Method for travel route planning

Abstract

To enable efficient agricultural processing of an unknown processing area.SOLUTION: A method for travel route planning includes: manually controlling and passing an agricultural machine through a primary area (23) of a processing area (20) while automatically identifying position data of the agricultural machine (10); at least partially automatically identifying a primary boundary which surrounds the primary area (23) on the basis of, the position data; defining a secondary outer boundary of a secondary area (24), the secondary outer boundary being arranged outside the primary area (23) to meet two connection points, where a part of the primary boundary located between the connection points forms an inner boundary between the primary area (23) and the secondary area (24), the inner boundary, together with the secondary outer boundary, completely surrounding the secondary area (24); automatically identifying a travel route for processing at least the secondary area (24); and automatically generating control data for controlling the agricultural machine (10) during the field processing, the control data representing an identified travel route.SELECTED DRAWING: Figure 2

Description

[Technical field]

[0001] The present invention relates to a method for planning a driving route according to the preamble of claim 1, a computer system for planning a driving route according to the preamble of claim 15, an agricultural machine according to the preamble of claim 16, and a computer program product according to claim 17.

[0002] When treating an agricultural useful surface, for example a given treatment area, the agricultural machine gradually travels over the entire surface on several travel tracks. In the inner area, for example, parallel travel tracks can be used. To transition between the parallel travel tracks, a turning path is required that is traversed in the headland arranged on the edge side of the inner area. In the case of harvesters, for example combines or forage harvesters, the headland is usually treated, i.e. harvested, first and then the inner area. Unnecessary travel over untreated cultivated areas in the headland is thus avoided. It is basically known to plan travel paths for treatment areas with known dimensions, which can then be traversed manually or at least partially autonomously by the agricultural machine. In contrast, the treatment of fields with unknown dimensions represents a special challenge, which in practice regularly occurs. In this case, the necessary information on which the path planning can be based is currently missing.

[0003] The object of the present invention is to enable efficient agricultural treatment of unknown treatment areas.

[0004] The above problem is solved by a method with the features of the independent claim 1. Advantageous embodiments emerge from the dependent claims.

[0005] To this end, a method for planning a travel path for field treatment is provided, the method comprising the steps of: the primary area of ​​the processing area is traversed under manual control by the agricultural machine, during which the position data of the agricultural machine are automatically determined, A primary boundary enclosing the primary region is at least partially automatically determined based on the position data; a secondary outer boundary of the secondary region is defined, the secondary outer boundary being disposed outside the primary region, the secondary outer boundary abutting the primary boundary at two connection points, a portion of the primary boundary located between the connection points forming an inner boundary between the primary region and the secondary region, the inner boundary together with the secondary outer boundary completely surrounding the secondary region; A travel path for processing at least the secondary area is automatically determined, During the field processing, control data for controlling the agricultural machine are automatically generated, the control data representing the determined travel path; A method is provided.

[0006] The method is intended for planning a travel path for field treatment, in which a treatment area is treated by an agricultural machine. The agricultural machine can also be called an agricultural work machine. Such an agricultural machine explicitly includes pairs with a towing vehicle and at least one machine, such as a towed vehicle towed or drawn by the towing vehicle. The agricultural machine can be a harvester, in particular a forage harvester, a combine, a baler or a loading vehicle. However, it can also be a tractor, for example with a haymaker, a plow, a fertilizer spreader, a liquid fertilizer tank, etc. In the context of field treatment, the agricultural machine treats a treatment area, for example an arable land, a field or a part thereof, in which case the terms "field" and "arable land" are used synonymously for the sake of simplicity. In the treatment area, field treatments, such as, for example, plowing, fertilizing, mowing, haying, harvesting, etc., are carried out, which does not necessarily mean that the entire treatment area is treated, although this is desirable.

[0007] The method comprises the steps described below, which may not necessarily be performed in the order explicitly described. The temporal sequence of the two steps may be reversed from the order of their names. It is also possible for the two steps to be performed completely or partly in parallel.

[0008] According to one method step, a primary area of ​​the processing area is traversed by the agricultural machine under manual control, with the position data of the agricultural machine being automatically determined. The primary area is a part of the processing area, and in this case the term "primary area" is used primarily for the purpose of distinction and should not be interpreted as limiting. The primary area can also be traversed by the agricultural machine and, in this case, processed, among others. In this case, the agricultural machine is controlled manually, i.e. by control actions of a user. The user can be a driver who accompanies the agricultural machine. However, the user can also be a person who remotely controls the agricultural machine. The user ensures by manual control that the agricultural machine moves within the processing area and also within the primary area. Conversely, it can also be said that the primary area is defined by the movement of the agricultural machine in this method step. When passing through the primary area, the position data is automatically determined. The position data is generally data that corresponds to the position of the agricultural machine. This position can be an absolute position (for example GPS coordinates) or a relative position with respect to a known reference point. In particular, a GNSS receiver can be used to ascertain the current position of the agricultural machine. The position data is determined automatically, i.e. without human intervention. The position data can also be saved. These actions and further automatically performed actions can be performed by a computer system, the functions of which can of course be partially implemented by software. In addition to the planar or spatial position of the agricultural machine, the position data can also include information about the orientation of the agricultural machine, i.e. for example about the horizontal or azimuth angle. In addition to the position data, it can be determined whether processing is taking place at a particular position, i.e. whether the processing mechanism of the agricultural machine is operating. Insofar as processing takes place during passage, the primary area can be defined as the entire processed surface.

[0009] Furthermore, based on the position data, a primary boundary surrounding the primary area is at least partially automatically determined. That is, the position data is used to completely or partially determine the boundary of the primary area, which is referred to as the primary boundary in this specification. This is also performed automatically, for example by the computer system described above. In the case of partial determination, only one section of the primary boundary is determined, which is sufficient in some cases. The primary boundary can be derived directly from the position data. However, especially when processing the primary area, it is sensible to take into account the geometry of the agricultural machine, in particular the working width with which the processing mechanism acts on the processing area.

[0010] In a further step, a secondary outer boundary of the secondary region is defined, which is located outside the primary region, the secondary outer boundary abutting the primary boundary at two connection points, the part of the primary boundary located between the connection points forming an inner boundary between the primary region and the secondary region, the inner boundary together with the secondary outer boundary completely surrounding the secondary region. The definition of the secondary outer boundary can be performed on the user's side or automatically, as will be further explained below. The secondary outer boundary is a part of the secondary boundary completely surrounding the secondary region. In this case, the secondary outer boundary is always defined so that it is connected to the primary boundary at two connection points, i.e. so that it merges with the primary boundary. The primary boundary must be specified in the aforementioned step to a size such that at least these connection points are included in this primary boundary. The part of the primary boundary located between the connection points forms an inner boundary. The term inner boundary therefore derives from the fact that this inner boundary is located inside the processing region, while the secondary outer boundary (like the primary outer boundary of the primary region) limits the processing region to the outside. That is, the secondary region is adjacent to the inner boundary of the primary region, while the secondary region is adjacent to the periphery of the processing region with a secondary outer boundary. The secondary region is therefore not completely surrounded by the primary region. Therefore, when the primary region runs, a run around the secondary region never occurs completely. Therefore, the run of the primary region cannot completely determine or define the secondary boundary, i.e. the boundary of the secondary region. Therefore, a part of the secondary region, i.e. the secondary outer boundary, is defined in particular in this processing step, and is defined independently of the position data.

[0011] In a further step, preferably carried out after the above-mentioned steps, a travel path for processing at least the secondary region is automatically determined. The travel path is determined automatically, i.e., for example by the above-mentioned computer system, without the need for human thought activity for this purpose. The travel path is used to process at least the secondary region, which, although advantageously, is processed as completely as possible, does not have to be processed completely. When determining the travel path, of course, it is utilized that the secondary boundary, consisting of the inner boundary and the secondary outer boundary, is completely known. Thus, a travel path adapted to the geometry of the secondary region can be determined.

[0012] In a further method step, control data for controlling the agricultural machine during the field processing are automatically generated, the control data representing the determined travel path. The control data in any case includes information required for controlling the agricultural machine along the travel path. The agricultural machine can be controlled according to the control data during the field processing. The term "control" generally refers here to any intended influence on the orientation and / or movement state of the agricultural machine, i.e. for example steering, acceleration, braking, etc. Optionally, the control data may also partly relate to the control of other functions of the agricultural machine, for example the lift. In any case, the agricultural machine may be controlled based on the control data, ideally to follow the determined travel path. The form of the control data as well as its content may be selected differently, inter alia depending on the respective agricultural machine and possibly further components used to implement the method. In particular, the travel path may be represented by a more or less dense series of waypoints. The control data may include explicit steering instructions or may only include position information for the waypoints, where the agricultural machine identifies appropriate steering parameters to get from one waypoint to the next.

[0013] The method makes it easier to process initially unknown treatment areas. In this case, the primary area is first scouted, so to speak, by being traversed under manual control. However, since the primary area does not include a complete drive around the secondary area, the information about the secondary area, i.e. the extent of the secondary outer boundary, must be supplemented in other ways. The method can be advantageous in different scenarios. For example, from the viewpoint of wildlife protection, it may be desirable not to completely drive and / or treat the edge of the treatment area for the time being, but to leave an untreated area available for wildlife as a transition to a protected area outside the field. It is also conceivable that the user wants to partially define the treatment area by manually controlling the agricultural machine, while partially considering another definition to be more sensible. In some circumstances, the definition of the secondary outer boundary may be more time-saving and / or accurate than, for example, a manually controlled drive. Finally, the route planning is performed automatically, which saves the user, for example the driver, complex deliberation.

[0014] The travel path is preferably planned taking into account the actual characteristics of the agricultural machine that is required to carry out the field treatment. More precisely, the travel path is preferably specified specifically for the agricultural machine that is to carry out the treatment, taking into account the geometry and / or driving characteristics of the agricultural machine. In case the agricultural machine uses attachments during the field treatment that can be removed as required and replaced by other attachments, the term "geometry of the agricultural machine" also explicitly includes the geometry of the attachments. Geometric characteristics that can be taken into account are the length and width of the agricultural machine, the working width (i.e. the width of the area where the crop is actually harvested), but also the attitude of the drive means (wheels, possibly tracks). In order to achieve a more realistic representation of the agricultural machine, the agricultural machine can also be simulated, for example, as polygons. Usually, in combination with the consideration of the geometry, the driving characteristics of the agricultural machine can also be taken into account. The driving characteristics can include the maximum possible speed (in some cases a distinction should be made between forward and reverse driving), acceleration and / or braking capacity, minimum radius of curvature, maximum steering angle, speed of change of steering angle, etc. It is also possible to distinguish what type of ground the agricultural machine is moving on, i.e. for example uphill, flat or downhill, dry or wet, etc. By taking into account at least one of the abovementioned variables, on the one hand it is possible to fully utilize the driving technical possibilities of the agricultural machine, i.e. for example to utilize the smallest curve radii, and on the other hand it is possible to avoid planning driving maneuvers which possibly cannot be performed at all by the agricultural machine in question.

[0015] It is possible to determine exactly one travel path according to a defined scheme, and then control data are generated for this exact one travel path. The travel path determined in this way would not be the best travel path (or at best it would only be the best travel path by chance) due to the many different processing areas and the many other possible variables, regardless of what the "quality" of this travel path is judged on the basis of. In order to achieve an improvement on this, it is highly preferable that an automatic optimization of the travel path is performed, by identifying an optimal travel path from a plurality of possible travel paths according to a corresponding optimization criterion, and the control data representing the optimal travel path are automatically generated. That is to say, a plurality of possible travel paths, which are so to speak candidates for the optimal travel path, are taken into account, and which of these plurality of travel paths is optimal is checked based on the optimization criterion for the travel path optimization. The term "optimal travel path" should be understood to mean that this optimal path is the best travel path found according to the optimization criterion. In fact, a better travel path may exist in some circumstances, but this better travel path was not found, for example because not enough different travel paths were checked. The optimization is performed automatically, i.e. by machine, i.e. computer-assisted. When the term "automatically" is used here and hereinafter, this includes, inter alia, the possibility that the corresponding process is performed, in whole or in part, by software implemented on suitable hardware. This and further steps of the method can be performed, for example, by a Farm-Management-Informations system (FMIS).

[0016] Preferably, the optimization criterion is based at least in part on the optimization of an optimization value determined by summing up the contributions of several individual path segments of one travel path. The optimization value is often also referred to as an expense value or cost value, in which case the optimization of the optimization value consists in its minimization. However, depending on the type or definition of the optimization value, it is also conceivable that the optimization consists in maximizing the optimization value. In the simplest case, the optimization criterion consists in minimizing (or maximizing) the optimization value. That is to say, the travel path which optimizes the optimization value is optimal. However, as will be explained further below, it is also possible that other variables, for example further optimization values, are taken into account, whereby the optimization of one optimization value competes with other goal settings. In any case, the optimization value is determined by summing up the contributions of several individual path segments of one travel path. The optimization value W is

number

[0017] For each automatic optimization, different methods for finding or searching for an optimal solution can be used, in particular metaheuristic methods such as simulated annealing or genetic or evolutionary algorithms.

[0018] According to one embodiment, the optimization criterion is based at least partially on the minimization of the traveled distance. "At least partially" in this context means that the minimization of the traveled distance does not have to be the only goal, but other values ​​are sought to be minimized or maximized, forming a compromise that may differ, for example, from the minimization of only the traveled distance. In particular, it can be envisaged to minimize the total traveled distance traveled on the travel path. Alternatively, it can also be envisaged to minimize the total traveled distance required for the turn. The latter traveled distance required for the turn can be regarded as a so-called "non-productive" traveled distance.

[0019] Alternatively or additionally, the optimization criterion can be based at least in part on the minimization of the travel time. In this case, on the one hand, the travel time for the entire travel path can be used as a basis. However, the travel times required for transitions between all tracks and / or for turns can also be taken into account as a whole, which are also "non-productive" travel times. Minimization of travel time is not necessarily synonymous with minimization of travel distance for various reasons. Thus, for example, the processing area may have a gradient, which causes a travel path in one direction to be traveled more quickly than a travel path in the opposite direction. Even if the length of the travel path is the same, the number and individual lengths of the travel paths traveled "downhill" or "uphill" as a whole may be different, and therefore the travel times required for the travel path as a whole may also be different.

[0020] Alternatively or additionally, the optimization criterion can be based at least in part on the minimization of energy consumption. In this case, the expected energy consumption for the entire travel path is usually taken into account. This energy consumption depends on the entire travel distance, but also on other parameters. Thus, if the agricultural machine has to overcome, for example, a more or less steep gradient, the energy consumption can also depend on the orientation of the individual travel paths. The number and individual lengths of the travel paths that are traveled "downhill" or "uphill" as a whole can also be different.

[0021] Alternatively or additionally, the optimization criterion can also be based at least in part on the minimization of the untreated area. In this case, it can also be said to be the maximization of the coverage or "coverage". That is, in practice, it is desirable to treat as much of the treated area as possible, ideally 100%. In this case, it is also possible to differentiate between the untreated area in the inner area and the untreated area in the headland. In general, there may be a conflict between the minimization of the untreated area and the minimization of the travel time, the travel distance and / or the energy consumption, and therefore it may be meaningful to take the treated area into account together, in order to avoid, for example, that a reduction in the travel time is achieved at the expense of an insufficient coverage.

[0022] A further possibility is that the optimization criterion is based, at least in part, on the minimization of the area that is run over multiple times. Running over an area multiple times can, on the one hand, intensify soil compaction, which is undesirable, and on the other hand, can mean running over crops that are present in the soil, in which case the crops are contaminated, damaged, or even pushed into the soil. Furthermore, running over multiple times implies a less than optimal efficiency in some situations, since one run is sufficient for processing.

[0023] In some situations, the optimization criterion may consist in minimizing or maximizing individual variables or individual optimization values, for example minimizing the total distance traveled. Depending on the nature of the treatment area, the type of field treatment to be performed, the performance data of the agricultural machine, and other factors, the minimization or maximization of one optimization value may compete to some extent with the equally desirable minimization or maximization of another optimization value. In this case, optimizing each individual optimization value individually often does not lead to a satisfactory solution. Therefore, in one embodiment, the optimization criterion is based on optimizing a weighted combination of multiple optimization values. For example, one optimization value may be the untreated area, while another optimization value is the travel time. The optimization criterion may then consist in minimizing a sum, where one augend is proportional to the travel time and the other augend is proportional to the untreated area. The relative weights of each optimization value can be adapted by selecting an appropriate weighting or normalization factor. This sum may be calculated as:

number

[0024] Alternatively, the optimization criterion can be based on Pareto optimization of multiple optimization values. That is, a parameter set that optimizes the optimization value is searched (within the parameter range to be tested) until no other parameter set improves one of the optimization values ​​more than itself without worsening the other optimization values. For example, for a certain travel route, if there is no other travel route that provides a shorter travel time without providing a longer travel distance, and if there is no other travel route that provides a shorter travel distance without providing a longer travel time, then the travel route may be Pareto optimal in terms of travel time and distance.

[0025] According to one embodiment of the method, an inner area of ​​the treatment area and a headland arranged outside the inner area are defined, and a travel path is specified in such a way that the inner area is treated after the headland. Both the inner area and the headland are part of the treatment area and therefore have to be treated by the agricultural machine. The inner area preferably occupies a large part of the area of ​​the treatment area. The inner area is preferably traversed according to a plurality of parallel travel paths, which are preferably formed in a straight line and can be characterized by an orientation or treatment direction. To transition between the parallel travel paths, a turning path in the headland arranged on the edge side of the inner area is used. The headland is arranged on the edge side and may be partially limited by a primary outer boundary. The headland may be formed in a continuous manner or may have a plurality of separated partial surfaces. The headland may be arranged so as to surround the inner area or may leave part of the outer boundary open. This may be meaningful, for example, to leave open at least one refuge edge or refuge edge for wild animals. The inner area and the headland may be defined by the user or automatically. Since the headland should be utilized for the turning process when processing the inner area, the travel path is specified such that the inner area is processed after the headland. This includes the possibility that the headland has already been processed at the start of the route planning.

[0026] In the above-described embodiment, the travel path (both in the headland and in the inner area) can be characterized by a number of modification parameters, which are modified within the scope of the optimization in order to find the optimal travel path. In order to keep the calculation costs within an acceptable range, it makes sense to limit the number of modification parameters. Of course, the available computing power plays a role here. The limitation of the modification parameters also means that certain assumptions are made from the beginning about what basic characteristics the optimal travel path has. An advantageous assumption is that the travel path has a number of parallel travel paths in the inner area. The travel paths, which may also be called processing paths, correspond in this context to a part of the travel path of the agricultural machine in the inner area, so that by going around the individual travel paths, all or at least most of the surfaces of the inner area are gradually processed. The travel paths run parallel, i.e. the distance between two adjacent travel paths is constant over the entire length. However, in addition to the travel trajectories within the processing area, the turning paths within the headland when the agricultural machine transitions from one travel trajectory to another also belong to the travel path.

[0027] Insofar as it is a priori predefined that all the tracks in the inner area are parallel, they can be characterized as a whole by three parameters: orientation, positioning and order. In the case of linear tracks, the orientation corresponds to a horizontal or azimuth angle, which can also be said to be the direction in which all the tracks run. However, the orientation does not define the position of the individual tracks. This is done by positioning. One possibility is that the positioning for each track includes the two-dimensional coordinates of the points through which the respective tracks pass. Thus, the arrangement of the tracks is completely defined together with the orientation. However, in principle, the distance or lateral offset between adjacent tracks is the same for all tracks and corresponds to the track width. The track width can also be changed (within a given range) and thus be a further change parameter. Besides, if the track width is known, the position of all the tracks can be defined by a single point of one track, and the positioning can be expressed by the two-dimensional coordinates of this single point. Finally, the travel path can be characterized by the order in which the multiple travel trajectories (spatially defined by orientation and positioning) are traveled. This allows different driving strategies to be presented in which the agricultural machine transitions from one travel trajectory to the next one in space, or to the next one, or to the next one, etc. For example, if the processing area has an irregular shape, it may be meaningful for one sub-area to transition to the next travel trajectory, while in another sub-area to transition to the next one, etc. In terms of data volume, the order of N travel trajectories can be represented by an N-tuple, for example an N-dimensional vector.

[0028] As long as the three parameters mentioned above are changed, different strategies are possible. For example, a stepwise optimization can be performed. At the lowest optimization level, two parameters (e.g., positioning and orientation) can be kept constant, while a third parameter (e.g., order) is changed. At an intermediate optimization level, one parameter (e.g., positioning) is kept constant, while a second parameter (e.g., alignment) is changed, and for the third parameter, the optimal value identified at the lowest optimization level is used. At a higher optimization level, one parameter (e.g., positioning) is changed, while for the second and third parameters, the optimal values ​​found at the intermediate and lower optimization levels are used. Alternatively, at a higher optimization level, the two parameters (e.g., orientation and positioning) can be treated as equivalent change parameters, so that in some cases travel paths that differ in two parameters are compared. At a lower optimization level, the third parameter (e.g., order) is changed. In the case of headland processing, parallel driving paths can also be assumed in part, and the orientation of these parallel driving paths can then be oriented based on the borders of the processing area. However, particularly in the case of headland processing, very different steering processes are conceivable, particularly in the headland corner areas. Also in the headland, the deviation of the two parallel driving paths can be a meaningful change parameter.

[0029] One possible approach consists in performing manually controlled driving only in the area envisaged for the headland. The headland can thus be driven (and possibly processed) fully or partly under manual control. The remaining part of the headland (if necessary) and the inner area can then be processed based on the automatically determined driving path. In one embodiment, the headland is thus defined such that the primary area is entirely located within the headland, i.e. the primary area forms the headland or part of it. Conversely, the inner area is entirely located within the secondary area and possibly congruent with the secondary area.

[0030] When planning the travel route, the secondary region is taken into account in any case so that the secondary region can be completely or at least mainly processed. This is sufficient and efficient as long as the primary region is processed during travel. This is certainly an advantageous approach, but is not essential. Thus, in one embodiment, if it is determined that the primary region is at most partially processed during travel, a travel route is determined for processing the primary region and the secondary region. That is, the processing of the primary region, which is only partially or not performed at all, is followed or continued within the framework of the travel route. In this case, the travel route is preferably determined such that the primary region is at least mainly processed. Such a travel route may require that already processed parts of the primary region are traversed again in order to reach or exit the unprocessed parts.

[0031] The determination of the secondary outer boundary can be performed fully or partly automatically, i.e. for example by the computer system mentioned above. "Automatically" in this sense means that the important processes are not performed by human thought activity but by a machine. In this case, some of the possibilities listed below can be combined with each other, by ensuring the reliability of the determination by two independent methods, or by determining part of the secondary outer boundary by one method and other parts by the other method.

[0032] In one embodiment, the secondary outer boundary is automatically identified as a straight line between the end points of a concave section of the primary boundary, the end points forming a connection point. The primary boundary, or at least the identified part of the primary boundary, has a concave section, which can also be said to be a setback. Viewed from outside the primary area, the above-mentioned section is generally concave. This section is limited by an end point. "Concave" in this context should be understood as the antonym of "convex", in the sense that the connection line between the points of the concave section lies at least partially outside the primary area. A concave section can therefore have a convex or projecting subsection. In any case, the concave section and its end points are selected such that the straight line between the end points lies completely outside the primary area. This straight line is determined as the secondary outer boundary. That is to say, it is assumed that the primary area so to speak surrounds or partially surrounds the secondary area, and that the secondary outer boundary is a straight line. The latter is in reality a realistic assumption for many land parcels. This embodiment is particularly well suited if, during manually controlled travel, one side of the field should remain untreated in order to keep open an escape route for wild animals. In this case, the primary area can have a U-shape as a headland or as part of a headland, within which the secondary area is, so to speak, housed.

[0033] According to one embodiment, the secondary outer boundary is determined automatically, at least in part, based on stored geographical data of the periphery of the primary area. The "geographical data" can, for example, correspond to a satellite photograph showing the periphery of the primary area, whereby the primary area itself can be detected together. From the satellite photograph, for example by means of image recognition, the boundaries of the land parcel can be recognized. If these boundaries of the land parcel are combined with information in which land parcel the primary area is located, the secondary outer boundary can be derived therefrom. The geographical data can also be taken, for example, from a map, for example from a cadastral map. Even if the boundaries of the land parcel are essentially drawn on such a map, it is still necessary to automatically perform the correspondence between the primary area and the land parcel and to automatically recognize which boundary areas of the land parcel form the secondary outer boundary.

[0034] In a further embodiment, a sensing inspection of the periphery of the primary area is performed, and the secondary outer boundary is determined automatically, at least in part, based on the inspection. The sensing inspection is preferably based on a non-contact sensor, for example an optical sensor such as at least one camera. Such a camera may be arranged on a flyable drone controlled by the agricultural machine, so that an overview of the surroundings can be obtained in a short time. The acquired sensor data can be used to create an image of the surroundings, and then structures representing the secondary outer boundary are recognized (for example, using image recognition) within this image. For example, the borders of the cultivation area can be recognized, or soil coloration and / or soil structure within the treatment area can be identified, based on which the borders of adjacent areas can be identified.

[0035] In order, on the one hand, to be able to correct possible errors in the case of automatic identification, and, on the other hand, to better integrate the user into the automatically proceeding process, according to one embodiment, the automatically identified secondary outer boundary is defined after approval by the user. That is to say, the computer system is not able or authorized to define an automatically determined secondary outer boundary, which requires approval on the part of the user. The user can enter the approval via a suitable interface (e.g. mouse, keyboard, touch screen, microphone, etc.). If the user declines his approval, the user can be offered the possibility to modify the identified secondary outer boundary or to initiate another automatic identification.

[0036] It is also possible for the secondary outer boundary to be at least partially defined on the user's side. In this case, an input on the user's side is preferably envisaged. The definition on the user's side can be carried out, for example, on a screen, in such a way that a drawing of the surroundings, for example a satellite record or a cadastral map, is displayed to the user. In this case, the primary area and / or the determined primary boundary can preferably be displayed together. The user can then define the secondary outer boundary via an interface (for example mouse, keyboard, touch screen, etc.), for example by specifying individual points of the secondary outer boundary, which are then automatically connected. The definition on the user's side may also consist in a correction of the automatically determined secondary outer boundary.

[0037] Optionally, based on the position data, the primary outer boundary of the primary region can be automatically determined, which together with the secondary outer boundary of the secondary region completely surrounds the processing region. The primary outer boundary is that part of the primary boundary that does not form the inner boundary. The primary outer boundary is not adjacent to the secondary region, but is, so to speak, on the opposite side of the secondary region. The primary outer boundary together with the secondary outer boundary forms the entire outer boundary of the processing region. Furthermore, the primary outer boundary together with the inner boundary completely surrounds the primary region. As long as processing is performed during the run of the primary region, the primary outer boundary can be determined by determining which faces have been processed. From this, the primary boundary is obtained, and thus also the primary outer boundary, if the inner boundary is known. If processing is not performed, the primary boundary that would have been obtained, for example, if the processing mechanism had been operating at the same position data, can be used as a basis. The automatically determined primary outer boundary can be checked by the user and accepted or rejected. Identifying the primary outer boundary is important, on the one hand, when the primary region has not yet been completely processed. However, even if the primary area has been completely processed, the primary outer boundary and / or the overall outer boundary derived from the primary outer boundary can be stored for future use, so that during further processing (e.g. by other agricultural machinery), the geometry of the processing area is known from the start.

[0038] On the one hand, it is conceivable that the agricultural machine has a computer system (i.e. an evaluation and / or calculation unit) that can determine an optimal travel path based on sufficient information on the processing area and the possible evacuation area. However, in many cases, it is more efficient if the travel path is determined externally to the agricultural machine and the control data is generated externally and transmitted to the agricultural machine. The determination of the travel path and the generation of the control data are then carried out via a central system, in particular a central evaluation unit, which may even be accommodated stationarily in a building, which does not even have to be located in the vicinity of the processing area. The generated control data can be transmitted to the agricultural machine by wireless communication, which can then be used by the agricultural machine to navigate the travel path. Central external processing requires only small resources on the part of the agricultural machine in terms of computational capacity and memory space.

[0039] Control data for an autonomously traveling agricultural machine can be generated, and the agricultural machine autonomously performs field treatments based on the control data, i.e., as long as the agricultural machine is provided with the control data, the agricultural machine can navigate the determined travel path and perform the corresponding field treatments without human intervention. The agricultural machine can use different internal and / or external sensors for navigation, at least in part using the same sensors used to detect position data during manual controlled travel in the primary area. The agricultural machine can determine its current actual position based on the sensor data and compare it with a target position corresponding to an optimal travel path.

[0040] Control data for the agricultural machine steered by the driver can also be generated, so that control instructions for the driver can be generated based on the control data. In this case, the control data can be provided inside the agricultural machine and converted into control instructions. Alternatively, it is conceivable to create control instructions based on the control data from the outside and transmit them to the agricultural machine. It is also conceivable that the same agricultural machine is steered autonomously at one time and by the driver at another time. The control instructions can be output visually and / or acoustically. The control instructions can explicitly suggest to the driver how he should steer the agricultural machine or can display, for example, a target driving line on the screen, and the driver can orient his position based on this target driving line.

[0041] The above problem is a computer system for planning a travel path for field treatment, in which a treatment area is treated by an agricultural machine, and the computer system is configured to perform a process in which a primary area of ​​the treatment area is traversed by the agricultural machine while being manually controlled, and in this case, position data of the agricultural machine is automatically identified, configured to at least partially automatically identify a primary boundary surrounding the primary region based on the position data; a secondary outer boundary of the secondary region is defined, the secondary outer boundary being disposed outside the primary region, the secondary outer boundary abutting the primary boundary at two connection points, a portion of the primary boundary located between the connection points forming an inner boundary between the primary region and the secondary region, the inner boundary together with the secondary outer boundary completely surrounding the secondary region; configured to automatically determine a travel path for processing at least a secondary region; the system is configured to automatically generate control data for controlling an agricultural machine during the field processing, the control data being representative of the determined travel path, This is further solved by a computer system.

[0042] The above mentioned concepts have already been explained above with respect to the method according to the invention and therefore will not be explained again. The computer system comprises at least one computer, i.e. a computer, in particular a data processing unit. However, the computer system may also comprise further components, for example wireless and / or wired interfaces for unidirectional or bidirectional communication with other devices. Preferably, the computer system is configured to automatically detect position data and / or to determine a secondary outer boundary. Further advantageous embodiments of the computer system according to the invention correspond to advantageous embodiments of the method according to the invention.

[0043] In particular, the computer system may be an agricultural management information system arranged externally to the agricultural machine, for example arranged stationarily inside a building. The computer system may also be arranged in a mobile unit (laptop, tablet, smartphone, etc.) that displays control instructions for the driver or transmits control data (usually wirelessly) to the agricultural machine. Generally speaking, the computer system may be configured externally to the agricultural machine and may be configured to generate control data for transmission to the agricultural machine. The computer system may have an interface for data transmission to the agricultural machine and may be configured to transmit the control data to the agricultural machine (wired or wirelessly).

[0044] Alternatively, the computer system may be integrated into the agricultural machine, i.e. the computer system may be part of the agricultural machine and may be located inside the agricultural machine. The computer system may in any case be partly realized by software. Regardless of whether the computer system is part of the agricultural machine or not, the computer system may be configured to control the agricultural machine according to the control data.

[0045] According to a second alternative, the present invention provides an agricultural machine equipped with a computer system for planning travel paths for field treatment, the computer system being adapted to determine, when a treatment area is treated by the agricultural machine, a primary area of ​​the treatment area is traversed manually controlled by the agricultural machine, with the position data of the agricultural machine being automatically determined: configured to at least partially automatically identify a primary boundary surrounding the primary region based on the position data; a secondary outer boundary of the secondary region is defined, the secondary outer boundary being disposed outside the primary region, the secondary outer boundary abutting the primary boundary at two connection points, a portion of the primary boundary located between the connection points forming an inner boundary between the primary region and the secondary region, the inner boundary together with the secondary outer boundary completely surrounding the secondary region; configured to automatically determine a travel path for processing at least a secondary region; the system is configured to automatically generate control data for controlling an agricultural machine during the field processing, the control data being representative of the determined travel path, Agricultural machinery will also be provided.

[0046] Preferably, the computer system is configured to automatically detect the position data and / or to determine the secondary outer boundary. Here again, the preferred embodiments of the agricultural machine according to the invention correspond to the preferred embodiments of the method according to the invention.

[0047] According to the invention there is further provided a computer program product comprising program code means enabling a computer system to carry out the above mentioned steps. The computer program product thus comprises software implementing the above mentioned steps on the hardware of the computer system. The computer program product may be in the form of a data medium on which the software, i.e. the program code means (volatile or non-volatile), is stored. The data medium may in this case be permanently embedded in the computer system and / or may be embeddable.

[0048] The invention will be explained below on the basis of the drawings, which are merely illustrative and are not intended to limit the general idea of ​​the invention. [Brief description of the drawings]

[0049] [Figure 1] FIG. 2 shows a plan view of a first processing area, an agricultural machine and a computer system for driving path planning according to the invention during a first phase of the method according to the invention; [Diagram 2] FIG. 2 is a plan view of the treatment area of ​​FIG. 1 during a second phase of the method, together with agricultural machinery. [Diagram 3] FIG. 2 is a plan view of the treatment area of ​​FIG. 1 during a second phase of the method, together with agricultural machinery. [Figure 4] FIG. 2 shows a plan view of the second treatment area and the agricultural machine during a first phase of the method according to the invention. [Diagram 5] FIG. 2 is a plan view of the third treatment area and the agricultural machine during a first phase of the method according to the invention. [Figure 6] 3 is a flowchart of a method for driving route planning according to the present invention.

[0050] FIG. 1 shows a plan view of a treatment area 20, i.e. a plan view of a field adjacent to an evacuation area 30, for example a forest area. An agricultural machine 10, for example a combine harvester, and a computer system 1 according to the invention for travel path planning are shown in a highly schematic manner, where the computer system 1 is arranged outside the agricultural machine 10, for example in a building that may be far away from the treatment area 20. The computer system 1 may be formed by or may be part of an agricultural management information system. The computer system 1 has an interface, not shown here, for wireless data transmission to the agricultural machine 10. The agricultural machine 10 performs a field treatment in the treatment area 20, for example mowing grass. The evacuation area 30 is considered as a possible evacuation zone for wild animals that escape from the treatment area 20 when the field treatment is performed.

[0051] Initially, the boundaries of the processing area 20 are not known to the computer system 1. The boundaries of the processing area 20 are detected in two stages. In a first stage, the agricultural machine 10 passes through the primary area 23 and processes this area 23 during its passage. That is to say, the cultivated area, which is indicated by the hatched area in the figure, is harvested in the primary area 23. This processing is carried out on the edge side of the cultivated area, while the evacuation edge 28 adjacent to the evacuation area 30 is left open. Meanwhile, position data is detected, for example via a GNSS receiver inside the agricultural machine 10, and transmitted to the computer system 1. The position data can also include data regarding the orientation of the agricultural machine 10, as well as data revealing whether a harvester has been operated at a particular position of the agricultural machine 10. The position data can be three-dimensional, so that for example a height profile of the primary area 23 can be detected. By association with data describing the geometry of the agricultural machine 10, the shape of the processed primary area 23 can be determined with good precision. Here, the computer system 1 automatically determines the primary boundary G1 of the primary region 23. In this case, it is identified that the primary boundary G1 has a concave section K. Assuming that the surface located within the concave section K belongs to the processing region 20, the end point of the concave region K is the connection point P A These connection points P A The secondary area 24 is defined by selecting the straight line between the primary outer boundary A1 and the connecting point P2 of the secondary area 24. This decision, which is made automatically on the part of the computer system 1, can be accepted or rejected by the user. After the secondary outer boundary A2 is defined, the primary boundary G1 is connected to the primary outer boundary A1 by the connecting point P2. A and an inner boundary I located between the primary region 23 and the secondary region 24. The inner boundary I is interposed between the primary region 23 and the secondary region 24, while the primary outer boundary A1 and the secondary outer boundary A2 together completely surround the treatment region 20.

[0052] For the now completely determined treatment area 20, an optimal driving route F is determined that provides the best possible treatment strategy for the previously untreated area.opt is required to determine. The corresponding planning and optimization is carried out automatically by the computer system 1. In order to be able to carry out realistic planning, the computer system 1 has data on the geometry of the agricultural machine 10, in particular on the effective working width of the agricultural machine 10, and on the driving properties of the agricultural machine 10, such as the maximum speed of the agricultural machine 10 when driving forward and in reverse, the minimum turning radius of the agricultural machine 10, etc. Furthermore, performance data of the agricultural machine 10 may be known depending on the soil properties, the gradient or other factors, such as the speed which depends thereon, or fuel consumption, etc.

[0053] For the optimization, an optimization criterion is defined. This optimization criterion may be, for example, minimizing the total distance traveled, the total travel time, the fuel consumption, or the untreated area. It is also possible to define several sub-criteria, between which there may be some degree of conflict. On the one hand, it is possible to aim at minimizing a weighted combination of several different optimization values ​​(e.g. distance traveled, travel time, etc.), and on the other hand, it is also possible to perform a Pareto optimization with respect to several different optimization values.

[0054] Within the processing area 20, an inner area 21 and a headland 22 are defined. The latter headland is arranged in a U-shape at the edge of the inner area 21, leaving free an escape edge 28. The decision to subdivide into inner areas 21 and headland 22 may be performed on the part of a user or automatically by the computer system 1.

[0055] Each travel path F1-F3 is defined in such a way that after the headland 22 the inner area 21 is processed. Figure 2 shows a state in which the agricultural machine 10 has completely processed the headland 22, whereas the inner area 21 has not yet been completely processed. The inner area is then harvested in several parallel travel tracks, as shown in Figure 3, which are connected by turns in the headland 22. Processing is here carried out starting at the end of the inner area 21 that is located opposite the escape edge 28.

[0056] The computer system 1 identifies a plurality of possible driving paths F1-F3 and selects the best driving path F1-F3 from the plurality of possible driving paths F1-F3 according to an optimization criterion. opt Then, the optimal driving route F opt The control data D corresponding to the above is generated and transmitted to the agricultural machine 10.

[0057] 4 shows a further embodiment in which the primary region 23 is formed in an L-shape. Moreover, the primary region 23 has not yet been fully harvested. Due to the L-shape, the secondary outer boundary A2 is connected to the connection point P AIf the primary boundary G1 is defined as a straight line between the primary boundary G1 and the secondary boundary A2, an erroneous result would occur. In this case, the computer system 1 can access geographical data, for example satellite photos, and can identify the boundary of the processing area 20 by means of image recognition. Once the primary boundary G1 is known, the secondary outer boundary A2 can be automatically determined therefrom. Again, this automatic determination can be accepted or rejected on the part of the user. Another possibility can consist in a sensing inspection of the periphery of the primary area 23 being carried out. This can be carried out, for example, by a flyable drone (not shown) equipped with a camera. On the basis of the sensing inspection, the boundary of the processing area 20 can be identified, again by means of image recognition, and on this basis the secondary outer boundary A2 can be determined. A further possibility consists in the secondary outer boundary A2 being completely defined by user input. For this purpose, for example a satellite image can be displayed to the user, on which the primary boundary G1 of the primary area 23 is drawn. The user can define the extent of the secondary outer boundary A2 via a suitable interface, for example a touch screen.

[0058] In the example according to Fig. 4, the primary area 23 has not yet been completely processed. A second travel path F2 is therefore identified in order to process the primary area 23 and the secondary area 24.

[0059] FIG. 5 shows a third example in which a rectangular primary area 23 is provided. In this case, the concave area of ​​the primary boundary G1 does not exist. Here too, the secondary outer boundary A2 can be determined and defined by one of the methods described in connection with FIG. 4. Then, here too, the inner area 21 and the headland 22 are determined. In contrast to the first two examples, in this case the primary area 23 is not located completely in the headland 22, but also extends on the inner area 21. The third travel path F3, which is partially shown, is planned in such a way that the remaining part of the headland 22 is processed first and only afterwards the remaining inner area 21 is processed.

[0060] The flow of the method for route planning, more particularly route optimization, according to the invention will now be described on the basis of the flow chart of FIG. 6. In a first step S100, the primary area 23 is traversed, the agricultural machine 10 being manually controlled and position data being determined. In a next step S110, the computer system 1 determines the primary boundary G1 using the position data. In a next step S120, it is checked whether the user wants to define the secondary outer boundary A2 himself. If yes, a definition on the part of the user is carried out in step S130. If no, in step S140, the secondary outer boundary A2 is automatically determined by one of the methods described above. In step S150, the user is asked whether he accepts the secondary outer boundary A2 thus determined. If no, the method proceeds to step S130. If the user accepts, in step S160, the automatically determined secondary outer boundary is defined. In step S170, the inner area 21 and the headland 22 are defined. This is followed by the actual route optimization in step S180. In step S190, the routes F1 to F3 are identified or planned. In step S200, the optimal route F opt has already been found, which is usually negative for the first driving path F1, so the method returns to step S190 where a further driving path is planned.

[0061] In general, different parameters can be changed for the individual travel paths F1-F3, for example the positioning, orientation and sequence of the travel trajectories in the inner area 21. In FIG. 3 the travel trajectories run parallel to the escape edge 28, but other orientations may be chosen. The positioning can also be changed, which can result in, for example, the inner area 21 being less completely covered, but therefore less travel trajectories being required. Finally, the sequence can be changed.

[0062] In step S200, the optimal travel route F optIf it is determined that the optimal travel route F has been found, the optimization ends. opt In the case where the agricultural machine 10 is required to be controlled by a driver, the control data D may correspond to instructions for the driver, based on which the driver drives the agricultural machine 10 along an optimal travel path F opt When the agricultural machine 10 is traveling autonomously, the control data D may include explicit travel and / or steering commands for the systems of the agricultural machine 10. In step S220, the control data D is wirelessly transmitted to the agricultural machine 10 as shown in FIG.

[0063] In order to enable the computer system 1 to carry out the presented method, the required software can be provided as a computer program product having program code means or program code for implementing the method on the hardware of the computer system, for example as a mobile or embedded data medium.

[0064] According to an alternative not presented, the computer system 1 can also be integrated in the agricultural machine 10. In this case, the control data D are directly available in the agricultural machine 10.

Claims

1. A method for planning a travel path for field treatment, the method comprising: an agricultural machine (10) treating a treatment area (20); A primary area (23) of the processing area (20) is traversed by the agricultural machine under manual control, while position data of the agricultural machine (10) is automatically determined (S100); Based on the position data, a primary boundary (G １ ) is at least partially automatically identified (S110); A secondary outer boundary (A) of the secondary region (24) located outside the primary region (23). ２ ) is defined (S130, S160), and the secondary outer boundary (A ２ ) is the two connection points (P Ａ ) at the primary boundary (G １ ) and the primary boundary (G １ ) among the connection points (P Ａ ) forms an inner boundary (I) between the primary region (23) and the secondary region (24), said inner boundary (I) being connected to said secondary outer boundary (A ２ ) together completely surrounding said secondary region (24); A travel path (F) for treating at least the secondary region (24) １ ~F ３ ) is automatically identified (S190), During the field processing, control data (D) for controlling the agricultural machine (10) is automatically generated (S210), and the control data (D) is generated based on the specified travel route (F １ ~F ３ ), method.

2. Multiple possible driving paths (F １ ~F ３ ) according to the corresponding optimization criterion, the optimal driving path (F ｏｐｔ ) is specified (S130), and the travel route (F １ ~F ３ ) is automatically optimized (S180), The optimal travel route (F ｏｐｔ ) is automatically generated (S210), The method of claim 1.

3. An inner region (21) of the processing region (20) and a headland (22) arranged outside the inner region (21) are defined (S170), The travel route (F １ ~F ３ ) is specified so that the inner region (21) is processed after the headland (22), 3. The method according to claim 1 or 2.

4. The headland (22) is defined such that the primary region (23) is located entirely within the headland (22).

4. The method according to any one of claims 1 to 3.

5. If it is determined that the primary area (23) is at most partially treated during the passage, the travel path (F) for treating the primary area (23) and the secondary area (24) is determined. １ ~F ３ ) is identified, 5. The method according to any one of claims 1 to 4.

6. The secondary outer boundary (A ２ ) is the primary boundary (G １ ) is automatically identified as a straight line between the end points of a concave section (K) of (S140), The termination point is the connection point (P Ａ ) to form 6. The method according to any one of claims 1 to 5.

7. The secondary outer boundary (A ２ ) are automatically identified (S140) based, at least in part, on stored geographic data surrounding said primary region (23); 7. The method according to any one of claims 1 to 6.

8. An inspection is performed by sensing the periphery of the primary area (23); The secondary outer boundary (A ２ ) is automatically identified based, at least in part, on the inspection (S140).

8. The method according to any one of claims 1 to 7.

9. The automatically determined secondary outer boundary (A ２ ) is defined after approval by the user (S160); 9. The method according to any one of claims 1 to 8.

10. The secondary outer boundary (A ２ ) is defined at least in part by the user (S130); 10. The method according to any one of claims 1 to 9.

11. Based on the position data, a primary outer boundary (A １ ) is automatically identified, The primary outer boundary (A １ ) is the secondary outer boundary (A ２ ) together completely surrounding the treatment area (20), 11. The method according to any one of claims 1 to 10.

12. The agricultural machine (10) is externally controlled to determine the travel path (F １ ~F ３ ) was identified, The control data (D) is generated externally (S210) and transmitted to the agricultural machine (10) (S220); 12. The method according to any one of claims 1 to 11.

13. Control data (D) for an autonomously traveling agricultural machine (10) is generated; The agricultural machine (10) autonomously performs the field treatment based on the control data (D).

13. The method according to any one of claims 1 to 12.

14. Control data (D) for a driver-steered agricultural machine (10) is generated; This allows generating control instructions for the driver based on the control data (D).

14. The method according to any one of claims 1 to 13.

15. A computer system (1) for planning a travel path for field treatment, the computer system (1) comprising: an agricultural machine (10) that processes a treatment area (20); the agricultural machine (10) manually controls passage through a primary area (23) of the treatment area (20); and, in this case, position data of the agricultural machine (10) is automatically identified (S100); Based on the position data, a primary boundary (G １ ) at least partially automatically (S110), A secondary outer boundary (A) of the secondary region (24) located outside the primary region (23). ２ ) is defined (S130, S160), and the secondary outer boundary (A ２ ) is connected to two connection points (P Ａ ) at the primary boundary (G １ ) and the primary boundary (G １ ) among the connection points (P Ａ ) forms an inner boundary (I) between the primary region (23) and the secondary region (24), said inner boundary (I) being connected to said secondary outer boundary (A ２ ) together completely encircling said secondary region (24), A travel path (F) for treating at least the secondary region (24) １ ~F ３ ) is automatically identified (S190), During the field processing, control data (D) for controlling the agricultural machine (10) is automatically generated (S210), and the control data (D) is generated based on the specified travel path (F １ ~F ３ ), Computer system (1).

16. An agricultural machine (10) having a computer system (1) for planning a travel path for field treatment, the computer system (1) being configured to, when a treatment area (20) is treated by the agricultural machine (10), and a primary area (23) of the treatment area (20) is manually controlled by the agricultural machine (10) to pass through, and at that time, position data of the agricultural machine (10) is automatically identified (S100), Based on the position data, a primary boundary (G １ ) at least partially automatically (S110), A secondary outer boundary (A) of the secondary region (24) located outside the primary region (23). ２ ) is defined (S130, S160), and the secondary outer boundary (A ２ ) is connected to two connection points (P Ａ ) at the primary boundary (G １ ) and the primary boundary (G １ ) among the connection points (P Ａ ) forms an inner boundary (I) between the primary region (23) and the secondary region (24), said inner boundary (I) being connected to said secondary outer boundary (A ２ ) together completely encircling said secondary region (24), A travel path (F) for treating at least the secondary region (24) １ ~F ３ ) is automatically identified (S190), During the field processing, control data (D) for controlling the agricultural machine (10) is automatically generated (S210), and the control data (D) is generated based on the specified travel path (F １ ~F ３ ), Agricultural machinery (10).

17. A computer program product comprising program code means enabling a computer system (1) to carry out the steps according to claim 15.

Images