METHOD FOR OPTIMIZED DETERMINATION OF MACHINE PARAMETERS OF AN AGRICULTURAL WORK MACHINE

Description

[0001] The present invention relates to a method for the optimized determination of machine parameters of an agricultural machine according to the preamble of claim 1 and to an agricultural machine configured for the optimized determination of machine parameters of the agricultural machine according to the preamble of claim 19.

[0002] The focus here is on agricultural machinery of all kinds. This includes tractors, especially agricultural tractors, but also harvesting machines such as combine harvesters and forage harvesters.

[0003] Agricultural machinery is adapted to its specific task by means of a multitude of machine parameters. These parameters include, for example, engine speed, threshing drum speed, power take-off torque, the gap of a grain cracker, driving speed, and the like. Various approaches exist for optimizing these machine parameters to achieve specific goals. This optimization can be carried out by the user, primarily the driver, either manually or in conjunction with a driver assistance system. It can also be fully automated.

[0004] In a known approach, the user can specify strategy parameters for the driver assistance system, for example, by selecting a predefined strategy or weighting several competing optimization goals. The driver assistance system then translates these strategy parameters into optimized machine parameters using an application rule, which can, for example, be map-based.

[0005] EP 2 401 904 A2 describes how, for current agricultural machinery, optimized machine parameters are determined from a user's strategic specifications (in particular the selection and optimization criteria) via an application instruction (in particular the characteristic curves in combination with the control units). It is also known that the application instruction can, in principle, be adapted to the local context, for example by consulting an external advisor who contributes expert knowledge.

[0006] From EP 2 687 922 A2, a further development of this principle is known, in which the application rule, there the characteristic curve fields (short: characteristic fields), is successively improved in the field during the execution of an agricultural work task by visiting different working points in order to take into account the influence of the local context (weather, climate, crop types, soil types, etc.) on the theoretical characteristic curve fields.

[0007] From EP 3 180 974 A1, a method for the optimized determination of machine parameters of an agricultural working machine according to the preamble of claim 1, and an agricultural working machine equipped for the optimized determination of machine parameters of the agricultural working machine according to the preamble of claim 19 are known.

[0008] A problem with successively optimizing machine parameters for the local context is that this process takes time, for example, 30 minutes, to determine or parameterize the characteristic maps to such an extent that optimized machine parameters can then be derived from these maps for various changes, such as those in the field composition. To achieve this, the agricultural machine may, for example, cycle through several non-optimized operating points, i.e., combinations of machine parameters, in order to identify support points for the characteristic map(s). During this optimization routine, the agricultural machine therefore operates with changing and suboptimal settings of its machine parameters.

[0009] It is a challenge to achieve optimized machine parameters faster and / or more efficiently.

[0010] The invention is based on the problem of designing and further developing the known method in such a way that further optimization is achieved with regard to the aforementioned challenge.

[0011] The above problem is solved by the features of the characterizing part of claim 1.

[0012] The fundamental consideration is that the optimization routine can be shortened or even eliminated by using optimized strategy parameters that have already been optimized in a similar local context. These strategy parameters can be used, in particular, to start the optimization routine closer to the optimized machine parameters for the specific local context. Compared to previous optimization routines that use the same initial strategy parameters regardless of the local context, this results in a significant improvement. Furthermore, by analyzing many already optimized machine parameters, it becomes possible to better adapt agricultural machinery to the local context, especially for machines that cannot make these adjustments themselves.For example, a table with optimized machine parameters can be provided to them, depending on sensor measurement data, user specifications and / or other machine parameter settings, from which the driver assistance system then extracts the optimized machine parameters.

[0013] Specifically, it is proposed that the driver assistance system use as initial strategy parameters strategy parameters that have already been optimized by at least one agricultural machine in a similar local context.

[0014] In a particularly preferred embodiment according to claim 2, a control arrangement is provided which includes a database of stored strategy parameters. From these strategy parameters, the control arrangement selects suitable strategy parameters depending on the respective context and transmits them to the driver assistance system. The larger the database of strategy parameters becomes, the higher the probability of being able to determine strategy parameters that are already very well suited to the local context. For this purpose, for example, mean values ​​from many strategy parameters of agricultural machinery used in similar contexts can be used as initial strategy parameters.

[0015] In a further preferred embodiment according to claim 3, the agricultural machine performs an optimization routine during the execution of the agricultural task, thereby successively optimizing the initial strategy parameters. An advantage of this is that this optimization routine can be shortened by the improved initial strategy parameters and / or can at least be started with more optimal machine parameters during the optimization routine. The machine parameters thus optimized are preferably transmitted back to the control system.

[0016] According to claim 4, it can be provided that the user also intervenes in the optimization of the agricultural machinery and thus acts as a kind of human

[0017] A sensor and a human optimization routine function together. Together with the driver assistance system's optimization routine, jointly optimized strategy parameters can be determined, also based on the user's expert knowledge. These optimized strategy parameters can then be transmitted to the control unit.

[0018] Claim 5 specifies a preferred embodiment of the strategy and, in particular, the application procedure, in which at least one characteristic map is used to determine the optimized machine parameters. In the simplest case, these can be read directly from the characteristic map depending on the input parameters. However, according to claim 6, the strategy can, in particular, include a cost function that, as part of the application procedure, weights the strategy specifications. The objective of the cost function can be to minimize predefined costs, with the minimum of the cost function indicating where the machine parameters are optimal for the given strategy specifications. An input or change to the strategy specifications can be reflected in a change in the weights of the cost function.

[0019] Claim 7 specifies one possible configuration of the optimization routine. Accordingly, the driver assistance system can access reference points of the characteristic map by actually setting the corresponding machine parameters on the agricultural machine and measuring the resulting influences on the strategy specifications, directly or indirectly, using the sensor arrangement. The coefficients and / or a shift of a characteristic map can be determined from these reference points. Due to the improved initial strategy parameters, it is particularly possible in the optimization routine to only shift the characteristic map(s) and make no or only minor changes to the coefficients. As already mentioned, the optimization routine can be dependent on the initial strategy parameters. This configuration is the subject of claim 8.Accordingly, the support points can be selected depending on the initial strategy parameters and / or the optimization routine can be shortened. This results in an overall more efficient execution of the agricultural task and higher user acceptance.

[0020] Claim 9 relates to the preferred embodiment in which the initial strategy parameters comprise optimized machine parameters and these are, in particular, initially set on the agricultural machine until the optimization routine has progressed at least a certain distance. This ensures a more efficient execution of the agricultural task, especially at the start of the work.

[0021] The initial strategy parameters can include strategy presets and / or initial strategy parameters of the application instructions. For example, it may be intended that strategy presets set by other users in similar local contexts are used initially. The user can subsequently modify these presets. Such user-configuration of strategy presets is the subject of claim 11.

[0022] Claim 12 specifies a preferred embodiment of the application rule. Accordingly, the user interface, in the form of strategy specifications, can remain constant, while "under the hood" it leads to a different result with regard to the optimized machine parameters. This is a description of the particularly preferred functionality of the strategy, especially the application rule.

[0023] Claim 13 specifies a preferred embodiment of the control arrangement. According to this, it is located externally to the agricultural machinery and is accessible via the internet.

[0024] Claims 14 to 17 specify preferred embodiments of the local context or content of the local context data used to determine similarity. According to claim 14, the initial strategy parameters can, in particular, be passed on from day to day on the same agricultural machine. In the simplest case, the agricultural machine thus starts the next day with the initial strategy parameters already optimized the previous day. However, the time interval can also be longer, in particular an entire season.

[0025] Claim 15 relates to the possibility of transferring initial strategy parameters between different agricultural machines on the same field or in a spatially close field. In particular, initial strategy parameters are transferred between agricultural machines. This is useful, for example, if one of the agricultural machines starts performing the agricultural task later or is itself unable to determine optimized strategy parameters. The probability of a similar local context is higher for fields that are close together.

[0026] The similarity of the local contexts can be determined according to claim 16 using a similarity measure that specifically considers the climate zones of the respective local contexts. It has been found that agricultural machinery in different climate zones exhibits differently optimized machine parameters, even with the same strategy settings, and that sometimes different strategy settings are used. This influence of climate zones is not taken into account by existing driver assistance systems, which always use the same initial strategy parameters, in particular the same initial map. However, the initial optimization of the machine parameters can be improved particularly efficiently using a big data approach. For example, the control system can store strategy settings for many agricultural machines organized by climate zone, and their mean values ​​can be used as initial strategy settings.

[0027] According to claim 17, the climate zones can be subdivided into similar climate zones according to the Köppen-Geiger classification.

[0028] The proposed method is particularly advantageous for combine harvesters. It has been shown that the dependence on climatic zones leads to significantly different optimized machine parameters for combine harvesters. This embodiment is the subject of claim 18.

[0029] According to a further teaching as claimed in claim 19, which has independent significance, an agricultural working machine is set up for the optimized setting of machine parameters of the agricultural working machine.

[0030] It is essential that the driver assistance system is configured to use initial strategy parameters that have already been optimized by at least one agricultural machine in a similar local context.

[0031] Reference may be made to all statements regarding the proposed procedure.

[0032] The invention will now be explained in more detail with reference to a drawing that merely illustrates exemplary embodiments. The drawing shows Fig. 1 the proposed procedure in abstract form and Fig. 2 a) the investigation routine and b) the optimization routine.

[0033] The proposed solution can be applied to a wide range of agricultural machinery, especially harvesting machines. This includes combine harvesters, forage harvesters, tractors, and similar equipment.

[0034] Such agricultural machinery 1 is used for a wide variety of agricultural tasks. An example shows Fig. 1 a harvesting activity.

[0035] During these agricultural tasks, a wide variety of machine parameters can be set on the agricultural machinery 1. These machine parameters are machine parameters in the narrow sense, such as engine speed or the position of a throttle valve. Settings of the rear linkage and similar components are also included. However, these settings can generally be adjusted indirectly by the driver assistance system 2, for example, by generating control signals for a dedicated control unit of a working unit and transmitting them to that unit.

[0036] These machine parameter settings have a strong influence on the outcome of the agricultural task. For example, it is possible to harvest a field quickly or, conversely, to achieve high harvest quality.

[0037] The embodiment shown in the figures, which is preferred in this respect, relates to a method for the optimized determination of machine parameters of an agricultural working machine 1.

[0038] It should be noted that the optimization of machine parameters is a process in which an optimization goal exists in principle, but will by no means always be achieved. Optimized machine parameters 3 are therefore neither optimal nor necessarily always more optimal than before. They are merely the result of an optimization. Furthermore, not all target variables of the optimization are necessarily detectable by sensors, and in particular, not always directly detectable, but may only be indirectly derivable.

[0039] The agricultural machine 1 has a driver assistance system 2 and a sensor array 4. As in Fig. 1 As schematically indicated, the agricultural work machine 1 performs an agricultural work task in a local context.

[0040] The local context encompasses all influencing factors that are fundamentally present at the location where the agricultural task is performed, i.e., locally, and that affect the outcome of the agricultural task. However, where the term is used here, the local context refers only to that portion of these influencing factors that is also known to the agricultural machine 1, in particular the driver assistance system 2, and / or the control arrangement 5, which will be explained later. These influencing factors can be measured, but can also be obtained, for example, from a weather service. Here, the local context preferably includes the weather and / or climate and / or crop type and / or variety and / or harvesting conditions such as straw or grain moisture and / or selected plant protection measures and / or ripening stage and / or soil condition. Additionally, machine equipment can be taken into account.GPS data can be used to determine the local context.

[0041] The driver assistance system 2 now determines optimized machine parameters 3 for the agricultural machine 1 for carrying out the agricultural work task in a determination routine 6 using a parameterizable strategy 7. This process is abstract in Fig. 2 a) This is shown and will be explained further below.

[0042] Strategy 7 comprises strategy specifications 8, an application rule 9, and the optimized machine parameters 3. More generally, strategy 7 can be parameterized by strategy parameters 10 to adapt it to a local context, as will be explained later.

[0043] In the determination routine 6, the driver assistance system 2 uses the strategy specifications 8 as input parameters of the application rule 9 in order to determine the optimized machine parameters 3 as output parameters of the application rule 9. The optimized machine parameters 3 are optimized with respect to the strategy specifications 8.

[0044] To determine the optimized machine parameters 3 for a new agricultural task, the driver assistance system 2 configures strategy 7 with initial strategy parameters at the start of the agricultural task. These initial strategy parameters can be adjusted during the execution of the agricultural task.

[0045] Strategy specifications 8 can include abstract strategies 7 and / or weightings of quality criteria. The abstract strategies 7 preferably include a machine-friendly strategy 7 with low wear of the autonomous agricultural machine 1 and / or an eco-mode with low energy consumption but longer operating time and / or fast processing with higher fuel consumption, and / or higher harvest quality with increased time expenditure and / or high throughput with lower harvest quality. The quality criteria preferably include competing quality criteria for which the user, as described in Fig. 2 Three quality criteria are shown as examples, and their weighting can be graphically adjusted. The graphical display visualizes the competition between the quality criteria. The optimization is therefore a multi-objective optimization. Here, and preferably, the strategy parameters are defined by the user.

[0046] The definition of each quality criterion is of particular importance here. Preferably, each quality criterion is defined in general terms by a target specification for the optimization or adjustment of a work process parameter. In the simplest case, the term "optimization" can encompass the maximization or minimization of the respective work process parameter. The term "adjustment" means that the respective work process parameter should assume a specific value, and the driver assistance system 2 optimizes the machine parameters so that this value is achieved, if possible. The quality criteria can preferably be selected from the list including "threshing yield," "broken kernel percentage," "separation losses," "cleaning losses," "threshing unit drive slippage," "fuel consumption," "throughput," "cleanliness," and "straw quality." A work process parameter could, for example, be a driving speed or a harvesting throughput.The work process parameters are determined by the interaction of machine parameters with local conditions.

[0047] It is essential that the driver assistance system 2 uses strategy parameter 10 as an initial strategy parameter, which has already been optimized by at least one agricultural machine 1 in a similar local context. The initial strategy parameters can have been optimized by one agricultural machine 1, or strategy parameter 10 from several agricultural machines 1 can be linked together.

[0048] Initial strategy parameters are strategy parameters 10 that are used at the beginning of the agricultural work task to parameterize strategy 7. In a preferred embodiment, the initial strategy parameters comprise initial coefficients of characteristic fields 12 from application rule 9, as will be explained later. These initial coefficients are used to parameterize initial characteristic fields 12. The initial characteristic fields 12 are used at the beginning of the agricultural work task. Therefore, the term "initial" in this context always refers to the beginning of an agricultural work task.

[0049] Instead of the agricultural machine 1 starting with a standard configuration, it uses initial strategy parameters that have a higher chance of leading closer to and faster to an optimum of the machine parameters.

[0050] The strategy parameters 10 can include strategy specifications 8, coefficients 11 of characteristic fields 12, and the like. To determine the similarity, a similarity measure can be defined, or, as will be explained later, climate zones can be used.

[0051] The optimized machine parameters 3 are set by the driver assistance system 2 on the agricultural machine 1 and used to carry out at least part of the agricultural work task.

[0052] Fig. 1 This shows that a control arrangement 5 for distributing optimized initial strategy parameters is provided here, and preferably. This control arrangement 5 can, in principle, be part of the agricultural machinery 1, in particular the driver assistance system 2. Here, however, and preferably, it is arranged externally.

[0053] The preferred procedure using a tax order 5 is explained below.

[0054] The control arrangement 5 includes a database 13 with stored strategy parameters 10 and associated data relating to local contexts. The database 13 can be part of the control arrangement 5 or located externally. The stored strategy parameters 10 were successively optimized by driver assistance systems 2 of agricultural machinery 1 in optimization routines 14 during the execution of agricultural tasks in local contexts, based on initial strategy parameters, to adapt to the respective local context. This data can be historical and / or current, and can originate from the local area or from anywhere in the world.

[0055] Here, and preferably, the driver assistance system 2 determines context data of the local context of the agricultural work task, in particular by means of the sensor arrangement 4, and transmits this data to the control unit 5. Additionally or alternatively, the control unit 5 can also determine the context data itself, in particular from GPS data of the agricultural machine 1. The determination can also be limited to a simple selection. Here, and preferably, however, average values ​​of a multitude of initial strategy parameters from similar contexts are calculated, or these are otherwise linked and transmitted as initial strategy parameters. When calculating the average values, the initial strategy parameters from similar contexts can be weighted, preferably depending on the reliability of the data and / or a similarity index.Additionally or alternatively, several initial strategy parameters, which are particularly far apart and thus potentially form local maxima, can be tried out.

[0056] The control arrangement 5 determines initial strategy parameters adapted to the local context based on a comparison of the received context data with the context data from database 13 and transmits these to the driver assistance system 2 of the agricultural working machine 1.

[0057] The driver assistance system 2 uses the transmitted initial strategy parameters as initial strategy parameters.

[0058] Furthermore, it is preferably provided that the agricultural machine 1, during the execution of the agricultural work task, particularly at the beginning of the execution of the agricultural work task, determines sensor data relating to the achievement of the strategy specifications 8 in an optimization routine 14 controlled by the driver assistance system 2 using the sensor arrangement 4, and that the driver assistance system 2 successively optimizes the initial strategy parameters based on the sensor data in the optimization routine 14 and thus further optimizes the optimized machine parameters 3.

[0059] In this way, the initial strategy parameters are used as the starting point of optimization routine 14, resulting in better results at the beginning of optimization routine 14. It also allows optimization routine 14 to be shortened.

[0060] Additionally, the transfer of initial strategy parameters and their use in later phases of the agricultural task can be provided, analogous to the described use. In particular, the agricultural machine 1 can compare its own performance with the performance of other agricultural machines 1 and, depending on the performance, reuse initial strategy parameters from another agricultural machine 1. Subsequently, an optimization routine 14 can optionally be executed again. In the event of a sensor failure that makes the execution of an optimization routine 14 difficult or impossible, initial strategy parameters from other agricultural machines 1 can be used regularly, for example, when the local context changes.

[0061] It may be provided that the driver assistance system 2 subsequently transmits the optimized strategy parameters 10 and the context data to the control arrangement 5.

[0062] It can also be provided that the user adjusts at least one, preferably several, of the strategy parameters 10 after and / or during the optimization routine 14, preferably that the driver assistance system 2 transmits the jointly optimized strategy parameters 10 and the context data to the control arrangement 5.

[0063] Of particular interest is the possibility of supplementing the optimization by the driver assistance system 2 by also recording how the user reacts to this optimization. This allows errors or shortcomings in the optimization routine 14 to be detected. Furthermore, these can be automatically corrected, if necessary, by enabling the control arrangement 5 to access this jointly optimized data in the future.

[0064] Here, the user preferably sets the strategy parameters and / or optimization criteria. However, it is also possible that the user can intervene more deeply in the system and even adjust individual machine parameters.

[0065] The proposed solution is particularly suitable in the case where strategy 7 includes at least one characteristic map 12 that maps relationships between the machine parameters and the strategy specifications 8, especially in the form of mathematical functions. Such characteristic maps 12 are found in the Figure 1 and 2The diagram illustrates that it is difficult, if not impossible, to mathematically represent the functionality of an agricultural machine 1 at all conceivable operating points and taking all interrelationships into account. However, it is possible to model some operating points and interrelationships and, within certain limits, achieve good optimization results. Using support points 15, which will be explained later, an initial characteristic map can be adapted to the local context. Here, and preferably, several characteristic maps 12 are provided, which, for example, represent individual working units of the agricultural machine 1.

[0066] Using application rule 9, the driver assistance system 2 determines the optimized machine parameters 3 based on the strategy specifications 8 from the map 12 or maps 12. In particular, the driver assistance system 2 simply reads the optimized machine parameters 3 from the maps 12.

[0067] The characteristic map 12 or maps 12 have or have coefficients 11 that parameterize the mathematical functions. These are determined or corrected here, preferably from the support points 15.

[0068] As an alternative to one or more characteristic fields 12, a tabular representation of empirically determined relationships can also be provided, which replaces the characteristic fields 12. In this case, it is particularly advantageous if the tabular representation is adapted to the context. The tabular representation is not necessarily further optimized in an optimization routine 14.

[0069] In the Fig. 2 In the embodiment shown and thus preferred, strategy 7 comprises a cost function 16. The cost function 16 weights the strategy specifications 8 with weights 17. Fig. 2Figure 16 shows such a cost function purely as an example and in a simplified form. The weights can correspond to the weightings of the quality criteria mentioned above or be derived from them, but can also be more or less independent of them.

[0070] By means of the application rule 9, the driver assistance system 2 determines the optimized machine parameters 3 from the characteristic map 12 or the characteristic maps 12 such that the cost function 16 aims for a predefined goal, in particular the goal of minimizing the costs of the cost function 16.

[0071] The weights 17 are chosen such that minimal costs of the cost function 16 achieve a maximization of the achievement of the strategy objectives 8.

[0072] Furthermore, it is preferably provided that the driver assistance system 2 sets 14 different machine parameters in the optimization routine, which form the support points 15 of the map 12 or maps 12 and determines the coefficients 11 and / or at least one shift of a map 12 from the support points 15.

[0073] At the start of an agricultural task, the driver assistance system 2 has little information about the local physical influences. Therefore, and preferably at this stage, the driver assistance system 2 sets some machine parameters to non-optimal values ​​in order to measure the effect. These settings form the reference points 15 from which the characteristic map 12 or maps 12 are then determined. This process can take some time, which is why the improved initial strategy settings 8 have a significant effect on the work result, especially during this initial phase.

[0074] Here, and preferably, the duration of the optimization routine 14 can be shortened, in particular by reducing the number of support points 15. For this purpose, it can be provided that at least one, preferably several, coefficients 11 of the characteristic map 12 or maps 12 initially derived from the strategy specifications 8 are no longer changed, but rather the characteristic map 12 is only shifted.

[0075] Specifically, it can be provided that the optimization routine 14 is dependent on the initial strategy parameters, preferably that the driver assistance system 2 selects the support points 15 depending on the initial strategy parameters, and / or that the optimization routine 14 is shortened depending on the initial strategy parameters compared to an optimization routine 14 based on non-context-dependent initial strategy parameters.

[0076] For example, a characteristic map 12 can be reduced in size, since the probability of larger deviations of the initial operating point, i.e. the initial optimized machine parameters 3, from the theoretical optimum is likely to be lower.

[0077] It is also conceivable that in the optimization routine 14 a suitable characteristic field 12 or suitable characteristic fields 12 is or are selected from several predefined characteristic fields 12 based on the support points 15.

[0078] In one variant, it is provided that the initial strategy parameters include optimized machine parameters 3.

[0079] Preferably, the driver assistance system 2 initially sets the optimized machine parameters 3 of the initial strategy parameters as optimized machine parameters 3 on the agricultural machine 1 and only during or after the optimization routine 14 sets the machine parameters 3 optimized in the determination routine 6 on the agricultural machine 1.

[0080] Thus, the agricultural machine 1 starts directly with suitable machine parameters, for example first recording measurement data at this operating point and then successively varying the machine parameters to determine further support points 15.

[0081] Alternatively, the machine parameters determined from application instruction 9 are also set as optimized machine parameters 3 initially and / or essentially throughout the entire execution of the agricultural work task.

[0082] Furthermore, it is preferably provided here that the initial strategy parameters include strategy specifications 8 and / or initial strategy parameters of the application rule 9, preferably that the initial strategy parameters of the application rule 9 include weights 17 of the cost function 16 and / or coefficients 11 of the characteristic maps 12.

[0083] Alternatively, the initial strategy parameters can include other cost functions 16 or other characteristic fields 12 in the form of other mathematical functions of the characteristic fields 12.

[0084] Preferably, the strategy specifications 8 do not only include optimized machine parameters 3 and / or not only strategy specifications 8. Even if optimized machine parameters 3 are transmitted, the driver assistance system 2 determines optimized machine parameters 3 during the execution of the agricultural work task and sets them accordingly.

[0085] Furthermore, it is preferably intended that the strategy settings 8 are selected by the user, in particular at a terminal of the agricultural machine 1. This process is visualized Fig. 2 a) based on competing strategic guidelines 8.

[0086] Here, and preferably, the driver assistance system 2 is generally designed such that the same strategy settings 8 lead to different optimized machine parameters 3, depending on the strategy parameters 10. Preferably, the same strategy settings 8, depending on the initial strategy parameters, in particular the weights 17 of the cost function 16, also lead to different optimized machine parameters 3 even with theoretically identical characteristic maps 12. It is not necessary that these variants are actually implemented; rather, they are descriptions of the functionality.

[0087] As already mentioned, it is preferably intended that the control arrangement 5 is located externally to the agricultural machine 1. The control arrangement 5 can be formed by one or more servers and communicate with the agricultural machine 1 via the Internet. Additionally or alternatively, a large number of strategy parameters 10, originating from other, in particular similar, agricultural machines 1, can be stored in the database 13.

[0088] The control arrangement 5 can also be located on the agricultural machine 1 or be divided between the agricultural machine 1 and an external part. In the latter case, long-term strategy parameters 10, in particular climate-dependent strategy parameters 10, can be stored locally, and short-term strategy parameters 10 can be stored externally. In the event of a connection failure, the local strategy parameters 10 can be used.

[0089] Similar agricultural machinery 1 are those that exhibit fundamentally extremely similar behavior, i.e., in particular, agricultural machinery 1 of the same type with the same equipment.

[0090] The following section explains the sources from which the initial strategy parameters 8 can originate, whether or not they are transmitted via the control arrangement 5. The initial strategy parameters may have been optimized by the same agricultural machine 1 in the same or a similar local context. Specifically, the initial strategy parameters may have been optimized by the same agricultural machine 1 within a maximum of 14 days prior to the current day, or they may have been optimized in a previous season, particularly the same season. It is therefore possible that one and the same agricultural machine 1 may use its own optimization results as initial strategy parameters 8 in the future.As an alternative to the season, comparable periods of a harvest cycle or harvest year can also be used, especially if the distribution of sowing, harvesting, etc. varies between different locations.

[0091] The variants mentioned above can occur for different agricultural work tasks with the same agricultural machine 1, in particular depending on which initial strategy parameters are available and / or how similar the local contexts are or how the initial strategy specifications 8 are transferred between agricultural machines 1, possibly indirectly via the control arrangement 5.

[0092] Here, and preferably, it is provided that the current local context and the local context of the initial strategy parameters used, in particular those transmitted by the control arrangement 5, is the same field or a nearby field, preferably that the initial strategy parameters were optimized by another agricultural working machine 1.

[0093] This case is particularly interesting if an agricultural machine 1 later drives onto a field or cannot perform an optimization routine 14.

[0094] Furthermore, it is preferably provided here that the similarity of the local contexts is determined via a similarity measure, in particular by the control arrangement 5, preferably that the similarity measure takes into account, in particular primarily takes into account, climate zones of the respective local contexts.

[0095] It was found that climate zones have a significant influence on the function of agricultural machinery. This influence can be attributed to differences in soil, crop composition, and also to the behavior of the agricultural machinery under varying environmental conditions. It was also discovered that manual operators select different machine parameter settings in different climate zones.

[0096] The term "primary" here means that only initial strategy specifications 8 where the context data relate to the same climate zone are eligible.

[0097] Here, and preferably, initial strategy parameters of agricultural machinery 1 in the same climatic zones are used, preferably those that are currently performing an agricultural task or have done so within the last 24 hours. Individual initial strategy parameters or linked initial strategy parameters, in particular averaged initial strategy parameters, can be used.

[0098] Furthermore, it is and preferably intended that the climate zones are subdivided into similar climate zones according to the Köppen-Geiger classification.

[0099] Particularly preferred is that the agricultural machine 1 is a combine harvester, preferably that the optimized machine parameters 3 are the threshing drum speed and / or the gap width of the threshing drum.

[0100] According to a further teaching, an agricultural machine 1 is proposed to be configured for the optimized determination of machine parameters of the agricultural machine 1, wherein the agricultural machine 1 has a driver assistance system 2 and a sensor arrangement 4, wherein the agricultural machine 1 is configured to perform an agricultural work task in a local context, wherein the driver assistance system 2 is configured to determine optimized machine parameters 3 for the agricultural machine 1 for performing the agricultural work task in a determination routine 6 by means of a parameterizable strategy 7, wherein the strategy 7 comprises strategy specifications 8, an application rule 9 and the optimized machine parameters 3, and wherein the strategy 7 is parameterizable by strategy parameter 10 to adapt to a local context.wherein the driver assistance system 2 is configured to use the strategy specifications 8 as input parameters of the application rule 9 in the determination routine 6 in order to determine the optimized machine parameters 3 optimized with respect to the strategy specifications 8 as output parameters of the application rule 9, and wherein the driver assistance system 2 is configured to parameterize the strategy 7 with initial strategy parameters at the beginning of the execution of the agricultural work task in order to determine the optimized machine parameters 3.

[0101] Essential according to this further teaching is that the driver assistance system 2 is set up to use strategy parameter 10 as initial strategy parameter, which has already been optimized by at least one agricultural machine 1 in a similar local context.

[0102] This agricultural machine 1 is used here and preferably in the proposed method and is therefore particularly equipped for use in the proposed method.

[0103] Reference may be made to all statements regarding the proposed procedure. Reference symbol list

[0104] 1 Agricultural machine 2 Driver assistance system 3 Optimized machine parameters 4 Sensor arrangement 5 Control arrangement 6 Determination routine 7 Strategy 8 Strategy specifications 9 Application instructions 10 Strategy parameters 11 Coefficients 12 Characteristic map 13 Database 14 Optimization routine 15 Support points 16 Cost function 17 Weights

Claims

1. Method for the optimized determination of machine parameters of an agricultural working machine (1), wherein the agricultural working machine (1) has a driver assistance system (2) and a sensor assembly (4), wherein the agricultural working machine (1) performs an agricultural task in a local context, wherein the driver assistance system (2) determines in a determination routine (6) machine parameters (3) optimized by means of a parameterizable strategy (7) for the agricultural working machine (1) to perform the agricultural task, wherein the strategy (7) comprises strategy specifications (8), an instruction for use (9) and the optimized machine parameters (3), and wherein the strategy (7) can be parameterized by strategy parameters (10) for adapting to a local context, wherein the driver assistance system (2) uses the strategy specifications (8) in the determination routine (6) as input parameters of the instruction for use (9) in order to determine the optimized machine parameters (3), optimized with respect to the strategy specifications (8), as output parameters of the instruction for use (9) and wherein the driver assistance system (2) parameterizes the strategy (7) with initial strategy parameters at the beginning of performing the agricultural task in order to determine the optimized machine parameters (3), characterized in that the driver assistance system (2) uses as initial strategy parameters strategy parameters (10) which have already been optimized by at least one agricultural working machine (1) in a similar local context.

2. Method according to Claim 1, characterized in that a control assembly (5) for allocating optimized initial strategy parameters is provided, in that the control assembly (5) has a database (13) with stored strategy parameters (10) and associated data on local contexts, in that the stored strategy parameters (10) have been successively optimized by driver assistance systems (2) of agricultural working machines (1) in optimization routines (14) during the performance of agricultural tasks in local contexts from initial strategy parameters for adaptation to the respective local context, in that the driver assistance system (2) determines context data of the local context of the agricultural task, in particular by means of the sensor assembly (4), and transmits them to the control assembly (5), in that, on the basis of a comparison of the received context data with the context data from the database (13), the control assembly (5) determines initial strategy parameters adapted to the local context and transmits them to the driver assistance system (2) of the agricultural working machine (1) and in that the driver assistance system (2) uses the transmitted initial strategy parameters as initial strategy parameters.

3. Method according to Claim 1 or 2, characterized in that, during the performance of the agricultural task, in particular at the beginning of the performance of the agricultural task, the agricultural working machine (1) determines in an optimization routine (14), controlled by the driver assistance system (2), by means of the sensor assembly (4) sensor data relating to the achievement of the strategy specifications (8) and in that, on the basis of the sensor data, the driver assistance system (2) successively optimizes the initial strategy parameters in the optimization routine (14) and thus further optimizes the optimized machine parameters (3), preferably in that the driver assistance system (2) transmits the thus optimized strategy parameters (10) and the context data to the control assembly (5).

4. Method according to Claim 3, characterized in that the user adjusts at least one, preferably more than one, of the strategy parameters (10) after and / or during the optimization routine (14), preferably in that the driver assistance system (2) transmits the thus jointly optimized strategy parameters (10) and the context data to the control assembly (5).

5. Method according to one of the preceding claims, characterized in that the strategy (7) comprises at least one characteristic map (12) which depicts relationships between the machine parameters and the strategy specifications (8), in particular in the form of mathematical functions, in that, by means of the instruction for use (9), the driver assistance system (2) determines, in particular reads off, the optimized machine parameters (3) on the basis of the strategy specifications (8) from the characteristic map (12) or the characteristic maps (12), preferably in that the characteristic map (12) has coefficients (11) which parameterize the mathematical functions.

6. Method according to Claim 5, characterized in that the strategy (7) comprises a cost function (16), in that the cost function (16) weights the strategy specifications (8), in that, by means of the instruction for use (9), the driver assistance system (2) determines the optimized machine parameters (3) from the characteristic map (12) or the characteristic maps (12) in such a way that the cost function (16) aims for a predefined target, in particular the target of minimizing costs of the cost function (16).

7. Method according to one of Claims 3 to 6, characterized in that, in the optimization routine (14), the driver assistance system (2) sets various machine parameters which form interpolation points of the characteristic map (12) or the characteristic maps (12) and determines from the interpolation points the coefficients (11) and / or at least one shift of a characteristic map (12).

8. Method according to one of Claims 3 to 7, characterized in that the optimization routine (14) is dependent on the initial strategy parameters, preferably in that the driver assistance system (2) chooses the interpolation points (15) depending on the initial strategy parameters, and / or in that, depending on the initial strategy parameters, the optimization routine (14) is shortened compared to an optimization routine (14) on the basis of non-context-dependent initial strategy parameters.

9. Method according to one of the preceding claims, characterized in that the initial strategy parameters comprise optimized machine parameters (3), preferably in that the driver assistance system (2) initially sets the optimized machine parameters (3) of the initial strategy parameters as optimized machine parameters (3) on the agricultural working machine (1) and only in the course of or at the end of the optimization routine (14) sets the machine parameters (3) optimized in the determination routine (6) on the agricultural working machine (1).

10. Method according to one of the preceding claims, characterized in that the initial strategy parameters comprise strategy specifications (8) and / or initial strategy parameters of the instruction for use (9), preferably in that the initial strategy parameters of the instruction for use (9) comprise weights (17) of the cost function (16) and / or coefficients (11) of the characteristic maps (12).

11. Method according to one of the preceding claims, characterized in that the strategy specifications (8) are selected by the user, in particular at a terminal of the agricultural working machine (1).

12. Method according to one of the preceding claims, characterized in that, depending on the strategy parameters (10), the same strategy specifications (8) lead to different optimized machine parameters (3), preferably in that, depending on the initial strategy parameters, in particular weights (17) of the cost function (16), the same strategy specifications (8) lead to different optimized machine parameters (3) even when there are theoretically identical characteristic maps (12).

13. Method according to Claim one of Claims 2 to 12, characterized in that the control assembly (5) is located externally to the agricultural working machine (1), in particular in that the control assembly (5) is formed by one or more servers and communicates with the agricultural working machine (1) via the Internet, and / or in that a multiplicity of strategy parameters (10) originating from other, in particular identical, agricultural working machines (1) are stored in the database (13).

14. Method according to one of the preceding claims, characterized in that the initial strategy parameters have been optimized by the same agricultural working machine (1) in the same or a similar local context, preferably in that the initial strategy parameters have been optimized by the same agricultural working machine (1) within at most 14 days before the current date, or in that the initial strategy parameters have been optimized at a time of year in the past, in particular the same time of year.

15. Method according to one of the preceding claims, characterized in that the current local context and the local context of the initial strategy parameters that are used, in particular those transmitted by the control assembly (5), is the same field or a nearby field, preferably in that the initial strategy parameters have been optimized by another agricultural working machine (1).

16. Method according to one of the preceding claims, characterized in that the similarity of the local contexts is determined by way of a similarity measure, in particular by the control assembly (5), preferably in that the similarity measure considers, in particular primarily considers, climatic zones of the respective local contexts.

17. Method according to Claim 16, characterized in that the climatic zones are divided into similar climatic zones according to Köppen-Geiger classification.

18. Method according to one of the preceding claims, characterized in that the agricultural working machine (1) is a combine harvester, preferably in that the optimized machine parameters (3) the rotational speed of the threshing drum and / or the gap width of the threshing drum19. Agricultural working machine configured for the optimized determination of machine parameters of the agricultural working machine (1), wherein the agricultural working machine (1) has a driver assistance system (2) and a sensor assembly (4), wherein the agricultural working machine (1) is configured to perform an agricultural task in a local context, wherein the local context is formed by influencing variables which exist at the location where the agricultural task is performed and have an influence on the result of the agricultural task and are known to the agricultural working machine, wherein the driver assistance system (2) is configured to determine in a determination routine (6) machine parameters (3) optimized by means of a parameterizable strategy (7) for the agricultural working machine (1) to perform the agricultural task, wherein the strategy (7) comprises strategy specifications (8), an instruction for use (9) and the optimized machine parameters (3), and wherein the strategy (7) can be parameterized by strategy parameters (10) for adapting to a local context, wherein the driver assistance system (2) is configured to use the strategy specifications (8) in the determination routine (6) as input parameters of the instruction for use (9) in order to determine the optimized machine parameters (3), optimized with respect to the strategy specifications (8), as output parameters of the instruction for use (9) and wherein the driver assistance system (2) is configured to parameterize the strategy (7) with initial strategy parameters at the beginning of performing the agricultural task in order to determine the optimized machine parameters (3), characterized in that the driver assistance system (2) is configured to use as initial strategy parameters strategy parameters (10) which have already been optimized by at least one agricultural working machine (1) in a similar local context.