Patent Application
20180122020
The present disclosure generally relates to techniques for obtaining accurate measurements of spatial parameters. More specifically, but not by limitation, the present disclosure relates to correcting post calibration bias in systems configured to measure agronomic parameters.
There are a wide variety of agricultural machines currently in use. Such machines include combine harvesters (combines), planting machines, tillage machines, and nutrient applicators, among others. Many agricultural machines operate not only to perform certain machine functionality, but also to obtain information about the operation being performed.
To obtain this information, machines may use one or more sensors during the operation. These sensors may be calibrated at the time of manufacture or at some time before, during, or after the operation is performed.
Calibration generally refers to a method of accounting for inaccuracies in data measurements. To calibrate a system with one or more sensors, for example, a measurement value is compared to a known value of accuracy to determine a difference between the two. The determined difference is then used to adjust the system so that future data measurements are more accurate.
The discussion above is merely provided for general background information and is not intended to be used as an aid in determining the scope of the claimed subject matter.
A sensing system bias is reduced across a first agricultural machine and a second agricultural machine. A collection of agronomic data is accessed, that is indicative of an estimated crop yield. The collection that is accessed, for example, includes at least a first set of data sensed by the first agricultural machine and a second set of data sensed by the second agricultural machine. In addition, the first and second sets of data can be scaled based on a yield correction factor. A bias between the scaled first set of data and the scaled second set of data is determined, and a smoothing operation is performed on the scaled first and second sets of data. For example, performing the smoothing operation can include generating a calibration correction factor based on the determined bias, removing the bias between the scaled first set of data and the scaled second set of data to obtain a corrected set of crop yield data, and using the calibration correction factor in sensing the first set of data on the first agricultural machine and the second set of data on the second agricultural machine.
This Summary is provided to introduce a selection of concepts in a simplified form that are further described below in the Detailed Description. This Summary is not intended to identify key features or essential features of the claimed subject matter, nor is it intended to be used as an aid in determining the scope of the claimed subject matter. The claimed subject matter is not limited to implementations that solve any or all disadvantages noted in the background.
Agricultural machines may use systems to capture information indicative of an agronomic parameter. Agronomic parameters may include a measurable form of information that relates to the properties of the agricultural operation being performed. For instance, machines may monitor agronomic parameters such as the amount of crop that is harvested, the amount of crop planted or a depth at which planting occurs, the depth of a soil tillage, or the amount or type of nutrients supplied, among others. As such, the term “agronomic parameter” used herein may refer to any measurable form of information such as, but not limited to, crop yield, planting depth, soil moisture level, and crop nutrient properties, among others.
As an example, agricultural combines harvest crops in a field. While harvesting, the combine may use a sensing system that includes a grain yield monitor, which measures a mass flow of crop with an impact-based mass flow sensor or another type of sensor. As such, the sensing system monitors the amount of crop that is being harvested from the field relative to the location at which it was harvested. For purposes of discussion, crop yield generally refers to the yield of crop per unit area of land that is cultivated and is typically measured in tons per hectare (t/ha) or bushels per acre (bu/ac). It is important that crop yield is accurate, as this information can provide valuable insights as to how specific areas of land performed with respect to a certain crop. In addition, crop yield can provide insights into the performance of the agricultural machine and the sensing system.
However, there may be inaccuracies in the measurements of mass flow, and thus the actual amount of crop that is harvested will differ from the sensed amount. It is also noted that similar measurement inaccuracies may occur in other sensing systems that obtain information indicative of any of the agronomic parameters mentioned above, or others.
In an attempt to ensure that sensed measurements are accurate, sensing systems may be calibrated. In typical systems, a calibration is used to adjust a sensed value based on a determined difference between the sensed value and a value that is known to be accurate (or more accurate).
In order to calibrate sensing systems such as a grain yield monitoring system, a series of manual steps is performed by an operator. Operators are often required to start and stop harvesting to manually record accurate measurements for comparison to the sensed measurements. Upon stopping a harvesting operation, an operator may initiate a calibration sequence and briefly continue to harvest a portion of the field in order to obtain sensor information that is indicative of an estimated weight of the harvested crop. Then, the operator moves the harvested crop to an accurate scale to obtain a ground truth weight. The operator then manually enters the ground truth weight, such as a calculated weight, into the calibration system and the system uses the weight to determine a deviation from the sensed crop yield to the actual crop yield. This provides a calibration deviation for a single load. Multiple deviations for multiple loads are often required to obtain an average deviation. The deviation can be used to calibrate the system.
However, this method of calibrating a sensing system may have deficiencies. For one, it can be time consuming. It generally requires the operator to postpone harvesting, and to perform multiple calibration field passes (to obtain multiple loads). In addition, this particular method can result in data inaccuracy. For example, inaccuracies may arise from the operator manually entering crop yield information, using a variety of different machines (e.g. multiple combines harvesting in the same field), variations in an accurate measurement device (e.g. a scale), and a variety of other factors.
Further, typical calibration systems do not allow a calibration adjustment to be applied to previously obtained (e.g. historical) data. This can result in calibration induced offsets in a single machine. That is, these will be an offset in the data values collected pre-calibration (the historical data) and those values collected post-calibration. There may also be calibration induced offsets between multiple machines. For instance, it may be beneficial for an operator to use several machines in a single field as this can decrease the amount of time that it takes to harvest a crop. However, as a calibration is performed on each individual machine, there may be variances between the total estimated crop yield amongst all machines as the sensors on each machine may vary from one machine to the other. Typical calibration operations may not apply the calibration adjustment to other machines.
Thus, there is a need for a system that accurately calibrates sensing systems and automatically corrects any post-calibration deficiencies. It will be noted that some examples of the present disclosure include a system that reduces the amount of manual steps involved in performing a calibration, decreases the operator time that is required to perform a calibration sequence, reduces calibration-induced offsets within a single machine, reduces machine-to-machine calibration bias, and provides more accurate sensed data, which leads to more accurate agronomic maps, improved agronomic decision making, and improved operational performance.
In one example, user interface 108 includes operator input mechanisms and output mechanisms. The output mechanisms can be mechanisms that convey information to operator 166, such as visual display devices, audio devices, haptic feedback devices, etc. In one example, user interface 108 interacts with data visualization system 148 to produce a variety of output mechanisms that are indicative of monitored operations, which will be discussed in further detail below. The operator input mechanisms can include a wide variety of different mechanisms that can be actuated by operator 166 to control and manipulate various systems and subsystems (e.g. controllable subsystems 112) of agricultural machine 100. The operator input mechanisms, for instance, can include levers, steering wheels, pedals, joysticks, buttons, keypads, touch sensitive display devices, and user input mechanisms on user interface displays, among a wide variety of other input mechanisms.
Control system 110 may receive sensor signals from sensors 122 and generate control signals to control the various controllable subsystems 112. It is shown in
Operational parameter monitoring system 118 illustratively identifies an operational parameter associated with each of the sensor signals that is received from sensors 122 and provides that information to control system 110, so that control system 110 can accommodate various levels of signal variability obtained by sensors 122. Sensors 122 can include sensors that are configured to determine operational parameters such as grain mass flow, soil moisture, planting depth, tillage depth, among a variety of others. Sensors 122 may also include a variety of other sensors such as a machine state sensor. It is illustratively shown that operational parameter monitoring system 118 also includes geospatial system 120. Geospatial system 120 includes at least one of a global positioning system (GPS) receiver, a LORAN system, a dead reckoning system, a cellular triangulation system, or other positioning system. In one example, geospatial system 120 is configured to associate signals obtained by sensors 122 with a geographical location, such as a location within a field. As such, a variety of different spatial parameter data may be obtained by sensors 122, geospatial system 120, and identified by system 118. Embodiments described herein may also be configured to perform loss sensing, such as grain loss sensing.
Operational parameter monitoring system 118 further illustratively includes yield monitoring logic 124. Yield monitoring logic 124 is configured to use sensors 122, such as grain mass flow sensors, and geospatial system 120, to estimate a crop yield at various locations in an operation environment (e.g. a field with a planted crop that is harvested). As such, yield monitoring logic 124 may use information obtained by a sensing system to estimate a yield density for a harvesting session. Operational parameter monitoring system 118 may also include moisture monitoring logic 126, and planting depth monitoring logic 128, among other monitoring logic 130. Moisture monitoring logic 126 may use soil moisture sensors (e.g. sensors 122) to estimate a moisture content of soil at various locations in the operation environment. Similarly, planting depth logic 128 may use a planting implement depth sensor (e.g. sensors 122) to estimate an average planting depth at various locations in the operation environment. Other monitoring logic 130 may be configured to monitor any other agronomic parameters discussed herein and use the sensed parameter information in association with the geospatial information.
It is also noted that the information obtained by operational parameter monitoring system 118, along with the other components of machine 100, may be stored in a variety of locations including, but not limited to, memory 104 and/or data store 114. Further, agricultural machine 100 may be in communication with one or more agricultural machines 162 and 164, and/or remote systems 160 over network 158. Network 158 may be any of a wide area network (WAN), local area network (LAN), or a wireless local area network (WLAN), among others.
Agricultural machines 162 and 164 may perform the same or a similar operation as that being performed by agricultural machine 100. For instance, agricultural machines 100, 162, and 164 may include combines that are performing a harvesting operation in different spatial regions of the same field. As such, it may be beneficial to utilize information from each machine in determining accurate measurements and correcting a single machine bias or multi-machine bias. Remote systems 160 may include any other systems relevant to an agricultural operation that is being performed or an agronomic parameter that is being monitored. For instance, remote system 160 includes a remote agricultural management system, other agricultural machines, and an imaging system such as an aerial imaging drone, among others.
In the example shown in
Pre-processing logic 142 is configured to prepare the obtained data for use with bias correction logic 144. At some point, it may be beneficial for the data that is aggregated to be filtered based on one or more filter criteria. The filter criteria may be defined by one or more pieces of filter logic. As such, pre-processing logic 142 includes a calibration data filter 168. Calibration data filter 168 illustratively includes path redundancy logic 170. Path redundancy logic 170 may filter the parameter data obtained by data aggregator 140 by removing any sensor signals from redundant geospatial locations. For example, during harvesting, a combine or agricultural machine 100 may perform a single pass across a field to harvest a crop in a row. In some instances, an operator may be required to perform an additional pass over that same row to harvest any crops that were missed on the first pass. Similarly, an operator may be required to pass over an already harvested row to reach a desired location within the field (e.g., the operator is not harvesting when re-passing over the already harvested row). The information from the subsequent pass can be filtered out.
In addition, or alternatively, pre-processing logic may perform a fusion operation that combines obtained information. For instance, calibration data filter 168 may determine similar sets of information and combine those similar sets to generate a fused set of obtained information.
Parameter data may also be filtered based on a machine state that is determined by machine state logic 172. Machine states can be indicative of the particular condition of a machine during an operation. For example, but not by limitation, machine states can include any of the following; an idle state, an idle to unload state, a field transport state, a road transport state, a harvesting state, a harvesting while unloading state, and headland turn state, among others. Some of the parameter data that is obtained by sensors 122 may not be relevant if the data is obtained during a certain machine state. For example, when an operator performs a headland turn, which is a turn at the end of a row pass to position the machine for a next pass on the adjacent row, a crop yield sensor (such as a grain mass flow sensor) may continue to obtain data during the turn. However, the data obtained during the turn may not include an indication of a crop being harvested as machine 100 may temporarily leave the harvestable area of the field. In addition, or alternatively, parameter data may be filtered based on an identified delay correction, such as a delay correction that is applied due to a change in machine speed that is inconsistent with the rate at which sensor information is captured. Any delay correction that is performed may be identified as a delay period that occurred while capturing sensor information. Machine state logic 172 may, accordingly, identify a set of spatial information that was obtained during a correction period. As such, machine state logic 172 may determine irrelevant parameter data based in part on an identified machine state, and remove said irrelevant data.
It is also contemplated that a variety of other filter logic 174 may be applied to filter the relevant parameter data in accordance with calibration data filter 168. For instance, other filter logic 174 may determine data points that are determined to be statistical outliers within a set.
In addition to filtering the obtained parameter data, pre-processing logic 142 may prepare the obtained parameter data in a variety of other ways. It is illustratively shown in
In the example of sensing crop yield as an agronomic parameter, parameter scaling logic 176 recalculates a yield from a sensed mass flow, machine speed, and header width (e.g. a component of controllable subsystems 112) to remove a yield upper limit. Upon removing a yield upper limit, the yield values may be scaled by parameter scaling logic 176 based on a determined yield correction factor. Once the yield correction factor is determined and the yield values are scaled, logic 176 may then calculate a new yield upper limit and truncate the determined yield values above the new yield upper limit. As such, scaling logic 176 may include a mechanism for determining a more accurate operational parameter by scaling the estimated parameters in relation to one another.
In one example, once the obtained parameter data is prepared by pre-processing logic 142, the prepared data is provided to bias correction logic 144 to reduce a bias determined between the prepared data. It is shown in
When multiple machines are harvesting in the same field, there may be a sensing system bias that occurs from machine to machine. In addition, when there is a single machine harvesting in a field, there may be a sensing system bias between the sensors on that machine. Conventional systems may exhibit a sensing system bias, either internally (single machine) or externally (multiple machines) that reflects a 10-50% difference between sets of sensed operational parameters. It is desirable to have a system that accounts for these determined offsets and provides accurate operational parameter outputs from each machine that is operating.
In one example, bias correction logic 144 provides a post-calibration algorithm that statistically calculates bias from machine to machine, or from sensor-to-sensor on an individual machine, to reduce the bias and output accurate parameter data. Operational parameter monitoring system 118 may obtain multiple sets of parameter data. In an operation where a single machine is used, each set of data is indicative of, for example, a set of parameter data that is obtained in a determined spatial region of the field. For instance, a first set corresponds to data obtained during a first harvesting pass (e.g. harvesting a first set of rows), while a second set corresponds to data obtained during a second harvesting pass (e.g. harvesting a second set of rows). Where multiple machines are used, each set of data that is obtained may be obtained by a different agricultural machine. For instance, a first set may correspond to data obtained by agricultural machine 100, while a second set corresponds to data obtained by agricultural machine 162, and a third set corresponds to data obtained by agricultural machine 164, etc. Each of the individual sets, regardless of whether one or more machines are used, may include parameter data from a localized region of the field. The sets may be compared to identify a bias between the sensing systems or individual sensors on a single machine, and therefore eliminate bias that occurs during measurement of those local regions.
In one example, bias correction logic 144 uses the scaled data sets, as scaled by parameter scaling logic 176. As such, the data sets have already been pre-processed to remove any identified statistical offsets within the sets. Bias correction logic 144 is then configured to analyze the sets with respect to one another, and thus determine a bias between the sets.
A variety of algorithms may be applied to the scaled parameter data with bias correction logic 144, and more specifically with smoothing logic 178. For example, smoothing logic 178 may apply a generalized additive model to smooth the parameter data in accordance with the determined latitude and longitude from geospatial system 120. Alternatively, but not by limitation, smoothing logic 178 may include a localized regression model that is executed to reduce a sensing system bias. In one example, smoothing logic 178 adjusts the yield value for each combine so that the average yield equals the original overall weighted average yield. It is also noted that smoothing logic 178 may be configured to perform a localized smoothing. For example, yield parameter data may be obtained for 1% of a total region of a field. Smoothing logic 178 will recognize that only a specific region of the field has been harvested and execute a smoothing algorithm on the data indicative of the 1% region. Upon executing the smoothing, a localized smoothing metric may be applied to data obtained from other localized regions within the field. As such, a result of performing smoothing on only a portion of the field can be used to smooth the remainder of the field data.
Multi-machine factor logic 180 is configured to determine a weighted offset value between various machines that are operating and obtaining operational parameters. For instance, it may be determined by multi-machine factor logic 180 that between each machine operating in a field, there is a sensing system bias of approximately 3% from machine-to-machine. As such, multi-machine factor logic 180 will determine a yield span across the machines and a deviation from one machine to another. The yield span and deviation may be used to generate a multi-machine factor that is applied with smoothing logic 178.
It is also noted that even when sensors (across a single machine or between multiple machines) have a relatively low bias (e.g., less than +/−2%), this does not guarantee that accurate parameter data will be determined by bias correction logic 144. For instance, there may be transient characteristics within the data. It may be beneficial to identify and remove these transient characteristics to provide a more accurate parameter output. For example, but not by limitation, when two combines (e.g. machines 100 and 164) are harvesting in a single field, each combine may change a forward speed depending on a machine state. The combines may need to slow down to perform a headland turn, move over rough terrain, pass over a slope or a particularly dense portion of the crop in the field, etc. When the forward speed of a combine changes while obtaining parameter data, there is generally a delay correction that is performed. For instance, there may be a delay of approximately 10 seconds between when geospatial data is obtained and when parameter data is obtained. As such, as speed is changed, the delay correction may be inconsistent until a consistent speed is reached. Of course, it is contemplated herein that delay correction can be identified and applied to a variety of channels in addition to a sensed change in speed. Regardless of the sensed channel, the need to correct delay may manifest as a need to attribute the sensed data (e.g. mass flow) to a corresponding geolocation. In conventional systems, these and other delay corrections may show up as inaccurate data parameter outputs when viewing a field map. Specifically, in some maps, these inconsistent delays may be visualized as specks or another visual inaccuracy on a field map. Artifact removal logic 182 is configured to identify and remove such artifacts that are indicative of outliers in the sensed operational parameter data.
Bias correction logic 144 also illustratively includes data verification component 184. In one example, data verification component 184 is configured to incorporate data that is provided by direct observation to further reduce the calibration bias. One example of a source of data that would be provided with a verification component 184 includes an indication of a ground truth. Ground truth may generally refer to an actual observed value of any of the parameters obtained by operational parameter monitoring system 118.
An example of ground truth for yield parameter data includes an actual measured weight of a sample of a harvested load. For instance, when harvesting a crop, the combine may place the harvested grain into a cart with one or more measurement scales (e.g. sensors 122). In another example, the agricultural machine 100 may be driven over a scale. The scales may obtain an indication of measured weight and verification logic 184 will use the indication in determining a calibration adjustment with logic 178. Two particular scenarios will now briefly be discussed. In one scenario, each combine obtains an accurate weight to determine the amount of grain that was harvested by that harvester. Data verification logic 184 then calculates a sum of the weights across all machines. Logic 184 may then compare the sum to the estimated mass flow as provided by yield monitoring logic 124 from each machine. In a second scenario, a single combine is used, and thus the single machine may obtain a final accurate weight measurement when all the field is harvested, or multiple accurate weight measurements that correspond to specific regions of the field. The measurement(s) will be reconciled by data verification component 184 and provided to smoothing logic 178 for reducing the calibration bias. It is noted that a variety of parameters in addition to crop yield may be utilized with data verification component 184.
Before describing these scenarios, it will be noted that
As similarly noted above, there may be several agricultural machines performing the same or similar operations in a field at the same or different times. Calibration of the sensors on each machine may initially occur at the time of manufacture. In addition, an operator 166 may perform manual calibrations with manual calibration component 134. Thus, as agricultural machine 100 performs an operation and operational parameter data is obtained with monitoring system 118, data is unique for each machine that is operating, and there is currently no way to reconcile that data across the machines to provide a consistent output that is representative of accurate parameter monitoring.
A calibration correction system, according to embodiments described herein, may operate in substantially real-time. It is noted that the term “real-time” used herein generally refers to operations that occur concurrently or with respect to one another without a great deviation between a time at which those operations are performed. The processing time does not negate the ability of the system to operate in real-time. As such, multiple machines may operate in the same field at the same time so sensor information and calibration correction is performed by the systems in substantially real-time.
At block 202, it is illustratively shown that multiple machines collect or have collected operational parameter data. Each machine may collect a variety of data that is included but not limited to machine identification, geospatial data, and parameter data. Multiple machines collecting machine identification information is generally indicated at block 216. Each of agricultural machines 100 may include a machine identifier stored at data store 114 and provided to remote systems 116 via communication system 108 and over network 158. In addition, as an operation is being performed, geospatial system 120 may obtain information such as, but not limited to, latitude and longitude at a variety of time points that are indicative of when the operation occurs. Collecting geospatial data with multiple machines is generally indicated at block 218. Depending on the operation that is being performed (such as a harvesting operation, planting operation, tillage operation, or nutrient monitoring operation, among others) each machine will collect parameter data that is indicative of the operation. Collecting parameter data is generally indicated at block 220 in
At block 204, it is illustratively shown that the method includes aggregating the multi-machine operational parameter data. Data aggregator 114 may aggregate the obtained data from operational parameter monitoring system 118 and/or data store 114 in preparation for performing data pre-processing. Aggregating the multi-machine parameter data may include aggregating individual data points into one or more sets, where the sets are defined by the machine that obtained the data, the location at which the data was obtained, the time at which the data was obtained, or other criteria. For instance, the first set of data is sensed by the first agricultural machine during a harvesting operation that is performed relative to a first portion of a harvesting environment, and the second set of data is sensed by the second agricultural machine relative to a second portion of the harvesting environment.
At block 206, the method includes performing data pre-processing. Performing data pre-processing can include executing any of the features discussed with respect to pre-processing logic 142 in
Performing data pre-processing may also include scaling the obtained data, which is generally indicated by block 226. In one example, scaling the obtained data includes scaling each of the first and second sets, individually, based on a determined correction factor corresponding to each of the sets. For instance, a first offset is determined for the data in the first set and a second offset is determined for the data in the second set, and the parameter correction factor is partially based on the determined first and second offsets. Further, the first offset may be indicative of a first average deviation in the first set, wherein the first average deviation is greater than a minimum deviation threshold that is used to perform the scaling. Similarly, the second offset may be indicative of a second average deviation in the second set, wherein the second average deviation is greater than a minimum deviation threshold that is used to perform the scaling.
At block 208 of method 200, it is illustratively shown that the method includes determining that there is a bias between the sensing systems of the multiple machines. In one example, determining there is a bias includes analyzing a deviation between the average parameter output between each machine. For example, calibration correction system 138 determines there is a significant bias between an average deviation of the scaled first set and an average deviation of the scaled second set. If the bias is above a threshold, calibration correction system 138 may determine that there is a significant bias between the machines and that a calibration correction is to be used.
Block 210 of method 200 includes performing a bias correction to smooth the operational parameter data. In one example, bias correction logic 144 executes one or more smoothing functions, one of which can include generalized additive smoothing, as indicated at block 228. Executing a generalized additive smoothing with smoothing logic 178 may include adjusting the obtained operational parameter values (e.g. in the scaled first and second sets) so that the average parameter value equals the original overall weighted average parameter value. In addition, performing a bias correction can include applying a multi-machine factor with multi-machine factor logic 180, which is generally indicated at block 232. The multi-machine factor 230 can include a weighted offset that is determined by multi-machine factor logic 180, which generally indicates an offset from the overall average between each agricultural machine. Further, performing a bias correction may include removing artifacts, as indicated at block 232. For example, artifact removal logic 182 removes transient characteristics in the obtained operational parameter data such as, but not limited to, inconsistencies that result from change of speed delay corrections, data outliers, among others. Performing a bias correction also illustratively includes verifying the data correction, as indicated at block 234. Verifying the data correction may include utilizing data verification component 184 to compare the estimated data correction to a known actual value of the operational parameter. For instance, the ground truth can be determined for a crop yield and that ground truth can be utilized to determine an additional correction offset between an actual crop mass that is harvested and the estimated crop mass.
At block 212, method 200 illustratively includes applying a bias correction to all the relevant machines collecting the operational parameter data. For example, applying the bias correction may include providing a calibration offset adjustment with bias correction logic 144 to calibration system 132. As such, the sensors 122, themselves, can be calibrated or block 212 may include calibration system 132 instructing sensor interface 136 to adjust the sensor signals provided by sensors 122 so any future operational parameter data that is obtained is consistent across all machines, relative to one another.
At block 214, method 200 includes outputting the corrected operational parameter data. Corrected parameter data may be output in a variety of ways. In one example, outputting the corrected parameter data includes providing a corrected field map with field map generator 130, as indicated at block 236. A corrected field map will generally provide a smooth visualization of the output parameter (e.g., crop yield) based on the location at which the operation was performed. This may provide a real-time view for operator 166 that indicates how the machine is performing during operation. Corrected field maps will generally be discussed in further detail below with respect to
Further, calibration correction system 138 may be configured to output corrected metrics, as indicated at block 240. The corrected metrics are indicative of the obtained parameter data as adjusted by bias correction logic 144. For instance, in the example of harvesting with a combine, corrected metrics 240 include an average yield compared to a percent deviation for each of the machines, and also a corrected yield compared to a deviation across an average of all of the machines. Further, outputting corrected operation parameter data can include outputting a calibration correction summary, as indicated at block 242. In one example, the calibration correction summary includes a visual output that compares the original parameter outputs to the corrected parameter outputs and the weighted offset that will be applied to future and/or historical parameter data. Metrics may also include an adjusted mass flow (kilograms per second) and an adjusted productivity (e.g. crop productivity in tons per hour), a combine average yield and a percent deviation from an overall average, along with a span that includes a calculated max percent standard deviation relative to a minimum percent standard deviation, among others. These are examples only.
As similarly discussed above with respect to method 200, a calibration correction system and the associated systems described herein may operate in substantially real-time with respect to a single machine. For one, a standalone, single machine may perform the method of reducing sensing system bias as a direct onboard operation. Second, but not by limitation, a single machine may use a calibration correction system where that machine is processing historical spatial information in combination with information from the single machine.
At block 302, it is illustratively shown that method 300 includes collecting or accessing a collection of operational parameter data. Collecting operational parameter data may include accessing geospatial data, as indicated at block 318, historical operational data, as indicated at block 320, and/or a current parameter data, as indicated at block 322.
Upon collecting or accessing the data, method 300 proceeds to aggregate the data that is associated with that single machine. This is generally indicated at block 304. As similarly discussed above with respect to
Data pre-processing may be performed on the obtained (and aggregated) parameter data, as indicated at block 306. It is noted that performing data pre-processing may include any of the pre-processing features discussed with respect to
At block 308, method 300 includes determining a bias between the first and second sets, such as between current and historical operational parameter data that is obtained with the single machine. Determining a bias may include calibration correction system 138 determining there is a bias between an average deviation of the scaled first set and an average deviation of the scaled second set. Of course, a bias may be determined in a variety of other ways as well.
At block 310, the method includes performing a bias correction to smooth the scaled first and second sets from the machine, such as the current and historical operational sets. The bias correction that is performed with respect to
Performing the bias correction may also include removing artifacts, as indicated at block 334, using artifact removal logic 182 to identify data outliers and transient characteristics in the parameter data. In addition, verifying a data correction is indicated at block 336, which may include using data verification component 184 to compare a ground truth (such as an actual yield as determined by a measurement with an accurate scale) to the scaled sets of parameter data.
At block 314, a bias correction is applied to all the relevant historical parameter data for the machine. Applying the bias correction to historical data may smooth all of the parameter data obtained by the single machine for a specific operation. This generates operational parameters that are accurate for the entire operation period.
At block 316, the corrected operational parameter data is output which may include any of: outputting a corrected field map and infographic, as indicated at blocks 338 and 340, respectively; corrected metrics, as indicated at block 342; and/or a calibration correction summary, as indicated at block 344; or any of the other outputs discussed with respect to
Prior to describing the operational control insights that are provided by calibration correction system 138, one example of an agricultural machine 100 will first be described.
In operation, and by way of illustration only, combine 400 illustratively moves through a field and in a direction generally indicated by arrow 446. As it moves, header 402 engages the crop that is harvested and gathers the crop towards cutter 404. Once the crop is cut, the crop is moved by a conveyor in feeder house 406 toward feed accelerator 408 which accelerates the crop into thresher 410. The crop is threshed by rotor 412 rotating the crop against concave 414. The threshed crop is moved by a separated rotor and separator 416 where some of the residue is moved by discharge beater 426 towards the residue subsystems 438. It can be chopped by residue chopper 440 and spread on the field by spreader 442. In other implementations the residue simply dropped in a windrow instead of being chopped and spread.
Grain falls to cleaning shoe (or cleaning subsystem) 418. Chaffer 422 separates some of the larger material from the grain and sieve 424 separates some of the finer material from the clean grain. Clean grain falls to an auger in clean grain elevator 430, which moves the clean grain upward and deposits it into clean grain tank 432. Residue can be moved from the cleaning shoe 418 by airflow generated by cleaning fan 420. That residue can be moved rearwardly in combine 400 toward the residue handling subsystem 438.
Cleaning shoe loss sensors 452 illustratively provide an output signal indicative of the quantity of grain loss by both the right and left sides of the cleaning shoe 418. In one example, sensors 452 are strike sensors which count grain strikes per unit of time (or per unit of distance traveled) to provide an indication of the cleaning shoe grain loss. The strike sensors for the right and left sides of the cleaning shoe can provide individual signals, or a combined or aggregated signal. It will be noted that sensors 452 can comprise only a single sensor as well, instead of separate sensors for each shoe.
Separator loss sensors 448 provides a signal indicative of grain loss in the left and right separators. The sensors associated with the left and right separators can provide separate grain loss signals or a combined or aggregate signal. As can be done using a wide variety of different types of sensors as well, it will be noted that separator loss sensors 448 may also comprise only a single sensor, instead of separate left and right sensors.
It will also be noted that the sensors described with respect to combine 400 (in addition to the sensors already described with respect to machine 100) can include other sensors as well. For instance, it may include a moisture sensor that is configured to sense the moisture level of the material that is passing through combine 400, and/or sense the moisture level of the soil that combine 400 passes over during operation. Combine 400 may also include a machine state sensor 462 that is configured to sense whether combine 400 is configured to drop the residue, drop a windrow, or perform another machine operation. They can also include cleaning shoe fan speed sensors that can be configured proximate fan 420 to sense the speed of the fan. They can include machine setting sensors that are configured to sense the various configurable settings on combine 400. They can also include a machine orientation sensor that can be any of a wide variety of different types of sensors that sense the orientation of combine 400. For instance, the sensed orientation may identify the orientation of the combine 400, the position of parts of combine 400 relative to other parts, or the position of the parts relative to the ground, etc. Another example of machine orientation sensors includes a sensor that senses the height of a header 402 above the ground. Further, crop property sensors can sense a variety of different crop properties such as crop type, crop moisture, and other crop properties. Other crop properties may include different grain constituents such as, but not by limitation, oil, starch, and protein properties. More particularly, the crop property sensors may sense characteristics of the crop as they are being processed by machine 400, for example as the crop is being passed through a grain elevator 430. One particular example of a crop property sensor includes a mass flow rate sensor 464 that senses the mass flow rate of a crop through elevator 430, or provides other output signals indicative of similar variables.
As such, the size and number of sensors used with the systems described herein may vary. In addition, or alternatively, inferred measurements may be obtained from virtual sensors. In one example, virtual sensors include combinations and series of combinations of communication between related inputs, commands, and actual sensors that, together, provide said inferred measurements.
To further illustrate, it can be generally seen near portions 502 and 504 that a high crop yield was originally sensed with monitoring system 118. To the contrary, it is generally shown in portions 506 and 508 that a relatively low crop yield was sensed by system 118. A sensed medium crop yield was generally sensed at the area represented by reference numeral 503. Such a high degree of variability as presented in a map can be confusing when the map is viewed by operator 166. Further, a group of transient characteristics is generally indicated by the cluster of specks within the circle referenced as numeral 501. Plot 500 provides little insight as to the performance of the crop as well as the performance of the machine in harvesting the crop.
Field plot 510 provides an accurate field map, as generated by field map generator 150, that allows an operator 166 to glean insights relating to the operation of agricultural machine 100 and the performance of the particular crop that is harvested. Of course, it is noted that map 510 may be also representative of any of the other previously discussed operational or agronomic parameters herein.
Other soil conditions may also be monitored by machine 600 or any other agricultural machine described herein. These other soil conditions may include, but are not limited to soil moisture, soil temperature, organic matter composition, soil nutrient levels, bulk soil density, and planter down pressure with respect to a ground surface, among others.
Other example agricultural machines (or any of the machines described above), in accordance with embodiments described herein, may include sprayers or applicators for applying, for instance, fertilizers, pesticides, or other nutrient formulas. Operational parameter monitoring may be performed to obtain information regarding the rate at which said formulas are applied. As such, the calibration correction system and methods described herein may be used accordingly in determining a proper calibration and reducing calibration bias for chemical application operations, and specifically for operations that include measuring the rate at which a formula is applied during operation. As such, the agronomic data that can be monitored and adjusted with embodiments of a calibration correction system discussed herein includes data associated with grain moisture, grain loss, grain quality, residue yield, residue quality, unthreshed properties, chemical application rate, soil moisture, soil temperature, organic matter composition, soil nutrient levels, bulk soil density, and planter down pressure with respect to a ground surface, among others.
The present discussion has mentioned processors and/or servers. In one example, the processors and servers include computer processors with associated memory and timing circuitry, not separately shown. They are functional parts of the systems or devices to which they belong and are activated by, and facilitate the functionality of the other components or items in those systems.
Also, a number of user interface displays have been discussed. They can take a wide variety of different forms and can have a wide variety of different user actuatable input mechanisms disposed thereon. For instance, the user actuatable input mechanisms can be text boxes, check boxes, icons, links, drop-down menus, search boxes, etc. They can also be actuated in a wide variety of different ways. For instance, they can be actuated using a point and click device (such as a track ball or mouse). They can be actuated using hardware buttons, switches, a joystick or keyboard, thumb switches or thumb pads, etc. They can also be actuated using a virtual keyboard or other virtual actuators. In addition, where the screen on which they are displayed is a touch sensitive screen, they can be actuated using touch gestures. Also, where the device that displays them has speech recognition components, they can be actuated using speech commands.
A number of data stores have also been discussed. It will be noted they can each be broken into multiple data stores. All stores can be local to the systems accessing them, all can be remote, or some can be local while others are remote. All of these configurations are contemplated herein.
Also, the figures show a number of blocks with functionality ascribed to each block. It will be noted that fewer blocks can be used so the functionality is performed by fewer components. Also, more blocks can be used with the functionality distributed among more components.
In the example shown in
It will also be noted that the elements of
Under other embodiments, applications can be received on a removable Secure Digital (SD) card that is connected to an interface 15. Interface 15 and communication links 13 communicate with a processor 17 (which can also embody processor 102 from
I/O components 23, in one embodiment, are provided to facilitate input and output operations. I/O components 23 for various embodiments of the device 16 can include input components such as buttons, touch sensors, optical sensors, microphones, touch screens, proximity sensors, accelerometers, orientation sensors and output components such as a display device, a speaker, and or a printer port. Other I/O components 23 can be used as well.
Clock 25 illustratively comprises a real time clock component that outputs a time and date. It can also, illustratively, provide timing functions for processor 17.
Location system 27 illustratively includes a component that outputs a current geographical location of device 16. This can include, for instance, a global positioning system (GPS) receiver, a LORAN system, a dead reckoning system, a cellular triangulation system, or other positioning system. It can also include, for example, mapping software or navigation software that generates desired maps, navigation routes and other geographic functions.
Memory 21 stores operating system 29, network settings 31, applications 33, application configuration settings 35, data store 37, communication drivers 39, and communication configuration settings 41. Memory 21 can include all types of tangible volatile and non-volatile computer-readable memory devices. It can also include computer storage media (described below). Memory 21 stores computer readable instructions that, when executed by processor 17, cause the processor to perform computer-implemented steps or functions according to the instructions. Processor 17 can be activated by other components to facilitate their functionality as well.
Computer 810 typically includes a variety of computer readable media. Computer readable media can be any available media that can be accessed by computer 810 and includes both volatile and nonvolatile media, removable and non-removable media. By way of example, and not limitation, computer readable media may comprise computer storage media and communication media. Computer storage media is different from, and does not include, a modulated data signal or carrier wave. It includes hardware storage media including both volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data. Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium which can be used to store the desired information and which can be accessed by computer 810. Communication media may embody computer readable instructions, data structures, program modules or other data in a transport mechanism and includes any information delivery media. The term “modulated data signal” means a signal that has one or more of its characteristics set or changed in such a manner as to encode information in the signal.
The system memory 830 includes computer storage media in the form of volatile and/or nonvolatile memory such as read only memory (ROM) 831 and random access memory (RAM) 832. A basic input/output system 833 (BIOS), containing the basic routines that help to transfer information between elements within computer 810, such as during start-up, is typically stored in ROM 831. RAM 832 typically contains data and/or program modules that are immediately accessible to and/or presently being operated on by processing unit 820. By way of example, and not limitation,
The computer 810 may also include other removable/non-removable volatile/nonvolatile computer storage media. By way of example only,
Alternatively, or in addition, the functionality described herein can be performed, at least in part, by one or more hardware logic components. For example, and without limitation, illustrative types of hardware logic components that can be used include Field-programmable Gate Arrays (FPGAs), Application-specific Integrated Circuits (e.g., ASICs), Application-specific Standard Products (e.g., ASSPs), System-on-a-chip systems (SOCs), Complex Programmable Logic Devices (CPLDs), etc.
The drives and their associated computer storage media discussed above and illustrated in
A user may enter commands and information into the computer 810 through input devices such as a keyboard 862, a microphone 863, and a pointing device 861, such as a mouse, trackball or touch pad. Other input devices (not shown) may include a joystick, game pad, satellite dish, scanner, or the like. These and other input devices are often connected to the processing unit 820 through a user input interface 860 that is coupled to the system bus, but may be connected by other interface and bus structures. A visual display 891 or other type of display device is also connected to the system bus 821 via an interface, such as a video interface 890. In addition to the monitor, computers may also include other peripheral output devices such as speakers 897 and printer 896, which may be connected through an output peripheral interface 895.
The computer 810 is operated in a networked environment using logical connections (such as a local area network—LAN, or wide area network WAN) to one or more remote computers, such as a remote computer 880.
When used in a LAN networking environment, the computer 810 is connected to the LAN 871 through a network interface or adapter 870. When used in a WAN networking environment, the computer 810 typically includes a modem 872 or other means for establishing communications over the WAN 873, such as the Internet. In a networked environment, program modules may be stored in a remote memory storage device.
It should also be noted that the different examples described herein can be combined in different ways. That is, parts of one or more examples can be combined with parts of one or more other examples. All of this is contemplated herein.
Example 1 is a method of reducing a sensing system bias across a first agricultural machine and a second agricultural machine, comprising:
Example 2 is the method of any or all previous examples, wherein scaling the first and second sets of data comprises:
Example 3 is the method of any or all previous examples, wherein scaling further comprises:
Example 4 is the method of any or all previous examples, wherein determining the first and second offsets comprises, respectively:
Example 5 is the method of any or all previous examples, wherein the first set of data is sensed by the first agricultural machine during a harvesting operation that is performed relative to a first portion of a harvesting environment, and wherein the second set of data is sensed by the second agricultural machine during a harvesting operation that is performed relative to a second portion of a harvesting environment.
Example 6 is the method of any or all previous examples, wherein performing the smoothing operation, according to the smoothing logic, comprises executing an additive smoothing function with respect to the estimated crop yield data.
Example 7 is the method of any or all previous examples, wherein performing the smoothing operation further comprises:
Example 8 is the method of any or all previous examples, wherein the at least one transient characteristic is indicative of a delayed speed correction.
Example 9 is the method of any or all previous examples, wherein performing the smoothing operation further comprises:
Example 10 is the method of any or all previous examples, wherein the indication of ground truth identifies an actual crop yield measurement.
Example 11 is a method of correcting a sensing system bias in an agricultural machine, comprising:
Example 12 is the method of any or all previous examples, wherein the first set includes agronomic data that is sensed during a first pass of the agricultural machine in an operation environment, and wherein the second set includes agronomic data that is sensed during a second pass of the agricultural machine in the operation environment.
Example 13 is the method of any or all previous examples, wherein performing the smoothing operation comprises executing at least one of:
Example 14 is the method of any or all previous examples, wherein the agronomic data is indicative of at least one of:
Example 15 is the method of any or all previous examples, wherein adjusting the first and second sets based on a parameter correction factor comprises:
Example 16 is the method of any or all previous examples, wherein the visual data representation comprises a corrected field map view.
Example 17 is the method of any or all previous examples, further comprising:
Example 18 is an agricultural machine comprising:
Example 19 is the agricultural machine of any or all previous examples, further comprising:
Example 20 is the agricultural machine of any or all previous examples, wherein the bias correction logic comprises a data verification component, and wherein the data verification component is configured to:
Although the subject matter has been described in language specific to structural features and/or methodological acts, it is to be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or acts described above. Rather, the specific features and acts described above are disclosed as example forms of implementing the claims.
