Method for regulating the operation of a machine for harvesting root crops

Abstract

The invention relates to a method for regulating the operation of a machine (2) for harvesting root crops (4) and / or for separating root crops (4) from other harvested material including inclusions (5), wherein at least one test image (8) of the harvested material is recorded by an especially optical image capture unit (6) and at least one operating parameter of the machine (2) is set, wherein the harvested material is moved forward in the conveying direction along a conveying path relative to a frame (12) by means of at least one conveying element (10), wherein at least one further test image (8) is recorded by at least one further, especially optical, image capture unit (6) offset in the conveying direction, and an evaluation device generates at least one operating parameter signal, in particular constituting a setting signal for a separation device, based on a first test data set generated based on or formed by the first test image (8) or based on at least another test data set generated based on or formed by the further test image, wherein at least one operating parameter of the machine (2), in particular the separation device, is set via the operating parameter signal, and the invention relates to a machine for harvesting root crops (4).

Description

Technical Field

[0001] The present invention relates to a method for regulating the operation of a machine for harvesting root crops and / or for separating root crops from other harvested material including foreign matter, and to a correspondingly designed machine. In the method, at least one inspection image of at least a portion of the harvest, which is moved forward relative to a machine frame by means of at least one conveyor element, is recorded by at least one optical image capture unit. Background Art

[0002] The inspection image shows the harvested material previously picked up by a machine for harvesting root crops. Here, a conveying element as part of the machine is used to move the harvested material forward in the machine. At least a portion of the harvested material is in direct contact with the conveying element.

[0003] Publication US 2018 / 0047177 A1 discloses a method in which recorded inspection images are used to calculate the speed of a conveyor element. The actual speed of the conveyor element is then adjusted based on the calculated speed.

[0004] Disadvantages of known methods of this type are that, depending on the harvesting conditions, they can result in significant damage to the root crops or a large amount of debris when the root crops are unloaded from the machine. Therefore, US 2018 / 0047177 A1 generally proposes, in addition, changing the harvesting rate or one or more machine configurations based on a server-based evaluation of three-dimensional harvest data recorded by the machine's sensors. Summary of the Invention

[0005] The object of the present invention is to provide a method in which the protection of root crops is improved while at the same time optimizing the overall machine performance.

[0006] This object is achieved by a method according to claim 1 and by a machine according to claim 33. Further advantages and details of the invention can be gathered from the dependent claims and the following description.

[0007] According to the invention, the object is achieved by a method in which at least one further, in particular optical, image detection unit offset in the conveying direction records at least one further test image, and the evaluation device generates at least one operating parameter signal, in particular constituting a setting signal for a separation device, based on a first test data set generated on the basis of or formed by the first test image and based on at least one further test data set generated on the basis of or formed by the further test image, and at least one operating parameter of the machine, in particular the separation device, is set via the operating parameter signal.

[0008] If an optical image capture unit detects a subsequent area of ​​the conveyor route and depicts it in a further detection image, the optical image capture unit is considered to be offset in the conveying direction. An image capture unit that monitors the conveying area is also referred to below as a measuring station. The machine according to the present invention particularly includes at least one separating device. Accordingly, the method according to the present invention can be used to control at least one separating device of a corresponding machine.

[0009] Because changes in operating parameters, particularly those of separating devices or conveyor elements located along a conveyor route, have an impact on the entire cleaning process and, in particular, on the separation performance of subsequent separating devices, it is advantageous according to the present invention to compare and, in particular, correlate the information contained in the test data sets of the image capture unit along the conveyor route. This correlation can occur as a simultaneous or temporally coordinated setting of operating parameters, and can also take into account dependencies between individual operating parameters, particularly the settings of the separating devices along the conveyor route. The corresponding rule sets are stored in a machine-specific and, accordingly, crop-specific manner in the corresponding evaluation device of the machine so that the evaluation device can automatically implement the desired settings during machine operation, i.e., at least largely and, in particular, completely without operator interaction. For example, if the separation performance of a separating device at the beginning of a conveyor route decreases, it may be necessary to adjust the subsequent separating devices more strictly, i.e., more aggressively, to achieve a greater separation of root crops from the soil, for example. In addition to the operating parameters of the separating devices, the digging depth or travel speed may also be one or more of the operating parameters to be adjusted.

[0010] In particular, the operating parameter signal is generated based on a comparative evaluation of the test data sets, so that the evaluation device takes into account information from at least two test images or test data sets when calculating the operating parameter signal and combines this information with one another for evaluation purposes.

[0011] This ensures, in particular, that the proportion of foreign matter decreases to the desired extent and, in particular, evenly over the path of the harvest through the machine. Consequently, excessive cleaning at the separation device can lead to overloading and excessive wear. At the same time, premature and excessive soil cleaning can, in particular, cause damage to the potatoes, particularly at the end of the screening chain, due to the subsequent lack of or reduced soil cushion. The same applies to other root crops, such as sugar beets. Similarly, removing unwanted foreign matter too late can lead to increased wear along the entire conveyor path. Furthermore, separation devices specifically provided for individual foreign matter may not be able to separate these foreign matter logistically in an ideal manner. This, in turn, leads to additional efficiency losses of the machine and a reduction in the overall digging performance. The method according to the invention ensures that operating parameters adapted to gentle, non-destructive, and simultaneously optimal harvesting and cleaning performance are automatically set.

[0012] According to the present invention, further test images show further areas of the conveying path, in particular of further conveying elements, offset in the longitudinal direction of the conveying path. Preferably, at least two test images are test patterns at the start or inlet and at the end or outlet of the separation device. For example, these could be sections of the screen belt or roller scraper in the conveying direction, in front and behind. Similarly, test images from areas upstream and downstream of the stone separation device can be particularly useful.

[0013] This machine is especially for the self-propelled or towed vehicle in operation that gathers in the harvest root crops, especially potato, beet, carrot or chicory.Alternatively, this machine also can be the stationary machine that is used for the root crops of harvested material and inclusion, for example clod, stone or soil separation.

[0014] During the method, a traveling or towing machine moves forward, in particular in the direction of a row, in particular a bank, of root crops to be harvested, and continuously removes the root crops from the ground as part of the harvest. After the harvest has been removed, at least a portion of the harvest, in particular the root crops and / or any debris, is at least partially moved forward relative to the machine frame by at least one conveyor element. In particular, the conveyor element is designed to be circumferential and to be in the form of a conveyor belt, preferably a screen belt, or a rotating star screen.

[0015] The possibly individually configurable separating elements of the separating device are part of the machine and preferably interact with one or more conveyor elements. Alternatively, the separating device is part of the conveyor element, is at least jointly formed by the conveyor element (e.g., in a screen belt equipped with a vibrating device), or jointly forms one or more conveyor elements (e.g., in a roller desludging machine). During operation, the crop is moved relative to the separating device, imparting a motion pulse to at least one component of the crop, in particular to root crops or foreign matter. The separating device, for example, in the form of a roller desludging machine, is provided with a rotating separating element in the form of a reversing roller, wherein the different components of the crop are at least not moved in the same direction by the separating device.

[0016] The optical image capture unit is particularly arranged in a fixed position on the machine above the corresponding conveyor element and is directed towards the conveyor element and thus, during operation, towards the stream of harvested material or its components, particularly root crops or inclusions, between the image capture unit and the conveyor element. The method according to the invention is particularly carried out during harvesting or separation by machine and is preferably repeated therein.

[0017] The test image is particularly a multidimensional image, preferably a two-dimensional image, depicting at least a portion of the harvested product with root crops, inclusions, and / or a conveyor element. Based on the test image recorded by the image capture unit, a test data set is either generated by the image capture unit or by the evaluation device. Alternatively, the test data set can be formed from the test image itself. This is particularly applicable to image capture units whose test images are already in a format suitable for subsequent analysis in the evaluation device. The test data set is particularly a data set that is provided, at least temporarily, in the system by processing, such as filtering and / or rendering in other ways, and whose information, such as color information, is evaluated in the evaluation device. For example, the test data set can be provided as an image file, a table, a matrix, or a vector field. The test image or the test data set provided by the image capture unit is transmitted by the image capture unit to the evaluation device. The optical image capture unit is particularly designed as a digital camera or video camera for recording the test image in two dimensions. If the subsequent processing of the image information in the evaluation device involves a test image, it can be referred to as a test data set in this context.

[0018] The evaluation device is used to evaluate the test data set. It includes at least one processor and can be used as a central processing unit or as a decentralized system comprising at least one memory and at least one processor located at various locations on the machine components. This is a local system so that any evaluation can be performed directly locally and the results made available directly.

[0019] Operating parameters are variables that relate to the geometry of the separating device or its separating elements, the position or orientation relative to the frame or the conveying element, the speed of the separating device or the drive power or the motor power. The operating parameters can be used to set the manner in which or the extent to which the separating device interacts with the harvest or at least one of its components. In particular, a change in the operating parameters can be used to change how many foreign matter remains in the root crops downstream of the separating device relative to the conveying path along which the root crops are to be conveyed within the machine. The operating parameters are in particular independent of the conveying speed of the conveying element for conveying at least the root crops, which are placed on the conveying element and move in the same direction as the conveying element.

[0020] The operating parameters define in particular how aggressively the separation device operates when separating the root crops and foreign matter. In the case of too low aggression, an excessive portion of the foreign matter is not separated from the root crops. In the case of too high aggression, not only the foreign matter but also the root crops are separated or damaged, thereby reducing the harvest. By generating a separation device setting signal and in particular by sending it to a separation device controller, the operating parameters are preferably set based on the part of the harvest imaged on the test image. The separation device controller in particular increases or decreases the operating parameters as a function of the separator setting signal. For this purpose, the separation device controller in particular outputs an electrical signal or changes the fluid pressure, wherein the separation device controller in particular is part of the same computing unit as the evaluation device.

[0021] This method allows for continuous optimization of the operation of a machine with a separating device. In particular, the screening can be continuously optimized depending on the load rate of the conveyor line, thereby achieving gentle treatment of the root crops along the entire conveyor line and efficient separation of foreign matter from the harvest.

[0022] In particular, the evaluation device evaluates the test data set for multiple isolating devices, locally on the machine or on a tractor vehicle directly connected to the machine, to provide the isolating device setting signal. This allows for almost instantaneous adjustments when undesirable isolating device states are detected, thus preventing blockages, damage, and reduced performance.

[0023] In an advantageous development of the method according to the invention, the evaluation device calculates at least one first component of the respective test image, formed from at least one image region. The at least one image region at least partially depicts a defined component of the harvest or the machine, and the operating parameter signal or operating parameters are set based on the corresponding component of the respective test image. Based on the first component, the component of the respective harvest component is determined, in particular in the monitored area of ​​the conveyor route, and is in particular set to be identical thereto.

[0024] The corresponding harvest components, ie the proportions of successively arranged monitored conveyor path regions resulting from the individual test images, can be compared with one another in a simple manner.

[0025] The setting of one or more operating parameters is carried out in particular as a function of the proportions of all or individual crop components or as a function of values ​​derived therefrom for the individual measuring locations, ie, the areas of the conveyor path captured by the individual image capture units.

[0026] Before calculating the first contribution, the components statistically represented by the first contribution are predefined. The test image and / or test data set is particularly divided into a plurality of image regions, preferably of equal size. The image regions that at least partially display the components together form the first contribution. In particular, the contribution at least partially displays the image regions of the components over the entire image region, wherein the first contribution is formed based on a ratio of the number of image regions or based on their total area.

[0027] The first share is a measure of the size of the image region describing the component and, therefore, a measure of the density of the component in the field of view of the image acquisition unit or the portion to be observed of the test image. The component is, in particular, at least partially a root crop component, whereby the first share at least approximately describes the density of the root crop. If at least 50% to 100% of the area of ​​the image region shows the component, the image region is particularly evaluated as depicting the component and is assigned to the first share. In particular, at least one image region can only be assigned to the first share by share or, preferably, to different shares in each case. This is particularly advantageous if, within the scope of a preferably model-based classification method, it is not possible to uniquely associate an image region with the corresponding component. In this case, the probability of association with the different shares is preferably determined. The image region is particularly preferably assigned to different shares by share or in part according to the probability. This allows the proportions of the components to be depicted more accurately.

[0028] By calculating at least the first fraction, the composition of the harvest is particularly characterized. Based on this, the corresponding operating parameters can be adjusted particularly advantageously, since the cleaning performance of the conveyor element or the separation device including the conveyor element is closely related to the composition of the harvest. In particular, the first fraction represents the density of foreign matter. Therefore, the operating parameters of the associated separation device can be modified by increasing the first fraction to achieve a greater separation effect or performance, thereby reducing the load on subsequent separation devices that may exhibit critical concentrations.

[0029] Preferably, the first portion is at least approximately the portion of the respectively affected harvest component, ie is therefore identical.

[0030] Preferably, at least one image region forming the first component is explicitly identified as a defined component, in particular based on a test sub-data set generated based on the image region. In particular, the image region is identified based on test values, preferably color information, contained in the test image and / or the test sub-data set. Color information includes, in particular, black and white, grayscale, and / or color channel values ​​in a color space.

[0031] Preferably, the test sub-datasets, test values, or color information are classified using a particularly model-based statistical classification method. Thus, if the result of the classification method corresponds to a defined component of the harvested product or machine, in particular, an image region is assigned to the first fraction. The classification method particularly utilizes a neural network, a random forest, a Bayesian classifier, a support vector machine, and / or a decision tree. By using this classification method, the calculation results of the first fraction, in particular the different fractions, are particularly stable and robust with respect to the composition of the harvested product.

[0032] Particularly preferably, the test values ​​or color information are aligned with one or more reference values ​​or reference areas, and based on this, the image area is either included in the first contribution or not. The reference image can preferably be captured by an optical image capture unit, just like the test image, wherein the user can, in particular, displayably mark different parts of the reference image as different components. This form of differentiation allows for particularly reliable identification of the relevant components on the test image. Particularly preferably, at least one of the test values ​​of a test sub-data set, in particular including color information, is compared with at least one reference value, and the image area is only included in the first contribution if at least one test value of the test sub-data set lies at least within an associated reference value range. The reference value range is particularly limited by a maximum value and a minimum value, wherein preferably, different test values ​​must lie within the respectively associated reference value range in order for the image area to be included in the first contribution.

[0033] In an advantageous embodiment of the present invention, upon input of exemplary image regions of a reference image to be included in the first contribution, the evaluation device automatically develops or further develops a model based on the classification method. Alternatively or additionally, upon input of exemplary image regions of a reference image that can be included in the first contribution, the evaluation device automatically calculates or modifies at least one reference value range. Thus, in particular, manual predefinition of reference values, reference value ranges, or models or their model parameters by the user is not necessary. Alternatively, it is sufficient to input at least one exemplary image region of a display component to start the evaluation device. The evaluation device automatically determines at least one reference value, at least one reference value range, or model or its model parameters based on the image region. This allows the evaluation device to be adapted to different applications as autonomously as possible. The greater the number of image regions input, the more precisely the reference values, reference value ranges, or models or their model parameters can be determined.

[0034] The method is particularly stable if the input image region displays components with different brightness and / or soil conditions. Therefore, the method can be used reliably under different application conditions. Particularly preferably, the evaluation device adapts at least one reference value or reference value range during repeated execution of the method, optionally with exemplary identification of the relevant components by an operator, from which training data for the algorithm can be generated.

[0035] In particular, the evaluation device automatically expands the reference data based on additional sensors, such as a brightness sensor for measuring the ambient brightness associated with a test data set recorded essentially simultaneously by the evaluation device. Alternatively or additionally, a user of the method, in particular a driver or operator of the machine or a machine coupled thereto, can manually mark at least one component on the visualized test image in order to expand the reference data of the evaluation device. Thus, based on a one-time user specification or based on data stored in the evaluation device, the evaluation device distinguishes between, for example, potatoes, leaves, stones, soil, and clods, and calculates the corresponding contributions.

[0036] Preferably, the method according to the invention is executed automatically after its start, in addition to the input of any existing training data in the form of component labels. Its control is simpler for the machine driver or operator.

[0037] Preferably, when evaluating the corresponding test image, the image regions forming the first component are additionally identified based on image sub-data sets generated from or formed from adjacent image regions. In particular, color information, particularly color information including black and white and / or grayscale values, contained in the test sub-data sets is used for this purpose. Thus, the image region is evaluated not only based on data associated with the image region, but also using additional data associated with surrounding image regions. This allows for the determination of brightness and / or color profiles, allowing for identification based on a broader data base.

[0038] Preferably, different image regions are weighted differently when calculating the first contribution. Consequently, the image regions contributing to the first contribution vary. This ensures that the first contribution is not calculated solely based on the stereo image of the test image, but rather that image regions showing components of the harvest further from the image capture unit are weighted more heavily than image regions showing components closer to the image capture unit. This allows for the creation of a cleaned stereo image of the first contribution, resulting in a particularly realistic image of the harvest composition on the conveyor element. This is particularly advantageous for comparing test images recorded from different perspectives along the conveyor route.

[0039] Preferably, the entire test image or a continuous test image portion is divided into partial image regions. In particular, the partial image regions each include the same number of pixels of the test image, preferably exactly one pixel. The test image is a portion or segment of the test image that includes multiple partial image regions. To calculate the first contribution, only the image region that represents the test image portion and that is part of the test image portion is considered. To this end, the test image portion is defined so that it depicts a sensitive area within the machine that is to be monitored. Therefore, the image region forming the first contribution includes, in particular, multiple partial image regions of the test image portion.

[0040] The test image or test image portion is particularly gridded into a plurality of partial image regions, each of which is preferably rectangular. Forming a partial image region by precisely one pixel provides a particularly large data base for evaluating the state of the harvest with respect to its individual components, thereby enabling particularly sensitive adjustment of the corresponding operating parameters. At the same time, the data sets provided by conventional 2D digital cameras, typically with a maximum of several million pixels, can be easily and quickly processed by evaluation devices equipped with one or more current processors.

[0041] Preferably, the image capture unit that records the corresponding test images of the paths that follow one another along the conveyor route includes a plurality of test image portions, for which the evaluation device calculates a first portion, in particular a plurality of portions, of the image area, wherein the test image portions (8A, 8B) preferably depict the harvest of different conveyor elements conveyed by the separation device. The test image portions particularly depict the same conveyor element or different sections of different conveyor elements. In particular, the test image portions depict sections of a conveyor element, one of which is arranged before the separation device or its separation element in the conveying direction and another of which is arranged after the separation device or its separation element. Alternatively, the test image portions depict different conveyor elements, which are alternative conveyor routes for different components of the harvest (e.g., a conveyor element for a preferably cleaned harvest stream of root crops and a conveyor element for sorted inclusions). Preferably, the composition of the harvest of the conveyor elements connected to the separation device and thus conveyed by it is determined once for the separated harvest and once for the harvest to be conveyed further. By calculating the first portion and the further portions for these different test image portions, a particularly comprehensive evaluation of the cleaning or separation performance along the entire conveyor route can be made.

[0042] Likewise, the test image portion depicted or present in the corresponding test data set can show a portion of the conveyor element before the separating or diverting element and a portion of the conveyor element after the separating or diverting element. If the image analysis indicates, for example, that a disproportionate amount of root crops is present in an undesirable region after the diverting element, the diverting element can be positioned differently, for example, lower above the conveyor element, thereby improving the diverting performance. The subsequent separating device can then be set more aggressively, i.e., more stringently, to optimize the overall machine, depending on the further setting of the operating parameters along the conveyor path, in order to process a larger quantity of root crops.

[0043] In another embodiment of the invention, the test image portion preferably shows different conveying elements after the separation device, in particular a conveying element for discharging the root crop mixture and a conveying element after the same separation device for discharging the inclusions. For these two test image portions, the respective proportions of the components of the root crops and the inclusions are preferably determined. Alternatively, different proportions are calculated for different test image portions. This makes it possible, for example, to compare the proportion of inclusions in the root crop mixture flow or the harvested flow with the proportion of root crops in the flow of sorted inclusions, and based on this, to adjust the separation elements included in the separation device, for example, with respect to their orientation relative to the conveying elements and / or with respect to their speed.

[0044] Preferably, the image area forming the first share shows the root crop or a part thereof, and the image area forming the second share shows the inclusion or a part thereof. Therefore, the evaluation device calculates at least two different shares for the corresponding test image. The evaluation device particularly preferably calculates at least four shares, a first share for the root crop, a second share for the soil, a third share for the stems and leaves, and a fourth share for damaged root crops. If necessary, at least one further share for stones and / or clods of earth can also be determined. In particular, the sum of the shares is ≤1. Alternatively, the first share can also be the inclusion share, the second share can also be the root crop share, and so on.

[0045] Alternatively or additionally, in another embodiment of the invention, at least two image capture units and at least two conveyor elements are provided, wherein the first image capture unit records a first test image of the harvest portion conveyed by the separation device by means of the first conveyor element, and the second image capture unit records another test image of the harvest portion conveyed by the separation device by means of the second conveyor element, and the separation device setting signal is generated based on at least one, preferably both, of the test data sets generated by forming the two test images or based thereon. The test data sets are evaluated, in particular with respect to the respective proportions, as described above or below.

[0046] The multiple contributions in the evaluation device's calculations allow for a more precise picture of the crop composition or the occupancy of the conveyor elements. This provides a precise picture of the crop composition for different areas of the conveyor path, allowing the evaluation device to accurately and harmoniously adjust the corresponding operating parameters.

[0047] As an alternative to identifying image regions based on limit values, all image regions of the corresponding test image or test image portion are associated with a share. Preferably, the degree of consistency between the test sub-data set calculated based on the image region and the reference sub-data set is evaluated, and each image region is associated with the share with the greatest consistency.

[0048] In an advantageous embodiment of the present invention, the corresponding operating parameter signal, in particular the separating system setting signal, is calculated from a plurality of, in particular temporally sequentially calculated, components, or at least one previously calculated component is included in the calculation or setting of the operating parameter. These measures allow the method according to the present invention to proactively set the operating parameters and to learn during operation.

[0049] In an advantageous embodiment of the present invention, at least one sensor transmits sensor data to the evaluation device, which is included in the calculation of the operating parameter signal. The sensor is, in particular, a sensor for measuring the thickness of the harvest layer on the conveyor element, in particular a touch sensor or ultrasonic sensor, a sensor for measuring the drive power, for example a pressure sensor for measuring the hydraulic oil pressure, and / or a speed sensor, in particular for measuring the speed of the conveyor element drive. In particular, the speed sensor is used to determine the slip of the conveyor element and transmit this slip to the evaluation device in the form of sensor data. The moisture sensor allows further information to be included in the calculation of the separation device setting signal or operating parameter signal.

[0050] Based on this additional information present in the sensor data, which goes beyond that provided by the inspection images, a significantly more precise picture of the cleaning situation of the evaluation device along the corresponding conveyor element exists, which in turn allows the corresponding operating parameters to be influenced better in coordination therewith.

[0051] Preferably, the evaluation device triggers an increase or decrease, in particular of multiple operating parameters, using different separating device setting signals. In particular, the evaluation device or separating device controller or a possible controller for setting further operating parameters comprises a three-point controller, a fuzzy controller, and / or a PID controller, thereby triggering an increase, decrease, or maintenance of at least one current operating parameter, alternatively to one another. An increase is triggered, in particular, if the possible contribution exceeds a predefined first threshold value, and if, in particular, the evaluation of further test images with further adaptation of the operating parameters is unfavorable for a further increase. If the corresponding contribution falls below a predefined second threshold value, a corresponding decrease is triggered, optionally taking into account the evaluation of further test images.

[0052] Preferably, the operating parameter is the distance between two conveying elements or the separation elements of a separating device or the separation device from the conveying element. In particular, the operating parameter is the distance between two conveying rollers of a roller table that rotate during operation. Alternatively, the operating parameter is the distance between a conveying element configured as a screen belt and a separation element configured as a deflecting roller, wherein the separation element extends transversely to the conveying element and causes the root crops to be deflected laterally from the conveying element. In this case, the deflecting roller rotates during operation about an axis of rotation that, when viewed from above, forms an angle of less than 90° relative to the conveying direction of the conveying element. Alternatively, the separation element is configured as a finger belt that rotates during operation, located above the conveying element, and whose outwardly projecting fingers engage with the harvested material disposed on the conveying element during operation. Alternatively, the separation element is configured as a non-rotating stripping device that is located above a thick leaf belt that interacts with the screen belt and causes the root crops to be stripped from the leaves accumulated on the thick leaf belt. The distance can be set in particular by means of a hydraulically or mechanically actuated adjusting device, whereby the aggressiveness of the separating element of the separating device in conjunction with the conveying element or the separating performance of the conveying element can be varied particularly easily.

[0053] Alternatively, the operating parameter or one of the operating parameters is the penetration depth of at least one cutting blade of the machine into the soil. This makes it possible to influence the amount of impurities in the harvest in a simple manner.

[0054] Alternatively or in addition to the above, the operating parameter or one of the operating parameters is the travel speed of the machine or the separating device or the separating element of the separating device, in particular the rotational speed or the rotational speed. In particular, the separating speed is the rotational speed of the aforementioned finger belt or the rotational speed of the aforementioned reversing roller. Alternatively, the separating speed is the rotational speed of an angled separating device that conveys the impurities upward during operation, for example, in the form of a thin leaf belt, the separating device being operated so that the impurities are conveyed upward as much as possible and the root crops are moved downward, opposite to the direction of movement of the section of the separating device facing the root crops.

[0055] Preferably, the operating parameter or one of the operating parameters is alternatively expressed as the angle of attack of the conveying element and / or the separating device, i.e., at least one separating element of the separating device. In particular, the operating parameter is the angle of attack of the separating device, known as the sliver lifter. The angle of attack varies the inclination of the conveying plane of the sliver belt of the separating device relative to the horizontal, thereby adjusting the aggressiveness of the separating device.

[0056] In an alternative advantageous embodiment of the present invention, an operating parameter causes a change in the air flow rate or air mass throughput per time unit. Here, for example, the motor power, represented by the motor speed, can be a corresponding operating parameter for a separation device that separates by air flow. The air, in turn, causes the separation of root crops and impurities, in particular, blowing stems and leaves out of the harvested stream and thus removing them. In this air separation device, which can also be fixedly mounted, the operating parameter can be the speed of the associated blower or the angle of attack of an associated unit in the form of an air guide plate, which, for example, divides the air flow into a main air flow and a transverse air flow.

[0057] In an advantageous embodiment of the method according to the invention, several of the aforementioned operating parameters are set, in particular, by means of identical or different operating parameter signals. To this end, rules can be stored in the evaluation device, which generate corresponding signals for the corresponding adjustable variables for the desired additional or minimum separation performance of the corresponding separation device.

[0058] Preferably, after triggering a change in an operating parameter, no further changes in operating parameters are triggered for a defined time period or a defined conveying route of the conveyor element. In particular, this only applies to the same operating parameter and / or at least one operating parameter of at least one separating device arranged downstream during operation. This ensures that the corresponding operating parameters, for example, of the separating elements, are not over-adjusted and that each change in operating parameters is based on a reliable data basis that already takes into account the preceding changes in operating parameters.

[0059] The signal for setting the separating device is preferably transmitted to the separating device controller by wire, in particular by means of a CAN bus or Ethernet, or wirelessly, wherein the separating device setting can preferably be activated beforehand by an operator via an input at an interface. This allows existing or at least established systems to be used for communication transmission for setting the separating element, and the reliability of the method is increased in particular by the operator receiving the obtained or to-be-implemented setting of the separating device in a displayed manner, in particular in the driver's cab, and activating it at an interface (so-called human interface device, HID) via a corresponding input.

[0060] To implement the method according to the present invention, more than two, preferably three to twelve, image capture units are preferably arranged along the conveyor path of the machine, each capturing one or more test images. The associated test data sets are evaluated in an evaluation device, possibly comprising multiple units, and used to set at least one operating parameter, but in particular, multiple operating parameters. This allows the separation device or excavation depth to be adjusted accordingly for optimal machine performance over the entire conveyor path. It should be understood that the evaluation device may comprise multiple evaluation units, in particular to facilitate timely evaluation of the data provided by the image capture units. To minimize operator burden, the setting of at least one operating parameter is preferably automated. For short conveyor paths, it may also be advantageous to display operating parameter signals in an easily understandable format for the operator, allowing the operator to perform the changes themselves. This display is preferably used solely for operator notification.

[0061] The method according to the present invention can be implemented particularly well if one or more, and preferably all, optical image capture units capture only one- or two-dimensional information. For example, this may involve line scan cameras or digital cameras equipped with two-dimensional sensors. Extensive testing has established that test images captured in two dimensions contain sufficient information for setting the corresponding operating parameters, particularly when information from depth sensors is omitted. Consequently, the algorithms used to evaluate the test image datasets are sufficiently fast to allow on-site evaluation of the acquired data without requiring an external server located remotely from the self-propelled machine or towed machine or its towing vehicle. It is therefore advantageous if an evaluation device evaluates the test dataset locally on the machine or a directly connected towing vehicle. Communication within the machine between the image capture unit and the evaluation device can occur wired. This can occur via a commonly existing CAN bus system, a similar machine network, or via a separate connection between the image capture unit and the evaluation device. Wireless transmission from one part of the system to another can also be used locally, if desired, with a wide variety of technologies available due to the short distances. The technologies (eg Bluetooth, W-LAN, ZigBee, NFC, Wibree or WiMAX, IDA, FSO) can also be used together with wired transmission.

[0062] To achieve a relatively simple and stable control, the method according to the present invention comprises an evaluation device that compares the proportions of the harvest components of conveyor elements arranged successively along the conveyor path, or values ​​derived therefrom, with the respective associated target values ​​and, based on this, generates at least one operating parameter signal. Thus, the optimal value for each conveyor path region imaged by the test image, or the optimal bandwidth for each proportion of the harvest under different conditions, can be stored in the system.

[0063] According to another advantageous embodiment of the present invention, preferably initially without generating an operating parameter signal or at least without automatically changing the operating parameters, the display unit shows the operator the specific individual portions of the harvest components of the conveyor elements arranged successively along the conveyor path, as described above or below, or values ​​derived therefrom, and the associated target values ​​described above or below. If at least one operating parameter signal has been generated, the operating parameters can be activated by the operator or, alternatively, can be directly set by the operator.

[0064] For example, based on a plurality of collected data, target values ​​for the crop composition, in particular the root crop portion or the trash portion, at various conveying locations along the conveyor route can be determined for each machine type. Accordingly, for optimal machine throughput, optimal target values ​​for the bandwidth of the portion at specific locations along the conveyor route can be empirically determined or preset based on different harvesting conditions. In particular, the target values ​​are specific to the root crops and trash and can be selected or preset by the operator, for example, during the operational preparation phase. Different harvesting conditions, such as dry, wet, rocky, or loamy soil, can also be preset for the target values ​​of a selected separating device, for example, along a conveyor route designed as a screen belt.

[0065] Thus, from among the generally adjustable operating parameters, the optimal separation performance and / or a correspondingly meaningful proportion of foreign matter can be pre-set and sought during operation via the conveyor line, preferably automatically. The individual operating parameter signals for the respective separation device or machine can be correlated using a simple and sufficiently intelligent algorithm. For example, such a control device can store information indicating when a separation device positioned earlier in the conveyor line results in a very strong cleaning of the product flow between the inlet and outlet, and the control module in the separation device can then pre-set whether further foreign matter can be conveyed there. Conversely, if a separation device detects that no significant cleaning has occurred between the inlet and outlet, the higher-level cleaning device can pre-set that a small loss of root crops can / must be tolerated in order to improve the overall cleaning performance of the machine. Alternatively or additionally, the control device can also be used to selectively set specific flow ratios at specific flow points in the machine. Thus, for example, the soil cushion on the screen chain can be increased by increasing the digging depth or the travel speed. In particular, at least one operating parameter, preferably a plurality of operating parameters, of the respective adjustable element of the machine is determined with the aid of a neural network, a random forest, a Bayesian classifier, a support vector machine or a decision tree.

[0066] In particular, such a control device can also store information on the extent to which foreign matter should be passed on to the corresponding separation system, or the extent to which product losses, for example in the form of potatoes or sugar beets, can be tolerated. These variables are important input variables for regulating specific individual separation capacity control devices and for calculating operating parameter signals.

[0067] In order to avoid dynamic and correspondingly sudden or frequent changes in the separation unit that could lead to excessive loads on the adjustable system, a range can be assigned to the corresponding target value within which the proportion or a value derived therefrom can be considered acceptable based on deviations from the target value. In this regard, a reasonable algorithm simply strikes a balance between optimal cleaning performance at the system and the consequent impact of the separation device and, if applicable, the sorting device.

[0068] Accordingly, the separating devices previously located in the conveyor route and the proportions associated therewith and originating from the corresponding test images or the values ​​derived therefrom can have different weights for determining the corresponding parameters.

[0069] According to another advantageous embodiment of the present invention, different parameter sets of theoretical values ​​are stored in the evaluation device, in particular satisfying the different conditions described above, and / or the evaluation device can preset different parameter sets so that matching theoretical values ​​exist for the corresponding digging or separation situations.

[0070] For checking the performance of the machine (cleaning), it is advantageous to store the operating parameter signal together with at least one component obtained in a subsequent operation or a value derived therefrom and in particular the associated operating parameter and store them in a database. This applies in particular to operating parameters whose recording allows the effects of operating parameter changes to be visualized later.

[0071] Typically, the control algorithm can be provided with preset or adaptable pause values ​​so that the machine is not continuously controlled. This pause value can also be selected based on the speed of the transported harvest. In particular, after triggering a change in an operating parameter, no further changes in the operating parameter are triggered until the inspection images recorded later along the transport route at least partially reveal the harvest that was picked up from the ground by the machine at the time of the triggering.

[0072] According to the present invention, the aforementioned object is also achieved by a machine for harvesting root crops and / or for separating root crops from other harvested material inclusions. The machine comprises a frame, a conveyor element, at least two, in particular optical, image capture units arranged sequentially along a conveyor path, in particular a separation device, a separation device, and an evaluation device. The machine is used to perform the methods described above or below. The optical image capture unit is particularly suitable for situations where it detects subsequent areas of the conveyor path and images them in other inspection images offset in the conveying direction.

[0073] The evaluation device preferably includes a graphics processor unit, in particular a processor unit based on a GPU (Graphics Processing Unit) or GPGPU (General Purpose Graphics Processing Unit) and / or an FPGA (Field Programmable Gate Array). This form of evaluation device allows for a particularly resource-efficient and, in particular, local evaluation of the test data set. It is understood that an evaluation device designed as an EDV device has other conventional means, such as for power supply, interfaces, and working memory.

[0074] In an advantageous embodiment of the present invention, the machine includes at least one sensor coupled to the evaluation device, in particular a touch sensor or ultrasonic sensor for measuring the thickness of the harvest layer on the conveyor element, a sensor for measuring the drive power, for example a pressure sensor for measuring the hydraulic oil pressure, and / or a rotational speed sensor arranged on the conveyor element. These sensors allow the calculation of a conveying speed signal based on the measured physical variables in addition to the movement characteristic data set, thereby significantly increasing the effectiveness of the variables calculated by the evaluation device and reducing their susceptibility to error. Similarly, a moisture sensor can also provide information that, within the scope of the evaluation device's analysis, helps to adjust one or more of the separation devices.

[0075] According to the methods described above or below, the conveyor route area captured by the corresponding inspection image is analyzed in at least one evaluation device. Preferably, only one central evaluation device is provided for evaluating the data from the image capture unit, but multiple independent evaluation devices can also be associated with the corresponding image capture unit. The evaluation device can then control the associated separating devices, particularly in coordination with the other evaluation devices. In this case, multiple individual units, such as processors, are present. Alternatively or additionally, the central evaluation device is responsible for generating the separating device setting signals and forwarding them to the machine control.

[0076] At least one of the image capture units is preferably configured so that the test image shows at least two alternative conveying paths for different crop components. This allows two conveyor elements to be monitored using the image capture unit, with each test image portion representing a different conveyor element or a section of the crop thereon. In particular, one of the conveyor elements is used to convey sorted debris, while the other conveyor element is used to convey cleaned root crops. This allows for a particularly comprehensive overview of the cleaning performance of the associated separation system.

[0077] One of the image detection units can preferably be arranged so that, during operation, the test image at least partially images at least two conveyor element sections separated by the separating element. In the figure, the conveyor element sections are separated by the separating element only by the test image and are respectively included by the conveyor element. The separating element is closer to the image detection unit than the conveyor element, and the conveyor element is therefore covered by the separating element on the test image. This positioning of the image detection unit makes it possible to calculate at least one first contribution for each of the two separate test image parts and thus directly evaluate the effectiveness of the separating element or the separating device. In particular, for this purpose, the composition of the harvest before reaching the separating element is compared with the composition of at least one portion of the harvest after passing through the separating element.

[0078] Preferably, at least one conveyor element is designed as a screen belt or a rake belt, which, during operation, extends in particular under at least one deflecting roller extending transversely to the conveyor element and deflecting the harvest laterally thereof. Alternatively, the conveyor element is designed as a star screen or conveyor rollers, wherein the conveyor rollers are in particular comprised by a roller table.

[0079] Alternatively or in addition to above-mentioned content, this machine is configured as the machine that is used to clean up and / or sort root crops.Here, this machine is especially fixedly operated during operation, promptly does not have the continuous local forward movement of machine and operates. BRIEF DESCRIPTION OF THE DRAWINGS

[0080] Further details and advantages of the present invention can be derived from the schematically illustrated exemplary embodiments described below. The accompanying drawings show:

[0081] Figure 1 A flowchart showing the method according to the present invention is shown.

[0082] Figure 2 A detailed diagram for determining the harvest composition in a monitored conveyor route area is shown,

[0083] Figure 3 shows a detailed diagram for determining an operating parameter signal,

[0084] Figure 4 A theoretical curve showing the relative composition of the harvested crop along the conveyor route,

[0085] Figure 5 A view showing a test image and its partial evaluation,

[0086] Figure 6 An example diagram showing the relative harvest composition over a monitored conveyor route,

[0087] Figure 7An example diagram shows the relative crop composition above a monitored conveyor route, which results from the method according to the invention,

[0088] Figure 8 An object according to the invention is shown,

[0089] Figure 9 and 10 Show the basis Figure 8 Different side views of an object,

[0090] Figure 11 Show the basis Figure 8 Partial view of the object and conveying elements,

[0091] Figure 12 Show the basis Figure 8 of the equipment Figure 11 Detailed view of the area shown in the middle section,

[0092] Figure 13 Shown from different perspectives Figure 12 objects,

[0093] Figure 14 Show the basis Figure 12 a view of the inspection image of the image detection unit,

[0094] Figure 15 Show the basis Figure 8 The separation device and image detection unit of the machine,

[0095] Figure 16 Shown from Figure 15 Schematic test image recorded from the viewing angle of the image acquisition unit shown,

[0096] Figure 17 Show the basis Figure 8 Another separation device of the machine with an image detection unit,

[0097] Figure 18 Shown from Figure 17 The test image is recorded and schematically illustrated from the viewing angle of the image capture unit shown.

[0098] Figure 19 Show the basis Figure 8 Another detailed view of the machine with another image detection unit,

[0099] Figure 20 Shown from the basis Figure 19 Schematic diagram of the inspection image observed from the viewing angle of the image detection unit,

[0100] Figure 21 A detail view of another device according to the invention is shown.

[0101] If applicable, identical or similarly acting components are provided with the same reference numerals. The individual technical features of the exemplary embodiments described below can also be realized in accordance with the invention in a development thereof in combination with the features of the previously described exemplary embodiments, but always in combination with the features of at least one of the independent claims.

[0102] Items listed in the list of figures are sometimes only partially shown in the individual figures. DETAILED DESCRIPTION

[0103] The method according to the invention is used to regulate the operation of a machine 2 for harvesting root crops 4 (see Figures 6 to 8 In the method, at least one test image 8 is recorded by at least one, in particular optical, image capture unit 6, which shows a harvested product including root crops 4 being moved forward relative to a frame 12 of the machine 2 by means of at least one conveyor element, initially generally designated by reference numeral 10. Furthermore, at least one further test image 8 is recorded by at least one further, in particular optical, image capture unit 6, which is offset in the conveying direction, and an evaluation device generates at least one operating parameter signal, in particular in the form of a separating device setting signal, based on a first test data set generated based on the first test image 8 and at least one further test data set generated based on the further test image, via which at least one operating parameter of the machine 2, in particular of the separating device, is set.

[0104] according to Figure 5 The image shown as test image 8 schematically illustrates only the parts relevant to the present invention, without any boundaries or limits. The image recorded by image capture unit 6, in particular a digital image, may contain additional information not shown in the image. Furthermore, the image shows some details from the evaluation device's analysis for visualization purposes.

[0105] In one embodiment according to the invention, from the material flow 1.1 provided by the excavation device and changing during the conveying route, an assessment of the composition of the harvest is carried out by means of the above-described method before the first separating element, alternatively already during a first separation immediately after receiving the harvest, for example on a first screen belt or roller desludging machine (block 1.2) ( Figure 1 ). In addition, the composition of the harvested stream can also be calculated again at other locations, in particular before the inlet and after the outlet of other separation devices (blocks 1.3 to 1.n). Preferably, detection units are present at least for the root crops 4 at the start and end of the transport route.

[0106] The evaluation determines the respective proportions A1_1 to A1_n, A2_1 to A2_n, A3_1 to A3_n, and A4_1 to A4_n of the root crops 4 and the foreign matter 5 in the form of stems, leaves, soil, and stones at the respective measuring locations 1 to n, i.e., the areas of the conveyor path captured by the image capture unit (blocks 1.4, 1.5, 1.6). Depending on the desired separation performance at the respective separation devices, the proportions of root crops 4 or foreign matter 5 are calculated relative to one another in the evaluation device (block 1.7) and preferably checked for deviations from the target values. From this, control variables are derived for the respective operating parameters, which are determined in block 1.8. Subsequently, operating parameters such as the travel speed, digging depth, and / or the separation device are set (block 1.9). From this, the new harvest throughput is determined (block 1.1).

[0107] exist Figure 2 The determination of the composition of the article flow is shown in more detail in FIG. Starting from the harvested article flux or article flow at the measurement location (block 1.1), a test image 8 is first recorded. In order to create a test data set, the relevant test image parts are then extracted (block 2.1). For this purpose, a mask or region of interest (ROI) can be predefined based on the position of the image detection unit (block 2.2), according to which the distances to be considered and not considered of the test image 8 are distinguished. Based on the relevant image segments of the test image 8 and the test data set available for processing, the proportion of the image area showing the individual harvested article components is now calculated (block 2.3). For this purpose, color information, which in particular includes black and white and / or grayscale values, can be evaluated. The values ​​can be obtained from a reference table or can also be preset by the operator (block 2.4). From this, the proportion A1 of the root crop, the proportion A2 of the soil, the proportion A3 of the stems and leaves, and the proportion A4 of the clods are respectively derived (block 2.5).

[0108] With the method according to the invention, the ratio between product and foreign matter present in the harvested stream is first determined individually for each measuring point along the conveyor line. This ratio is then compared with a theoretical value stored specifically for the respective measuring point in the evaluation device, so that deviations from the desired theoretical value can be determined at the respective monitored conveyor line region. Figure 3 , Box 3.1). The target values ​​or curves are also specific to the respective digging conditions and are constructed in relation to how the machine 2 is operated (see Figure 4 ).

[0109] In addition to determining the deviation from the target values, in block 3.2, the flow path between successive measurement points is evaluated to determine how the material flow components develop depending on the respective separation device settings. The flow path comprises, in particular, the separation devices or separation elements of the separation devices. Based on this, corresponding operating parameters are generated for the entire relevant flow path, taking into account the interaction of the corresponding setting variables (block 3.3). Thus, an optimal soil mat for protecting the root crops 4 can be implemented over the entire flow path, for example, at maximum transport performance.

[0110] The desired setpoint value can be specified in various ways, for example as a tabular value, a function curve or a matrix. Figure 4 Possible scenarios for any harvesting process are shown schematically and by way of example, which can be selected by the operator. The relative material flow composition is given on the y-axis, the conveyor route on the x-axis. The difference between the value of the curve and 100% corresponds to the relative share of root crops. The dash-dotted curve corresponds to balanced operation. Machine 2 performs uniform stripping, so that good product protection is achieved by a soil mat that is evenly reduced along the conveyor route. The lower dashed curve corresponds to a setting scenario in which inclusions are separated as early as possible, the relative share of root crops already rising rapidly at the beginning of the conveyor route. In order to achieve a higher travel speed / harvesting performance, the screening / separation performance is increased according to the presetting in the evaluation unit. Accordingly, the evaluation device reduces the travel speed and, for example, increases the separation performance of the first separation device.

[0111] The solid curve is the setting of the maximum digging performance, wherein so much harvest is picked up at the starting point of the conveying route that it is more strongly separated by the subsequent separation device than passively by the screen belt. In this maximum harvesting flow, the wear of the machine 2 increases.

[0112] The measuring locations distributed along the conveyor route and having the image capture unit 6 each represent a conveyor route segment, for which discrete segments of the conveyor route are monitored, for which discrete target values ​​are preferably also predefined, if necessary from the perspective of the conveyor route. Figure 4 It is derived from the curve.

[0113] Figure 5 In the upper part of the figure, a test image 8 is shown by way of example, which shows the transition from conveyor element 10a to conveyor element 10b. Root crops 4 and inclusions 5, which may include stones and stems and leaves, are located in the conveyor path area. The individual partial image areas 16 are checked for the presence of identical components based on a classifier defined in the algorithm training or predefined via a database, for example, a table with color information, particularly including black and white and / or grayscale values. Thus, based on the comparison of the corresponding image areas with the example in Figure 5The association of the individual shares shown in the lower left of FIG results in a share distribution of the individual shares of root crops and inclusions in the test image 8. Thus, A1 shows the share of root crops 4 in the test image or the corresponding test data set, A2 shows the share of soil, A3 shows the share of stems and leaves, and A4 shows the share of stones (not shown). Preferably, the association is performed based on the color information of the individual pixels, i.e., the image area 19 associated with a share corresponds in particular to a surface of the pixel. This can be similar to Figure 4 The values ​​obtained at each transport route area are shown along the transport route ( Figure 6 ). Based on the measured values, for example, starting from the corresponding flow composition at the individual measuring locations MS1 to MS5, the operating parameters are adapted, with the aim of handling the root crops 4 more gently on the conveyor line due to the larger soil mat. The corresponding adjustment of the preset lower screening at the beginning of the conveyor line is obtained, for example, according to Figure 6 The regulation in the evaluation device preferably takes into account that not all predetermined values ​​are reached exactly and simultaneously, so that a deviation Δ (=Delta) from the desired value is tolerated.

[0114] exist Figure 8 The arrangement of the optical image capture units 6 is disclosed in . The machine 2 according to the invention is designed as a trailed potato harvester, in which a plurality of conveyor elements 10 and their associated separating devices are held by means of a frame 12, which is only partially numbered. Along the conveyor path there are a plurality of image capture units 6 which pick up the harvested material, including the root crops 4, which is transported on the conveyor elements 10. The optical image capture units 6 form the individual measuring locations. In Figure 8 The positions of the image capture unit 6 shown in the figure are the area immediately after the digging device 29 (measuring location MS1), at the transition from the first conveyor element 10A in the form of a mesh belt to the second conveyor element 10B in the form of a mesh belt, and at the transition from the second mesh belt 10B to another conveyor element 10C comprising another separating device (measuring location MS3), which is additionally surrounded by a thick stem and leaf belt (measuring location MS2). Furthermore, at the outlet of the separating device, another image capture unit 6 monitors the conveyor element 10E leading to the sorting table (measuring location MS4), while simultaneously monitoring another conveyor element 10F, which is provided for the remnants of inclusions 5, particularly stones. Finally, another optical image capture unit 6 is located at the sorting table 45 (measuring location MS5).

[0115] The evaluation device can be positioned at any desired location, but preferably at a centrally accessible location near the selection table. Figure 8 The cable 12.1 visible in the drawing supplies information concerning the setting of the separating device or a travel speed signal to the towing vehicle.

[0116] exist Figure 9 and 10 The machine 2 shown in the side view in FIG illustrates the position of the optical image detection unit 6. In particular, the image detection unit 6 at the sorting table 45 can be arranged directly at the step leading to the storage chamber 33.

[0117] Figure 11 and Figure 12 The arrangement of an optical image detection unit 6 is shown, which is arranged on the frame side above the first descending step between the conveying element 10A and the conveying element 10B, with the field of view of the image detection unit pointing downward (measuring position 2). The light source 7 is used to illuminate the field of view to detect a sufficiently illuminated test image 8. The conveying element 10A is a screen belt, which has been screened out of a portion of the inclusions 5, in particular soil and / or clods, from the digging device 29 and is delivered to another conveying element 10B configured as a screen belt via the descending step. The conveying element 10B also has a thick stem and leaf belt, which is provided for separating stems and leaves present in potatoes or in the harvest. Accordingly, the stripping device 32 is provided across the width of the conveying element 10B.

[0118] The height H of the stripping device 32 above the conveying plane of the conveying element 10B can be set with the aid of an operating parameter signal which is configured as a setting signal for the separation device. This makes it possible to influence the separation performance of the separation device configured as a leaf belt. In addition, the relative speed of the screening belt with respect to the coarse leaf belt 43 can be set. Figure 12 For the purpose of overview, only the thick stem and leaf belt 43 is shown without the actual conveying element 10B in the form of a screening belt (see Figure 14 ).

[0119] exist Figure 14 The details are shown in Figure 13 : A test image 8 is obtained from the field of view of the optical image capture unit 6, which is indicated by the dashed line in FIG. Based on the test data set provided by the test image, the above-described evaluation is performed based on the proportion of detected and classified objects.

[0120] Starting from the conveyor element 10B, the harvest still present is transferred to another conveyor element 10C in the conveying direction 1C. A separating device in the form of a plurality of rotating reversing rollers 24 arranged one above the other is associated with the other conveyor element. The harvest ( Figure 15 ).

[0121] To vary the separation performance, the distance H between the conveyor element 10C and the lower deflecting roller 24 can be adjusted and thus presents a configurable operating parameter. If necessary, further distances between the individual deflecting rollers 24 can be varied in terms of their relative spacing to increase the degree of deflection between the leaves and stems drawn between the deflecting rollers 24 or a possible separation function. Alternatively or additionally, the variation in the separation performance or deflection results from the adjustable rotational speed of the deflecting rollers 24.

[0122] Similarly, the height H of the lower end of the separating device designed as a finger belt 26.1, which is associated with the conveyor element 10D, can be set as one of several operating parameters. The height H describes the distance between the fingers 26 and the upper edge of the conveyor element designed as a rake belt. Furthermore, the angle of attack of the finger belt 26.1 relative to the perpendicular to the conveying plane of the conveyor element can be configured to be adjustable. The same applies to the rotational speed of the finger belt 26.1.

[0123] Figure 15 The image detection unit 6 (measurement site MS3) shown in FIG generates Figure 16 , wherein the relevant test image portion 8A in the present exemplary embodiment is limited by filtering or masking. To monitor the performance of the separation system, in this case the separation performance of the reversing roller 24, a test image portion 8B can also be selected, which is located behind the reversing roller 24 as viewed in the conveying direction 1C. In particular, the area before the reversing roller 24 is monitored for the travel speed setting. A test data set is then generated from the corresponding test image portion 8A.

[0124] If the associated theoretical value for test image portion 8A indicates that the separation performance of the upstream or illustrated separation device is too low, the separation device can be set to a more aggressive level depending on the further specifications of the upstream and downstream separation devices. Alternatively, if the corresponding proportion in test image portion 8B or a value derived therefrom indicates an excessively high separation performance, for example due to an excessive proportion of inclusions 5 in the form of clods behind the deflecting roller 24, the spacing H between the deflecting roller 24 and the conveying element can be reduced, and the separation device can be set to a less aggressive level, the excessive proportion at least partially also serving to protect the potatoes on the subsequent conveying path.

[0125] exist Figure 17 and Figure 18 In FIG. 1 , another optical image detection unit 6 is shown, which is arranged in the region of the conveyor belts 10C and 10D. Figure 6 An image detection unit 6 can be used. In particular, the image detection unit is used to monitor the separation and reversing device formed by the reversing roller 24. A light source 7 is also associated with the monitoring unit to better illuminate the monitored area.

[0126] Another optical image detection unit 6 and the associated light source 7 are arranged above the sorting table to observe the conveying element 10E and the conveying element 10F ( Figure 19 ). With the help of the mask, select Figure 20 Test image sections 8A and 8B depicted in the test image 8 of FIG. These test image sections monitor, on the one hand, a conveyor element 10E with a conveying direction 1E for transporting away root crops 4, and, on the other hand, a conveyor element 10F with a conveying direction 1F for transporting away debris 5 in the form of stones and clods of earth, as a separate conveying path. The previously described evaluation is used to check whether the proportion of root crops on conveyor element 10F is too large. If this is the case, the method according to the present invention adjusts the upstream separation device more stringently or more aggressively, depending on the control specifications for the entire conveying path. This separation device is located above conveyor element 10D, which is designed as a rake belt and, in particular, is provided as a finger belt, with fingers 26 (illustrated by way of example and dashed lines), although in the illustrated illustration, they are arranged behind a cover 40 located in front of it. For example, the distance between fingers 26 and conveyor element 10D is reduced to allow more harvested root crops to be transported via the associated chute 41 onto conveyor element 10E. If too many foreign objects in the form of stones are detected on the conveyor element 10E, the rotational speed of the deflecting roller 24 can be changed, for example, so that smaller impulses applied to the foreign objects better deflect any stones in the direction of the conveyor element 10F. The foreign objects then slide better on the slide 42 onto the conveyor element 10F.

[0127] Figure 21 The arrangement of measuring points MS1 to MS5 in the schematically illustrated conveyor path of a machine 2 designed as a sugar beet harvester is shown. An optical image capture unit is located on a roller table 10M after the digging device and at the ends of a conveyor element 10N designed as a screen belt (measuring points MS1 and MS2). Another optical image capture unit 6 monitors, in particular, a conveyor element 10P designed as a star screen (measuring point MS3). Similarly, the subsequent conveyor element 10Q designed as a star screen is monitored, as is the conveyor element 10R designed as an endless elevator (measuring points MS4 and MS5).

Claims

1. A method for regulating the operation of a machine (2) for harvesting root crops (4) and / or for separating root crops (4) from other harvested material including inclusions (5), wherein at least one inspection image (8) of the harvested material is recorded by an optical image detection unit (6) and at least one operating parameter of the machine (2) is set, wherein the harvested material is moved forward in a conveying direction along a conveying path relative to a frame (12) by means of at least one conveying element (10), characterized in that At least one further optical image capture unit (6) offset in the conveying direction records at least one further test image (8), the further test image (8) showing a further conveying area of ​​a further conveying element (10) offset in the longitudinal direction of the conveying path, and an evaluation device generates at least one operating parameter signal based on a first test data set generated on the basis of the first test image (8) or formed by the first test image and based on at least one further test data set generated on the basis of the further test image (8) or formed by the further test image, the at least one operating parameter of the machine (2) being set via the operating parameter signal, the operating parameter signal being set based on A comparative analysis of the first test data set with the further test data set results in the evaluation device calculating at least one first portion (A1) of the corresponding test image (8) formed by at least one image area (19), wherein the at least one image area (19) at least partially images a defined component of the harvest or the machine (2), and setting the operating parameter signal based on the corresponding portion (A1), wherein the corresponding harvest component is determined based on the first portion (A1), wherein the foreign matter portion decreases to a desired extent over the distance of the harvest through the machine, thereby achieving optimal separation performance and / or a corresponding meaningful foreign matter portion during operation via the conveying route.

2. The method according to claim 1, characterized in that The machine (2) is a separation device.

3. The method according to claim 1, characterized in that The operating parameter signal is designed as a separating device setting signal.

4. The method according to any one of claims 1 to 3, characterized in that The image region (19) forming the at least one first component (A1) is explicitly identified as a defined component of the harvest or the machine (2) based on a corresponding inspection sub-data set generated based on the image region (19).

5. The method according to claim 4, characterized in that The image region (19) forming the at least one first component (A1) is explicitly identified as a defined component of the harvest or the machine (2) based on at least one color information item contained in a test sub-data set.

6. The method according to claim 5, characterized in that The test sub-data set is classified by a model-based statistical classification method, and the image region (19) is only assigned to the first portion (A1) if the result of the classification method corresponds to a defined component of the harvest or the machine (2).

7. The method according to claim 6, characterized in that At least one test value included in the test sub-data set is classified by a model-based statistical classification method, and the image area (19) is only assigned to the first portion (A1) if the result of the classification method corresponds to a defined component of the harvest or the machine (2).

8. The method according to claim 6, characterized in that The color information is classified by a model-based statistical classification method, and the image region (19) is only assigned to the first portion (A1) if the result of the classification method corresponds to a defined component of the harvest or the machine (2).

9. The method according to claim 7, characterized in that At least one test value of the test sub-data set is compared with at least one reference value, and an image region (19) is only assigned to the first contribution (A1) if at least at least one test value of the test sub-data set lies within a corresponding reference value range.

10. The method according to claim 9, characterized in that The color information is compared with at least one reference value, and the image region (19) is only assigned to the first contribution (A1) if at least at least one test value of the test sub-data set lies within a corresponding reference value range.

11. The method according to claim 6, characterized in that Upon input of an exemplary image region (19) of a reference image that can be included in the first contribution (A1), the evaluation device automatically further develops the model on which the classification method is based and / or automatically calculates or changes at least one reference value region.

12. The method according to any one of claims 1 to 3, characterized in that When calculating the first component (A1), different image regions (19) are weighted differently.

13. The method according to any one of claims 1 to 3, characterized in that The entire corresponding test image (8) or a consecutive corresponding test image portion (8A) is divided into partial image regions (16), each of which includes the same number of pixels of the test image (8).

14. The method according to claim 13, wherein: The partial image regions each comprise exactly one pixel.

15. The method according to any one of claims 1 to 3, characterized in that The test image (8) comprises a plurality of test image portions, for each of which the evaluation device calculates a first portion (A1) of an image region (19), wherein the test image portions image the harvest of different conveyor elements conveyed away by the separating device.

16. The method according to claim 15, characterized in that The evaluation device calculates a plurality of portions of the image region (19) for each of the test image parts.

17. The method according to any one of claims 1 to 3, characterized in that The image region (19) forming the first component (A1) shows the root crop (4) or a part thereof, and the image region forming the second component shows the foreign body (5) or a part thereof.

18. The method according to any one of the preceding claims 1 to 3, characterized in that The operating parameter signal is calculated as a function of a plurality of temporally sequentially calculated components or values ​​derived therefrom, or at least one precalculated component is included in the calculation of the operating parameter signal.

19. The method according to any one of claims 1 to 3, characterized in that At least one further sensor for measuring the thickness of the harvest layer on the conveyor element, a sensor for measuring the drive power, a rotational speed sensor and / or a moisture sensor transmits sensor data to the evaluation device, which are included in the calculation of the separation device setting signal.

20. The method according to claim 19, characterized in that The at least another sensor is an ultrasonic sensor or a touch sensor.

21. The method according to any one of claims 1 to 3, characterized in that The evaluation device triggers an increase or decrease of the operating parameter by means of a corresponding operating parameter signal.

22. The method according to claim 21, characterized in that After triggering the operating parameter change, no further operating parameter changes are triggered for a defined time period or a defined conveying path of the conveyor element (10).

23. The method according to claim 2, characterized in that The operating parameter or one of the operating parameters is the distance (H) between two conveyor elements (10) or the distance (H) of a separating element of the separating device from the conveyor element (10) or a further conveyor element.

24. The method according to any one of claims 1 to 3, characterized in that The operating parameter / one of the operating parameters is the digging depth and / or the travel speed.

25. The method according to claim 2, wherein The operating parameter / one of the operating parameters is the separation speed of a separation element or of the separation device.

26. The method according to claim 25, characterized in that The separation speed is a rotational speed or a rotational speed.

27. The method according to claim 2, wherein The operating parameter / one of the operating parameters is the angle of attack (α) of the conveying element (10) or of the separating device.

28. The method according to any one of claims 1 to 3, characterized in that The operating parameter / one of the operating parameters is the drive power and / or the motor power and / or the angle of attack of the associated unit.

29. The method according to claim 2, wherein The separation element setting signal is transmitted to the separation element controller by wire; or wirelessly transmitted to the separation element controller, wherein the separation device setting can be enabled in advance by an operator via input on an interface.

30. The method according to claim 29, wherein The decoupling element setting signal is transmitted to the decoupling element controller via a CAN bus or Ethernet.

31. The method according to any one of claims 1 to 3, characterized in that More than two image capture units (6) capture a plurality of test images (8) along a conveying path of the machine and evaluate the associated test data sets in the evaluation device and use them to set a plurality of operating parameters.

32. The method according to claim 31, characterized in that Three to twelve image capture units (6) capture a plurality of test images (8) along a conveying path of the machine and evaluate the associated test data sets in the evaluation device.

33. The method according to any one of claims 1 to 3, characterized in that The operating parameter signal is represented to an operator and / or is automatically used to set the at least one operating parameter.

34. The method according to any one of claims 1 to 3, characterized in that The optical image capture unit only captures one-dimensional or two-dimensional information.

35. The method according to any one of claims 1 to 3, characterized in that The evaluation device evaluates the test data set locally on the machine (2) or on a directly coupled tractor.

36. The method according to any one of claims 1 to 3, characterized in that The evaluation device compares individual crop components or proportions of the values ​​derived therefrom of the crop components of conveyor elements (10) arranged successively along the conveyor path with the respectively associated target values ​​and generates the at least one operating parameter signal based thereon.

37. The method according to claim 36, wherein The stated theoretical values ​​are specific to root crops and inclusions.

38. The method according to claim 36, characterized in that Different parameter sets for target values ​​are stored in the evaluation device and / or different parameter sets can be predefined for the evaluation device.

39. The method according to any one of claims 1 to 3, characterized in that The operating parameter signals and the corresponding portions of the test images (8) obtained in subsequent runs are stored and archived in a database.

40. The method according to any one of claims 1 to 3, characterized in that After triggering an operating parameter change, no further operating parameter changes are triggered until the harvest picked up from the ground by the machine (2) at the time of triggering is at least partially shown by a test image (8) recorded later along the conveying route.

41. A machine for harvesting root crops (4), the machine comprising at least one frame (12), a conveying element (10), at least two optical image detection units (6) and an evaluation device arranged in sequence along a conveying route, and the machine is used to perform a method according to any one of claims 1 to 40.

42. The machine according to claim 41, characterized in that The conveying line has a separation device.

43. A machine according to claim 41 or 42, characterised in that The evaluation device includes a graphics processor unit and / or an FPGA-based processor unit.

44. The machine according to claim 43, characterized in that The graphics processor unit is a GPGPU.

45. The machine according to claim 41, characterized in that At least one sensor coupled to the evaluation device is provided for measuring the thickness of the harvest layer on the conveyor element (10), a sensor for measuring the drive power, a rotational speed sensor arranged on the conveyor element (10) and / or a moisture sensor.

46. ​​The machine according to claim 45, characterized in that The sensor for measuring the thickness of the harvest layer on the conveyor element (10) is an ultrasonic sensor or a touch sensor.

47. The machine according to claim 41, characterized in that The conveying element is a screen belt, a rake belt, a star screen, a conveyor roller (10T), a reversing roller, a stem and leaf belt, a finger belt, a vibration or beating device or an air conveying or air discharge device.

48. The machine according to claim 47, characterized in that The screen belt is a screen belt comprised by an endless elevator.

49. The machine according to claim 47, characterized in that The conveying roller is a conveying roller surrounded by a roller table (10M).

50. The machine according to claim 41, characterized in that At least one of the image capture units (6) is designed such that the inspection image (8) displays at least two alternative transport routes for different harvest components.

51. The machine according to claim 50, characterized in that The inspection image (8) shows at least a conveying path for the root crops (4) and a conveying path for the foreign objects (5).

Images