum turning radius constraint comprises the following steps: ， wherein, Rmin is the minimum turning radius of the vehicle, Rmin is the minimum turning radius of the vehicle.

Images