A yield measuring method and device using multi-sensor information fusion

Abstract

The application discloses a yield measuring method and device based on multi-sensor information fusion, which comprises the following steps: using a harvester to measure yield for the first time, wherein the sensors used in the first yield measurement include an acceleration sensor, a non-contact sensor and a contact sensor; obtaining yield measurement data and processing the yield measurement data to obtain a first yield measurement value time sequence measured by the non-contact sensor, a second yield measurement value time sequence measured by the contact sensor, a first acceleration measured by the acceleration sensor at the same time as the non-contact sensor, and a second acceleration measured by the acceleration sensor at the same time as the contact sensor; constructing a time sequence prediction model according to the first yield measurement value time sequence and the second yield measurement value time sequence; measuring the yield of the grain to be measured for the second time, and selecting a non-contact sensor yield measurement mode, a contact sensor yield measurement mode and a time sequence prediction model yield measurement mode according to the comparison result of the first acceleration, the second acceleration and an acceleration threshold.

Description

Technical Field

[0001] This invention relates to the fields of smart agriculture, agricultural sensing and information technology, and in particular to a method and apparatus for yield measurement by multi-sensor information fusion. Background Technology

[0002] The yield monitoring system installed on grain combine harvesters can accurately collect crop yield information and draw yield distribution maps while harvesting crops, helping farm managers to formulate precise planting and production management plans such as fertilization and sowing. Among these, accurate crop yield measurement is the core of all yield-based decisions on fertilization and sowing, such as sowing and fertilization decisions for farmland fertility management zones based on historical yields, and soil testing and formula fertilization decisions based on yield. Therefore, improving the measurement accuracy of grain flow sensors is particularly important.

[0003] Due to its simple structure, convenient installation, and low cost, the impulse-type grain flow sensor is currently the most widely used, mature, and extensively researched method for yield measurement. Most commercial yield monitoring systems abroad currently use this grain flow sensor. Photoelectric sensors calculate grain volume by measuring the thickness of grain on each scraper of a scraper conveyor, representing a non-contact grain flow measurement technology. Photoelectric sensors are easy to install and calibrate, making them a commonly used volumetric flow sensor, and commercially available photoelectric yield measurement sensors are already available. However, the vibration of the combine harvester, particularly under special operating conditions such as machine foundation vibration, uneven field surfaces causing bumps, tilting, turning at the edge of the field, and emergency braking, has the most significant impact on both types of sensors, thus increasing measurement errors. Summary of the Invention

[0004] To address the issue of yield measurement errors caused by the vibration of the combine harvester under special operating conditions such as machine vibration, uneven field ground causing bumps, tilting, turning at the edge of the field, and emergency braking, which affect sensor measurements, this invention discloses a yield measurement method based on multi-sensor information fusion, comprising the following steps:

[0005] The first yield measurement was conducted using a harvester. The sensors used in the first yield measurement included an acceleration sensor installed on the body of the harvester, a non-contact sensor installed at a first position of the grain conveying mechanism of the harvester, and a contact sensor installed at a second position of the grain conveying mechanism.

[0006] Acquire yield measurement data and process the yield measurement data to obtain a time series (p1, p2, ..., p) of the first yield measurement value measured by the non-contact sensor. i ,…), and the time series of the second production measurement values ​​(q1,q2,…,q) measured by the contact sensor. iThe acceleration sensor measures the first acceleration simultaneously with the non-contact sensor and the second acceleration simultaneously with the contact sensor.

[0007] Based on the first production measurement time series (p1, p2, ..., p...), i ,…) and the time series of the second yield measurement (q1,q2,…,q i Construct time series forecasting models;

[0008] A second yield measurement is performed on the grain to be tested. Based on the comparison results of the first acceleration, the second acceleration, and an acceleration threshold a0, the yield measurement methods of the non-contact sensor, the contact sensor, and the time series prediction model are selected.

[0009] In one embodiment of the method described above, the grain conveying mechanism employs a scraper-type elevator.

[0010] In one embodiment of the above method of the present invention, the first position of the grain conveying mechanism is the middle part of the scraper-type elevator, and the second position of the grain conveying mechanism is the grain outlet of the scraper-type elevator.

[0011] In one embodiment of the method described above, there are multiple scrapers between the non-contact sensor and the contact sensor, and the first production measurement value time series (p1, p2, ..., p...) i One of the first production measurements p) i The corresponding time series of the second production measurement values ​​(q1, q2, ..., q) i The second production measurement q in (,…) n+i The first production measurement value p i The first acceleration a corresponds to the first acceleration among the plurality of first accelerations. i The second production measurement value q n+i The second acceleration a corresponds to one of the plurality of second accelerations. n+i .

[0012] In one embodiment of the method described above, the first yield measurement time series (p1, p2, ..., p) used when constructing the time series prediction model i ,…) and the time series of the second yield measurement (q1,q2,…,q i (,...) is measured when the acceleration is less than the acceleration threshold a0.

[0013] In one embodiment of the method described above, the step of constructing a time series prediction model is based on constructing an autoregressive (AR) model.

[0014] In one embodiment of the method described above, the step of constructing an autoregressive AR model further includes:

[0015] The autoregressive AR model is as follows:

[0016]

[0017] Grain yield value on one of the scrapers i = 1, ..., N

[0018] N is the length of the time series of the first production measurement value, and P is the model order. For model parameters, ∈ n+1 This represents the model residuals.

[0019] In one embodiment of the method described above, the determination of model parameters... The steps further include:

[0020] Model parameters are calculated using the least squares method. Least squares estimation Make the model residuals ∈ n+1 The sum of squares is the smallest.

[0021] In one embodiment of the method described above, the step of determining the model order P further includes:

[0022] AIC(P)=NInσ a 2 +2P

[0023] Where, σ a 2 To pass the model residual ∈ n+1 The calculated variance of discrete white noise, AIC(P), is a function of the model order P;

[0024] The model order P is obtained by minimizing the value of AIC(P).

[0025] In one embodiment of the method of the present invention, the step of selecting the non-contact sensor, the contact sensor, and the time series prediction model measurement method based on the first acceleration, the second acceleration, and an acceleration threshold a0 further includes:

[0026] When the harvester is operating smoothly, select the non-contact sensor and / or contact sensor yield measurement method;

[0027] When the harvester is not operating smoothly, the yield measurement method of the time series prediction model is selected.

[0028] In one embodiment of the method described above, the step of selecting the non-contact sensor and / or contact sensor yield measurement method when the harvester is operating smoothly further includes:

[0029] When the first acceleration a i ≤Acceleration threshold a0<Second acceleration a n+i At that time, the grain yield value measured on the scraper by the non-contact sensor is selected as the scraper grain yield value y. i , that is, y i =p i ;

[0030] When the second acceleration a n+i ≤Acceleration threshold a0<First acceleration a i At that time, the grain yield value measured on the scraper by the contact sensor is selected as the scraper grain yield value y. i , that is, y i =q n+i ;

[0031] When the first acceleration a i The second acceleration a n+i When the acceleration threshold a0 is less than or equal to the current grain yield value on the scraper, the average of the grain yield values ​​measured by the non-contact sensor and the contact sensor is selected as the grain yield value y. i ,Right now

[0032] In one embodiment of the method of the present invention, the step of selecting the time series prediction model for yield measurement when the harvester operation is unstable further includes:

[0033] When the first acceleration a i The second acceleration a n+i When the acceleration threshold a0 is reached, the yield time-series data f obtained by the yield measurement method selected during the stable operation of the harvester is acquired. i ;

[0034] The time series prediction model is used to predict the current grain yield on the scraper, resulting in the scraper grain yield value y. i ,Right now

[0035] In one embodiment of the method described above, the time series prediction model is used to predict the current grain yield value on the scraper to obtain the scraper grain yield value y. i The steps further include:

[0036] The time series prediction model is used to predict the current grain yield on the scraper, resulting in the scraper grain yield value y. i The production time series data f can be added.i This will allow for the estimation of subsequent scraper grain yield.

[0037] In one embodiment of the method described above, the step of acquiring and processing the yield measurement data further includes acquiring multiple location information for associating with the results of the second yield measurement to provide location-based yield services and applications.

[0038] In one embodiment of the method described above, the non-contact sensor is a photoelectric sensor, and the contact sensor is an impulse sensor.

[0039] The present invention also discloses a multi-sensor information fusion production measurement device, used to perform the steps of any of the above methods, including:

[0040] The first yield measurement module is used to perform the first yield measurement using a harvester. The sensors used in the first yield measurement include an acceleration sensor installed on the body of the harvester, a non-contact sensor installed at a first position of the grain conveying mechanism of the harvester, and a contact sensor installed at a second position of the grain conveying mechanism.

[0041] The yield measurement data acquisition and processing module is used to acquire and process the yield measurement data to obtain a time series (p1, p2, ..., p) of the first yield measurement value measured by the non-contact sensor. i ,…), and the time series of the second production measurement values ​​(q1,q2,…,q) measured by the contact sensor. i ...), the acceleration sensor measures a first acceleration simultaneously with the non-contact sensor, and a second acceleration simultaneously with the contact sensor;

[0042] The time series forecasting model building module is used to construct a time series forecasting model based on the first yield measurement value (p1, p2, ..., p...). i ,…) and the time series of the second yield measurement (q1,q2,…,q i Construct time series forecasting models;

[0043] The second yield measurement module is used to perform a second yield measurement on the grain to be measured. Based on the comparison results of the first acceleration, the second acceleration, and an acceleration threshold a0, it selects the non-contact sensor yield measurement method, the contact sensor yield measurement method, and the time series prediction model yield measurement method.

[0044] The present invention also includes a multi-sensor information fusion yield measurement system, comprising an acceleration sensor installed on the body of the harvester, a non-contact sensor installed on the grain conveying mechanism of the harvester, a contact sensor installed on the grain conveying mechanism, a storage unit and a control unit, wherein the control unit is connected to the acceleration sensor, the non-contact sensor, the contact sensor and the storage unit, and further includes the aforementioned device connected to the control unit.

[0045] The present invention also includes a harvester, comprising a body mounted on a chassis, a drive unit and a grain conveying mechanism, and the aforementioned multi-sensor information fusion yield measurement system.

[0046] The present invention also includes a storage medium for storing a computer control program for performing the steps of any of the above methods.

[0047] Based on the above, the multi-sensor information fusion yield measurement method, device, system and harvester disclosed in this invention uses two yield measurement sensors to measure the yield at the same scraper at different times, and selects different scraper yield measurement methods according to the vibration amplitude of the harvester body, thereby overcoming the influence of increased harvester vibration on the measurement accuracy of the yield measurement sensor at a certain time and improving the yield measurement accuracy.

[0048] Furthermore, location-based production services and applications can be provided by combining location information obtained during production measurement with the production measurement values. Simultaneously, the multi-sensor information fusion production measurement method can be applied to various grain conveying mechanisms suitable for measurement using photoelectric sensors and impulse sensors. Attached Figure Description

[0049] Figure 1 This is a schematic block diagram of a multi-sensor information fusion method for yield measurement in one embodiment of the present invention.

[0050] Figure 2 This is a schematic diagram of a grain conveying mechanism used to measure grain yield in one embodiment of the present invention.

[0051] Figure 3 This is a schematic block diagram of a production measurement device for multi-sensor information fusion in one embodiment of the present invention.

[0052] Figure 4 This is a schematic block diagram of a multi-sensor information fusion production measurement system according to an embodiment of the present invention.

[0053] Figure 5 This is a schematic block diagram of a harvester according to one embodiment of the present invention.

[0054] In the attached figures, the following labels are used:

[0055] 1: Grain transport agencies

[0056] 2: Grain transport direction

[0057] 3: Grains to be tested

[0058] 4: Scraper

[0059] 5: Non-contact sensors

[0060] 5': Photoelectric sensor

[0061] 6: Contact Sensors

[0062] 6': Impulse sensor

[0063] 7: Accelerometer

[0064] 8: Scraper type elevator

[0065] 10: Multi-sensor information fusion production measurement device

[0066] 11: First Production Testing Module

[0067] 12: Production Measurement Data Acquisition and Processing Module

[0068] 13: Time Series Forecasting Model Building Module

[0069] 14: Second Production Measurement Module

[0070] 100: Multi-sensor information fusion production measurement system

[0071] 110: Control Unit

[0072] 120: Storage unit

[0073] 200: Harvester

[0074] 210: Chassis

[0075] 220: Body

[0076] 230: Drive unit Detailed Implementation

[0077] The technical solution of the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments, so as to further understand the purpose, solution and beneficial technical effects of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0078] It should be noted that in this specification, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0079] Certain terms are used in this specification and the appended claims to refer to specific components or parts. Those skilled in the art will understand that users or manufacturers may use different names or terms to refer to the same component or part. This specification and the appended claims do not distinguish components or parts by differences in name, but rather by differences in function.

[0080] In this invention, the terms "upper," "lower," "left," "right," "front," "rear," "top," "bottom," "inner," "outer," "middle," "vertical," "horizontal," "lateral," and "longitudinal" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. These terms are primarily for the purpose of better describing the invention and its embodiments, and are not intended to limit the indicated devices, elements, or components to having a specific orientation, or to be constructed and operated in a specific orientation.

[0081] Furthermore, in addition to indicating direction or positional relationship, some of the aforementioned terms may also have other meanings. For example, the term "above" may also be used in certain situations to indicate a dependency or connection. Those skilled in the art can understand the specific meaning of these terms in this invention based on the specific circumstances.

[0082] Furthermore, the terms "installation," "setup," "equipped with," "connection," and "linked" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral structure; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium, or an internal connection between two devices, components, or parts. Those skilled in the art can understand the specific meaning of these terms in this invention based on the specific circumstances.

[0083] Please refer to Figure 1 and Figure 2This invention discloses a yield measurement method based on multi-sensor information fusion, comprising the following steps:

[0084] The first yield measurement was conducted using a harvester. The sensors used in the first yield measurement included an acceleration sensor 7 installed on the body of the harvester, a non-contact sensor 5 installed at the first position of the grain conveying mechanism 1 of the harvester, and a contact sensor 6 installed at the second position of the grain conveying mechanism 1.

[0085] Acquire and process the yield measurement data to obtain a time series (p1, p2, ..., p) of the first yield measurement value measured by the non-contact sensor 5. i ,…), and the time series of the second production measurement values ​​(q1,q2,…,q) measured by the contact sensor 6. i The acceleration sensor 7 measures the first acceleration at the same time as the non-contact sensor 5, and measures the second acceleration at the same time as the contact sensor 6;

[0086] In addition, during the first yield measurement, the location information of multiple harvesters during operation was collected by a positioning antenna installed on the center line of the harvester's roof.

[0087] Based on the first production measurement time series (p1, p2, ..., p...), i ,…) and the time series of the second yield measurement (q1,q2,…,q i Construct time series forecasting models;

[0088] A second yield measurement is performed on the grain to be tested 3. Based on the comparison results of the first acceleration, the second acceleration, and an acceleration threshold a0, the yield measurement methods of the non-contact sensor 5, the contact sensor 6, and the time series prediction model are selected.

[0089] In one embodiment of the method described above, the non-contact sensor 5 is a photoelectric sensor 5', and the contact sensor 6 is an impulse sensor 6'.

[0090] In one embodiment of the method described above, the grain conveying mechanism 1 adopts a scraper-type elevator 8.

[0091] Those skilled in the art should know that the above-described scraper-type elevator is only one example of a grain conveying mechanism 1, and can also be applied to other harvesting devices suitable for installing photoelectric sensors 5' and impulse sensors 6' for measurement.

[0092] In one embodiment of the above method of the present invention, the first position of the grain conveying mechanism 1 is the middle of the scraper-type elevator 8, and the second position of the grain conveying mechanism 1 is the grain outlet of the scraper-type elevator 8.

[0093] In one embodiment of the method described above, there are multiple scrapers 4, for example, n scrapers 4, between the non-contact sensor 5 and the contact sensor 6, and the first production measurement value time series (p1, p2, ..., p... i One of the first production measurements p) i The corresponding time series of the second production measurement values ​​(q1,q2,…,q) i The second production measurement q in (,…) n+i The first production measurement value p i The first acceleration a corresponds to the first acceleration among the plurality of first accelerations. i The second production measurement value q n+i The second acceleration a corresponds to one of the plurality of second accelerations. n+i .

[0094] In one embodiment of the method described above, the first yield measurement time series (p1, p2, ..., p) used when constructing the time series prediction model i ,…) and the time series of the second yield measurement (q1,q2,…,q i ,…) are measured when the acceleration is less than the acceleration threshold a0.

[0095] In one embodiment of the method described above, the step of constructing a time series prediction model is based on constructing an autoregressive (AR) model.

[0096] In one embodiment of the method described above, the step of constructing an autoregressive AR model further includes:

[0097] The autoregressive AR model is as follows:

[0098]

[0099] Grain yield value on one of the scrapers 4 i = 1, ..., N

[0100] N is the length of the time series of the first production measurement value, and P is the model order. For model parameters, ∈ n+1 The model residual is the difference between the actual data and the predicted data.

[0101] In other words, the above time series forecasting model states that the output at the next time step is related to the outputs of the previous P times, where P is the model order, and the degree of correlation is determined by... This means that the output value at the next moment is predicted based on the previous P related output values.

[0102] In one embodiment of the method described above, the determination of model parameters... The steps further include:

[0103] Model parameters are calculated using the least squares method. Least squares estimation Make the model residuals ∈ n+1 The sum of squares is the smallest.

[0104] In one embodiment of the method described above, the step of determining the model order P further includes:

[0105] AIC(P)=NInσ a 2 +2P

[0106] Where, σ a 2 To pass the model residual ∈ n+1 The calculated variance of discrete white noise, AIC(P), is a function of the model order P;

[0107] The model order P is obtained by minimizing the value of AIC(P).

[0108] In one embodiment of the method of the present invention, the step of selecting the non-contact sensor 5, the contact sensor 6, and the time series prediction model production measurement method based on the first acceleration, the second acceleration, and an acceleration threshold a0 further includes:

[0109] When the harvester is operating smoothly, select the non-contact sensor 5 and / or the contact sensor 6 to measure the yield.

[0110] When the harvester is not operating smoothly, the time series prediction model is selected for yield measurement.

[0111] In one embodiment of the method described above, the step of selecting the non-contact sensor 5 and / or the contact sensor 6 for yield measurement when the harvester is operating smoothly further includes:

[0112] When the first acceleration a i ≤Acceleration threshold a0<Second acceleration a n+i At that time, the grain yield value measured by the non-contact sensor 5 on the current scraper 4 is selected as the scraper grain yield value y. i , that is, y i =p i ;

[0113] When the second acceleration a n+i ≤Acceleration threshold a0<First acceleration a iAt that time, the grain yield value measured by the contact sensor 6 on the current scraper 4 is selected as the scraper grain yield value y. i , that is, y i =q n+i ;

[0114] When the first acceleration a i The second acceleration a n+i When the acceleration threshold a0 is less than or equal to the value of the grain yield on the scraper 4, the average value measured by the non-contact sensor 5 and the contact sensor 6 is selected as the scraper grain yield value y. i ,Right now

[0115] In one embodiment of the method of the present invention, the step of selecting the time series prediction model for yield measurement when the harvester operation is unstable further includes:

[0116] When the first acceleration a i The second acceleration a n+i When the acceleration threshold a0 is reached, the yield time-series data f obtained by the yield measurement method selected during the stable operation of the harvester is acquired. i ;

[0117] The time series prediction model is used to predict the grain yield value on scraper 4 to obtain the scraper grain yield value y. i ,Right now

[0118] In one embodiment of the method described above, the time series prediction model is used to predict the grain yield value y on the current scraper 4. i The steps further include:

[0119] The time series prediction model is used to predict the current grain yield on scraper 4 to obtain the scraper grain yield value y. i The production time series data f can be added. i This will allow for the estimation of subsequent scraper grain yield.

[0120] In addition, when determining the scraper output y i Simultaneously, the harvest location information is acquired to obtain the scraper output information of the harvester at a specific location at a specific time, for subsequent applications. For example, each output data point corresponds to a location, and these yield measurement data containing location information can be used to obtain the output of a certain area of ​​the plot, perform output data visualization with geographical location, and conduct more output applications based on location-based services (LBS).

[0121] like Figure 3 As shown, the present invention also discloses a multi-sensor information fusion production measurement device 10, used to perform the steps of any of the above methods, including:

[0122] The first yield measurement module 11 is used to perform the first yield measurement using a harvester. The sensors used in the first yield measurement include an acceleration sensor 7 installed on the body of the harvester, a non-contact sensor 5 installed at the first position of the grain conveying mechanism 1 of the harvester, and a contact sensor 6 installed at the second position of the grain conveying mechanism 1.

[0123] The yield measurement data acquisition and processing module 12 is used to acquire yield measurement data and process the yield measurement data to obtain a time series (p1, p2, ..., p) of the first yield measurement value measured by the non-contact sensor 5. i ,…), and the time series of the second production measurement values ​​(q1,q2,…,q) measured by the contact sensor 6. i ...), the acceleration sensor 7 measures the first acceleration simultaneously with the non-contact sensor 5, and the second acceleration simultaneously with the contact sensor 6;

[0124] Time series forecasting model building module 13 is used to construct a time series of the first yield measurement values ​​(p1, p2, ..., p...). i ,…) and the time series of the second yield measurement (q1,q2,…,q i Construct time series forecasting models;

[0125] The second yield measurement module 14 is used to perform a second yield measurement on the grain 3 to be measured. Based on the comparison results of the first acceleration, the second acceleration, and an acceleration threshold a0, it selects the yield measurement method of the non-contact sensor 5, the yield measurement method of the contact sensor 6, and the yield measurement method of the time series prediction model.

[0126] like Figure 4 As shown, the present invention also includes a multi-sensor information fusion yield measurement system 100, including an acceleration sensor 7 installed on the body of the harvester, a non-contact sensor 5 installed on the grain conveying mechanism 1 of the harvester, a contact sensor 6 installed on the grain conveying mechanism 1, a storage unit 120 and a control unit 110. The control unit 110 is connected to the acceleration sensor 7, the non-contact sensor 5, the contact sensor 6 and the storage unit, and also includes the aforementioned device connected to the control unit 110.

[0127] The control unit 110 described above may include a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. The control unit 110 may also be a combination that implements computing functions, such as including one or more microprocessor combinations, a combination of a DSP and a microprocessor, etc.

[0128] like Figure 5 As shown, the present invention also includes a harvester 200, including a body 220 mounted on a chassis 210, a drive device 230 and a grain conveying mechanism 1, and also includes the above-mentioned multi-sensor information fusion yield measurement system 100.

[0129] This invention discloses a storage medium for storing a computer control program, the computer control program being used to execute the steps of any of the methods described above.

[0130] The aforementioned processor-executable computer program may be placed in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, register, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0131] Based on the above, the present invention discloses a method, apparatus, yield measurement system, and harvester for extracting yield tags based on harvesting paths. By combining two yield measurement sensors, the yield at four locations on the same scraper is measured at different times. Based on the vibration amplitude of the harvester body, different yield measurement methods for the four scrapers are selected to overcome the influence of increased harvester vibration at a certain time on the measurement accuracy of the yield measurement sensor and improve the yield measurement accuracy.

[0132] In summary, the present invention may have many other embodiments. Without departing from the spirit and essence of the present invention, those skilled in the art can devise various corresponding changes and modifications based on the present invention, but these corresponding changes and modifications should all fall within the protection scope of the patent application of the present invention.

Claims

1. A method for yield measurement based on multi-sensor information fusion, characterized in that, Includes the following steps: The first yield measurement was conducted using a harvester, and the sensors used in the first yield measurement included an acceleration sensor installed on the body of the harvester, a non-contact sensor installed at a first position of the grain conveying mechanism of the harvester, and a contact sensor installed at a second position of the grain conveying mechanism. Acquire yield measurement data and process the yield measurement data to obtain a time series of the first yield measurement value measured by the non-contact sensor. , , ..., ...), and the time series of the second production measurement value measured by the contact sensor (...), , ,…, The acceleration sensor measures the first acceleration simultaneously with the non-contact sensor and the second acceleration simultaneously with the contact sensor. Based on the time series of the first production measurement ( , , ..., ...) and the second production measurement time series ( , ,…, Construct a time series prediction model; A second yield measurement was performed on the grain to be tested, based on the first acceleration, the second acceleration, and an acceleration threshold. Based on the comparison results, the non-contact sensor production measurement method, the contact sensor production measurement method, and the time series prediction model production measurement method are selected; among them, The grain conveying mechanism adopts a scraper-type elevator; The first position of the grain conveying mechanism is the middle of the scraper-type elevator, and the second position of the grain conveying mechanism is the grain outlet of the scraper-type elevator; There are multiple scrapers between the non-contact sensor and the contact sensor, and the first production measurement value time series ( , , ..., One of the first production measurements (…) Corresponding to the second production measurement time series ( , ,…, The second production measurement value in (,…) The first production measurement value The first acceleration corresponding to the plurality of first accelerations The second production measurement value The second acceleration corresponding to the plurality of second accelerations ; The first production measurement time series used in constructing a time series prediction model ( , , ..., ...) and the second production measurement time series ( , ,…, (,...) is when the acceleration is less than the acceleration threshold. Measured in time; The step of constructing a time series prediction model is based on constructing an autoregressive AR model; The step of constructing an autoregressive AR model further includes: The autoregressive AR model is: Grain yield value on one of the scrapers , The length of the time series of the first production measurement value. The model order is... For model parameters, This represents the model residuals.

2. The method as described in claim 1, characterized in that, Determine the model parameters The steps further include: Model parameters are calculated using the least squares method. Least squares estimation This makes the model residuals The sum of squares is the smallest.

3. The method as described in claim 2, characterized in that, Determine the order of the model The steps further include: in, To pass through model residuals The calculated variance of discrete white noise, Model order The function; By making The model order is obtained when the value is minimized. .

4. The method as described in claim 3, characterized in that, Based on the first acceleration, the second acceleration, and an acceleration threshold The step of selecting the non-contact sensor, contact sensor, and time series prediction model for yield measurement further includes: When the harvester is operating smoothly, select the non-contact sensor and / or contact sensor yield measurement method; When the harvester is not operating smoothly, the yield measurement method of the time series prediction model is selected.

5. The method as described in claim 4, characterized in that, The step of selecting the non-contact sensor and / or contact sensor for yield measurement when the harvester is operating smoothly further includes: When the first acceleration At that time, the grain yield value measured on the scraper by the non-contact sensor is selected as the scraper grain yield value. ,Right now ; when At that time, the grain yield value measured on the scraper by the contact sensor is selected as the scraper grain yield value. ,Right now ; when At that time, the average of the current grain yield values ​​measured by the non-contact sensor and the contact sensor is selected as the grain yield value of the scraper. ,Right now .

6. The method as described in claim 5, characterized in that, The step of selecting the time series prediction model for yield measurement when the harvester operation is unstable further includes: when At the same time, the yield time-series data obtained by the selected yield measurement method during the stable operation of the harvester is acquired. ; The time series prediction model is used to predict the current grain yield on the scraper to obtain the scraper grain yield value. ,Right now .

7. The method as described in claim 6, characterized in that, The time series prediction model is used to predict the current grain yield on the scraper to obtain the scraper grain yield value. The steps further include: The time series prediction model is used to predict the current grain yield on the scraper to obtain the scraper grain yield value. The aforementioned production time series data can be added. This will allow for the estimation of subsequent scraper grain yield.

8. The method as described in claim 1, characterized in that, The process of acquiring and processing the yield measurement data also includes acquiring multiple location information, which is used to associate with the results of the second yield measurement to provide location-based yield services and applications.

9. The method according to any one of claims 1 to 8, characterized in that, The non-contact sensor is a photoelectric sensor, and the contact sensor is an impulse sensor.

10. A multi-sensor information fusion yield measurement device, used to perform the steps of the method as described in any one of claims 1 to 9, characterized in that, include: The first yield measurement module is used to perform the first yield measurement using a harvester. The sensors used in the first yield measurement include an acceleration sensor installed on the body of the harvester, a non-contact sensor installed at a first position of the grain conveying mechanism of the harvester, and a contact sensor installed at a second position of the grain conveying mechanism. The yield measurement data acquisition and processing module is used to acquire yield measurement data and process the yield measurement data to obtain a time series of the first yield measurement value measured by the non-contact sensor. , , ..., ...), and the time series of the second production measurement value measured by the contact sensor (...), , ,…, ...), the acceleration sensor measures a first acceleration simultaneously with the non-contact sensor, and a second acceleration simultaneously with the contact sensor; The time series forecasting model building module is used to build a time series of the first yield measurement value ( , , ..., ...) and the second production measurement time series ( , ,…, Construct time series forecasting models; The second yield measurement module is used to perform a second yield measurement on the grain to be tested, based on the first acceleration, the second acceleration, and an acceleration threshold. Based on the comparison results, the non-contact sensor production measurement method, the contact sensor production measurement method, and the time series prediction model production measurement method are selected.

11. A multi-sensor information fusion yield measurement system, comprising an acceleration sensor mounted on the body of a harvester, a non-contact sensor mounted on the grain conveying mechanism of the harvester, a contact sensor mounted on the grain conveying mechanism, a storage unit, and a control unit, wherein the control unit is connected to the acceleration sensor, the non-contact sensor, the contact sensor, and the storage unit, characterized in that, It also includes the apparatus of claim 10 connected to the control unit.

12. A harvester, comprising a body mounted on a chassis, a drive unit, and a grain conveying mechanism, characterized in that, It also includes the multi-sensor information fusion production measurement system as described in claim 11.

13. A storage medium for storing a computer control program, characterized in that, The computer control program is used to perform the steps of the method as described in any one of claims 1 to 9.

Images