METHOD FOR CONTROLLING AN AGRICULTURAL HARVESTING CAMPAIGN

Description

[0001] The present invention relates to a method for controlling an agricultural harvesting campaign according to claim 1 and to a self-propelled agricultural working machine for carrying out such a method according to claim 18.

[0002] The term "self-propelled agricultural machine" is to be interpreted broadly in this context. This includes not only harvesting machines such as combine harvesters and forage harvesters, but also tractors and similar vehicles.

[0003] The known method for controlling a harvesting operation in the form of a combine harvester (DE 10 2016 118 651 A1), on which the invention is based, uses machine-local field information provided by the sensors of a work machine to optimize harvesting process chains. The optimization is based on field information from an environment detection device assigned to the respective work machine. Using the environment detection device, crop properties such as crop density or grain moisture can be determined and processed as part of the optimization.

[0004] Another method for controlling a harvesting campaign is known, for example, from EP 2 174 537 A1.

[0005] The optimized control of an entire harvest campaign, in which predetermined harvesting actions are to be carried out on a field contingent, i.e. on a field area consisting of one or more fields, within a campaign period, represents a complex challenge depending on the optimization criteria, which can only be partially mastered with the known methods.

[0006] The invention is based on the problem of designing and developing the known method in such a way that the optimization of the control of an agricultural harvest campaign is improved with low control complexity.

[0007] The above problem is solved by a method having the features of claim 1.

[0008] The first and most important consideration is that the control of the harvest campaign is structured into different control-technical application levels. This distributes the control of the harvest campaign across the application levels, keeping the control-technical complexity of each application level low.

[0009] It is also essential that information is generated at each application level and that the generated information is made available to all other application levels. The generation of information can be, in a narrow sense, the data processing of input data to produce output data, or, in a broad sense, the receipt of data from a data source. With this consistent availability of field information across all application levels, all the information required for efficient optimization of harvest campaign control is always available.

[0010] Finally, it is essential that the generated information includes remote sensing field information. This demonstrates the particular advantage of the availability of information at all application levels. The generation, i.e., reception, of remote sensing field information is usually more complex than the generation of local information, such as local environmental information. This is especially true when the remote sensing field information is obtained from an external service provider.

[0011] The inventive design concerns the structuring of the application levels into a coarse application level, a fine application level, and a machine application level. This three-tiered division has proven particularly easy to implement in control systems with regard to the generation of processing sequences, harvesting actions, and machine control data.

[0012] In the further preferred embodiment according to claim 2, the harvesting campaign is controlled by a control system in different control routines, which, with a suitable definition of the control routines, as will be explained later, is accompanied by a corresponding reduction in complexity. The term "control routine" is to be understood as a routine that can run sequentially and / or in parallel with other routines and that can be implemented in software in the usual way. The control routines run on a control system that can be equipped with distributed control hardware in a manner to be explained later.

[0013] As mentioned above, harvest control requires not only machine-local information, but also information from remote data sources. Such information is referred to herein as remote sensing field information. Ideally, both types of information are continuously available in all control routines.

[0014] Preferably, in at least one of the control routines, georeferenced remote sensing field information from data sources remote from the machine fleet is processed, wherein in at least one of the control routines, georeferenced live field information from machine-specific data sources, in particular sensor data sources, of the machine fleet is processed, and wherein at least part of this field information is processed in at least two control routines (claim 3).

[0015] Furthermore, according to claim 4, it has been found to be advantageous to define a total of at least three control routines, each of which is assigned to one of the above application levels and which can run at least for the most part simultaneously.

[0016] In detail, with regard to the definition of the control routines, it is proposed that the processing sequence of the harvesting actions over the campaign period be generated in a coarse control routine based on the above field information, that harvesting process chains be generated in a fine control routine based on the field information for implementing the processing sequence, and that machine control data be generated during harvesting operation in a machine control routine based on the field information for implementing the harvesting process chains.

[0017] The basic control routine therefore only concerns the chronological order in which the various harvest actions should be processed over the campaign period. A harvest action can involve harvesting an entire field or harvesting a field zone (which will be explained later).

[0018] The detailed control routine concerns the generation of the harvesting process chains required for the execution of the above harvesting actions. This involves the definition of subprocesses such as harvesting subprocesses, loading subprocesses, and the like. Finally, the machine control routine concerns the generation of specific machine control data for controlling the work machines.

[0019] By structuring the control routines into the coarse control routine, the fine control routine and the machine control routine, the complexity with regard to the control of an entire agricultural harvest campaign is reduced, so that the increase in complexity associated with the processing of the remote sensing field information and the machine-local field information itself is compensated to a certain extent.

[0020] This also enables the processing of both of the aforementioned types of field information in all three control routines, where appropriate. For example, the coarse control routine can also access the machine-local field information to check whether the live field information is consistent with the remote sensing field information. If necessary, the processing sequence can be adjusted or the remote sensing field information can even be corrected during harvesting.

[0021] Likewise, it is conceivable that in the detailed control routine, for example when defining sub-processes of overloading, the crop density of an entire field is derived from the remote sensing field information, so that a correspondingly long-term planning of the sub-processes of overloading is possible.

[0022] Finally, in addition to the live field information, the remote sensing field information can also be used in the machine control routine, for example to estimate whether fluctuations in the machine-local field information are likely to be just "outliers" or a systematic trend.

[0023] From all of the above, it follows that the proposed method provides a particularly good basis for optimizing the control of an agricultural harvest campaign, which allows for a particularly high level of robustness, especially in view of the expected redundancies between the remote sensing field information and the live field information.

[0024] The proposed method can be operated on a distributed system hardware according to claim 4, which allows an adaptation of the control system to the respective control system boundary conditions.

[0025] The further preferred embodiment according to claim 5 relates to the division of the field contingent assigned to the harvest campaign into georeferenced field zones, each of which has a uniform property, and the definition of the field zones is stored in the control system. This particularly simplifies the generation of the processing sequence in the coarse control routine by considering the field zones as a unit in the relevant planning steps.

[0026] The machine control data are always generated during harvesting operations within the machine control routine, as proposed in claim 7. However, the generation of the processing sequence within the coarse control routine and the generation of the harvesting process chains within the fine control routine are preferably also generated at least before harvesting operations. Furthermore, all control routines can run in an overlapping manner.

[0027] The equally preferred embodiments according to claims 8 and 9 relate to the storage of already generated, optimized harvesting process chains or machine control data in the control system, specifically with an explicit reference to the respective field zone. Thus, it is provided that the field zones are assigned not only a property, but also optimized harvesting process chains or machine control data. This demonstrates that structuring the field contingent into field zones is accompanied by a considerable simplification of the relevant planning steps. This is particularly the case if, as proposed in claim 10, the harvesting of the field zones is provided as a unit.

[0028] A time-of-day-dependent change in the crop properties of the field contingent as a whole or of the field zones is covered by claims 11 to 13. For this purpose, the remote sensing field information and / or the live field information and / or the field zones are each referenced to the time of day. For the field zones in particular, this means that the division of the field contingent into field zones according to claim 12 must be carried out and stored separately for different times of day. With the latter, preferred embodiments, a control of the agricultural harvest campaign optimized for the respective time of day can be created without excessively increasing the complexity of the associated planning steps.

[0029] The further preferred embodiment according to claim 14 takes into account the fact that remote sensing field information is often not available in absolute values, so that calibration of this information is always necessary.

[0030] According to claim 15, the machine control routine preferably provides for the field zones to be displayed to the machine operator, so that the operator has access to the information underlying the processing sequence. The definition of the field zones allows for a particularly clear, particularly graphical, display.

[0031] The further preferred embodiments according to claims 16 and 17 relate to advantageous variants for the generation of the remote sensing field information on the one hand and the live field information on the other hand.

[0032] The invention will be explained in more detail below with reference to a drawing which merely represents an exemplary embodiment. Fig. 1 a graphical representation of an agricultural harvesting campaign and Fig. 2 a graphical representation of the control routines used to control the agricultural harvesting campaign according toFig. 1 underlie.

[0033] The proposed solution can be applied to a wide range of agricultural machinery 1. In the Fig. 1 In the scenario indicated, the work machines 1 are assigned to a machine fleet 2. The work machines 1 are, for example, combine harvesters, transport vehicles, balers, tractors or the like.

[0034] The proposed method is used to control an agricultural harvest campaign. The term "harvest campaign" is defined here as meaning that, during the harvest campaign, predetermined harvesting actions E 1 -E n are carried out on a field contingent 3 assigned to the harvest campaign within a campaign period t K by several agricultural working machines 1 of the machine fleet 2. The term "field contingent" refers to a field or a plurality of fields that are to be processed within the campaign period t K. In the Fig. 2 In the scenario shown, the field contingent 3 comprises two fields 3a, 3b, which are directly adjacent to each other at a boundary line 4. In principle, however, it can also be provided that the fields 3a, 3b of a field contingent 3 are arranged at least partially separately from each other.

[0035] The campaign period tK can vary depending on the prevailing conditions. Preferably, the campaign period tK represents a harvest period for the crop in question, which typically extends over several months.

[0036] To control such an agricultural harvest campaign, the working machines 1 must be deployed in the most optimal chronological order possible, using the most optimal machine control data possible. Extensive field information is available for this purpose, which can be georeferenced as well as time-of-day referenced. This means that the relevant field information assumes different values ​​depending not only on location but also on time.

[0037] In order to be able to convert the above-mentioned complex control task into simple sub-control tasks, it is proposed that the control of the harvest campaign be carried out at different control-related application levels. Information required for the implementation of control tasks is generated at each application level. Generally speaking, the generated information is provided to all other application levels. The generated information includes, at a minimum, remote sensing field information, which will be explained later. The resulting reduction in complexity in the control of the harvest campaign has already been explained above.

[0038] Here, and preferably, three application levels are defined, which is not to be understood as limiting. In a coarse application level, the processing sequence A of the harvesting actions E 1 -E n over the campaign period t K is generated and made available to the other application levels. In a detailed application level, harvesting process chains P 1 -P n are generated and in turn made available to the other application levels. Finally, in a machine application level, machine control data M 1 -M n are generated and made available to the other application levels. This means that the various application levels ultimately represent planning levels in which the harvesting campaign is planned with varying degrees of abstraction.

[0039] In the preferred implementation of the proposed method, each application level is assigned a control routine in which the part of the harvesting campaign control allocated to the respective application level is carried out. Preferably, a coarse control routine is assigned to the coarse application level, a fine control routine to the fine application level, and a machine control routine to the machine application level.

[0040] Preferably, the harvesting campaign is controlled by a control system 5, preferably in the various control routines 6 to 8. As indicated above, the term "control routine" is to be understood as meaning that a control process of some kind is executed, which is particularly software-based on the system hardware of the control system 5, which will be explained later. The control routines 6-8 can run sequentially and / or at least partially overlapping, i.e., simultaneously.

[0041] As also indicated above, control routines 6-8 have access to extensive field information for processing the control tasks. In this case, it is assumed that at least two basic types of field information are available. First, it is intended that at least one of the control routines 6-8 will process georeferenced remote sensing field information 9 from data sources 10 remote from the machine fleet 2. This refers to field information that is stored, i.e., stored, in data sources 10 remote from the machine fleet.

[0042] Furthermore, it is provided that in at least one of the control routines 6-8, georeferenced live field information 11 from machine-specific data sources, in particular sensor data sources such as the sensors 12, of the machine fleet 2 is processed. The live field information 11 here and preferably is the field information that is generated and provided locally on the machine, i.e., by the machine's own sensors 12, during ongoing harvesting operations. This can be machine-internal field information such as the layer height, the grain moisture content, or the like of the crop flow passing through the working machine 1. Alternatively or additionally, the live field information 11 can also be environmental information such as the crop density or the crop moisture content in the immediate vicinity of the working machine 1.

[0043] What is interesting about the proposed solution is the fact that at least some of this field information is available in at least two control routines. This means that at least some of the remote sensing field information 9 and the live field information 11 is available in at least two control routines 6-8. This is particularly noteworthy because, for example, in the coarse control routine 6, not only the remote sensing field information 9 but also the live field information 11 can be accessed, even though the planning of the processing sequence A of the harvesting actions E 1 -E n does not actually require the highly dynamic nature of the live field information 11. However, it has been recognized that the live field information 11 can be used, for example, to adapt the remote sensing field information 9 to the actual conditions in the environment of the respective work machine 1, as will be explained below.

[0044] In general, it can even be provided that each control information available in one control routine 6-8 is available in all other control routines 6-8. This consistency in the availability of field information across all control routines 6-8 and thus across all application levels allows for optimization of the harvest campaign, taking into account all field information relevant to the individual case.

[0045] The remote sensing field information 9 includes field information relating to a field contingent 3 assigned to the harvest campaign. Such field information includes, in particular, biomass maps, grain mass maps, NKB mass maps (NKB non-grain components), ripening maps, yield maps, swath placement maps, soil moisture maps, dew maps, weed maps, weather information, or the like. This remote sensing field information 9 is generated or provided by the relevant system hardware 13, 14 from satellites, drones, remote databases 15, or the like.

[0046] The live field information 11, on the other hand, is generated or provided by the machine-side sensors of at least one work machine 1, here and preferably several work machines 1, of the machine fleet 2.

[0047] The proposed structuring of the control routines 6-8 is of particular importance in this case. Firstly, a coarse control routine 6 is provided in which, based on the field information 9, 11, the processing sequence A of the harvesting actions E 1 -E n is generated over the campaign period t K . To control this processing sequence A, harvesting process chains P 1 -P n are generated in a fine control routine 7 based on the field information 9, 11, which relate to the sub-processes of the harvesting actions E 1 -E n underlying the processing sequence A. These include, for example, overloading sub-processes, transport sub-processes, or the like. Finally, during harvesting, machine control data M 1 -M n for controlling the work machines 1 are generated in a machine control routine 8 based on the field information 9, 11. The machine control data is data that is required for the operation of the work machine 1.These include, for example, driving machine settings such as the target forward speed and the target tire pressure or unit settings such as threshing unit settings, attachment settings or the like.

[0048] For the preferred case where all field information 9, 11 is constantly available in all control routines 6-8, the possibility of optimizing the control of the agricultural harvest campaign with particularly high dynamics arises because the dynamic live field information 11 is now consistently processed in all control routines 6-8. Because a certain redundancy exists between the remote sensing field information 9 and the live field information 11, the above optimization is highly robust, since incorrect information does not necessarily lead to incorrect control.

[0049] With the proposed solution, the control of the harvest campaign can be closely aligned with current and dynamically changing conditions. According to the proposal, a prediction of a quality characteristic can also be made over a day based on the remote sensing field information 9 and the live field information 11, which can be further refined continuously if necessary, particularly based on the live field information 11.

[0050] The proposed method can be implemented using very different hardware on which the respective control routines 6-8 run. Preferably, at least a portion of the control system 5, here and preferably the coarse control routine 6 and the fine control routine 7, runs or runs on a machine-remote system hardware 13. Additionally, it is preferably the case that at least a portion of the control system 5, here and preferably the machine control routine 8, runs on a machine-side system hardware 14. In this respect, the control system 5 is a hardware-distributed control system.

[0051] The machine-side system hardware 14 can, in principle, be configured as part of an already provided driver assistance system of the respective work machine 1. However, it is also conceivable that this machine-side system hardware 14 is implemented in addition to an existing machine control system or an existing driver assistance system.

[0052] Fig. 1shows a further interesting variant of the proposed solution, in which the field quota 3 assigned to the harvest campaign is divided into georeferenced field zones 16, in particular to simplify the generation of the processing sequence A of the harvesting actions E 1 -E n. Specifically, it is proposed here that in the coarse control routine 6, the field quota 3 assigned to the harvest campaign is divided into georeferenced field zones 16 depending on at least one property derived from the remote sensing field information 9 and / or live field information 11, in each of which one property is uniformly present. For example, such a field zone 16 can be defined by the grain moisture content of the grain crop to be harvested being within a predetermined moisture range. Fig. 1 the identically defined field zones 16 have been provided with correspondingly identical hatching.

[0053] By dividing field allocation 3 into georeferenced field zones of the same crop property, field allocation 3 can be easily structured with regard to the respective crop property of interest. A general advantage of this is the fact that harvesting operations can be carried out differently for the different field zones. If field zones of the same crop density are defined, working machines 1 with different throughputs can be distributed among field zones 16 of different crop density, for example, in such a way that an overall optimal machine utilization is achieved.

[0054] The machine control data are generated within the machine control routine 8, preferably during harvesting. In contrast, the processing sequence within the coarse control routine 6 and the harvesting process chains P 1 -P n within the fine control routine 7 are generated at least before harvesting. In a particularly preferred embodiment, the processing sequence and the harvesting process chains P 1 -P n are continuously optimized, i.e., also during harvesting, with regard to the current field information 9, 11.

[0055] The definition of the above field zones 16 not only simplifies the structuring of the harvesting operation, but also enables the easy reusability of optimization results already achieved for the respective field zone 16. Specifically, it is preferably provided that, within the framework of the fine control routine 7, optimized harvesting process chains P 1 -P n are generated according to an optimization strategy. Alternatively or additionally, it can be provided that, within the framework of the machine control routine 8, optimized machine control data M 1 -M n are generated according to an optimization strategy. It is now essential that the optimized harvesting process chains P 1 -P n and / or the optimized machine control data M 1 -M n can be stored together with the respectively assigned field zone 16 as remote sensing field information 9 in the control system 5.This fundamentally allows for the multiple use of optimization results for spatially separated field zones 16 with the same properties. In particular, the optimization results generated on a first work machine 1 can be used for another work machine 1. Only the first-mentioned work machine 1 needs to be equipped with a control system for the optimization, while the other work machine 1 only needs to be equipped with a rudimentary control system.

[0056] In this context, it is generally provided that, within the framework of the fine control routine 7 and / or within the framework of the machine control routine 8, the remote sensing field information 9 assigned to a field zone 16 is used as the basis for generating the harvesting process chains P 1 -P n and / or the machine control data M 1 -M n for this field zone 16. As explained above, this remote sensing field information 9 can be optimization results, for example in the form of optimized aggregate settings or the like.

[0057] A particularly simple generation of the processing sequence A is achieved by defining the processing sequence A of the harvesting actions E 1 -E n in the coarse control routine 6 such that the field zones 16 are each harvested as a unit. This uniform treatment of the field zones 16 within the control routines results in a particularly significant reduction of the complexity mentioned above.

[0058] Given that the crop properties in question here, in particular grain moisture, are subject to time-dependent, particularly time-of-day, changes, it is preferably provided that the processing sequence is generated in the coarse control routine 6 with a time reference, in particular a time-of-day reference, whereby the remote sensing field information 9 and / or the live field information 11 are correspondingly time-of-day referenced for this purpose. In this example, the term "time-of-day referenced" means that, as part of the generation of the processing sequence, the harvesting actions E 1 -E n are assigned a time stamp that also includes a time of day for the processing of the respective harvesting action E 1 -E n. For this purpose, it is particularly advantageous if the respective field information 9, 11 is time-of-day referenced, as indicated above.

[0059] As shown in the illustration Fig. 1As can be seen, here and preferably also the field zones 16 are time-of-day referenced. For this purpose, the rough control routine 6 divides the field quota 3 assigned to the harvest campaign into field zones 16 separately for different times of day and stores them in the control system.

[0060] The above-mentioned, time-of-day referenced processing sequence is shown, for example, by the fact that in the coarse control routine 6, in the event that at least some of the field zones 16 assume changing property values, here and preferably grain moisture values, at different times of the day, the processing sequence A is generated over the course of the day with regard to the changing property values. Fig. 1 This is indicated by the fact that the processing sequence A, depending on the expected processing time, provides for the harvesting of the next field zone, which preferably has an identical field zone definition.

[0061] A particularly advantageous feature of the proposed solution is the fact that remote sensing field information 9 and live field information 11 can be present simultaneously, for example, when processing sequence A is generated within the coarse control routine 6. Here, a comparison can be made between the remote sensing field information 9 and the live field information 11, so that the outliers in the field information mentioned above are ignored. In general, it can be provided that at least part of the remote sensing field information 9 is calibrated based on the live field information 11. This is also appropriate given that remote sensing field information 9 is often not available in absolute values, but only in relative values.

[0062] The high level of clarity of the aforementioned field zones 16 can also be advantageously utilized by the machine operator. For this purpose, the field zones 16 are preferably transferred in the machine control routine 8 to the respective machine control system of at least some of the work machines 1 and displayed by the respective machine control system. This allows the operator of the respective work machine 1 to check whether the division of the field zones 16 with regard to the respective crop property corresponds to the actual conditions or not. Based on this, the operator can intervene in the control system of the work machine 1 if necessary.

[0063] According to a further teaching of independent significance, the agricultural work machine 1 is claimed as such for carrying out a method according to the proposal explained above. Reference is made to all relevant statements. List of reference symbols

[0064] 1Working machine 2Machine fleet 3Field contingent 3a, 3bFields E 1 -E n Harvesting actions 4Boundary line 5Control system 6Coarse control routine 7Fine control routine 8Machine control routine 9Remote sensing field information 10Remote data source 11Live field information 12Sensors P 1 -P n Harvesting process chains M 1 -M n Machine control data 13Remote system hardware 14Machine-side system hardware 15Remote database 16Field zones

Claims

1. A method for controlling an agricultural harvesting campaign by means of a control system (5) in which, on a field allotment (3) associated with the harvesting campaign, prespecified harvesting actions (E1 - En) are executed within a campaign time period (tK) by a plurality of agricultural working machines (1) of a fleet of machines (2), wherein the control of the harvesting campaign is implemented by means of the control system in different control engineering application levels in which information is respectively generated, wherein the respective generated information is made available to all other application levels and wherein the generated information comprises remote sensing field information, wherein in a coarse application level, the execution sequence (A) for the harvesting actions (E1 - En) over the campaign time period (tK) is generated and made available to the other application levels, in that in a fine application level, harvesting process chains (P1 - Pn) are generated and made available to the other application levels, and in that in a machine application level, machine control data (M1 - Mn) are generated and made available to the other application levels.

2. The method according to claim 1, characterized in that a respective control routine (6-8) is associated with the application levels, in which control routine the portion of the control of the harvesting campaign allocated to the respective application level is implemented, preferably in that a coarse control routine (6) is associated with the coarse application level, in that a fine control routine (7) is associated with the fine application level and in that a machine control routine (8) is associated with the machine application level.

3. The method according to claim 2, characterized in that in at least one of the control routines (6-8), georeferenced remote sensing field information (9) from data sources which are remote with respect to the machines of the fleet of machines are processed, in that in at least one of the control routines (6-8), georeferenced live field information (11) from machine-specific data sources of the fleet of machines (2), in particular sensor data sources (12), are processed, and in that at least a portion of this field information (9, 11) is processed in at least two control routines (6-8), preferably in that all of the control information which is available in a control routine (6-8) is available in all of the other control routines (6-8).

4. The method according to claim 2 and claim 3, characterized in that in a coarse control routine (6), the execution sequence (A) for the harvesting actions (E1 - En) over the campaign time period (tK) is generated on the basis of the field information (9, 11), wherein in a fine control routine (7), in order to control the execution sequence (A), harvesting process chains (P1 - Pn) are generated on the basis of the field information (9, 11), and in that in a machine control routine (8) which runs during the harvesting operation, in order to control the harvesting process chains (P1 - Pn), machine control data (M1 - Mn) are generated on the basis of the field information (9, 11).

5. The method according to claim 4, characterized in that at least a portion of the control system (5), in particular the coarse control routine (6) and the fine control routine (7), runs or run on system hardware (13) which is remote from the machines, and in that at least a portion of the control system (5), in particular the machine control routine (8), runs on machine-side system hardware (14).

6. The method according to claim 4 and according to claim 5 if appropriate, characterized in that in the coarse control routine (6), as a function of at least one property of the material which is derived from the remote sensing field information (9) and / or live field information (11), the field allotment (3) associated with the harvesting campaign is divided into georeferenced field zones in which a respective property of the material is consistent, and in that the definition of the field zones is stored in the control system (5).

7. The method according to claim 4 and according to claim 5 or claim 6 if appropriate, characterized in that the machine control data (M1 - Mn) are generated in the context of the machine control routine (8) during the harvesting operation and also in that the execution sequence in the context of the coarse control routine (6) and the harvesting process chains (P1 - Pn) in the context of the fine control routine (7) are generated at least before the harvesting operation.

8. The method according to claim 4 and according to one of claims 5 to 7 if appropriate, characterized in that in the context of the fine control routine (7), optimised harvesting process chains (P1 - Pn) are generated in accordance with an optimisation strategy, and / or in that in the context of the machine control routine (8), optimised machine control data (M1 - Mn) are generated in accordance with an optimisation strategy, and in that the optimised harvesting process chains (P1 - Pn) and / or the optimised machine control data (M1 - Mn) together with the respective associated field zone (16) are stored in the control system (5) as the remote sensing field information (9).

9. The method according to claim 4 and according to one of claims 5 to 8 if appropriate, characterized in that in the context of the fine control routine (7) and / or in the context of the machine control routine (8), the remote sensing field information (9) associated with a field zone (16) form the basis of the generation of the harvesting process chains (P1 - Pn) and / or the generation of the machine control data (M1 - Mn) for this field zone (16).

10. The method according to claim 4 and according to one of claims 5 to 9 if appropriate, characterized in that in the coarse control routine (6), the execution sequence (A) for the harvesting actions (E1 - En) is defined in a manner such that the field zones (16) are respectively harvested as a single entity.

11. The method according to claim 4 and according to one of claims 5 to 10 if appropriate, characterized in that in the coarse control routine (6), the execution sequence (A) is generated in a time-referenced manner, in particular with reference to the time of day, and in that to this end, the remote sensing field information (9) is with reference to the time of day.

12. The method according to claim 4 and according to one of claims 5 to 11 if appropriate, characterized in that the field zones (16) are with reference to the time of day and in that in the coarse control routine (6), the division of the field allotment (3) associated with the harvesting campaign into field zones (16) is implemented separately for different times of the day and is stored in the control system (5).

13. The method according to claim 4 and according to one of claims 5 to 12 if appropriate, characterized in that in the coarse control routine (6), in the case in which at least a portion of the field zones (16) assume varying property values for the material, in particular grain moisture values, at different times of the day, the execution sequence (A) is generated over the course of the day taking the varying values in the property of the material into consideration.

14. The method according to claim 4 and according to one of claims 5 to 13 if appropriate, characterized in that at least a portion of the remote sensing field information (9) is calibrated on the basis of the live field information (11).

15. The method according to claim 4 and according to one of claims 5 to 14 if appropriate, characterized in that in the machine control routine (8), the field zones (16) are transmitted to the respective machine controller of at least a portion of the working machines (1) and displayed by means of the respective machine controller.

16. The method according to claim 4 and according to one of claims 5 to 15 if appropriate, characterized in that the remote sensing field information (9) comprises field information relating to the field allotment (3) associated with the harvesting campaign, in particular biomass maps, grain mass maps, non-grain constituent mass maps, ripening maps, yield maps, swathe deposit maps, ground moisture maps, dew maps, weed maps, weather information or the like, which are provided by the data sources which are remote from the machines, in particular from satellites, drones, databases which are remote from the machines, or the like.

17. The method according to claim 4 and according to one of claims 5 to 16 if appropriate, characterized in that the live field information (11) is generated by the machine-side sensors (12) of at least one working machine (1), preferably a plurality of working machines (1), of the fleet of machines (2).