EA042191B1 - METHOD OF ADJUSTING OPERATION OF MACHINE FOR HARVESTING ROOT CROPS - Google Patents

Description

The invention relates to a method for controlling the operation of a machine for harvesting root crops and/or separating root crops from the rest of the heap, as well as to a corresponding machine. When implementing this method, at least one optical recording device takes (registers) at least one control image of at least one part of the heap, moved by at least one transporting body relative to the machine frame.

The control image shows a heap that was previously picked up (picked up) by a root harvester or loaded into it. In this case, the transporting body as part of the machine serves to move the heap within the machine, and at least part of the heap is in direct contact with the transporting body during its movement.

US 2018/0047177 A1 discloses a method in which a captured reference image is used to calculate the speed of a transport organ. The actual speed of the conveying member is then corrected based on this calculated speed.

The disadvantage of this known method, which has a similar purpose, is that, depending on the conditions of their harvesting, significant damage may be caused to the root crops, or a large amount of impurities may be contained among the root crops unloaded from the machine. Therefore, in addition, in US 2018/0047177 A1, it is generally proposed, depending on the results of the server analysis of three-dimensional heap data obtained by the sensors of the machine, to change the harvesting speed or one or more machine configurations.

The objective of the present invention is to develop a method that provides more careful handling of root crops and the separation of the largest possible proportion of impurities from root crops.

According to the invention, this problem is solved in the method of the above purpose, in which the data analysis device, on the basis of a control data set generated using the reference image or formed by it, generates a separation device control signal for automatically adjusting at least one operating parameter of the separation device of the machine .

The machine is a self-propelled or trailed vehicle for harvesting root crops, in particular potatoes, beets, carrots or chicory. When implementing the method, the machine moves, in particular, in the direction of rows, in particular beds of harvested root crops, and continuously captures a pile of root crops, removing it from the ground. After capturing the heap, at least one part of it, in particular root crops and/or impurities, is at least partially moved by at least one transporting body relative to the machine frame. The conveying body is, in particular, circulating and is made in the form of a belt, preferably screening, conveyor or is made in the form of a rotating screening star.

Alternatively, the machine may also be a machine for separating the root crops from impurities contained in their heap, such as lumps, stones or soil.

The separating device, with separating elements that can be individually adjusted in one way or another, is part of the machine and preferably cooperates with one or more conveying elements. Alternatively, the separating device is part of, at least partly formed by, the conveying body (for example, in the case of screen conveyors equipped with agitators) or partly forms one or more conveying bodies (for example, in the case of roller soil separators). During operation, due to the movement of the heap relative to the separating device, at least one component of the heap, in particular root crops or impurities, is affected by a movement impulse. The separating device, for example in the form of a roller soil separator, is provided, in particular, with rotating separating elements in the form of rollers, wherein the different components of the heap are moved by the separating device at least not in the same direction.

The optical recording device is fixedly mounted on the machine, in particular above the transporting body, and is directed to the transporting body, and accordingly to the flow of the heap or its component, in particular root crops or impurities, supplied during operation, located between the shooting device and the transporting body. The method according to the invention is carried out in particular during harvesting or the separation of the heap by the machine and is preferably repeated in this case.

The reference image is, in particular, a multi-dimensional, preferably two-dimensional, image, which displays at least one part of the heap with root crops and impurities and/or a transport organ. Based on the reference image taken by the camera, either the camera itself or the data analysis device generates a control data set. Alternatively, the reference data set may be formed by the reference image itself. This applies, in particular, to imaging devices that record control images immediately in a format suitable for subsequent analysis in the analysis device.

- 1 042191 data. A control data set is, in particular, a data set generated by processing, for example, filtering and/or other types of display, and at least temporarily present in the system, whose information, for example color information in the form of one or more color characteristics, is analyzed in the device. data analysis. Such a data set may be contained, for example, in the form of a graphic file, a table, a matrix, or a vector field. The control image or the control data set created immediately in the imaging device is transferred from the imaging device to the data analysis device. The optical pickup device is designed, in particular, as a digital photo or video camera for capturing a reference image in two dimensions or as a single-line camera. If, in the following, reference image is referred to in connection with the processing of graphical information in the data analysis device, the reference data set may be referred to in this context.

The data analysis device is used to analyze (evaluate) the control data set. The data analysis device contains at least one processor and is made either in the form of a central processing unit or in the form of a decentralized system that includes at least one processor and at least one memory device located in different places on the machine components. Thus, we are talking about a local system, which allows you to process data directly at the place of their receipt and immediately (directly) provide the results of such processing for their use.

The operating parameter is a variable relating to the geometry of the separating device or its separating element, to its position or orientation relative to the frame of the machine or conveying element, to the speed of the separating device and/or to the power of the engine. Through the operating parameter, it is possible to regulate how or to what extent the separating device interacts with the heap or at least one of its components. In particular, by changing the operating parameter, it is possible to vary how much impurities will remain with the root crops behind the separating device, in relation to the transport path along which the root crops must be transported within the machine. The operating parameter is in particular independent of the speed of the conveying member which serves to deliver at least the root crops when they lie on the conveyor member and move in the same direction as the conveyor member.

In accordance with the invention, the operating parameter is how aggressively (ie vigorously) the separating device acts in separating root crops and impurities. If the aggressiveness is too low, a significant part of the impurities from the root crops is not separated. If the aggressiveness is too high, not only impurities are damaged or separated, but also root crops, as a result of which the yield is reduced. By generating the control signal of the separating device, and in particular by sending it to the control unit of the separating device, the operating parameter is adjusted, preferably in accordance with the displayed reference image in a heap or part of a heap. On the basis of the control signal of the separating device, the control unit of the separating device, in particular, increases or decreases the operating parameter. To this end, the control unit of the separating device preferably sends a signal, in particular an electrical signal, to the actuator. The control unit of the separating device, in particular by means of an electrical signal, corrects, in particular, the pressure in the hydraulic system, the pressure in the pneumatic system, the current, voltage, force and/or torque necessary to drive or adjust the separating device or the operating parameter. In this case, the control unit of the separating device is, in particular, part of the same computing device as the data analysis device.

The method proposed in the invention achieves continuous optimization of the operation of the separating device. In particular, it is possible to continuously optimize the aggressiveness of the action of the separating body or separating device, and thereby achieve gentle handling of root crops and at the same time effective separation of impurities from the heap.

In particular, in order to issue a control signal for the separating device, the data analysis device locally analyzes the control data sets in the machine or in the tractor directly coupled to it. Due to this, when an undesirable state is detected in the separating device, it is possible to create the necessary control action almost instantly, which, accordingly, makes it possible to avoid jams or damage to root crops.

In a preferred embodiment of the method according to the invention, the data analysis device calculates at least one first portion of the reference image formed by at least one image area. At least one area of the image at least partially displays a given component of the heap or machine. On the basis of the first share, in particular, the cleaning parameter is calculated.

Before calculating the first fraction, the component statistically represented by the first fraction is specified. The reference image and/or reference data set is divided in particular into a plurality of image areas, preferably areas of the same size. Areas of the image that at least partially show the component collectively form the first lobe. Said proportion is in particular the proportion of these image areas at least partially showing a component, among all image areas, the first share being obtained by the ratio of the number of image areas or their total areas.

The first share is a measure of the size of the image areas that display the component, and, accordingly, a measure of the density (density) of this component in the field of view of the shooting device or the considered part of the control image. The component, in particular, is at least partially the component of the heap attributable to root crops, whereby the first proportion at least approximately indicates the concentration of root crops. The area of the image is regarded as representing the component and referred to the first lobe, in particular when the component is visible over at least 50 to 100% of its area. Alternatively, the proportion can be obtained by adding the individual pixels representing the component. In particular, at least one area of the image can also be related to the first share only in a certain proportion (proportional), or preferably can be related respectively partially to different shares. This is advantageous in particular if, within the framework of a classification method, preferably based on a model, it is not possible to unambiguously assign an image area to one respective component. In this case, the probabilities of assigning the image area to different proportions are preferably determined. In a particularly preferred case, according to these probabilities, the image areas are assigned to different proportions proportionally or partially. Thanks to this, the relationships of the components to each other are displayed even more accurately.

By calculating the first share, a parameter is calculated that characterizes, in particular, the composition of the heap. The operating parameter is adjusted on this basis in a particularly advantageous manner, since the efficiency of the cleaning effected by the conveying member or by the separating device containing the conveying member is highly dependent on the composition of the heap. If the first fraction characterizes the concentration of impurities in the heap, with an increase in the first fraction, it is preferable to vary the operating parameter in order to achieve a greater separating effect or separation efficiency. The cleaning parameter is preferably calculated using at least the first fraction or equal to the first fraction.

The at least one image area constituting the first lobe is preferably identified as showing the predetermined component, in particular based on the control subset generated using this image area. In particular, this area of the image is identified on the basis of the reference value contained in the reference image and/or included in the reference data subset, preferably color information. Color information includes, in particular, the values of black and white, gray and/or color channels of the color space.

The control subset, control value or color information is preferably classified by a statistical classification method, in particular based on a model. Accordingly, the image area is referred to the first share, in particular, when the result of the classification method is correlated with a given component of the heap or machine. The classification method uses, in particular, a neural network, a random forest algorithm, a Bayesian classifier, a support vector machine, and/or a decision tree. Thanks to the application of the classification method, the result of the calculation of the first share, in particular the different shares, is particularly robust and informative in relation to the composition of the heap.

In a particularly preferred embodiment, the control value, or color information, is compared with one or more reference (reference) values or reference ranges, and on this basis, the image area is either assigned or not assigned to the first lobe. The reference image is preferably captured (registered) like the reference image by an optical pickup device, the user having to mark, in particular, different parts of the reference image as showing different components. This form of differentiation enables a particularly reliable identification of the respective component in the reference image. It is particularly preferred to compare at least one of the reference values of the reference subset, which includes, in particular, color information, with at least one reference value, and the image area is assigned to the first lobe, in particular when at least at least one control value of the control subset of data is within the corresponding range of the reference values. This range of reference values is limited, in particular, by the upper and lower extreme values, and in order to assign the image area to the first lobe, the different reference values must lie in their respective reference value ranges.

In a preferred embodiment of the invention, when the reference image areas to be assigned to the first lobe are entered, the data analysis device automatically builds or automatically improves the model underlying the classification method. Alternatively, or in addition, when the reference image regions to be assigned to the first beat are entered, the data analysis device automatically calculates or modifies at least one range of reference values. Thus, in particular, reference values, ranges of reference values, or the model or its parameters may at least not be completely manually specified by the user. Instead, it is sufficient to enter at least one sample area of the image showing the component to start the data analysis device. Using this area of the image, the data analysis device automatically determines at least one reference value, at least one range of reference values, or the model or its parameters. In this way, the data analysis device can be configured almost autonomously for various applications. In this case, the greater the number of entered areas of the image, the more accurately it is possible to determine the reference values, or ranges of reference values, or the model or its parameters.

The method according to the invention is particularly robust if the input image areas show the component under different brightness and/or soil conditions. Thus, this method can also be reliably used in various application conditions. It is particularly preferred that during the repetition of the method, the data analysis device adapts at least one reference value or ranges of reference values, if necessary, with the identification of relevant components by the operator as relevant examples or samples, based on which data can be displayed for training the algorithm.

In particular, by using additional sensors, such as brightness sensors for measuring the brightness of the environment, which the data analysis device assigns to the reference data sets almost simultaneously with the acquisition of the corresponding images, the data analysis device automatically expands the amount of reference data. Alternatively or in addition, the user of the method, i.e. in particular the driver or operator of the harvester or associated machine, has the ability to manually mark at least one component on the rendered reference images to expand the reference data of the data analysis device. Thus, on the basis of information once entered by the user or data stored in the data analysis device, said device can distinguish between, for example, potatoes, leaves, stones, loose soil and clods, and can calculate the corresponding proportions.

The method according to the invention is preferably carried out automatically after it has been started, with the exception of the input of possibly available training data in the form of component marks. This makes it easier for the driver or operator of the machine to control.

The image areas constituting the first lobe are preferably further recognized, or identified, based on subsets of image data created by or formed by respectively adjacent image areas. For this, again, in particular, the color information contained in the control data subsets, in particular including black and white and/or gray values, is used. Thus, the image areas are not evaluated by means of data relating to these areas alone, but other data relating to the surrounding image areas are additionally used for this. This makes it possible to construct distributions of luminance and/or color indices and thus to perform recognition of image areas based on a wider information base.

When calculating the first beat, different areas of the image are preferably assigned different weights (weighting factors), i. e. different areas of the image are weighted differently. Thus, the contribution of the image areas that form the first lobe is different. This makes it possible not only to calculate the first fraction by means of a reference image in perspective, but, in particular, to assign higher weights to areas of the image that show a component of the heap that is more distant from the camera than to areas of the image that show a component that is closer to the camera. device. In this way it is possible to determine the first beat with perspective correction, i.e. purification from the influence of the perspective of shooting, and thereby obtain a picture of the composition of the heap located on the transporting body, which is especially close to reality.

The entire reference image, or a connected part of the reference image, is preferably divided into partial image areas. Such partial image areas contain in particular the same number of reference image pixels, preferably exactly one pixel each. The above part of the reference image is its part or fragment containing(s) a plurality of partial areas of the image. In order to calculate the first fraction, in particular, only those areas of the image showing this fraction are taken into account which belong to the indicated part of the reference image. To this end, a part of the reference image is set, in particular, so as to display important and to be monitored areas within the machine. Thus, the image area forming the first lobe contains, in particular, several partial image areas of one part of the reference image.

The reference image, or part of the reference image, in particular, is screened, or grid-divided, into a plurality of sub-image areas, each of which is preferably rectangular. If the partial areas of the image are formed by exactly one pixel, this makes it possible to create a particularly large database for analyzing the state of the heap with respect to its individual components and thus provide a particularly sensitive adjustment of the operating parameter. At the same time, this makes it possible to easily process data volumes produced by conventional two-dimensional digital cameras, which are usually a maximum of several million pixels, for a data analysis device equipped with one or more modern processors, in conditions close to real time.

The reference image preferably contains several parts, for each of which the data analysis device calculates the first share, in particular several parts in image areas, and the parts of the reference image preferably display a heap on various conveying organs that take it away from the separating device. Thus, these parts of the reference image show, in particular, different sections of the same conveying element, one of which is located in the feed direction in front of the separating device or its separating element, and another behind the separating device or its separating element. Alternatively, portions of the reference image show various transport organs representing alternative routes for supplying the various components of the heap (eg, transport organ for peeled root crops, transport organ for sorted contaminants). The calculation of the first fraction for these different parts of the reference image makes it possible to evaluate in a particularly comprehensive manner the effectiveness of cleaning the roots or separating the components of the heap with a suitable separating device. In particular, the first fraction of the heap flow moving towards the separating device can be compared with the first fraction of the diverted root crop heap flow moving away from the separating device, and thereby the efficiency of the separating device can be determined. As an alternative, it is preferable to determine the composition of the heap located on the conveying organs adjacent to the separating device, and therefore diverting the heap from the separating device, once for the separated heap and once for the heap fed further. Depending on the efficiency of the separating device, in particular, the operating parameter is adjusted. Similarly, portions of the reference image displayed or present in the respective reference data sets may show the portion of the conveying member located in front of the separating or diverting member of the separating device and the portion of the conveying member positioned behind the separating or deflecting member. If, as a result of image analysis, it is determined that too large parts appear in the undesirable area behind the deflecting organ, for example, root crops, this deflecting organ can be installed differently, for example, lower above the transporting organ, which increases the separation efficiency of the heap components.

In a preferred embodiment of the invention, parts of the reference image preferably show different conveying elements behind the separating device, in particular one conveying element for removing a bunch of root crops and one conveying element for removing impurities, located behind the same separating device. In this case, for both parts of the control image, preferably determine the first proportion of the component, such as root crops. Alternatively, different proportions are calculated for different parts of the reference image. This allows, for example, to compare the proportion of impurities in the discharged flow of a heap of root crops with impurities with the proportion of root crops in the stream of sorted impurities and, based on this comparison, to regulate the operation of the separating device included in the separating device in relation to its position relative to the transporting organs or in relation to its speed.

The image areas forming the first lobe preferably show root crops or parts thereof, and the image areas forming the second lobe show impurities or parts thereof. Thus, the data analysis device calculates at least two different proportions. In a particularly preferred embodiment, the data analysis device calculates at least four shares, including the first share for root crops, the second share for plant impurities, in particular haulm, the third share for soil, and the fourth share for damaged roots. The sum of the shares, in particular, is less than or equal to one (<1). As an alternative, the first proportion may also be the proportion of impurities, the second proportion may be the proportion of roots, and so on. As an addition or alternative, the fractions for lumps and/or stones are preferably calculated.

Alternatively or in addition, in yet another embodiment of the invention, at least two filming devices and at least two transporting elements are provided, the first control image of the part of the heap removed from the separating device by means of the first transporting element is taken by the first filming device, the second shooting device another reference image of the part of the heap removed from the separator by means of the second conveying member is taken, and the separator device control signal is generated on the basis of at least one of two reference data sets formed by reference images or generated on their basis, preferably on the basis of both such reference data sets. In this case, control data sets are analyzed as described above or discussed below, in particular with respect to the respective shares.

- 5 042191

The use of several proportions when performing a calculation by the data analysis device allows you to get a more accurate picture of the composition of the heap or the workload of the transporting body. As an alternative to identifying areas of the image by means of boundary values, all areas of the reference image or parts of the reference image are assigned, if necessary, to a certain proportion. In this case, the degree of correspondence of the control data subsets calculated by the image regions and the reference data subsets is preferably assessed, and each image region is assigned to the proportion for which this correspondence is greatest.

In a preferred embodiment of the invention, the cleaning parameter is determined by deviating the first fraction from a threshold calculated by the data analysis device. In particular, the threshold value characterizes the optimal load of the conveying body, and the deviation from its predetermined value causes, or initiates, a change in the operating parameter. The cleaning parameter is based, in particular, on several shares, and preferably additional data, in particular data from sensors.

In a preferred embodiment of the invention, the control signal of the separating device is calculated using a plurality of cleaning parameters, in particular calculated sequentially in time, or when calculating the cleaning parameter, at least one previously calculated cleaning parameter is used. In particular, a running average value of the cleaning parameter is calculated, which is taken as a basis for determining the control signal of the separating device, or the curve (change profile) of the cleaning parameter is smoothed, in particular by means of a low-pass filter. Thanks to these measures, the method according to the invention becomes particularly noise-immune and can therefore be carried out in a particularly robust manner.

In a preferred embodiment of the invention, at least one additional sensor is used, which transmits to the data analysis device data used in calculating the control signal of the separating device. Such a sensor can be, in particular, a sensor, preferably tactile or ultrasonic, for measuring the thickness of the heap layer on the conveying body, a power sensor, made, for example, in the form of a pressure sensor for measuring the pressure of a hydraulic fluid, and/or a speed sensor, in particular for measuring the rotational speed of the drive of the conveying body. In particular, the slippage of the conveying member is determined by means of the rotational speed sensor, which is transmitted to the data analysis device in the form of corresponding data from the sensor. The use of a humidity sensor makes it possible to use additional information when calculating the control signal of the separating device.

Thanks to this additional information available in the data received from the sensors and beyond the information obtained from the reference image, the data analysis device has a much more accurate picture of the situation with the cleaning of root crops in the area of the transport organ, which, in turn, allows to provide influence on the operating parameter even better according to this situation.

By means of various control signals of the separating device, the data evaluation device preferably initiates either an increase or a decrease in the operating parameter. In particular, the data analysis device or the separation device control unit contains a controller with a fuzzy logic system, a proportional-integral-derivative or a three-position controller, which either increase or decrease the operating parameter or maintain its current value in a mutually exclusive manner. An increase in the operating parameter is initiated, in particular, if the cleaning parameter is above a predetermined first threshold value, and a decrease is correspondingly initiated if the cleaning parameter is below a predetermined second threshold value.

The operating parameter is preferably the distance between the two conveying members, by which the conveying members can function together as a separating device, or the distance from the separating member of the separating device or the separating device itself to the conveying member. In particular, the operating parameter is the distance between the two transport rollers of the roller table that rotate during operation, between which the soil is sieved. As an alternative, the operating parameter is the distance from the conveying body, made in the form of a screening conveyor, to the separating body, made in the form of a deflecting roller, extending across the transporting body and providing deviation of root crops away from the transporting body. In this case, the deflecting roller during operation rotates around the axis of rotation, which, when viewed from above on the transporting body, passes at an angle of less than 90° to the direction of movement of the heap by the transporting body. Alternatively, the separating member is made in the form of a finger conveyor circulating during operation, which is located above the conveying member and whose protruding fingers comb through the heap located on the conveyor member during operation. Again, as an alternative, the separating element is made in the form of a cleaning device that does not circulate during operation, located above the grain conveyor

- 6 042191 of haulm, interacting with the screening conveyor, and providing separation of root and tuber crops from the haulm located on the conveyor for large haulm. The aforementioned distance is in each case regulated, in particular, by a hydraulic or mechanical actuating device, which makes it particularly easy to change the aggressiveness of the separating element of the separating device in cooperation with the conveying element, or the separation efficiency of the conveying elements.

As an alternative, the operating parameter is the penetration depth of at least one lifter of the machine into the soil. This allows you to easily influence the amount of impurities in the heap.

As an alternative or in addition to the above operating parameters, the operating parameter is the speed of movement, in particular circulation or rotation, of the separating device or the separating element of the separating device. In particular, such a travel speed is the circulation speed of the pin conveyor described above or the rotation speed of the deflector roll described above. As an alternative, the speed of movement is the circulation speed of an angled separating device that during operation feeds the impurities upwards and is made, for example, in the form of a haulm conveyor, driven in such a way that as much of the impurities as possible move upwards and the root crops roll down down against the direction of movement of the section of the separating device facing them.

As an alternative, the operating parameter is preferably the installation angle of the conveying element, the separating device, i. e. at least one separating body of the separating device. In particular, the operating parameter is the installation angle of the separating device, also called the fine haulm elevator. Through the angle of installation, the inclination of the feed plane of the conveyor for fine haulm of the separating device relative to the horizontal is changed and thus the aggressiveness of the separating device is regulated.

In an alternative preferred embodiment of the invention, the effect on the operating parameter causes a change in the air flow rate or air mass flow, i.e. the mass of air passing per unit of time, and the corresponding separating device operates using the air flow. In this case, the corresponding operating parameter may be the power of the engine, expressed, for example, in terms of engine speed. At the same time, the air flow, in turn, ensures the separation of root crops and impurities, in particular, blows plant impurities out of the flow of a heap of root crops and thereby removes these impurities. In such an air separator, which can also be used in particular as a fixed installation, the operating parameter can preferably be the speed of the respective fan or the installation angle of the respective baffle assembly, which, for example, divides the air flow into the main air flow and the cross air flow.

In a preferred embodiment of the method according to the invention, several of the aforementioned operating parameters are set, in particular by means of the same separator control signal or different separator device control signals. To this end, a control mechanism can be provided in the data analysis device, which, for the desired increased or reduced separation effect in the respective separation device, outputs appropriate signals for the respective adjustable (adjustable) values.

After initiating a change in the operating parameter for a predetermined period of time or for a given section of the conveying body, a further change in the operating parameter is preferably not initiated. This concerns, in particular, only one and the same operating parameter and/or at least one operating parameter of at least one separating device located downstream during operation. This avoids overshooting the separating element and ensures that any change in the operating parameter has a solid information base, already taking into account the previous change in the operating parameter.

The control signal of the separator device is preferably transmitted to the control unit of the separator device via wired communication, in particular via CAN bus or Ethernet technology, or via wireless communication, and preferably the permission to control the separator device is previously entered by the operator via an interface. This makes it possible to use existing or at least proven data transmission systems for transmitting information when adjusting the separating body, and the reliability of the implementation of the method is increased, in particular, due to the fact that instead of automatically regulating the operation of the separating device, an indication is provided to the operator, in particular, in the driver's seat. cabin, the calculated or required regulatory action on the operation of the separating device, which he must authorize by entering the appropriate indication using the interface.

The task in accordance with the invention is also solved in a machine for harvesting root crops and / or for separating root crops from impurities contained in their heap. The machine contains a frame of the machine, at least one transporting body, at least one opti

- 7 042191 Physical surveying device, separating device and data analysis device. At least one optical recording device is installed with the possibility of capturing at least one reference image of at least one part of the heap moved by means of the conveying member relative to the machine frame, and transmitting the reference image or a reference data set generated using it to the data analysis device for carrying out method according to one of the previous paragraphs.

The data analysis device preferably comprises a graphics processor device, in particular a general purpose graphics processor or graphics processor and/or a user programmable gate array processor device. Such an embodiment of the data analysis device makes it possible to analyze the reference data set in a particularly resource-saving manner and in particular locally. It goes without saying that the data analysis device, made in the form of an electronic computer, also contains other typical means, such as power supply means, interfaces and random access memory.

In a preferred embodiment of the invention, the machine has at least one sensor associated with a data analysis device, in particular a tactile or ultrasonic sensor for measuring the thickness of the heap layer on the conveying member, a sensor for measuring drive power, such as a pressure sensor for measuring hydraulic fluid pressure, and /or a speed sensor located on the conveying body. Thanks to this sensor, the speed signal of the conveying organ can also be calculated on the basis of measured physical quantities, which significantly increases the information content of the parameters calculated by the data analysis device and reduces their error proneness. Additionally, a moisture sensor can be used to provide information taken into account during data processing by the data analysis device to control the operation of one or more separating devices.

The machine according to the invention preferably has several imaging devices, each of which, during operation, takes at least one reference image of the same transport organ. As an alternative, the machine according to the invention preferably has several imaging devices, each of which, during operation, takes at least one reference image of different conveying elements, in particular those conveying material from the same separating device. Alternatively, one of the two filming devices can be directed towards the retraction area of a separating device formed, for example, by an air separator. The presence of several filming devices allows you to track the composition of the heap, in particular, the nature of the change in the first share along the transport path, in particular, on different parts of the transport path behind the separating device of the machine. In this way, in particular, on the basis of different first shares, it is possible to regulate the speeds of movement of the various conveying organs. Accordingly, according to the method described above or discussed below, in at least one data analysis device, analysis is performed for sections of the transport path that fall into the corresponding reference images. Although in the preferred case only one central data analysis device is provided for the analysis of data from the surveying devices, it is also possible for the respective surveying devices to be assigned their own data analysis devices. In this case, such data analysis devices can control the separating devices assigned to them, in particular in concert with other data analysis devices. As an alternative to or in addition to these options, the central data evaluation unit is responsible for generating control signals for the separating device and forwarding them to the machine control system.

The filming device is preferably positioned in such a way that the reference image shows at least two alternative feed paths within the entire conveyance path for the various components of the heap, in particular root crops and impurities. As a result, two conveying members can be monitored with a single recording device, wherein different parts of the reference image show the respective portions of the different conveying members or the heap transported on them. In particular, one of the transporting organs is designed to supply sorted impurities, and another transporting body is designed to supply cleaned root crops. This makes it possible to obtain a particularly comprehensive picture of the cleaning efficiency and thus also of the loading of the conveying member and/or the separating device containing the conveying member.

The filming device is preferably positioned so that the reference image taken by it during operation at least partially shows at least two portions of the conveying member separated by at least one separating device, in particular the separating member of the separating device. These sections of the conveying element are separated by the separating element of the separating device only in their representation by the control image and belong to the respective conveying element. The separating body is closer to the filming device than the transporting body, as a result of which the latter is covered by the separating body in the control image. This location of the camera allows

- 8 042191 calculate at least one first fraction for each of the two separate parts of the reference image and thereby directly evaluate the efficiency of the separating body and the corresponding separating device. In particular, for this purpose, the composition of the heap before reaching the separating element is compared with the composition of at least one fraction of the heap after passing through the separating element.

The conveying element is preferably in the form of a screening conveyor or a pintle conveyor, which during operation passes, in particular, under at least one deflecting roller extending across the conveying element and deflecting the heap to the side from it. As an alternative, the conveying element is in the form of a screening star or a conveying roller, the conveying roller being part of the roller table in particular.

As an alternative to or in addition to the embodiments described above, the machine according to the invention is designed as a machine for separating root crops from the rest of the heap, i. from other components of a heap of root crops. Moreover, in a preferred embodiment of the invention, such a machine is operated, in particular, stationary (as a stationary installation), i.e. without continuous movement of the machine in space during its operation.

In another preferred embodiment of the invention, the machine is a potato or beet harvester.

Other details and advantages of the invention are disclosed in the following examples of its implementation, schematically explained by the drawings.

In FIG. 1 shows a block diagram of the implementation of the method according to the invention;

in fig. 2 is a flowchart of a corresponding process for obtaining a control signal of a separating device;

in fig. 3 is a flow chart of a corresponding cleaning parameter processing process;

in fig. 4 is a flowchart of a corresponding control process of the separating device;

in fig. 5 shows a view of the reference image and its partial analysis;

in fig. 6 - the object of the present invention;

in fig. 7 and 8 show the subject matter of the invention shown in FIG. 6 in various lateral projections;

in fig. 9 is a fragment of the object of the invention shown in FIG. 6, with a transport body;

in fig. 10 is a partial view of the part shown in FIG. 9 area of the device shown in FIG. 6;

in fig. 11 is a view of the one shown in FIG. 10 areas in a different perspective;

in fig. 12 is a view of a reference image captured by the pickup device shown in FIG. 10;

in fig. 13 shows the separating device of the machine shown in FIG. 6, with a filming device;

in fig. 14 is a schematic view of a reference image captured from the point of view of the one shown in FIG. 13 filming devices;

in fig. 15 shows another separating device of the machine shown in FIG. 6, with a filming device;

in fig. 16 is a schematic view of a reference image taken from the point of view of the one shown in FIG. 15 filming device;

in fig. 17 is a view of another fragment of the machine shown in FIG. 6, with another filming device;

in fig. 18 is a schematic view of a reference image viewed from the point of view of the pickup device shown in FIG. 17;

in fig. 19 is a separate view of another device according to the invention;

in fig. 20 is another separate view of another device according to the invention.

Functionally identical or similar elements are designated as appropriate by the same reference numbers. Individual technical features of the embodiments of the invention described below, also in combination with the features of the examples described above, may characterize particular embodiments of the present invention, but always at least in combination with the features of one of the independent claims. The objects shown in the list of drawings are in some cases only partially shown in separate drawings.

Proposed in the invention, the method is used to regulate the operation of the machine 2 for harvesting root crops 4 (see Fig. 6-8). When implementing this method, at least one optical recording device 6 removes (registers) at least one control image 8 showing a heap containing root crops 4 and moving relative to the frame 12 of the machine 2 by means of at least one transporting body, generally designated by the number 10.

The reference image 8 is transmitted to the data analysis device, which, on the basis of the reference data set generated using the reference image or formed by it, generates a separating device control signal for adjusting at least one operating parameter of the separating device of the machine 2. Images illustrating in the drawings, reference images only schematically show parts relevant in the context of the invention without any borders or borders. Camera images, in particular digital images, may contain other information not shown in the drawings. Such information can be masked or filtered out, for example, in the camera itself or during the construction or processing of the control data set.

In one embodiment of the present invention, starting from the heap flow 1.1, the analysis of the composition of the heap (block 1.2) before the first separating element (Fig. 1) is performed in the manner described above. In addition, the composition of the heap flow is recalculated (blocks 1.3 and 1.4) before the inlet of another separating device and after its outlet. Finally, the composition of the heap stream is re-determined before entering the third separating device. From here, the corresponding shares of A1 and A2 of root crops and impurities are obtained (blocks 1.6, 1.7, 1.8 and 1.9). Depending on the required separation efficiency in individual separators, the data analysis device jointly subjected to computational processing (block 1.10) individual shares A1 and A2 of root crops and impurities for the respective separators. Then, the operating parameters of the respective separating devices are tuned (block 1.11) to optimize the efficiency, or productivity, of the respective separating device.

The definition of the separation device control signal is shown in more detail in FIG. 2. First, to build a control data set, its relevant parts are extracted from the control image 8 (block 2.1). To do this, based on the position of the shooting device, a mask or region of interest (ROI) (block 2.2) can be pre-set, with the help of which a distinction is made between the considered and unaccounted areas of the control image. Now, on the basis of the relevant area of the control image and the control data set available for processing, the calculation of the proportions of image areas showing the individual components of the heap is performed (block 2.3). For this, in particular, color information can be analyzed. These values can be taken from a lookup table, or they can be set by service personnel (Box 2.4).

Based on the definition of the threshold value (block 2.5), deviations of the calculated shares from the threshold value are calculated (block 2.6). The threshold value is, for example, the ideal value for the proportion considered in the respective case (eg root crop, 1st admixture, 2nd admixture). Then the resulting deviations are passed through a low-pass filter (block 2.7) for smoothing. In this case, the filter time constant specified in block 2.8 is used. Then, based on the smoothed deviation values for individual positions along the transport path and the corresponding shares, the cleaning parameter RS is calculated (block 2.9). In this case, for simplification, the controller parameters can be used, according to which, for example, the cleaning parameter RS_1 (A1) in the first controlled transport path for the share of A1 in the form of root crops is set equal to one, if the deviation is too large, which means that on the considered control image or its part there are too many impurities. With sufficiently small deviations from the ideal value, the cleaning parameter RS for the A1 fraction can be set to zero. Based on the cleaning parameter RS_1 (A1) determined according to block 2.9, it is possible to generate, for example, by means of a three-position controller, a control signal for the separating device (block 2.10).

Thus, for example, by means of the method according to the invention, the relations between the products present in the heap flow before the separating device and further behind the separating device, for example, located on the conveying body for the removal of products, i.e. root crops (4), and a transporting body for removing impurities (5). In the event that both of these transporting organs are controlled by one chamber, separate areas of interest are set for the transporting organ that removes impurities and the transporting organ that removes root crops. Depending on whether there are too many potato tubers (root tubers 4) on the conveying body that removes impurities or whether there are too many impurities on the conveying body that removes products, the parameters of the separating device that affect the separation threshold are adjusted accordingly. Thus, in particular, it is possible to set higher or lower the fingers of the finger conveyor or brush conveyor and/or it is possible to slow down or accelerate the movement of the finger or brush conveyor.

In one implementation of a three position control system, considered as an example and illustrated in FIG. 3, the clearing parameters RS_1 (A2) to RS_n (A2) are first dealt with, i.e. cleaning parameters for A2 fractions describing impurities (Box 3.1). In this case, as an example, we are talking about the consideration of the proportions of impurities contained in a heap in the form of stones in front of and behind a separating device or in front of and behind a transport path containing several separating devices. According to the above principle, the cleaning parameters are set to zero or one. Then, in block 3.2, it is checked whether the sum of the clearing parameters RS_i (A2) (for i=1...m) is equal to zero. If the answer is negative, in block 3.3, the last

- 10 042191 cleaning parameters, and, if necessary, related data on the composition of the heap in the relevant provisions. If the check performed in block 3.5 shows that the change in the settings of the separating device has been made for a long time, a degressive (reducing aggressiveness) signal of the operating parameter is generated for the corresponding separating device in block 3.6. The clearing parameters recalled from memory can also be taken into account. The separating device or separating devices are adjusted by setting it / them to work less aggressively (block 3.7) so that less root crops remain in the flow of impurities or heaps passing through the transport path. The last clear parameter or last clear parameters RS_i (A2) are stored in 3.4, if necessary, together with the corresponding shares of A2 (block 3.15).

If the sum of the cleaning parameters RS_i (A2) (when i=1...m) is equal to zero, then in block 3.8 the cleaning parameters RS_i (A1) related to the shares of A1, for example, the shares of root tubers, i.e. correspondingly recall these clearing parameters from memory. Then it is checked whether the sum of these cleaning parameters is again equal to zero (block 3.9). If so, the last clearing parameters RS_i (A1) are also stored back in memory 3.4 (block 3.10). No change in operating parameters is required, and a neutral operating parameter signal is issued or no operating parameter signal is issued (block 3.11). If the sum of the clearing parameters relating to the shares of A1 is not equal to zero, then in block 3.12 the old values are recalled from the memory and in block 3.13 it is checked whether sufficient time has elapsed since the last operating parameters were changed. If this is the case, in block 3.14 an operating parameter is output to set the separating devices for more aggressive operation.

To implement the settings obtained in block 3.7 in Fig. 3, you need to control the actuators of the separating devices. For this, according to FIG. 4 in the controller 4.1, based on the settings obtained in blocks 3.6, 3.11 and 3.14 and taking into account the current adjustments (4.2) of the corresponding separating device, changes in operating parameters depending on the separating device are determined. For this purpose, for example, the speeds of the elements of the separating devices, rotational speeds, distances or angles of inclination are set. Based on them, in block 4.3, control actions are set for the actuators of the corresponding separating devices, by means of which the separating devices are regulated (block 4.4).

In FIG. 5 in its upper part, by way of example, is a reference image 8 showing the transition from the conveying member 10a to the conveying member 10b. In this section of the transport path there are root crops 4 and impurities 5, which may include stones and vegetable impurities (tops). In accordance with the classifiers determined during the training of the algorithm or given by the database, for example, a table with color information, for example, in the HSV format, the individual partial areas 16 of the image are checked for the presence of the same components. Thus, by associating the respective areas of the image with the individual lobes shown as an example at the bottom left of FIG. 5, the distribution of individual proportions of root crops 4 and impurities 5 in the control image 8 is obtained. Thus, A1 shows the proportion of root crops 4 in the control image or in the corresponding control data set, A2 is the proportion of haulm and A3 is the proportion of stones. Such a mapping is preferably performed on the basis of the color information of the individual pixels, i.e. the area 19 of the image, correlated with a certain proportion, corresponds in particular to the area of the pixel. The cleaning parameter, generally referred to as RS, is based, by way of example, again preferably on the deviation of the first share A1 from the threshold value R, which indicates the optimal distribution of the shares of root crops 4 in the observed location of the transport path. For example, when the deviation from the cleaning value is greater than or equal to 50%, the cleaning parameter RS is set to one (RS=1), and when the deviation from the cleaning value is less than 50%, it is set to zero (RS=0). These values are then respectively stored or processed within the further implementation of the method according to FIG. 1-4.

The machine 2 according to the invention is made according to FIG. 6 in the form of a trailed potato digger, in which, by means of a frame 12, which is only partly indicated, several conveying elements 10 and also the separating devices associated with them are fixed. Along the transport path there are several filming devices 6, which remove a heap moving on the transporting organs 10, containing root crops 4. Indicated in FIG. 6, the installation positions of the filming devices 6 are the transition from the first conveying body 10A, made in the form of a screening conveyor, to the second conveying body 10B, made in the form of a screening conveyor and additionally surrounded by a conveyor for large haulm, and the transition from this second screening conveyor 10B to an additional conveying body 10C containing an additional separating device. In addition, at the exit of this separating device, the conveying element 10E leading to the sorting table is controlled by another filming device 6, which simultaneously removes the additional conveying element 10F, which is provided for transporting residues of impurities 5, in particular stones.

The data analysis device may be located in any centrally accessible location, but preferably located near the sorting table. Vehicle speed signal or

- 11 042191 data concerning the adjustment of the separating devices can be transmitted by the data analysis device to the tractor, for example by means of the control shown in FIG. 6 cables 12.1.

The machine shown in Fig. 7 and 8 in side views may be provided with optical imaging devices 6 in other positions. Thus, additional filming devices can also be located directly in the region of the digging device 29 or the drop stage leading to the hopper 33 .

In FIG. 9 and 10 show the installation of an optical camera 6, which is located on the frame above the first fall stage between the transport body 10A and the transport body 10B and the field of view of which is directed downwards. Illumination of this field of view in order to register a sufficiently illuminated reference image 8 is provided by a light source 7 . The conveying body 10A is a screening conveyor, which already sifts out some of the impurities 5, in particular soil and/or clods, from the heap coming from the digging device 29, and transfers the heap through the fall stage to another conveying body 10B, made in the form of a screening conveyor. This transport body 10B further has a large haulm conveyor provided for separating the haulm present on the potato tubers or in the heap. Accordingly, cleaning devices 32 are located across the width of the conveying body 10B.

The height H of the cleaning device 32 above the feed plane of the conveying member 10B is adjustable by means of a separating device control signal. This allows you to influence the separation efficiency in the separating device, made in the form of a haulm conveyor. In addition, the relative speed of the sieve conveyor and the haulm conveyor 43 can be adjusted. In FIG. 10, for clarity, only the haulm conveyor 43 is shown, while the actual conveying body 10B, made in the form of a screening conveyor (see Fig. 14), is not shown.

In FIG. 12 shows in detail the reference image 8 taken by the field of view optical pickup 6, which is indicated by a dotted line in FIG. 11. Based on the control image 8 created from this control image or the control data set formed by it, the above-described analysis operations are performed based on the deviations of the proportions of the captured and classified objects from the threshold values R.

From the transporting body 10B, the heap still on it is transferred to an additional transporting body 10C, which has a feed direction 1C. It is given a separating device in the form of several rotating deflecting rollers 24 located one above the other. The impulse created by it moves the heap in the direction of the transporting body 10D (Fig. 13).

The distance H between the conveying member 10C and the lower deflector roller 24 can be adjusted to vary the separation efficiency of the heap components, and is thus a tunable (adjustable) operating parameter. If necessary, other distances between the individual deflection rollers 24 can also be varied in order to control the intensity of the deflection or possible separation function, in which the haulm is drawn in between the deflection rollers 24. Alternatively or in addition to this, the variation in the separation efficiency or deflection is due to the possibility of adjusting the rotation speeds deflecting rollers 24.

The height H of the lower ends of the fingers 26 of the separation device, made in the form of a finger conveyor 26.1 and related to the transport body 10D, can also be adjusted as one of several operating parameters. This height H describes the distance between the pins 26 and the upper edge of the conveyor made in the form of a pintle conveyor. In addition, the installation angle of the finger conveyor 26.1 relative to the normal to the supply plane of the transporting body can be adjustable. The same applies to the circulation speed of the finger conveyor 26.1.

Shown in FIG. 13, the filming device 6 produces the image shown in FIG. 14 is a reference image 8, in which, by means of filtering or masking, a significant (relevant) part 8A of this reference image is set in the considered embodiment of the invention. To control the efficiency of the separating device, in this case the separation efficiency of the deflection rollers 24, you can additionally select another part 8B of the control image, which is located in the feed direction 1C behind the deflection rollers 24. In particular, to control the separating device, the area in front of the deflection rollers 24 is controlled In this case, the reference data set is obtained from the corresponding portion 8A of the reference image.

If the corresponding cleaning parameter RS for part 8A of the control image results in too low separation efficiency of the separation device located in front of this image or depicted on it, the separation device can be adjusted to more aggressive operation. If, on the other hand, the cleaning parameter in part 8B of the control image indicates a too high separation efficiency, for example, due to the presence behind the deflection rollers 24 of too high proportions of impurities 5 in the form of lumps, which are at least partially still

- 12 042191 are needed to ensure a gentle feeding of potatoes in the subsequent transport path, the distance H from the deflecting rollers 24 to the transport body 10 can be reduced and thus the separating device can be adjusted to a less aggressive operation.

Another optical pickup device 6 located in the area of the conveyor belts 10C and 10D is shown in FIG. 15 and 16. This camera 6 can be used in addition to the camera shown in FIG. 14, or as an alternative to it. In particular, it serves to monitor the operation of the separating and deflecting device formed by the deflecting rollers 24. This video monitoring device is also provided with a light source 7 for better illumination of the monitored area.

Another optical filming device 6 with a light source 7 attached to it is located above the sorting table and faces the transport body 10E and the transport body 10F with its field of view (Fig. 17). By masking in the one shown in FIG. 18 of the reference image 8, parts 8A and 8B of the reference image are selected, with the help of which, on the one hand, the feed path formed by the conveying body 10E with the feed direction 1E for diverting the root crops 4 is controlled, and on the other hand, another feed path formed by the conveying member 10F with feed direction 1F to remove impurities 5 in the form of stones and/or lumps. By performing the above-described analysis operations, it is checked whether the proportions of root crops 4 on the conveying body 10F are not too large. If so, then by means of the method proposed in the invention, the separating device located in front of it is adjusted to work harder. This separating device is located above the conveying body, made in the form of a needle conveyor 10D and, in particular, is provided with fingers 26 as a finger conveyor, shown by way of example in dotted lines, but in the drawing located behind the canopy 40 located in front of them. For example, in order to divert more harvest in the form of root crops 4 to the transporting body 10E along the corresponding slope 41, the distance between the fingers 26 and the transporting body 10D is reduced. If too many impurities 5 in the form of stones and/or lumps are found on the conveying element 10E, it is possible, for example, to reduce the rotation speed of the deflecting rollers 24 so that a smaller impulse acting on the impurities 5 provides a better deflection of possible stones in the direction of the conveying element 10F. Then the impurities 5 will better move down the slope 42 to the conveying body 10F.

In FIG. 19 shows the application of the method according to the invention in another separating device, made in this case in the form of a fine haulm elevator. Accordingly, shown in FIG. 19, reference image 8 shows an inlet conveyor 10K with feed direction 1K, which leads via a drop stage to a haulm conveyor 30 designed as a fine haulm elevator. This haulm conveyor moves any fine haulm that may be present in the haulm flow further in the direction 31, while potato tubers and heavier impurities in the haulm flow that fall on the haulm conveyor 30, due to the inclined position of the haulm conveyor 30, determined by the adjustable installation angle α, fall into clearance between conveyor body 10K and fine haulm elevator. If, in parts 8A and 8B of the control image, the proportions of root crops 4 and impurities 5 go beyond the reference values, then, for example, by adjusting the installation angle α or changing the circulation speed of the haulm conveyor 30, the separating device can be adjusted for more aggressive or less aggressive operation, so that how to achieve the required separation efficiency.

In the case of the roller cultivator shown in FIG. 20 and preferably used in the field of warehouse technology, the distance H between the individual conveying elements 10T is adjustable. Thus, the transport body 10T directly forms the separating device. In separate parts 8A and 8B of the reference image 8, the proportion of impurities is preferably calculated and used to control the efficiency of the separating device. At the same time, due to the use of a wide-angle lens (fisheye type) in the filming device 6, the above-described perspective correction is performed during image processing.

In other alternative examples, the machine may be, for example, a self-propelled beet harvester or cleaning line used in a potato or beet warehouse.

Claims (31)

CLAIM 1. A method for regulating the operation of a machine (2) for harvesting root crops (4) and / or for separating root crops (4) from the rest of the heap containing impurities, during which at least one optical imaging device (6) removes at least one reference image (8) of at least one part of the heap moved by means of at least one transporting body (10, 10A, 10B, 10C, 10D, 10E, 10F, 10K, 10T) relative to the frame (12) of the machine, characterized in that based on the control data set generated using the control image (8) or formed by it, the data analysis device expressed - 13 042191 The separating device control signal is activated to automatically adjust at least one operating parameter of the separating device of the machine (2), the operating parameter setting how aggressively the separating device acts in separating root crops and impurities. 2. The method according to claim 1, characterized in that the data analysis device analyzes the control data sets locally in the machine (2) or in a tractor directly coupled to it. 3. The method according to claim 1 or 2, characterized in that the data analysis device calculates at least one first fraction (A1) of the reference image (8) formed by at least one area (19) of the image, at least partially displaying the specified component of the heap or machine (2), and on the basis of the first share (A1) calculate, in particular, the parameter (14) cleaning. 4. The method according to claim 3, characterized in that at least one area (19) of the image forming the first share (A1) is identified as showing a given component of the heap or machine (2), preferably based on the generated using this area ( 19) an image of the control subset of data, in particular at least one element of color information contained therein. 5. The method according to claim 3 or 4, characterized in that the control subset of data, in particular at least one control value included in it, preferably color information, is classified by a statistical classification method, in particular based on the model, and the area (19) images are assigned to the first share (A1), in particular, when the result of the classification method is correlated with a given component of the heap or machine (2). 6. The method according to one of claims 3 to 5, characterized in that at least one reference value of the reference subset of data, in particular color information, is compared with at least one reference value (R) and the image area (19) is assigned to the first fraction (A1), in particular when at least one control value of the control subset of data is within the corresponding range of reference values. 7. The method according to claim 5 or 6, characterized in that when the reference image areas (19) to be assigned to the first lobe (A1) are entered, the data analysis device automatically improves the model underlying the classification method and/or automatically calculates or modifies at least one range of reference values. 8. A method according to one of claims 3 to 7, characterized in that, when calculating the first fraction (A1), different areas (19) of the image are assigned different weights. 9. Method according to any one of claims 3 to 8, characterized in that the entire reference image (8) or the connected part (8A) of the reference image is divided(a) into partial image areas (16) containing, in particular, the same number of pixels reference image (8), preferably exactly one pixel. 10. The method according to one of claims 3 to 9, characterized in that the reference image (8) contains several parts (8A, 8B), for each of which the data analysis device calculates the first share (A1), in particular several parts (A1 , A2, A3) in the areas (19) of the image, and parts (8A, 8B) of the control image preferably display a heap on various conveying organs that take it away from the separating device. 11. A method according to one of claims 3 to 10, characterized in that the image areas (19) forming the first lobe (A1) show root crops (4) or parts thereof, and the image areas forming the second lobe (A2) show impurities (5) or parts thereof. 12. The method according to one of claims 3 to 11, characterized in that the cleaning parameter (14) is determined by means of the deviation of the first fraction (A1) from the threshold value (R) calculated by the data analysis device. 13. The method according to one of claims 3 to 12, characterized in that the control signal of the separating device is calculated using a plurality of cleaning parameters (RS), in particular calculated sequentially in time, or when calculating the cleaning parameter (RS) using at least one previously computed cleanup parameter (RS). 14. The method according to one of the previous paragraphs, carried out with at least two filming devices (6) and at least two transporting bodies (10), characterized in that the first filming device (6) takes the first control image of the part of the heap removed from separating device by means of the first conveying element (10), the second recording device (6) takes another control image of the part of the heap removed from the separating device by means of the second conveying element (10), and the separating device control signal is generated based on at least one of the two control data sets formed by reference images or generated from them. 15. The method according to one of the preceding claims, characterized in that at least one sensor, in particular an ultrasonic or tactile sensor, is used to measure the thickness of the heap layer on the transport body, a sensor to measure the drive power, a humidity sensor and/or a speed sensor , which transmits to the data analysis device the data used in the calculation - 14 042191 separating device regulation signal. 16. The method according to one of the preceding claims, characterized in that, by means of various control signals of the separating device, the data analysis device initiates either an increase or decrease in the operating parameter. 17. The method according to claim 16, characterized in that after initiating a change in the working parameter for a given period of time or for a given section of the transporting body (10, 10A, 10B, 10C, 10D, 10E, 10F, 10K, 10T), a further change in the working parameter is not triggered. 18. A method according to one of the preceding claims, characterized in that the operating parameter is the distance (H) between two conveying organs (10T) or the distance from the separating member (32) or separating device to the conveying member or additional conveying member (10C). 19. Method according to one of the preceding claims, characterized in that the operating parameter is the speed of movement, in particular circulation or rotation, of the separating body (24, 26, 30) or separating device. 20. A method according to one of the preceding claims, characterized in that the operating parameter is the installation angle (α) of the conveying body (10) or separating device. 21. The method according to one of the preceding paragraphs, characterized in that the operating parameter is the engine power and/or the installation angle of the respective unit. 22. The method according to one of the preceding claims, characterized in that the control signal of the separating device is transmitted to the control unit of the separating device via wired communication, in particular via CAN bus or Ethernet technology, or via wireless communication, and preferably the control enable of the separating device is previously entered operator through the interface. 23. Method according to one of the preceding claims, characterized in that the control signal of the separating device and/or the operating parameter is displayed to the operator. 24. A method according to one of the preceding claims, characterized in that the optical imaging devices register only one-dimensional or two-dimensional information. 25. Machine (2) for harvesting root crops (4) and/or for separating root crops (4) from the rest of the heap containing impurities, containing a frame (12) of the machine, at least one transporting body (10), at least one an optical recording device (6), a separating device and a data analysis device, wherein at least one optical recording device (6) is installed with the ability to capture at least one reference image (8) of at least one part of the heap moved by means of the transporting body ( 10) with respect to the frame (12) of the machine, and transferring the control image (8) or the control data set generated using it to the data analysis device for implementing the method according to one of the previous paragraphs. 26. Machine (2) according to claim 25, characterized in that the data analysis device comprises a graphics processor device, in particular a general purpose graphics processor, and/or a processor device based on a user-programmable gate array. 27. The machine (2) according to claim 25 or 26, characterized in that it contains at least one sensor associated with the data analysis device, in particular an ultrasonic or tactile sensor for measuring the thickness of the heap layer on the conveying body (10), a sensor for measurement of drive power, located on the conveying body (10) speed sensor and/or humidity sensor. 28. Machine (2) according to one of claims 25-27, characterized in that it contains several imaging devices (6), each of which during operation takes at least one control image (8) of the same conveying body (10), or one and the same separating device, or different conveying elements (10), in particular, diverting heap components from the same separating device, or different separating devices. 29. The machine (2) according to one of claims 25-28, characterized in that the filming device (6) is located so that the control image (8) taken by it during operation at least partially displays at least two sections of the conveying body separated by at least one separating body (24, 32). 30. The machine (2) according to one of claims 25 to 29, characterized in that the filming device (6) is located in such a way that the control image (8) shows at least two conveying organs (10) of the machine (2) forming alternative feed paths for various heap components, in particular one feed path for root crops (4) and one feed path for impurities. 31. The machine (2) according to one of claims 25-29, characterized in that the conveying body (10) is made in the form of a screening conveyor (10A, 10B, 10E), or in the form of a needle conveyor (10C, 10D), which during the operating time passes, in particular, under at least one deflecting roller (24) extending across the conveying body (10) and deflecting the heap from it, either in the form of a screening star (10P, 10Q, 10S), or in the form of a conveying roller ( 10T), in particular, in - 15 042191 as part of a roller table (10M), or as a digging tool.