A control method of an intelligent sugarcane seeder

Abstract

The application discloses a control method of an intelligent sugarcane seeder, and relates to the technical field of agricultural planting. The method comprises the following steps: obtaining seeding parameters one and two by respectively performing initial presetting on the seeding depth and spacing of the same and different seeding areas of a target area divided according to soil conditions, and verifying the seeding parameter two; establishing a correlation prediction model according to historical seeding data; according to the distribution of the different seeding areas, counting a pre-processing difference variable between the seeding parameters one and two to predict a comprehensive influence amplitude; if the comprehensive influence amplitude presents an upward trend or a stable trend, it is judged that the preset of the seeding parameter two is reasonable; if the comprehensive influence amplitude presents a downward trend, a reasonable seeding parameter three is preset and verified according to the pre-processing difference variable; and the seeding parameters one and three are input into the sugarcane seeder for parameter control. The control method of the intelligent sugarcane seeder can improve the efficiency of seeding and the growth quality of sugarcane.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of agricultural planting, and particularly relates to a control method of an intelligent sugarcane seeder. BACKGROUND

[0002] With the progress of science and technology, the degree of agricultural mechanization is higher and higher. In the field of sugarcane planting, various agricultural machinery such as tractors, seeders and the like have been widely used. This provides basic equipment support for intelligent regulation and control of sugarcane seeding depth and spacing.

[0003] In the traditional sugarcane seeding process, farmers need to spend a lot of time and effort to adjust the seeding depth and spacing. They usually need to estimate by experience, and the operation process is relatively cumbersome and inefficient. The intelligent regulation and control system can automatically seed according to the set parameters, without manual adjustment one by one, greatly saving manpower and time, and the current regulation and control of the seeding depth and spacing of the intelligent seeder often plant the whole planting field with uniform depth and spacing, without adjusting the seeding parameters of different depths and different spacings according to different soil condition information of different blocks, so as to cause low planting efficiency. SUMMARY

[0004] In order to overcome the deficiencies of the prior art, the present application provides a control method of an intelligent sugarcane seeder.

[0005] The control method of the intelligent sugarcane seeder provided by the present application comprises the following steps:

[0006] Step S1, the same seeding area and the different seeding area divided according to the soil condition of the target area needing to be seeded with sugarcane are respectively subjected to initial presetting of the seeding depth and spacing to obtain seeding parameters one and seeding parameters two, and the seeding parameters two are verified to output parameter verification information;

[0007] Step S2, according to the parameter verification information, the distribution of the different seeding area is counted, and historical seeding data is obtained, and an association prediction model is established according to the historical seeding data;

[0008] Step S3, according to the distribution of the different seeding area, a preprocessed difference value between the seeding parameters one and the seeding parameters two is counted, and a comprehensive influence amplitude is predicted according to the preprocessed difference value and the association prediction model;

[0009] Step S4, if the comprehensive influence amplitude presents an upward trend or a stable trend, it is judged that the preset of the second sowing parameter is reasonable, if the comprehensive influence amplitude presents a downward trend, an adjusted sowing parameter is preset according to the preprocessed difference variable, and the adjusted sowing parameter is tested through the correlation prediction model to correspondingly screen out a third sowing parameter from the adjusted sowing parameter, the first sowing parameter and the third sowing parameter are input into the sugarcane sowing machine for parameter control, and a sowing control result is output.

[0010] Preferably, the target area requiring sugarcane sowing is detected to obtain soil characteristic information, and the target area is divided into a same sowing area and a different sowing area according to the soil characteristic information.

[0011] The first sowing parameter is obtained by initially presetting the sowing depth and spacing parameters of the same sowing area according to the soil conditions in the soil characteristic information, the second sowing parameter is obtained by initially presetting the sowing depth and spacing parameters of the different sowing area according to the soil conditions in the soil characteristic information, and the second sowing parameter is verified to output sowing parameter verification information.

[0012] Preferably, according to the sowing parameter verification information, if the distribution position of the different sowing area is the edge of the target area and adjacent to two sowing point positions in the same sowing area, distribution case one is output, if the distribution position of the different sowing area is the edge of the target area and adjacent to three sowing point positions in the same sowing area, distribution case two is output, and if the distribution position of the different sowing area is the middle of the target area and adjacent to four sowing point positions in the same sowing area, distribution case three is output.

[0013] Preferably, historical sowing data of sowing in a historical period is obtained, and a correlation factor to be tested is evaluated according to the historical sowing data, and a difference variable to be tested is counted between the sowing depth and spacing parameters of the same sowing area and the different sowing area under different soil conditions.

[0014] According to the correlation factor to be tested, a weight ratio one or a weight ratio two or a weight ratio three is counted according to the weight ratio of the correlation influence degree of each of the distribution case one or the distribution case two or the distribution case three in the historical period, and the weight ratio one or the weight ratio two or the weight ratio three is combined to form a weight ratio to be tested.

[0015] According to the historical sowing data, the correlation factor to be tested, the difference variable to be tested and the weight ratio to be tested, a correlation prediction model is established.

[0016] Preferably, a parameter difference variable value between the seeding parameter one of the same seeding area in the distribution case one or the distribution case two or the distribution case three and the seeding parameter two of the different seeding area in the distribution case one or the distribution case two or the distribution case three is obtained by statistics, and the preprocessed difference variable value is obtained;

[0017] According to the preprocessed difference variable value, a weight ratio is selected from the to-be-tested weight ratios to obtain a preprocessed weight ratio, and an association factor is selected from the to-be-tested association factors according to the soil characteristic information and the preprocessed difference variable value to obtain a preprocessed association factor;

[0018] The soil characteristic information, the preprocessed weight ratio, the preprocessed association factor and the preprocessed difference variable value are input into an association prediction model to obtain a comprehensive influence amplitude.

[0019] Preferably, if the comprehensive influence amplitude presents an upward trend or a stable trend, it is judged that the preset reasonable seeding parameter two is obtained, and the seeding parameter one and the seeding parameter two are input into a sugarcane seeder to perform parameter control, and a seeding control result is output.

[0020] Preferably, if the comprehensive influence amplitude presents a downward trend, an adjusted seeding parameter is preset according to historical seeding data, the to-be-tested difference variable value and the preprocessed difference variable value, and the adjusted seeding parameter contains at least one seeding parameter value.

[0021] A preset difference variable value is obtained by statistics of a parameter difference variable value between the seeding parameter one of the same seeding area in the distribution case one or the distribution case two or the distribution case three and the adjusted seeding parameter of the different seeding area in the distribution case one or the distribution case two or the distribution case three.

[0022] Preferably, the soil characteristic information, the preprocessed weight ratio, the preprocessed association factor and the preset difference variable value are input into an association prediction model to obtain a to-be-tested influence amplitude.

[0023] According to a maximum value existing in the to-be-tested influence amplitude, a most reasonable seeding parameter is selected from the adjusted seeding parameter to obtain a seeding parameter three, and the seeding parameter one and the seeding parameter three are input into a sugarcane seeder to perform parameter control, and a seeding control result is output.

[0024] Compared with the prior art, the present application has the following characteristics and beneficial effects:

[0025] The soil characteristic information of the target area needing to be planted with sugarcane is detected, so that the preset of the corresponding plantable sugarcane depth and spacing parameters is carried out according to the soil characteristic information of different areas, so as to avoid the setting of the uniform sugarcane planting depth and spacing parameters in the traditional technology, ignoring the difference of the sugarcane planting conditions caused by different soil characteristic information in different areas, and further affecting the final growth quality of the sugarcane. The depth analysis of the planting depth and spacing is carried out mainly for the different planting areas, so as to preset the most reasonable planting depth and spacing, and the influence of the same planting area on the different planting area is considered, such as promoting the growth quality of the sugarcane in the different planting area, and at the same time, the growth quality of the sugarcane in the same planting area is not affected. The influence degree of the different planting area is different due to the different distribution positions of the different planting area, so three distribution conditions are judged, one is that the different planting area is located at the edge of the target area and adjacent to two planting points in the same planting area, one is that the different planting area is located at the edge of the target area and adjacent to three planting points in the same planting area, and the other is that the different planting area is located in the middle of the target area and adjacent to four planting points in the same planting area. Subsequently, the statistical information of the above three distribution conditions is distinguished, that is, the comprehensive influence amplitude of the above distribution conditions is predicted, so as to further judge whether the preliminary preset planting depth and spacing of the different planting area are reasonable according to whether the comprehensive influence amplitude presents an upward trend or a stable region or a downward trend. If the judgment result is unreasonable, the adjustment is carried out to achieve the effect of verifying the result as reasonable. Through the above different information processing method and verification processing method, the scientific researchability analysis of the historical data resources is fully carried out, and the efficiency of the whole planting parameter regulation work and the growth quality of the whole target area sugarcane are improved. BRIEF DESCRIPTION OF DRAWINGS

[0026] Figure 1 is a step block diagram of a control method of an intelligent sugarcane planter mainly embodied by the embodiment. DETAILED DESCRIPTION

[0027] The present application will be further described in detail below in combination with the following embodiments.

[0028] REFERENCE Figure 1 A control method of an intelligent sugarcane planter, the method comprising the following steps:

[0029] Step S1, the same planting area and the different planting area divided according to the soil conditions of the target area needing to be planted with sugarcane are respectively subjected to initial preset of planting depth and spacing to obtain planting parameters one and planting parameters two, and the planting parameters two are verified to output parameter verification information.

[0030] Step S2, according to the sowing parameter verification information, the distribution of the different sowing area is counted, and the historical sowing data is obtained, and the correlation prediction model is established according to the historical sowing data.

[0031] Step S3, according to the distribution of the different sowing area, the parameter difference between the sowing parameter one and the sowing parameter two is counted to obtain the pretreatment difference value, and the comprehensive influence amplitude is predicted according to the pretreatment difference value and the correlation prediction model.

[0032] Step S4, if the comprehensive influence amplitude presents an upward trend or a stable trend, it is judged that the preset of the sowing parameter two is reasonable, if the comprehensive influence amplitude presents a downward trend, the adjustment sowing parameter is preset according to the pretreatment difference value, and the sowing parameter three is selected from the adjustment sowing parameter by the correlation prediction model test, the sowing parameter one and the sowing parameter three are input into the sugarcane seeder for parameter control, and the sowing control result is output.

[0033] Specifically, the soil characteristic information of the target area to be sowed is detected, so as to preliminarily preset the sowing depth and spacing parameters according to the soil characteristic information of different areas, so as to avoid the setting of uniform sowing depth and spacing parameters in the traditional technology, ignoring the difference of sowing conditions caused by different soil characteristic information in different areas, and further affecting the final growth quality of sugarcane. The sowing depth and spacing are analyzed differently for part of the different sowing areas, so as to preset the most reasonable sowing depth and spacing, and considering the correlation influence of the same sowing area on the different sowing area, such as promoting the growth quality of sugarcane in the different sowing area, while not affecting the growth quality of sugarcane in the same sowing area. The correlation influence degree is different for different distribution positions of the different sowing area, so three distribution conditions are judged, one is that the different sowing area is located at the edge of the target area and adjacent to two sowing points in the same sowing area, one is that the different sowing area is located at the edge of the target area and adjacent to three sowing points in the same sowing area, and the other is that the different sowing area is located in the middle of the target area and adjacent to four sowing points in the same sowing area. The information of the above three distribution conditions is counted, that is, the comprehensive influence amplitude of the above distribution conditions is predicted, so as to further judge whether the sowing depth and spacing preliminarily preset for the different sowing area are reasonable according to the upward trend or stable region or downward trend of the comprehensive influence amplitude. If the judgment result is unreasonable, adjust to achieve the effect of reasonable verification result. Through the above different information processing mode and verification processing mode, the historical data resources are fully analyzed scientifically, and the efficiency of the whole sowing parameter control work and the quality of the whole target area sugarcane growth are improved.

[0034] The specific step S1 includes the following sub-steps:

[0035] The target area where sugarcane needs to be planted is detected for soil condition to obtain soil characteristic information, and the target area is divided into a same planting area and a different planting area according to the soil characteristic information.

[0036] The same planting area is planted with sugarcane according to the soil characteristic information to obtain a first planting parameter, and the different planting area is planted with sugarcane according to the soil characteristic information to obtain a second planting parameter, and the second planting parameter is verified to output planting parameter verification information.

[0037] Specifically, the soil characteristic information includes soil PH value (such as nitrogen, phosphorus, potassium, trace elements, and organic matter content), soil texture (such as sandy loam, loam, and clay), soil moisture content, and soil pest and disease conditions (such as diseases and insect pests). The same planting area and the different planting area (if the same planting area is T and the different planting area is C, the coverage area of T is larger than that of C, and the soil characteristic information of T is different from that of C, such as a critical value for differential analysis, if the critical value for differential analysis is not reached (for example, the comprehensive difference value of the above soil information reaches the critical value, which is obtained from historical analysis data statistics), the target area is directly processed with a unified planting depth and spacing parameter), the first planting parameter and the second planting parameter (for example, the sugarcane planting depth and spacing in T are set to (h1, j1), and the second planting parameter is set to (h2, j2), which is matched according to the historical record information matching table), and the planting parameter verification information (in traditional technology, the sugarcane planting parameter is often set with a unified depth and spacing parameter, without considering the preset of different planting parameters, so in order to reduce the misjudgment probability, the subsequent verification processing is performed on the different second planting parameter).

[0038] The specific step S2 includes the following sub-steps:

[0039] According to the sowing parameter verification information, if the distribution position of the difference sowing area is the edge of the target area and adjacent to two sowing point positions in the same sowing area, distribution case one is output, if the distribution position of the difference sowing area is the edge of the target area and adjacent to three sowing point positions in the same sowing area, distribution case two is output, and if the distribution position of the difference sowing area is the middle of the target area and adjacent to four sowing point positions in the same sowing area, distribution case three is output.

[0040] The historical sowing data of sowing in a historical period is acquired, and according to the historical sowing data, an influence degree value of the difference sowing area under different soil conditions affected by the correlation of the same sowing area is evaluated to obtain a to-be-tested correlation factor, and a parameter difference variable value of the sowing depth and interval between the same sowing area and the difference sowing area under different soil conditions is counted to obtain a to-be-tested difference variable value.

[0041] According to the to-be-tested correlation factor, a weight ratio one or a weight ratio two or a weight ratio three is counted according to the weight ratio of the respective correlation influence degree of the distribution case one or the distribution case two or the distribution case three in the historical period, and the weight ratio one or the weight ratio two or the weight ratio three is combined to obtain a to-be-tested weight ratio.

[0042] According to the historical sowing data, the to-be-tested correlation factor, the to-be-tested difference variable value and the to-be-tested weight ratio, a correlation prediction model is established.

[0043] Specifically, as distribution condition one (refers to C is located at the right angle edge of the target area: at this time, two seeding points (if d1, d2) in T are in adjacent position relationship, that is, d1, C, d2 can be connected to form a shape of a right triangle), distribution condition two (refers to C is located at the straight edge of the target area: at this time, three seeding points (if d1, d2, d3) in T are in adjacent position relationship, that is, d1, C, d3 are located on a straight line, and d1, d2, C, d3 can be connected to form a shape of an isosceles triangle), distribution condition three (refers to C is located at the middle position of the target area: at this time, four seeding points (if d1, d2, d3, d4) in T are in adjacent position relationship, that is, d1, d2, d3, d4 can be connected to form a shape of a parallelogram, and C is located at the middle position of d1, d2, d3, d4), historical seeding data (contains soil feature detection record information of each seeding area in different historical periods if G, seeding parameter record data, soil feature change condition of sugarcane in different growth stages (for each seeding area), quality condition of sugarcane in different growth stages (for each seeding area), difference value of seeding depth and spacing between different seeding parameter areas, influence degree weight of different seeding area in each of the above three distribution conditions), to be measured correlation factor (for example, if two parts of the seeding area are taken from the historical seeding data for data illustration: if they are L1 area and L2 area, the soil feature information of L1 area and L2 area is not the same, if the soil condition comprehensive evaluation result of L1 sugarcane in growth stage D (if it is mature period, the growth stages of sugarcane include seedling stage, tillering stage, elongation stage (jointing stage), mature period, dormancy period) is G1, and the corresponding sugarcane growth quality in growth stage J is M1, if the soil condition comprehensive evaluation result of L2 sugarcane in growth stage D (also mature period) is G2, and the corresponding sugarcane growth quality in growth stage J is M2, the absolute value of G1-G2 and the absolute value of M1-M2 are taken to obtain a ratio B1, if the preset judgment threshold value under ideal conditions is B2-B4 (that is, used to judge whether the correlation influence between L1 and L2 is promoting influence or weakening influence), the difference between B2-B1 is the to-be-measured correlation factor y1, which has a symbol of “-”, that is, weakening influence, if the absolute value of G1-G2 and the absolute value of M1-M2 are taken to obtain a ratio B5, the difference between B5-B4 is the to-be-measured correlation factor y2, which has a symbol of “+”, that is, promoting influence, here, L1 and L2 areas are analyzed, and the same analysis is performed on multiple different areas (refers to the area with soil feature difference), and the data set is Y), to-be-measured difference value (for example, the seeding depth and spacing of L1 are (H1, J1), and the seeding depth and spacing of L2 are (H2, J2), (H1, J1)-(H2, J2) is (H0, J0),J0) Perform the difference degree comprehensive statistics: superimpose the values of H0 and J0, if it is n1, it is the difference variable to be tested, here is analyzed between L1 and L2 regions, and so on, the same analysis is carried out between multiple different regions (referring to the region where the soil characteristics differ), if the data set is N, the weight ratio to be tested (if here only one distribution situation one is taken as an example, and only L1 and L2 of distribution situation one are taken as an example, the average value statistics of the influence degree of the same sowing area on the difference sowing area belonging to distribution situation one existing in different historical periods is carried out, if it is P1, the average value statistics of the influence degree of the same sowing area on the difference sowing area belonging to distribution situation two existing in different historical periods is carried out, if it is P2, the average value statistics of the influence degree of the same sowing area on the difference sowing area belonging to distribution situation three existing in different historical periods is carried out, if it is P3, then the historical weight ratio of distribution situation one is P1 / (P1+P2+P3), if it is q1, the historical weight ratio of distribution situation two is P2 / (P1+P2+P3), if it is q2, the historical weight ratio of distribution situation three is P3 / (P1+P2+P3), if it is q3, here is under the condition of one to be tested difference variable, and so on, the statistics is carried out under the condition of multiple to be tested difference variables in history, if the data set is Q, the correlation prediction model is established (if the soil characteristic difference between L1 and L2: G1-G2 is u1, here is analyzed between L1 and L2 regions, and so on, the same analysis is carried out between multiple different regions (referring to the region where the soil characteristics differ), if the data set is U, if the unified conversion ratio of N and U is a / A, (N / U)*a / A is the additional error value, then Z=P*Y+(N / U)*a / A is established, wherein Z refers to the comprehensive correlation influence degree value, Z has "+" and "-" symbols, "+" represents that the comprehensive correlation influence degree value is in an upward trend, "-" represents that the comprehensive correlation influence degree value is in a downward trend, if the value is 0, it represents that the comprehensive correlation influence degree value is in a stable trend).

[0044] The specific step S3 includes the following sub-steps:

[0045] The parameter difference variable between the sowing parameter one of the same sowing area in distribution situation one or distribution situation two or distribution situation three and the sowing parameter two of the difference sowing area in distribution situation one or distribution situation two or distribution situation three is statistically obtained to obtain the pre-processing difference variable.

[0046] According to the pre-processing difference variable, one weight ratio is selected from the to-be-tested weight ratio to obtain a pre-processing weight ratio, and according to the soil characteristic information and the pre-processing difference variable, one correlation factor is selected from the to-be-tested correlation factor to obtain a pre-processing correlation factor.

[0047] The soil characteristic information, the pretreatment weight ratio, the pretreatment correlation factor and the pretreatment difference variable are input into the correlation prediction model to test to obtain a comprehensive influence amplitude.

[0048] Specifically, the pretreatment difference variable (if n2, here the difference variable between the obtained parameters C and d1, and the difference variable between the obtained parameters C and d2 are integrated), the pretreatment weight ratio (if the position distribution of the current different sowing area is distribution condition one, the pretreatment weight ratio is q1), and the comprehensive influence amplitude (if the soil characteristic difference between the current different sowing area and the same sowing area is u2, according to the soil characteristic information, u2 and n2 are counted, if the corresponding matching C and d1 are obtained from Y, the correlation factor y1 is a symbol “-”, and the correlation factor y2 between C and d2 is a symbol “+”, then Z=P*Y+(N / U)*a / A, that is, Z=q1*(y1+y2)+(n2 / u2)*a / A, if the result is z1).

[0049] The specific step S4 includes the following sub-steps:

[0050] If the comprehensive influence amplitude shows an upward trend or a stable trend, it is judged that the preset sowing parameter two is reasonable, and the sowing parameter one and the sowing parameter two are input into the sugarcane planter to perform parameter control, and a sowing control result is output.

[0051] If the comprehensive influence amplitude shows a downward trend, the historical sowing data, the to-be-tested difference variable and the pretreatment difference variable are used to preset an adjusted sowing parameter, and the adjusted sowing parameter contains at least one sowing parameter value.

[0052] The parameter difference variable between the sowing parameter one of the same sowing area in distribution condition one or distribution condition two or distribution condition three and the adjusted sowing parameter of the different sowing area in distribution condition one or distribution condition two or distribution condition three is counted to obtain a preset difference variable.

[0053] The soil characteristic information, the pretreatment weight ratio, the pretreatment correlation factor and the preset difference variable are input into the correlation prediction model to test to obtain a to-be-tested influence amplitude.

[0054] According to the maximum value existing in the to-be-tested influence amplitude, one of the most reasonable sowing parameters is selected from the adjusted sowing parameter to obtain a sowing parameter three, and the sowing parameter one and the sowing parameter three are input into the sugarcane planter to perform parameter control, and a sowing control result is output.

[0055] Specifically, if the sign of z1 is "+" or the value of z1 is 0, it indicates that the correlation between T and F has a promoting effect, and thus the same sowing area and the different sowing area of the sowing machine can be controlled according to the first preset sowing parameter and the second sowing parameter. If the sign of z1 is "-", it indicates that the correlation between T and F has a weakening effect, and thus the second sowing parameter needs to be re-set and verified. If the test difference variable n2, n4, n5, n6 has a promoting effect, the pre-processed difference variable n2 is removed from n2, n4, n5, n6. If the test difference variable n3, n4, n5, n6 has a promoting effect, and the difference between n3 and n2 is less than the removal critical difference x, n3 is removed from n3, n4, n5, n6, and the adjusted sowing parameter is n4, n5, n6. If the test effect amplitude is z2, z3, z4, the sowing control result is that, if z3 is the maximum value, it indicates that n5 corresponds to the best promoting effect between T and F, and n5 is the third sowing parameter. The same sowing area and the different sowing area of the sowing machine can be controlled according to the first sowing parameter and the third sowing parameter.

[0056] The above are preferred embodiments of the present application, and are not intended to limit the protection scope of the present application. Any equivalent changes made according to the structure, shape, and principle of the present application should be covered within the protection scope of the present application.

Claims

1. A control method of an intelligent sugarcane planter, characterized by, The method comprises the following steps: Step S1, for the target area needing sugarcane planting, the same planting area and the different planting area divided according to the soil conditions, the initial preset of the planting depth and the interval is carried out to obtain the planting parameters one and the planting parameters two, and the planting parameters two are verified, and the parameter verification information is output; Step S2, according to the planting parameter verification information, the distribution of the different planting area is counted, and the historical planting data is obtained, and according to the historical planting data, the correlation prediction model is established; Wherein, S2 step includes: According to the planting parameter verification information, if the distribution position of the different planting area is the edge of the target area and adjacent to two planting point positions in the same planting area, distribution one is output, if the distribution position of the different planting area is the edge of the target area and adjacent to three planting point positions in the same planting area, distribution two is output, if the distribution position of the different planting area is the middle of the target area and adjacent to four planting point positions in the same planting area, distribution three is output; The historical planting data of planting in the historical period is obtained, and according to the historical planting data, the influence degree value of the correlation of the different planting area under different soil conditions is evaluated to obtain the test correlation factor, and the parameter difference value of the planting depth and the interval between the same planting area and the different planting area under different soil conditions is counted to obtain the test difference value; According to the test correlation factor, the weight ratio of the correlation influence degree of each distribution one or distribution two or distribution three in the historical period is counted to obtain the weight ratio one or weight ratio two or weight ratio three, and the weight ratio one or weight ratio two or weight ratio three is combined to obtain the test weight ratio; According to the historical planting data, the test correlation factor, the test difference value and the test weight ratio, the correlation prediction model is established; Step S3, according to the distribution of the different planting area, the parameter difference value between the planting parameters one and the planting parameters two is counted to obtain the pretreatment difference value, and according to the pretreatment difference value and the correlation prediction model, the comprehensive influence amplitude is predicted; Wherein, S3 step includes: The parameter difference value between the planting parameters one of the same planting area in the distribution one or distribution two or distribution three and the planting parameters two of the different planting area in the distribution one or distribution two or distribution three is counted to obtain the pretreatment difference value; According to the pretreatment difference value, one weight ratio is selected from the test weight ratio to obtain the pretreatment weight ratio, and according to the soil feature information and the pretreatment difference value, one correlation factor is selected from the test correlation factor to obtain the pretreatment correlation factor; The soil feature information, the pretreatment weight ratio, the pretreatment correlation factor and the pretreatment difference value are input into the correlation prediction model for testing to obtain the comprehensive influence amplitude; Step S4, if the comprehensive influence amplitude presents an upward trend or a stable trend, it is judged that the preset of the second sowing parameter is reasonable, if the comprehensive influence amplitude presents a downward trend, an adjusted sowing parameter is preset according to the pretreatment difference variable, the adjusted sowing parameter is tested through the correlation prediction model, a third sowing parameter is correspondingly screened out from the adjusted sowing parameter, the first sowing parameter and the third sowing parameter are input into the sugarcane seeder for parameter regulation, and a sowing regulation result is output. The step S4 comprises: The soil characteristic information, the pretreatment weight ratio, the pretreatment correlation factor and the preset difference variable are input into the correlation prediction model for testing to obtain a to-be-tested influence amplitude; According to the maximum value existing in the to-be-tested influence amplitude, a most reasonable sowing parameter is correspondingly screened out from the adjusted sowing parameter to obtain a third sowing parameter, the first sowing parameter and the third sowing parameter are input into the sugarcane seeder for parameter regulation, and a sowing regulation result is output.

2. The control method of the intelligent sugarcane planter according to claim 1, wherein, The step S1 comprises: The soil characteristic information is obtained by detecting the soil conditions of the target area to be sowed with sugarcane, the target area is divided into the same sowing area and the different sowing area according to the soil conditions, the same sowing area is larger than the different sowing area; The first sowing parameter is obtained by initially presetting the sowing depth and spacing parameters of the same sowing area according to the soil conditions in the soil characteristic information, the second sowing parameter is obtained by initially presetting the sowing depth and spacing parameters of the different sowing area according to the soil conditions in the soil characteristic information, and the second sowing parameter is verified to output sowing parameter verification information.

3. The control method of the intelligent sugarcane planter according to claim 1, characterized by, The step S4 further comprises: If the comprehensive influence amplitude presents an upward trend or a stable trend, it is judged that the preset of the second sowing parameter is reasonable, and the first sowing parameter and the second sowing parameter are input into the sugarcane seeder for parameter regulation, and a sowing regulation result is output.

4. The control method of the intelligent sugarcane planter according to claim 1, wherein, The step S4 further comprises: If the comprehensive influence amplitude presents a downward trend, an adjusted sowing parameter is preset according to the historical sowing data, the to-be-tested difference variable and the pretreatment difference variable, the adjusted sowing parameter comprises at least one sowing parameter value; The parameter difference variable between the first sowing parameter of the same sowing area in the distribution situation one or the distribution situation two or the distribution situation three and the adjusted sowing parameter of the different sowing area in the distribution situation one or the distribution situation two or the distribution situation three is counted to obtain a preset difference variable.

Images