METHOD AND ARRANGEMENT FOR THE COMPUTER-AID PROCESSING OF AN ELECTRONIC PROFIT CARD

Description

[0001] The invention relates to a method and an arrangement for the computer-aided processing of an electronic yield map in which signals generated by one or more sensors cooperating with harvested crops during the harvesting of a field in a harvesting machine are georeferenced and entered with associated time data. State of the art

[0002] Agricultural harvesting machines are typically equipped with yield mapping equipment. This includes a sensor to record the properties (especially throughput and / or constituents such as moisture content) of the harvested crop within the machine and a positioning device, usually a receiver of signals from a satellite-based navigation system (GNSS). Based on the measured crop properties and the recorded position, yield maps are created, which record, location-specifically, where crops with which properties were harvested in a field. These yield maps can be used for precision agricultural purposes, such as fertilization planning for the next season or for accounting purposes.Furthermore, they can be used for predictive control of the harvesting machine during a subsequent harvesting process or to supply self-learning plant growth models with learning data.

[0003] Due to the design of the harvesting machine, there is a time delay between picking up a harvested plant from its growing location and its interaction with the sensor. In a combine harvester, for example, the plant or its grain must first pass through the header, the inclined conveyor, the threshing and separating unit, the cleaning system, and the grain elevator before it interacts with a typical impact plate sensor at the grain elevator outlet. This transit time is on the order of more than 10 seconds. Similar, though not quite as long, are the time delays in forage harvesters, where yield measurement takes place at the pre-compression rollers or in the discharge chute. Similarly, due to the crop's transit time through the harvesting machine, yields are still measured even when no more plants are being harvested but are still being processed within the machine.Similar problems exist with other harvesting machines, such as cotton pickers and sugar cane harvesters, where harvesting and sensing of the crop are spatially and therefore temporally separated.

[0004] In the prior art, this time delay is assumed to be constant and taken into account when evaluating the measured values ​​(EP 0 960 558 A1, EP 1 180 925 B1, P. Reitz et al., Investigations on a particular yield mapping system for combine harvesters, Computers and Electronics in Agriculture, February 1, 1996, pages 137 to 150). More recent approaches use the speed of the crop conveying system (EP 3 008 990 A2) or predictive sensors (EP 3 772 269 A1) for this purpose and for assigning yields across the width of the header.

[0005] Another possibility, used in the prior art, is to take this time delay into account only during post-processing of the yield maps by generally advancing the start and end times of the yields measured during a pass over the field by predetermined times (JL Ping et al., Processing of Yield Map Data, Precision Agriculture 6 (2005), pages 193 to 212), or to derive these time delays from the yield curves by identifying the stand edges based on rising or falling signals (SO Chung et al., Determining Yield Monitoring System Delay Time with Geostatistical and Data Segmentation Approaches, Transactions of the ASAE Vol. 45 (4), (2002), pages 915-926), or to compare the yield maps with aerial photographs (R. Goncalves Trevisan et al., Improving yield mapping accuracy using remote sensing, accessed on April 5, 2023). https: / / www.preprints.org / manuscript / 201901.0287 / v1).

[0006] In practice, the time delay depends, among other things, on field (topography) and plant characteristics (maturity, grain size, throughput), machine configuration (rotor tines, grates, header type, i.e., lateral conveying by screw or belt) and machine settings (travel speed, rotor speed, setting of skids in the separator, resulting return, etc.), some of which vary over time or depending on location. Task

[0007] Time delays are relatively high, especially with combine harvesters. Furthermore, they depend on various parameters, such as crop characteristics (moisture) and the respective throughput, as mentioned above. A blanket approach therefore leads to errors in the yield maps. Using additional sensors to measure throughput time or an aerial image increases the effort, and retrospectively determining the time delays based on the yield maps still involves uncertainties, at least in assigning the measured yields to specific locations. This is because the yield curve alone can only provide an approximate time when a crop edge was crossed, but not its precise location, as information about the location of the corresponding crop edge is lacking.

[0008] The object underlying the invention is seen as being to provide an improved method and arrangement compared to the prior art for the computer-aided processing of an electronic yield map, in which signals generated by one or more sensors cooperating with harvested crops in a harvesting machine during the harvesting of a field are georeferenced and entered with associated time data, which does not have the aforementioned disadvantages or at least to a reduced extent. Solution to the task

[0009] This problem is solved according to the invention by the teaching of claims 1 and 10, wherein further claims list features which advantageously develop the solution further.

[0010] A method and arrangement for the computer-aided processing of an electronic yield map, in which signals generated by one or more sensors interacting with harvested crops in a harvesting machine during the harvesting of a field are georeferenced and entered with associated time data, comprises the following steps or a computer system programmed for their execution: (a) Identifying the location of one or more headlands or other points with a crop edge in the field using the yield map; (b) Identifying the location of one or more crop edges separating the headland or other point from a part of the field where the harvester has completed at least one pass and picked up crop, using the yield map; (c) Identifying the time or position of the start and / or end of crossing the crop edge while driving the pass using the yield map; (d) Identifying the time or position of the start and / or end of the detection of crop picking by the sensors while driving the pass using the sensor signals recorded on the yield map.(e) Identifying a time delay or spatial offset between the identified time or position of crossing the stand edge and the identified time or position of the start and / or end of crop pickup, and (f) correcting the yield map based on the identified time delay or spatial offset.

[0011] In this way, without the need for additional sensors, the time delay between the pickup of the crop and its sensing is detected and compensated for solely based on the yield map and the track traversed by the harvesting machine recorded therein, along with the associated sensor signals and timestamps. For this purpose, the position of the crop edge at the beginning and / or end of the track is identified. Furthermore, the sensor signals are used to determine where along the track crop was first and / or last picked up. Based on the corresponding timestamps, the time delay or the associated spatial offset is determined, and the yield map is corrected accordingly.

[0012] The crop edge lies at the edge of a headland or another point in the field where a crop edge exists. This could be, for example, a swath intentionally cut through the middle of the field to divide it into beds or to extend the procedure described here to areas beyond the headland. Alternatively, it could be an area where, for other reasons, no crop is present, such as because none was planted there or because topographical factors (watercourse, etc.) prevent crop growth. In the case of a swath, the corresponding location and crop edge can be identified by the path taken by the harvester during harvesting (namely, to create the swath), while an area without crop can be identified by the harvester's path and / or the associated yields.

[0013] The harvesting machine can be a combine harvester or any other harvesting machine that takes in harvested material, e.g. a forage harvester or a baler.

[0014] The time delay caused by the crop's travel time between being picked up by a harvesting header and reaching the sensor leads to a distortion of the yield map, as the sensor signals are only recorded and entered into the yield map with a delay. This distortion of the yield map is compensated for in step (f) such that the generated yield map contains the sensor values ​​for the respective location where the respective plant or its crop components were picked up by the harvesting machine interacting with the sensor.

[0015] The time delay or spatial offset can be determined as a general rule for the entire field, or separately for each lane, or for a portion of the lanes.

[0016] In step (d) the beginning and / or end of the harvesting of the crop can be detected by a change in the sensor signals, in particular by the sensor signals exceeding or falling below a certain limit.

[0017] The sensor can detect the throughput and / or moisture and / or other components of the harvested crop.

[0018] The said procedure can be carried out during or after the harvesting process, in particular by a computer unit of the harvesting machine or a computer unit located away from it, to which the yield map is transmitted wirelessly or by means of a data carrier. Example of implementation

[0019] The drawings illustrate an embodiment of the invention, which is described in more detail below. It shows: Figure 1 a combine harvester with a measuring device for throughput measurement, Figure 2a depiction of a field at harvest time, and Figure 3 a flowchart for generating a yield map.

[0020] The Figure 1 Figure 1 shows a self-propelled harvesting machine in the form of a combine harvester 10 with a chassis 12, which is supported on the ground by driven front wheels 14 and steerable rear wheels 16 and is moved by these. The wheels 14 and 16 are set in motion by means of drive mechanisms (not shown) to move the combine harvester 10, for example, across a field to be harvested. In the following, directional terms such as front and rear refer to the direction of travel V of the combine harvester 10 during harvesting, which is shown in the Figure 1 runs to the left.

[0021] A detachable harvesting header 18, in the form of a cutting unit, is attached to the front end of the combine harvester 10. During harvesting, it harvests crops such as grain or other threshable cereals from the field and feeds them upwards and backwards through an inclined conveyor assembly 20 to an axial threshing unit 22. The mixture, containing grains and impurities, passes through threshing concaves and grates in the axial threshing unit 22 and enters a cleaning unit 26. The grain cleaned by the cleaning unit 26 is conveyed by a grain auger 42 to a grain elevator 48 and from there, via a transition housing 50, to a filling auger 52, which transports it into a grain tank 28. The cleaned grain from the grain tank 28 can be unloaded by a discharge system with a transverse auger 30 and a discharge conveyor 32.The aforementioned systems are driven by an internal combustion engine and controlled and operated by an operator from a driver's cab 34. The axial threshing unit 22 is only one embodiment and could be replaced by a tangential threshing drum with a subsequent straw walker or axial separating devices.

[0022] The cleaning device 26 comprises, in a manner known per se, an upper sieve 44 and a lower sieve 46, which are subjected to an airflow by a blower 40, with the airflow passing through the sieves to the rear and upwards. The size of the sieve openings and the speed of the blower 40 can be changed in a manner known per se by an automatic cleaning setting or by the operator from the driver's cab 34. The mixture discharged from the rear end of the upper sieve 44 is distributed on the field by a chaff spreader or a straw chopper, and the mixture discharged from the rear end of the lower sieve is fed by a return conveyor to a further threshing operation, either by a separate secondary thresher or the axial threshing unit 22.

[0023] The grain discharged by the grain elevator 48 impacts a deflector plate 54, which is pivotally mounted at its upstream end about an axis that can be pivoted transversely to the forward direction against a defined (spring) force. The plate's position is detected by a sensor 56 used for throughput measurement. The position of the combine harvester 10 is determined by a positioning device 60, which receives signals from satellites of a global navigation system (GNSS, such as GPS, Galileo, etc.) and calculates the current position of the combine harvester 10 based on these signals, possibly using local correction signals (DGPS or RTK). The grain elevator 48 is also equipped with a moisture sensor 62 for measuring the moisture content of the grain. This sensor can be located in a bypass or directly within the grain elevator 48.

[0024] During harvesting, a computer 58 stores the position recorded by the positioning device 60, along with the corresponding (clock) times provided by the positioning device 60 and the associated signals from sensor 56 and moisture sensor 62 (georeferenced). The moisture sensor 62 could also be replaced or supplemented by another (NIR) sensor for detecting any constituents. Sensors 56 and 62 can also be located at any other point on the combine harvester 10 to record the throughput and / or constituents of the harvested crop and can be based on any measurement principles. The signals from the positioning device 60 can also be used for the automatic steering of the combine harvester 10.

[0025] The Figure 2Figure 1 shows a schematic representation of a field 100 during harvest. The combine harvester 10 first travels around the outer perimeter of the field, traversing paths 102, 104, 106, and 108. The areas 110 and 112 harvested while traveling paths 102 and 106 serve as headlands when harvesting the central part 114 of field 100. The combine harvester 10 turns on these headlands as it gradually harvests individual tracks 116 across field 100. Unlike the illustration, more paths than the ones shown (102 to 108) can initially be traveled to create headlands 110 and 112, the width of which corresponds to two, three, or more times the width of the header 18. Furthermore, lanes 116 would typically be laid parallel to the longer side of field 100 in order to reduce the number of unproductive turning operations.Furthermore, tracks 116 driven in succession do not have to be directly adjacent to each other; tracks 116 can be skipped and harvested later, which simplifies the turning process.

[0026] The georeferenced signals from sensors 56 and 62, stored by the computer unit 58, contain data on throughput (sensor 56) and moisture and / or other constituents of the harvested crop (sensor 62). These signals do not correlate directly, either temporally or spatially, with the point at which the respective crop was picked up. This is due to the travel time of the harvested crop between being cut by the cutter bar 62 of the header 18, passing through the header 18, the inclined conveyor 20, the axial threshing unit 22, the cleaning unit 26, the grain auger 42, and the grain elevator 48, until it reaches sensors 56 and 62. This travel time can depend on a multitude of factors and is constantly changing. A similar time delay occurs when exiting the field. Furthermore, the position determination unit 60 is spatially separated from the cutter bar 62, but this can be corrected relatively easily by recalculating the position (see Figure 58).EP 0845 198 A1 and EP 0 970 595 A1).

[0027] To solve or at least reduce the aforementioned problem of time delays, the computer system 58, or any other computer, particularly one located away from the combine harvester 10, to which the georeferenced yield data are transmitted wirelessly or via a data carrier, proceeds according to the flowchart of the Figure 3 The input values ​​for the procedure are the georeferenced data from sensors 56 and / or 62, which, together with the associated position data and timestamps acquired by the positioning device 60, serve as input values. Optionally, a map of the field can be supplied to the computer 58, in which the boundaries of field 100 are defined as a polygon or in any other data format.

[0028] In a first step 302, the location of the headland 110, 112 is determined. Based on the tracks 102 to 108 and 116 driven during harvesting of the field, as well as the known width of the harvester header 18, the computer 58 can recognize which of the tracks were initially driven and harvested and later used as headlands 110, 112, i.e., for turning. The map of field 100 can serve to define an outer boundary of field 100 within which the headland 110, 112 should lie.

[0029] In the following step 304, the boundaries of field 100 minus the headland 110, 112, i.e., the boundary coordinates of the central part 114 of field 100, where the harvested crop is transported on tracks 116, are determined. For this purpose, the headland 110, 112 is subtracted from the boundaries of field 100 known from the map or the tracks 102 to 108, leaving the central part 114.

[0030] In the following step 306, the positions of the crop edges 118, 120 are determined based on the position of the central section 114. These edges determine where the combine harvester 10, when driving along the tracks 116 on the central section 114, enters or exits the remaining crop. The crop edges 118, 120 generally run perpendicular to the width of the header 118 (depending on the shape of the field 100).

[0031] The following step 308 represents possible repetitions of the next steps 310 to 316 for one or more additional combine harvesters that may be used together with combine harvester 10 of the Figure 1 and 2 Field 100 has been processed. The following steps 310 to 316 will therefore be repeated for the other combine harvesters, if necessary.

[0032] In the following step 310, the times at which the combine harvester 10 entered and exited the crop in the central part 114 of field 100 on the individual tracks 116 are determined based on the known positions of the crop edges 118, 120. After step 310, it is thus known for each of the tracks 116 at what times the cutter bar 62 of the header 18 began and stopped cutting the crop and the header 18 conveyed the crop.

[0033] In the next step 312, the signals from sensors 56 and / or 62 are analyzed for each of the lanes 116 to determine at what point in time, upon entering the crop in the central part 114 of field 100, the signal first exceeds a predetermined threshold value, i.e., when entering lane 116 at the crop edge 118 or 120, and at what point in time, upon exiting the crop in the central part 114 of field 100, the signal last falls below a predetermined threshold value, i.e., when exiting lane 116, harvested crop is last recorded. At this point, any detection algorithms can also be used that, based on the temporal progression of the signals from sensors 56 and / or 62, recognize that harvested crop was first or last recorded along a lane 116, for example, by fitting a sigmoidal function (see MW).Veal, Enhanced Grain Crop Yield Monitor Accuracy through Sensor Fusion and Post-Processing Algorithms, 2006, University of Kentucky Doctoral Dissertations, 249 (https: / / uknowledge.uky.edu / gradschool diss / 249), there page 51ff).

[0034] The results of steps 310 and 312 make it possible in step 314 to determine the time delay (also referred to in the literature as delay time) between the cutting of the crop by the cutter bar 62 and the detection by the sensors 56 and / or 62 by comparing them with each other.

[0035] In step 316, the time delay determined in step 314 is used to correct the yield map. For example, if a specific time delay delta t occurs during the inbound and outbound phases in step 314, all sensor values ​​must be shifted forward by this time delay delta t. This is easily accomplished using the known positions and times, resulting in yield maps where the time delay is automatically and correctly compensated.

[0036] In steps 314 and 316, there are different ways to account for the time delays, which can vary considerably across the individual tracks 116. A first option is to determine a single, average time delay for the entire field 100. This approach can be useful if there are no significant changes in the combine harvester 10's settings, its driving speed, or the crop properties (yield, moisture) across the field. A second option is to consider the time delays for each of the tracks 116 separately, which can be useful if there are significant changes in the speed and / or settings of the combine harvester 10 and / or the crop properties (yield, moisture) across the field 100. A third option would be a statistical approach, i.e.,In step 314, one identifies those time delays that occur most frequently on the entire field 100 or in sub-areas thereof and uses these or their mean value in step 316 for the entire field 100, or for the respective sub-areas with similar time delays the corresponding time delays.

[0037] If field 100 was harvested by several combine harvesters 10, steps 308 to 316 can be repeated for the other combine harvesters. The yield maps of all combine harvesters can then be combined into a single yield map, if necessary after correction or calibration measures.

[0038] The steps of Figure 3Corrections to the yield map can also be made during the harvesting process by the computer unit 58 or the remote computer mentioned above. The yield map can therefore be corrected during its generation, i.e., during or at least after each pass 116 has been driven. The yield map can be displayed to the combine harvester operator 10 on a suitable user interface and stored in the computer unit 58 or any storage medium (cloud, etc.) for later use. It can then be used, for example, for predictive control of a combine harvester 10 during a later harvest, for fertilizer planning, or as a database for self-learning plant growth models.

[0039] Finally, it should be noted that theoretically, one could also compare the locations of the stand edges 118 and 120 with the locations where, according to the signals from sensors 56 and 62, crop was first or last picked up while driving along track 116, and simply shift the locations corresponding to the signals from sensors 56 and 62 back by the resulting spatial offset. However, this would have the disadvantage that any changes in the combine harvester's speed would cause errors in the yield map. With this approach, the time data is unnecessary and therefore does not necessarily need to be saved with the yield map.

Claims

1. Method for the computer-aided preparation of an electronic yield map, in which signals that are generated by one or more sensors (56, 62), which interact with crop in a harvesting machine during the harvesting of a field (100), are georeferenced and plotted with associated time data, having the following steps: (a) identifying the location of one or more headlands (110, 112) or a different place with a crop edge on the field (100) on the basis of the yield map, (b) identifying the location of one or more crop edges (118, 120), which separate the headland (110, 112) or the different place in the field from a part (114) of the field (100) on which the harvesting machine has travelled on at least one track (116) and picked up crop, on the basis of the yield map, (c) identifying the time or the position of the crossing of the crop edge (118, 120) when driving down the track (116) on the basis of the yield map, (d) identifying the time or the position of the start and / or end of the capture of the picking up of crop when driving down the track (116) by the sensors (56, 62) on the basis of the signals of the sensors (56, 62) plotted in the yield map, (e) identifying a time delay or a spatial offset between the identified time or the position of the crossing of the crop edge (118, 120) and the identified time or the position of the start and / or end of the picking up of crop, and (f) correcting the yield map on the basis of the identified time delay or the spatial offset.

2. Method according to Claim 1, wherein the harvesting machine is a combine harvester (10).

3. Method according to Claim 1 or 2, wherein the time delay is caused by the running time of the crop between the picking up by a harvesting header (18) and reaching the sensor (56, 62) and the corruption of the yield map caused by this time delay is compensated in step (f), so the yield map generated in step (f) contains the sensor values for the respective place at which the respective plant was picked up by the harvesting machine which interacts with the sensor (56, 62).

4. Method according to one of Claims 1 to 3, wherein the time delay or the spatial offset is determined as a whole for the entire field (100) or separately for each track (116) or in each case for a part of the tracks (116).

5. Method according to one of Claims 1 to 4, wherein, in step (d), the start and / or the end of the picking up of the crop is detected on the basis of a change in the signals of the sensor (56, 62).

6. Method according to Claim 5, wherein, in step (d), the start and / or the end of the picking up of the crop is detected on the basis of the signals of the sensor (56, 62) exceeding or falling below a limit.

7. Method according to one of the preceding claims, wherein the sensor (56, 62) senses the throughput and / or the moisture and / or other contents of the crop.

8. Method according to one of the preceding claims, wherein it is carried out during the harvesting process or thereafter.

9. Method according to one of the preceding claims, wherein it is carried out by means of a computer device (58) of the harvesting machine or a computer device that is remote therefrom.

10. Arrangement for the computer-aided preparation of an electronic yield map, in which signals that are generated by one or more sensors (56, 62), which interact with crop in a harvesting machine during the harvesting of a field (100), are georeferenced and plotted with associated time data, having a computer device (58) that is configured for carrying out a method according to one of the preceding claims.