Animal farm system with data processor programmed for generating a route

Abstract

An animal farm system comprising: an autonomous agricultural vehicle for performing an animal related action in an operating area, comprising at least one drive wheel under the control of a control device, arranged to control the at least one wheel to move the vehicle in accordance with control signals based on a route using a drive sensor system and at least one distance sensor to monitor a pose of the vehicle along the route with respect to a normal driving direction of the vehicle; and a data processing system comprising a processor programmed to carry out a method for generating a route, the method for generating the route comprising: - forming a computational representation of the agricultural operating area, comprising coordinates relating to boundaries of the agricultural operating area; - receiving user-input defining at least one displacement for the autonomous agricultural vehicle in the agricultural operating area, forming a list of route datapoints corresponding to associated vehicle poses in the agricultural operating area along the at least one displacement; - determining for the at least one displacement a risk of colliding with a boundary of the agricultural operating area by simulating the at least one displacement of the vehicle in the agricultural operating space; and - relaying the risk.

Description

[0001] ANIMAL FARM SYSTEM WITH DATA PROCESSOR PROGRAMMED FOR GENERATING A ROUTE

[0002] FIELD

[0003]

[0001] The invention relates to an animal farm system comprising: an autonomous agricultural vehicle adapted to perform an animal related action in an agricultural operating area, comprising at least one drive wheel under the control of a control device, arranged to control the at least one wheel to move the vehicle in accordance with control signals based on a route using a drive sensor system and at least one distance sensor to monitor a pose of the vehicle along the route with respect to a normal driving direction of the vehicle; and a data processing system comprising a processor programmed to carry out a method for generating a route. The invention further relates to a computer program for use with an animal farm system, a data processor and a user interface.

[0004] BACKGROUND

[0005]

[0002] Known autonomous agricultural vehicles, in particular vehicles for performing tasks directly related to the feeding livestock and / or the cleaning of their living space, are programmed in situ, i.e. while the vehicle is at the agricultural location it is going to be used. The location where the vehicle is going to be used depends on the specific vehicle and its tasks and can be one or a combination of a barn, shed, farmyard or farmland. The routes for these known autonomous agricultural vehicles are programmed step by step, wherein each step consists of giving an instruction to the vehicle, waiting for the vehicle to execute the instruction and then saving it. Any potential problems, such as the vehicle colliding with a structural item or exceeding operational limits, are checked visually and / or prevented manually before saving the instruction. This method of installing an autonomous agricultural vehicle allows for tailor-made installation, substantially independent of the particular type of vehicle and lay-out of the location and can therefore be reliably used for all types of autonomous agricultural vehicles.

[0006]

[0003] A downside of these methods of installation, however, is that installers need to be trained and experienced to correctly interpret data from the communication device in order to make the robot function correctly. Furthermore, these methods of installation even requires experienced installers to be on site for a long time, as the step-by-step programming of the route in situ with a vehicle can take more than a day. Moreover, the operation of the farm in the location where the vehicle is to be used is limited during the installation time. To ensure a safe working environment for the installer and accurate route planning, the space where route is planned for is preferably cleared of animals and other work activities.

[0007]

[0004] A goal of the invention is to make the installation of autonomous agricultural vehicles more efficient. SUMMARY OF THE INVENTION

[0008]

[0005] According to a first aspect ofthe invention, a solution is provided through an animal farm system comprising: an autonomous agricultural vehicle adapted to perform an animal related action in an agricultural operating area, comprising at least one drive wheel under the control of a control device, arranged to control the at least one wheel to move the vehicle in accordance with control signals based on a route using a drive sensor system and at least one distance sensor to monitor a pose of the vehicle along the route with respect to a driving direction of the vehicle; and a data processing system comprising a processor programmed to carry out a method for generating a route. The method for generating the route comprises: forming a computational representation of the agricultural operating area, comprising coordinates relating to boundaries of the agricultural operating area; receiving user-input defining at least one displacement for the autonomous agricultural vehicle in the agricultural operating area, forming a list of route datapoints corresponding to associated vehicle poses in the agricultural operating area along the at least one displacement; determining for the at least one displacement a risk of colliding with a boundary of the agricultural operating area by simulating the at least one displacement of the vehicle in the agricultural operating space; and relaying the risk. In this method, the step of determining a risk of colliding with a boundary comprises: calculating a minimum required clearance between the vehicle and the boundary at each route datapoint with respect to a preceding route datapoint, using an approximation of localization uncertainty gain caused by inaccuracy of the drive sensor system to calculate the minimum required clearance. As a result, a reasonably accurate risk of the vehicle colliding with any boundaries along the route can be determined while the route is being planned in the computational space. Having a computational representation of the intended agricultural operating space is sufficient for planning the route and could for example be generated based on available map data or a scan of the space. Through simulation of the displacement in the agricultural operating space, including the localization uncertainty that exists for the vehicle in the actual agricultural space, risks of the route not being reliably viable for the vehicle in real life can be determined during route planning without the vehicle actually having to move or even be present in the intended agricultural operating space. This reduces installation time on site compared to the currently known installation method, as the installer on site is not required to plan each detail ofthe route. Moreover, downtime ofthe intended agricultural operating space is limited during installation, i.e. since the route planning does not require the vehicle to move around in the agricultural operating space to check for viability the agricultural operating space can be mostly used as normal during installation.

[0009]

[0006] In an embodiment, the step of determining a risk of colliding with a boundary further comprises calculating if a boundary of the agricultural operating area is within a predetermined distance from the route datapoint, the predetermined distance being based on a sensing range of the at least one distance sensor, and, if a boundary is found to be within the predetermined distance, reducing the calculated minimum required clearance in dependence of a position and orientation of the boundary with respect to the vehicle pose indicated by the route datapoint.

[0010] This method for generating the routes includes simulating the minimum required clearance along the route, accounting for both increases and decreases therein due to localization uncertainties and certainties as would be experienced by the actual vehicle. As a result, a more accurate risk of the vehicle colliding with any boundaries along the route can be determined while the route is being planned in the computational space.

[0011]

[0007] The processor may be integrated in the vehicle, allowing the vehicle to automatically simulate potential routes generated based on user input while for example remaining at the charger location. Additionally or alternatively, the data processing system may be independent from the vehicle, such that the route can be generated independently from the physical location where the vehicle is (to be) used and may simply be uploaded to the vehicle. Hereto, the autonomous agricultural vehicle may further comprise a vehicle communication device and the data processing system may be configured to communicate with the vehicle through the vehicle communication device for sending the generated route to the vehicle. This allows route planning to be performed away from the intended agricultural operating space and for example to be performed from an office.

[0012]

[0008] According to an embodiment, the calculated minimum required clearance is cumulative along the sequence of route datapoints. Thus, the calculated minimum required clearance grows along the directions of the coordinate system that is used for the computational representation in dependence of a particular distance between two subsequent route datapoints and is not fixedly related to the length and width of the vehicle. As such, the calculated minimum required clearance mimics the actual circumstances for the vehicle, where localization uncertainty along the route increases with each distance traveled within the agricultural operating space until a position therein is identified which cannot be any other position.

[0013]

[0009] In an embodiment, the reducing of the calculated minimum required clearance in dependence of a position and orientation of the boundary is capped at a minimum localization uncertainty value based on a sensor uncertainty of the distance sensor. Thus the route simulation may account for sensor uncertainty of the sensors present in the actual vehicle to further increase the accuracy of the simulation.

[0014]

[0010] In an embodiment, the calculating if a boundary of the agricultural operating area is within a predetermined distance from the route datapoint comprises determining a sensor detection area along the movement between each two subsequent route datapoints based on the sensing range. As a result, the simulation only needs to account for obstacles that are indicated in the computational space that corresponds to the sensor detection area. Preferably this detection area is determined by projecting the sensing range of each vehicle sensor from the end- position of the displacement backwards along the length of the displacement. For example, the detection area may be determined by calculating the convex hull through for a projection of the sensing range at the route datapoint defining an end position of a displacement and the preceding route datapoint, corresponding to a start position of the displacement.

[0015]

[0011] In an embodiment, the boundaries of the agricultural operating area in the computational representation are included as a plurality of boundary-sections, each extending between two endpoints and each having a type indication indicating if the boundary-section is a non-detectable boundary-section or a detectable boundary-section for the distance sensor and wherein the calculated minimum required clearance is only reduced if the found boundary is a detectable boundary-section. In particular when the vehicle is provided with a number of different sensor types, the type indication in the computational space may be made sensor-specific, e.g. some boundary-sections are detectable to none of the sensors, some to a particular type of sensor and some are detectable to all. The type indication in the computational representation ensures that the calculated localization uncertainty is not unjustly reduced, e.g. ensures the simulation accurately mimics the actual localization behavior of the vehicle in-situ in the agricultural operating space.

[0016]

[0012] In an embodiment, the method for generating the route further comprises optimizing the computational representation of the agricultural operating area by merging boundary-sections of the same type indication if endpoints of two respective boundary-sections have the same coordinates and the two respective boundary-sections are parallel to one another. This reduction optimizes the computational space required to perform the simulation.

[0017]

[0013] In an embodiment, the calculating if a boundary is within the predetermined distance from the route datapoint further comprises comparing the sensor detection area with the computational representation of the agricultural operating area, such as by overlaying, and identifying all intersecting detectable boundary-sections and any associated endpoints thereof. An identified detectable boundary-section may be used to reduce the calculated minimum required clearance in a direction perpendicular to the normal driving direction and an identified endpoint may be used to reduce the calculated minimum required clearance in a direction parallel to the normal driving direction.

[0018]

[0014] The vehicle may be programmed to ignore obstacles having a relatively short predetermined length, such as to prevent the vehicle from inadvertently operating the controller based on sensor’s detecting animal-legs. To account for the vehicle being programmed as such in the simulation, in an embodiment, identified intersecting detectable boundary-sections having a length smaller than a predetermined minimum length are disregarded as identified intersecting boundary-sections. In this case, the length of the boundary-sections is preferably calculated in a direction parallel to the normal driving direction.

[0019]

[0015] In an embodiment, the identifying of all intersecting detectable boundary-sections and any associated endpoints thereof comprises identifying which detectable boundary-section and which endpoint is closest to the route datapoint, preferably when considered in a direction parallel to the normal driving direction, and the reduction of the calculated minimum required clearance is determined based on the identified closest detectable boundary-section and closest endpoint. This optimizes the number of calculations performed in the simulation in cases where multiple boundary-section and / or endpoints are found present within the sensor detection area, without significantly impacting the accuracy of the minimum required clearance.

[0020]

[0016] To more accurately calculate the minimum required clearance in situations where an end point is found in a sensor detection area away from the end position of a displacement, the method may further comprise determining a distance between the endpoint closest to the route datapoint and the endpoint, preferably in a direction parallel to the normal driving direction, and calculating a further gain in minimum required clearance using the approximation of localization uncertainty gain caused by inaccuracy of the drive sensor system and the determined distance.

[0021]

[0017] In an embodiment, the method for generating the route further comprises forming a computational representation of the autonomous agricultural vehicle defining a vehicle outline representing at least a maximum vehicle width and a maximum vehicle length with respect to the central point and a normal driving direction corresponding to an orientation of the vehicle indicated by a respective route datapoint. The step of determining a risk of colliding with a boundary then preferably comprises projecting the vehicle outline onto each route datapoint with the normal driving direction oriented in accordance with the associated vehicle pose and determining the presence of a boundary on or inside the vehicle outline. This makes for a relatively simple collision risk calculation, which furthermore is easy to understand when visualized, such as may be done for a user. Preferably, the vehicle outline representing at least a maximum vehicle width and a maximum vehicle length with respect to the central point and a normal driving direction represents at least a maximum vehicle width and a maximum vehicle length is increased by the minimum required clearance before determining the presence of a boundary on or inside the vehicle outline at a respective route datapoint.

[0022]

[0018] In an embodiment, the method for generating the route further comprises: obtaining a vehicle model corresponding to the autonomous agricultural vehicle, the vehicle model comprising a drive sensor model representing the localization uncertainty gain of the vehicle in at least one direction with respect to a driving direction and a distance sensor model representing the sensing range of the at least one distance sensor with respect to the central point and the driving direction, and optionally the minimum localization uncertainty value. The vehicle model may contain all relevant required information of a particular type of agricultural vehicle that may be provided in the animal farm system and uploaded to the processor / selected by the user when the method is being performed. As such, the processor may be provided as a more generic processor for use in animal farm systems with a variety of agricultural vehicles. In particular, a single processor might be used with multiple agricultural vehicles within a single farm system, or to plan routes for a multitude of farm systems.

[0023]

[0019] In an embodiment, the method for generating the route further comprises comparing the determined minimum required clearance to a predetermined maximum clearance value and relaying a localization risk if the minimum required clearance has a value equal to or greater than the predetermined maximum clearance value.

[0024]

[0020] In an embodiment, the autonomous agricultural vehicle is one of a manure removal vehicle comprising a manure collection device arranged to collect manure while moving along the route or a feed handling vehicle comprising a feed pusher and / or a feed providing outlet. These vehicles are typically used in animal farms to relieve the farmer of repetitive work while providing the animals still with a constant level of care and comfort. Manure vehicles are relatively hard to plan routes for, compared to for example feed vehicles, due to the vehicles having to clean the entire space. To clean the entire space, the vehicle’s route has to include sections wherein the vehicle is in contact with walls and other objects to allow these parts of the space to be cleaned. While for other vehicles the problem of potentially colliding with walls or other boundaries is strictly avoided and commonly remedied by simply planning the route with a larger margin around these boundaries, for manure removal vehicles the route needs to include sections wherein the vehicle purposefully interacts with the boundaries, such as moving along the boundary whilst maintaining close contact.

[0025]

[0021] In an embodiment, the step of relaying a risk comprises outputting a risknotification signal and / or message to a user, the signal and / or message optionally including a suggested adjustment of the displacement to remedy the risk. This triggers the user to check if the risk is acceptable and adjust the input defining the displacement if the risk is not acceptable. The risk notification may comprise advice related to the specific risk that is detected, providing the user with a better understanding of the vehicle’s limitations and guiding towards a solution that may be implemented by adjusting the route selection data.

[0026]

[0022] According to a second aspect of the invention, a computer program for use with an animal farm system according to the first aspect is provided. The computer program comprises instructions which, when the program is executed by a computer, cause the computer to: form a computational representation of the agricultural operating area comprising coordinates relating to boundaries of the agricultural operating area; form a list of route datapoints corresponding to associated vehicle poses in the agricultural operating area along the at least one displacement based on received user-input defining at least one displacement for the autonomous agricultural vehicle in the agricultural operating area; determine for the at least one displacement a risk of colliding with a boundary of the agricultural operating area by simulating the at least one displacement of the vehicle in the agricultural operating space; and relaying the risk, wherein the step of determining a risk of colliding with a boundary comprises calculating a minimum required clearance between the vehicle and the boundary at each route datapoint with respect to a preceding route datapoint, using an approximation of localization uncertainty gain caused by inaccuracy of the drive sensor system to calculate the minimum required clearance. The computer program may run independently from the farm system, enabling a user, such as a route planner, to plan routes for the vehicle of the farm system at a remote location from the agricultural operating space, such as for example from a remote office. As a result, the user can plan the route at any given time, independently from presence and / or operations at the farm where the farm system is (to be) operating. The simulation identifies potentially problematic routesections to the user, lowering the requirements of expertise for the user to be able to plan routes that may be reliably performed. A single computer program may be used for different (types of) vehicles and / or agricultural operating areas. It will be obvious that most, if not all, embodiments and advantages as described for the first aspect of the invention are also applicable to the second aspect of the invention.

[0027]

[0023] According to a third aspect of the invention, a data processing device is provided, containing the computer program of the second aspect. The same advantages as identified for the first and second aspect are applicable.

[0028]

[0024] According to a fourth aspect of the invention, a user interface for use with the computer program according to the second aspect or data processing device according to the third aspect is provided, adapted to visualise the computational representation of the agricultural operating area, preferably as a graphical representation, and a graphical representation of an autonomous agricultural vehicle therein, the user interface being adapted to receive user input defining at least one displacement of the autonomous agricultural vehicle within the agricultural operating area with respect to a start pose. The user interface thus visualises all features accounted for in the simulation for route planning, making the route planning for the user easy to comprehend. The user interface may be adapted to visualise a risk of colliding with a boundary at an associated route datapoint in the visualized computational representation, thereby enabling the user to easily identify what parts of the route need adjusting, and optionally how. Additionally or alternatively, the user interface may be adapted to visualize a localization risk if the minimum required clearance associated with a route datapoint has a value equal to or greater than the predetermined maximum clearance value. Such visualization signals to a user that the route as planned at that instance contains insufficient “certain” positions at which the localization uncertainty is reduced, thereby indicating a risk for the route being sufficiently robust to be continuously reliably performed by the actual vehicle in the intended agricultural operating space.

[0029] DRAWINGS

[0025] The invention will be explained below with reference to the drawings, which show non-limiting exemplary embodiments of the invention, in which identical reference numerals indicate identical or similar components, and in which:

[0030]

[0026] Figures 1A and 1 B conceptually illustrate application of embodiments of the invention;

[0031]

[0027] Figure 2 schematically shows a method for providing an autonomous agricultural vehicle with a route;

[0032]

[0028] Figure 3 schematically shows an example of an autonomous agricultural vehicle that may be provided with a route in accordance with the invention, and a schematic representation of vehicle data representing said vehicle for use in the route planning;

[0033]

[0029] Figure 4 shows a flowchart of method steps according to embodiments of the invention;

[0034]

[0030] Figures 5A and 5B show an example of boundaries in the computational representation, respectively as is and with an optimized amount of datapoints;

[0035]

[0031] Figure 6A - 6E each respectively schematically show a displacement for a vehicle with a particular boundary (section) of an agricultural operating area, together with a visual representation of an associated collision risk determination;

[0036]

[0032] Figure 7 schematically shows an exemplary sequence of displacements for a vehicle within a computational representation of an agricultural operating area, with associated localization uncertainties;Figures 8A, 8B and 8C each provide a schematic visualization of localization uncertainty adjusting or resetting under pre-defined conditions related to force sensing by the actual vehicle, wherein in Figs. 8A and 8B a displacement in which the normal direction of travel is substantially perpendicular with respect to the boundary and in Fig. 8C the displacement in the normal direction of travel is substantially parallel to the boundary;

[0037]

[0033] Figures 9A, 9B and 9C provide schematic visualizations of localization uncertainty adjusting using an algorithm representing a distance sensor;

[0038]

[0034] Figure 10 schematically shows another exemplary visual representation of simulating vehicle localization using an algorithm representing a distance sensor; and

[0039]

[0035] Figure 11 schematically shows an embodiment for a module to calculate a minimum required clearance for use in a route planning method.

[0040] DETAILED DESCRIPTION

[0041]

[0036] In the present detailed description, various embodiments of the system and method according to the present invention are described with reference to an autonomous agricultural vehicle in the form of a manure removal vehicle of a particular design and an animal farm system for use in an agricultural operating area designed for dairy animals, such as a barn. However, the present invention may equally be used with differently designed autonomous manure removal vehicles as well as with autonomous vehicles adapted to perform some other animal-related action, such as, but not limited to, obtaining and / or providing feed. Moreover, the present invention may equally be used in animal farm systems designed for other animals, such as meat cattle. In such cases, fewer, more or other animal related structures may be provided in the animal-related space and have to be accounted for during vehicle route planning and performance. Finally, the agricultural operating area may also consist or comprise an outdoor area, such as a farmyard or field in which the autonomous vehicle for example drives to gather feed or to dump manure.

[0042]

[0037] Figures 1A and 1 B conceptually illustrate application of embodiments of the invention. Figure 1A shows the manure removal vehicle 1 1 and Figure 1 B shows an external communication device 20.

[0043]

[0038] The manure removal vehicle 1 1 comprises a first drive wheel 19a, a second drive wheel 19b, a control unit 14 with a memory (not shown), a drive sensor system 15, a first distance sensor 16, a second distance sensor 17 and a vehicle communication device 18. The vehicle 11 can move manure across the floor with a scraping device and / or have a system that moves the manure from the floor into a tank for the vehicle, for example using a vacuum system. Thereto, it moves along at least one pre-planned route through the agricultural operating area under the control of the control device 14, which operates each of the drive wheels 19a, 19b of the vehicle 1 1 individually and in accordance with the pre-planned route. The control device 14 is operably connected to the drive sensor system 15, the distance sensors 16, 17 and the vehicle communication device 18. The vehicle 11 has a normal driving direction F as indicated.

[0044]

[0039] The drive sensor system 15 collects vehicle driving data and has at least a sensor for monitoring motor-speeds of a drive motor actuating a drive wheel and / or a directly proportional value such as motor-current and wheel revolutions. Further, the drive sensor system 15 has at least a sensor adapted to monitor a relative orientation and / or position, absolute or with respect to a previous orientation and / or position. Typical drive sensor system sensors are, but are not limited to, odometers and gyroscopes. The controller is adapted to generate control inputs corresponding to the pre-planned route, and to use the data collected from the drive sensor system 15 in this process.

[0045]

[0040] The vehicle’s distance sensors 16, 17 are distance sensors, which in the depicted manure vehicle design are oriented perpendicular to the vehicle’s normal driving direction F, in opposite directions from one another, for determining distances to objects, such as walls, in the agricultural operating area. The distance sensors may be one-dimensional distance sensors and / or two-dimensional sensors, such as time-of-flight sensors, that are for example based on laser or LED, and ultrasonic sensors. In the depicted vehicle, the distance sensors 16, 17 have a linear or conical sensing range substantially perpendicular to the normal driving direction F. The distance sensor(s) 16,17 are used to monitor the vehicles distance with respect to closest objects within range thereof within the agricultural operating area, which ideally consist of known objects also identified in the map, such as walls, to enable a localization of the vehicle 11 within the agricultural operating area being estimated by a localization module of the vehicle (not shown, may be comprised in the control device 14) with increased accuracy. Hereto, the localization module combines the data from the distance sensor(s) with at least a part of the vehicle driving data obtained by the vehicle drive sensor system 15 at a current pose along the route, as well as a localization of a preceding pose along the route. A pose is to be understood as defining both a position and an orientation of the vehicle with respect to the agricultural operating area.

[0046]

[0041] As indicated, alternative autonomous agricultural vehicle types and designs thereof are foreseen to be used within the scope of the inventive concept. Such alternatives may, for example, have fewer or more drive-wheels and / or have different outer dimensions and / or shapes. In particular, when the alternative autonomous agricultural vehicle is adapted for collecting or providing feed, the autonomous vehicle may have one or more of a feed-pusher, a hopper or tank and a feed-distributor. Moreover, all alternative autonomous agricultural vehicle types may have more and / or differently oriented distance sensors included, such as for example have a sensor with a sensing range that is at least partly oriented substantially parallel to the normal driving direction F and / or with a (semi-)circular sensing range.

[0047]

[0042] The external communication device 20 is shown as consisting of a processor 25 and a user interface system 35 with a display screen 21 for providing a graphical user interface. The user interface system 35 is shown as displaying a map 22 as well as having buttons 24. The map 22 is a visual representation of the computational representation of the agricultural operating area the vehicle is intended to operate in and comprises coordinates relating to boundaries of the agricultural operating area. The user interface system 35 is adapted for receiving user input for generating a pre-planned route for the vehicle. Hereto, a user may for example manually enter coordinates or movement displacements (e.g. a distance to move in a particular direction or a turn at a turning-rate). For example, the route may be manually input by inputting a series of displacements, each second displacement being defined with respect to an end pose of the first displacement, similarly to the route being defined in the known in-situ route programming. Additionally or alternatively, the user interface may allow the user to draw the at least one displacement in the visual representation of the map data, for example via touchscreen. The processor 25 has a memory (not shown) storing instructions, which, when executed, cause the processor to carry out the method of route planning further described with reference to Figures 2 and 4 below. The user interface system 35 may further visualize any information related to the method of route planning, such as route planning input and / or output indicating any risks impacting completion of the route. Preferably, the user-interface provides the user with a visual representation of the map data on which the computational representation is based and shows route datapoints corresponding to the provided input. Preferably, a system used to perform the method is adapted to correlate the input data to the computational representation based on the map data and determine datapoints that correspond to the input.

[0043] The user interface system 35 and processor 25 may be comprised in a single device, such as a computer. Alternatively, the processor 25 may be comprised in a device or system that is operably connectable to the user interface system 35, preferably in a wireless manner. For example, the processor 25 may be comprised in a server, that is either located on the farm itself or remotely and for example set up to provide a shared service such as via cloud. In a further alternative, the processor 25 may be comprised in the vehicle 11 . The operably connectable user interface system 35 may be a computer, tablet or smartphone, comprising the processor 25 or communicating with the processor 25 via an app, which may be web-based. It will be obvious to the skilled person that the display screen, regardless of the external communication device being a computer, a tablet or a smartphone, may be a touchscreen. As such, the displayed buttons 24 may be provided as physical buttons and / or touch-screen buttons that are part of the graphical user interface.

[0048]

[0044] The external communication device 20 can communicate with the vehicle 11 through the vehicle communication device 18, such as to send the pre-planned route that is to be stored in the memory of the vehicle and for the controller to actuate the vehicle accordingly. The pre-planned route depends on the particular vehicle, the vehicle task, the agricultural operating area and any specific user requirements.

[0049]

[0045] Figure 2 schematically shows a method for providing an autonomous agricultural vehicle with a route. The method comprises the steps S1 : Obtain vehicle data, S2: Obtain map data, S3: Plan route for vehicle and S4: Set route in vehicle controller.

[0050]

[0046] The map data obtained under step S2 is map data of the agricultural operating area the agricultural vehicle is intended to operate in, hence map data of an agricultural operating area. The agricultural operating area may consist or comprise of an indoor space such as a barn and / or an outdoor space such as a farm field or a farm’s yard. The map data includes boundaries of the space, such as any walls, fences, doors and optionally ditches, as well as relevant locations of vehicle related features such as charger location(s), fill location(s) and dump pit(s). The manner of obtaining the map data itself is not part of the invention. Various methods of obtaining map data are well known and include taking physical measurements using measuring tape or optical sensors, using verified building maps and using camera images. The map data may be input manually or automatically, such as via software using the building map, sensors or camera images. Moreover, the map data may be two-dimensional or three-dimensional.

[0051]

[0047] The step of planning the route for the vehicle S3 is carried out as a computer- implemented method 100 for making a route of the autonomous agricultural vehicle in the agricultural operating area. This enables the route planning to be carried out without a person having to be present in the agricultural operating area with the vehicle. The vehicle data from S1 and map data from S2 provide necessary input for the simulation of a route. The route planning may be carried out by a processor that is separate from the vehicle, such as in a computer device. A route that leads to a satisfactory simulation result may be exported from the computer- implemented method as output and saved in the vehicle controller in S4, such that the controller of the vehicle controls the vehicle to drive along this route in the agricultural operating area. This uploading to the controller may be performed at any time and is not limited to the vehicle already being installed at work location. Uploading the route to the controller of the vehicle may thus, for example, be performed before the vehicle is delivered to the agricultural operating area where it is intended to perform work. Throughout the remainder of the description, parts of the computer- implemented method enabling the simulation may be described in relation to the vehicle performing the resulting route(sections) in the agricultural operating area. To distinguish between the vehicle in the simulation and the vehicle working in the agricultural operating area, the latter will be referred to as “actual vehicle” and “actual agricultural operating area” from hereon.

[0052]

[0048] The step S1 of obtaining vehicle data comprises obtaining at least a minimum amount of data of the particular actual vehicle for which the method is carried out, necessary to generate a sufficiently representative computer model thereof. An example of an actual autonomous agricultural vehicle 10 and a schematic representation of obtained vehicle data S1 representing said vehicle for use in the simulation is shown in Figure 3. The particular autonomous agricultural vehicle 10 shown is a manure removal vehicle similar to the one described in reference to Figure 1 A. The vehicle may be reduced to a mere datapoint 1 , containing information defining an outer circumference of the vehicle with respect to the datapoint 1 , comprising at least a maximum length L and maximum width w, and a normal driving direction F thereof. Alternatively, a vehicle outline 2 corresponding to the outer circumference of the vehicle may be directly represented. In the remainder of the detailed description, the term vehicle outline 2 will be used, regardless of the vehicle outline 2 being directly represented in the vehicle model or if the information regarding the outer circumference is otherwise modelled. The datapoint 1 is preferably positioned on a longitudinal central axis of the vehicle, such that the vehicle outline 2 is substantially symmetrical with respect to the longitudinal central axis. Although the datapoint 1 may be regarded as the center of the vehicle, such that the vehicle outline 2 is at substantially equal distance therefrom, for ease of calculations it is preferred to have the longitudinal position of the datapoint 1 correspond to a longitudinal position of the drive wheel(s) of the vehicle. The vehicle outline 2 included in the depicted schematic representation is a simplification of the actual outer circumference of the vehicle, and may be included as such, or as a more accurate representation in the vehicle data for more accurate modelling and / or be included more accurately in a visualization for a user in the user interface. It is obvious that when a route is planned for an actual vehicle of a different type and / or design, the computer model will be generated accordingly and thus look different from the depicted schematic representation of Figure 3. Further data may be included in the vehicle data but is not shown in the schematic. This data may comprise one or more of: battery capacity, tank capacity, controller limits and localization sensor characteristics such as accuracy. Obviously, if desired, such information could also be included in a visualization in the user-interface in a number of known ways.

[0049] A flowchart of method steps according to embodiments of the computer implemented method 100 for planning a route for an autonomous agricultural vehicle is shown in more detail in Figure 4. The steps of the method 100, in order of being performed, are shown to be:

[0053] • form a computational representation of the agricultural operating area 101 ,

[0054] • the option to optimize the amount of endpoints per boundary(section) 102,

[0055] • receive at least one displacement input for execution by the vehicle within the computational representation 103,

[0056] • calculate a minimum required clearance for the vehicle along the at least one displacement using at least one boundary(section) 104,

[0057] • determine for at least one displacement a risk impacting completion of said displacement 105, and

[0058] • the option to adjust a displacement 106 and repeat the previous two steps.

[0059]

[0050] In the step of forming a computational representation of the agricultural operating area 101 , map data of the agricultural operating area is converted to a computational data set representing the actual agricultural operating area with its boundaries and any further features relevant to the actual vehicle. This computational representation may take any form that is usable for further computations and any visualizations, such as lists, matrices or graphs. Although the computational representation may make use of any type of coordinate system, use of a cartesian coordinate system is preferred due to being intuitive to understand and will be used from hereon in the description. The coordinates may be defined with respect to a predetermined start pose, e.g. [0;0], within the agricultural operating area, such as for example a lower left-hand corner of the map data or a pose of a charger for the vehicle.

[0060]

[0051] In the optional step of optimizing the amount of endpoints per boundary(section) 102, the amount of datapoints used to indicate the presence of a boundary in the graphical representation is reduced to only indicate endpoints of a particular type of boundary, in a particular direction. Depending on how map data is gathered and transformed into the graphical representation, each boundary(section), such as a wall, may initially be included as a series of interconnected boundary-points. These boundary-points are used in further calculations. By optimizing the number of boundary-points to only include relevant endpoints, the number of calculations may also be optimized, thereby reducing the computational power and / or memory space required. Moreover, a longer distance between endpoints of a boundary-section may better mimic physical measurements by the actual vehicle, which may for example be programmed to identify sensor values indicating a relatively short obstacle as likely being an animal leg and prompt the vehicle’s controller to take a specific animal-related action, such as for example pausing to let the animal pass.

[0061]

[0052] An illustration of the result of the optional step of optimizing the amount of endpoints per boundary(section) in the computational representation of the agricultural operating area is depicted in Figures 5A and 5B. In Figure 5A an exemplary set of map data is visualized, representing an imaginary section of an agricultural operating area. The map data shows boundaries as a series of lines 5a, 5b that each have endpoints 6. The series of lines has two line types, a first line type 5a representing a boundary with a first property and a second line type 5b representing a boundary with a second property. The first and second properties respectively result in the boundary not being detectable and being detectable to the vehicle’s distance sensor. Boundaries with a first property, for example, have a height and / or structure that is outside the range of the distance sensor. These property differences may be manually added to the map data, or automatically, such as for example when the map data is generated from a three- dimensional scan. During the method step of optimizing the amount of endpoints per boundary(section), lines of the same line type that are connected and parallel are reduced to a single line with one set of endpoints 6. The result of this reduction is shown in Figure 5B. It will be appreciated that this optional step is thus usually only carried out once, upon submitting new map data.

[0062]

[0053] In Fig. 4, in the step of receiving at least one displacement input for execution by the vehicle within the computational representation 103, a user provides route planning input resulting in at least one displacement. The at least one displacement forms the basis for the route planning. Based on the at least one displacement, one or more route datapoints are automatically determined within the computational representation, corresponding to the at least one displacement 103. Depending on the particular input, substantially all route datapoints may directly correlate to an input displacement. The input displacement starts at a starting pose defining a start route datapoint, which may be the last route datapoint from a preceding displacement or a (standard) starting location for the vehicle such as a datapoint corresponding to a pose of a charging station, and define a motion to an end point which is the route datapoint associated with the input displacement. Alternatively or additionally, additional route datapoints may be automatically generated in between the start and end route datapoint of the input displacement to more accurately define the route therebetween. The correlation between the amount of displacements received and the amount of route datapoints generated may depend on a distance covered through the at least one displacement and / or the particular type of displacement. For example, for accurate route determination, it may be predefined within the method to have a distance between route datapoints be no larger than a distance corresponding to 0.5 meters in the actual agricultural operating area when traveling in a straight line. For driving through a bend, this distance may be smaller, such as for example no larger than 0.1 meters or even 0.01 meters.

[0063]

[0054] The at least one displacement is provided via a user interface as described with reference to Fig. 1 B. Various methods of providing the input are foreseeable. For example, coordinates or movement displacements (e.g. a distance to move in a particular direction or a turn at a turning-rate) could be manually entered in input boxes. Preferably, the user-interface provides the user with a visual representation of the map data on which the computational representation is based and / or with the computational representation itself and shows the route datapoints corresponding to the provided input. Additionally or alternatively, the user-interface allows the user to directly draw the at least one displacement in the visual representation of the map data, for example via touchscreen. The system is adapted to correlate the input data to the computational representation and determine datapoints that are corresponding to the input. The system may comprise additional instructions based on which the aforementioned additional datapoints are automatically identified as route datapoints, such as for example a maximum distance between route datapoints, which may further depend on a direction or type of displacement (i.e. straight or turn).

[0064]

[0055] The at least one displacement may be an indication of a series of distances and turns for the vehicle to carry out in the order put in, each distance and turn resulting in at least one next route datapoint. The at least one displacement may also be an indication of a series of locations indicated in a coordinate system correlating to the computational representation and / or map data, with the order of the series of locations determining the route and corresponding to at least a part of the route datapoints. Alternatively, the at least one displacement may be a single location or sub-section within the map data (or computational representation based thereon) for the vehicle to drive to and / or in, optionally together with a further predefined requirement such as a time, duration or predefined threshold related to the work being carried out by the vehicle (cleanliness level, amount of feed present, etc.).

[0065]

[0056] The computational representation allows any series of datapoints to be initially assigned as route datapoints, such that further computational steps are to be carried out to verify that these route datapoints do not result in situations that cannot be reliably executed by the actual vehicle in the actual agricultural operating area. Therefore, after the determining of one or more route datapoints, the step of calculating a minimum required clearance along the at least one displacement for the vehicle 104 and the step of determining for the at least one displacement a risk impacting completion of said displacement 105 are performed (automatically, i.e. by a computer program carrying out the method). These steps may be performed after receiving each displacement independently, and / or be performed once a series of displacements is input, such as once a complete route is input.

[0066]

[0057] The step of determining for the at least one displacement a risk impacting completion of said displacement 105 comprises determining a risk that each displacement is physically impossible to execute as intended due to interaction with a fixed boundary of the agricultural operating area, e.g. being steered off course or having a collision that stops the vehicle. A method of assessing such risks is described in more detail in relation to Figs. 6A - 6E below. In the present method, this step of determining a risk impacting completion of displacement uses the minimum required clearance calculated for the vehicle. Said step of calculating a minimum required clearance for the vehicle 104 is explained in further detail with respect to Figs.

[0067] 7 - 10 below.

[0068]

[0058] If a risk is identified for one or more route datapoints of a displacement, the risk is relayed. The manner in which the risk is relayed may depend on the type of risk and the severity of impact of the risk on completing a displacement. At least part of the risks, such as a set of risks comprising displacements for the vehicle that always cause the route to be stopped prematurely, are communicated to the user via the user-interface, through a risk indicating message and / or a risk identifying icon in the visual representation of the map data with the route datapoints.

[0069]

[0059] Based on the type of risk and optionally the number of risks, the displacement may then be adjusted through the step of optionally adjusting the displacement 106. The step of adjusting the displacement 106 may be performed manually. Furthermore, and in particular when the method is performed with a relatively high number of route datapoints being generated automatically compared to the number of displacements received as input, an additional or alternative adjusting step may be included in the method, wherein one or more route datapoints corresponding to a particular risk are automatically adjusted, preferably within predetermined boundaries, or wherein the method returns a suggestion for adjustment to the user (for the user to accept before implementing as new route datapoint). Thus, the computer-implemented method 100 is an iterative process of assessing route viability for the vehicle through simulation, which may be partially automated to lower the risk of route completion failure.

[0070]

[0060] A manner of determining the risk for collision is further explained in reference to Figures 6A - 6E, which each respectively schematically show a displacement for a vehicle within a particularly shaped boundary 5 of an agricultural space. Figures 6A - 6D each show a displacement 3 along a differently shaped wall (section) 5, which respectively are straight, having a slightly sloped section, having a significantly sloped section and having a facing section. Figure 6E shows a turn displacement 3 around a corner of a wall-section. For displacements consisting of a substantially straight path, i.e. a path wherein the normal driving direction F of the vehicle uses the drive sensor system and is pointing in substantially the same direction for both route datapoints defining the start and end of the displacement, such as the displacement 3 shown in Figures 6A - 6D, the area in which the vehicle performs the motion is automatically, i.e. by the computer-implemented method, simplified to a displacement area 3’, the displacement area being a shape enclosing the vehicle outline 2 of the vehicle both at the start pose and the end pose of the displacement. The displacement area 3’ may be obtained by simply interconnecting the two vehicle outlines, such as depicted in Figures 6A and 6D, or a polygonal tightly enclosing the vehicle outlines may be used, such as the rectangular shape depicted in Figures 6B and 6C. A collision risk is then checked for determining if the displacement area 3’ intersects with any boundaries that are indicated in the computational graph. This is a relatively simple way of checking collision risk for each displacement as a whole, requiring relatively few computations.

[0061] The particular example in Figure 6A, where the wall follow displacement is performed along a straight wall section, does not return a collision risks for the displacement area 3’.

[0071]

[0062] In Figure 6B, the wall 5 contains a section that is at an angle with respect to the first part of the wall, causing a substantial intersection 7 between that section of the wall 5 and the displacement area 3’. Thus, a collision is calculated for the depicted displacement. The calculated collision may be further categorized, for example using subcategories such as described in NL2039046, of which the description is incorporated in the present description by reference. In the depicted situation only a boundary of the vehicle is calculated to move through the wall-section, and the normal moving forward direction F of the vehicle is substantially perpendicular to the wall-section, such that the risk of the collision is low to medium. In the actual agricultural space, the actual vehicle will likely be moved off course, following the angle of the section of the wall. Thus, the present displacement will not directly lead to a failure of route completion but might result in the vehicle steering off course if the route is not adjusted appropriately.

[0072]

[0063] In Figure 6C, the wall 5 contains a section that is at a more substantial angle with respect to the first part of the wall, also causing substantial intersection 7 between that angled section of the wall 5 and the displacement area 3’ such that a collision is calculated. The depicted situation in Figure 6C differs from the situation depicted in Figure 6B in that the angle of the wallsection is much steeper, such that that section of the wall cannot be considered substantially perpendicular and poses a more significant riskthatthe collision results in the actual vehicle being stopped by the collision. The computer-implemented method may be set-up to distinguish this more significant risk from the risk calculated for Figure 6C.

[0073]

[0064] In Figure 6D, the displacement area 3’ is intersected by a perpendicular wallsection, i.e. the wall-section is perpendicular to the normal driving direction F of the vehicle and the center of the vehicle 1 moves through the wall-section 5. This collision is categorized as the highest risk for the vehicle not being able to complete the route, i.e. using this route for the actual vehicle in the actual agricultural space will certainly lead to failure. Thus, the collision calculation will return a collision risk that may be classified as the highest risk.

[0074]

[0065] For displacements following a curved path, i.e. where the normal driving direction F of the vehicle is pointing in a different direction for the start and end route datapoints, such as the turn displacement 3 around the corner of the wall 5 depicted in Figure 6E, the vehicle outline 2 is simplified to a simplified shape 3”, such as the depicted rectangle having a length and a width substantially corresponding to a maximum length and maximum width of the vehicle. This simplified shape 3” is used to determine intersection with the wall-section 5 at the start, the end and at least one intermediate pose along the curved path as depicted. The number of intermediate poses along the bend, i.e. the spacing of automatically generated intermediate route datapoints in turns, are predetermined and based on a weighing of accuracy versus available and / or required computational power. In the situation shown in Figure 6E, the simplified shape 3” comparison with the boundary results in no calculated collision risk for the start and end poses of the curved path, but a collision risk 9 is found at least halfway through the displacement. The size of calculated overlap between the corner of the wall 5 and the simplified shape 3” may then be used to determine a severity of the collision, i.e. based on pre-set conditions. Additionally, or alternatively, if further intermediate poses were accounted for through additional sensors such as distance intermediate route datapoints, an orientation of the vehicle at the position of impact could be compared to the orientation of the wall at the position of colliding to determine the severity. This method of checking for collisions is also simple to implement and is more accurate than the method described for the substantially straight path traveling. Although optionally this method could therefore be used to check for collisions throughout all displacements, this requires more computational effort. Therefore, it is preferred to use this method only for the curved paths of displacements, i.e. for determining a collision risk for a turn displacement and / or curved-path sections of other displacements.

[0075]

[0066] Figure 7 schematically shows an exemplary sequence of displacements 3-1 , 3-2, 3-3, 3-4, 3-5, 3-6 for a vehicle within a map-section 200 of an agricultural operating area with boundaries 5, together defining a part of a route for the vehicle. The aforementioned visual representation of the map data and route datapoints as optionally provided in a user-interface could be similar to this schematic representation. The exemplary route section consists of a straight 3-1 , a turn 3-2, a wall follow 3-3, a parallel to the wall 3-4, another turn 3-5 and another wall follow 3-6. Although each displacement 3-1 , 3-2, 3-3, 3-4, 3-5, 3-6 is provided with an independent reference numeral to allow different displacements being defined, displacements may also simply be referred to by the general reference numeral 3. As mentioned previously in relation to Fig. 4, each of the displacements between route datapoints of the route section may correspond to individual displacement inputs or a lower number of displacement inputs may have been provided, with further intermediate route datapoints having been automatically generated. The depicted route section is thus a result of the aforementioned steps of receiving at least one displacement for execution by the vehicle within the computational representation 102 and determining one or more route datapoints within the computational representation corresponding to the at least one displacement 103. The term “displacement” is used during the remainder of this description to refer to a motion between two route datapoints, regardless of the associated route datapoints demarking a start and / or end pose of such a displacement being directly related to user input or are automatically generated.

[0076]

[0067] The actual vehicle uses the drive sensor system and additional sensors such as distance sensors fordetermining its (relative) position within the actual agricultural operating area, enabling the vehicle to follow the planned route. All sensor modalities have limitations that affect the localization accuracy. In particular, a distance and / or moving direction the actual vehicle is displaced in as controlled by the vehicle’s controller often does not fully correspond to the true physical distance and or direction the vehicle has moved. Exemplary causes are that the wheels slip or the gyro drifts. These problems lead to some degree of inaccuracy when following a planned route and have an associated potential risk that impacts completion of displacements. The impact of the cumulative localization inaccuracies on the route following performance of the actual vehicle is commonly reduced by including sufficient route positions in the route where a sensed location of the actual vehicle cannot be anything else than the true physical location seen in at least one direction of motion for the vehicle, such as for example at the charger location, when touching a fixed boundary such as a wall or fence, or when obtaining a measurement from a distance sensor. Thus, the actual vehicle is provided with one or more sensors and adapted to recalibrate its relative (e.g. sensed) position in the agricultural operating area based on detection of a boundary under a predetermined condition.

[0077]

[0068] Sensor certainties having a reducing effect on the calculated localization uncertainty may be:

[0078] • a route datapoint corresponding to a predefined vehicle related location in the agricultural operating area, such as at a charger, a dump pit or a filling station, or

[0079] • a route datapoint being at a predetermined distance and orientation from a predetermined type of boundary indicated in the computational representation, optionally under a predetermined condition.

[0080] The latter is particularly applicable along a route, i.e. route datapoints that are not a start or end point of a route. A route datapoint being at a predetermined distance and / or orientation from a predetermined type of boundary indicated in the computational representation, optionally under a predetermined condition, mimics the sensing of a boundary under a predetermined condition. The particular distance and orientation, as well as the predetermined condition under which a route datapoint falling within a predetermined range thereof, depends on the particular vehicle that is simulated, and its sensing capabilities. In the exemplary vehicle of Fig. 1 A, which is used for the remainder of this detailed description to further explain the computer-implemented method, the sensors used to recalibrate the position include both sensors detecting a physical touch in the normal driving direction F or a sideways direction thereto and sensors measuring a distance in the sideways direction. The sensors for detecting a physical touch may be provided as force sensors for sensing the actual vehicle touching a boundary with a side and / or with a front of the vehicle outline. Additionally or alternatively, the actual vehicle touching a boundary may be identified through the sensing of resistance in a particular direction by the aforementioned drive sensor system, directly or indirectly, i.e., the resistance may be measured as is or derived from another modality such as for example motor currents. It will be obvious that the method may easily be adjusted to alternative vehicle designs as previously indicated. These adjustments may for example see to the method only accounting for one type of sensors being present to recalibrate the position and / or the sensors being alternatively positioned, such as for example having a distance sensor oriented to sense in the normal driving direction F.

[0069] To ensure the computer-implemented method for planning routes 100 identifies risks that impact route completion with appropriate accuracy, i.e., accounting for localization inaccuracies without being overly cautious, the method includes calculating a minimum required clearance for each displacement. The calculation uses mathematical approximations that mimic the cumulative effect of the drive sensor system inaccuracies as well as sensor certainties that have a reducing effect on the inaccuracies. Hereto, the vehicle data for the vehicle further includes localization characteristics of the vehicle, comprising at least one gain factor representing a localization inaccuracy increase in at least one direction, such as the normal driving direction F, and one or more reduce factors and / or conditions corresponding to the sensing modalities present in the actual vehicle, preferably having one or more pre-set ranges for distance and / or orientation.

[0070] Thus for each move to a subsequent route datapoint a localization uncertainty is calculated to determine a required clearance, after which the risk of collision with the boundary (in this case wall) is checked. The minimum required clearance is proportional with the localization uncertainty along the route. Some examples of the calculated resulting localization uncertainty reduction are shown in Fig. 7 and include route datapoints corresponding to a physical position in the agricultural operating area where distance sensor identifies a cornerof a wall and a physical position where the vehicle drives along a wall while “touching” the wall.

[0081]

[0071] As stated previously, the actual localization uncertainty caused by drive sensor limitations is cumulative over the sequence of displacements. This cumulative localization uncertainty is simulated using a drive sensor model having a function that accounts for the whole route up until the present route datapoint, which, when using cartesian coordinates, may be expressed as: wherein the localization uncertainty in the x-direction Cx(for a current displacement is the result of the localization uncertainty of the previous displacement Cxi-1in x-direction plus a forward moving gain Gforwardtimes the distance driven in the x-direction A and a gain perpendicular to the direction of movement Gsidewaystimes the distance driven in y-direction Ay. The distance driven is simply the difference between the start coordinates (xn, yi-i) and the end coordinates (x, yi) of the displacement. Similarly, the localization uncertainty in the y-direction Cy(for a current displacement is the result of the localization uncertainty of the previous displacement Cyi-1in y- direction plus a gain Gforwardtimes the distance driven in the y-direction Ay and a gain perpendicular to the direction of movement Gsidewaystimes the distance driven in x-direction Ax. If a displacement comprises intermittent route datapoints along which a route-section curve for the displacement is generated, the distances in the x- and y-direction can be the sum of the differences between all x-positions and y-positions, respectively. For most vehicle types and models, the sideways gain will be larger than the forward moving gain Gforward< Gsideways. The forward moving gain GfOrwardis preferably a value between 0 and the sideways gain Gsideways. The gain factors Gforward, Gsidewaysmay be fixed values, or a range of values, which are for example dependent on vehicle driving speed. Alternative or additional functions relating localization uncertainty to a gain factor could be used. In particular, the above functions may be adapted to produce a fixed localization uncertainty for each displacement, adapted to include a predetermined minimum localization uncertainty value that is considered applicable for each displacement regardless of the distance moved and / or adapted to have a cap, i.e. a predetermined maximum localization uncertainty value.

[0082]

[0072] The exemplary visual representation of the localization uncertainty 4-1 , 4-2, 4-3, 4- 4, 4-5 is provided in Fig. 7 for displacements 3-0, 3-1 , 3-2, 3-3 and 3-4, together with a correspondingly inflated vehicle outline 2’-1 , 2’-2 for the first two depicted route datapoints 1-1 , 1- 2, indicating a minimum required clearance. The localization uncertainty 4-1 , 4-2, 4-3, 4-4, 4-5 is related to a distance by which the vehicle moves through the gain factor, thereby allowing the localization uncertainty being determined for each subsequent route datapoint. The localization uncertainty, visualized as a rectangle, is shown to increase proportionally with each distance moved in a particular direction Ax, Ay, such as is particularly clear for the first two depicted displacements indicating a straight 3-0 and a turn 3-1 . For the following displacements 3-2, 3-3 and 3-4, the localization uncertainty is also reduced in a predetermined manner in x- and / or y- direction, depending on a respective route datapoint 1-2, 1-3, 1-4, 1-5 being at a predetermined distance and / or orientation from the boundary 5. The inflated vehicle outline 2’-1 , 2’-2 may be obtained by projecting the vehicle outline onto each corner of the localization uncertainty 4-1 , 4- 2 and calculating the complex hull thereof, i.e. the calculated complex hull is the inflated vehicle outline.

[0083]

[0073] The straight displacement 3-0 in the x-direction assumes the uncertainty is minimal at the start of the straight displacement 3-0 and results in the localization uncertainty 4-1 increasing in the x-direction only, proportionally by a distance Ax moved to reach the route datapoint 1-1 marking the final position of the straight displacement 3-0. The localization uncertainty may be translated from the route datapoint to the associated representation of the vehicle outline 2, to result in a correspondingly inflated vehicle outline 2’-1. The correspondingly inflated vehicle outline 2’-1 may be included in the visualization for a user and is used to automatically check for potential collisions with boundaries. The turn displacement 3-1 has both a displacement in the same x-direction as well as a displacement in the perpendicular thereto y- direction, resulting in the associated localization uncertainty 4-2 increasing proportionally to distances Ay and Ax in both directions from the preceding localization uncertainty 4-1 . The route datapoint 1-2 is within a predetermined distance from a corner of the boundary-section of 5, such that the localization uncertainty in y-direction is reduced at this position. During the displacement to the subsequent route datapoint 1-3, the localization uncertainty in y-direction increases again, while the particular displacement is associated with a wall-touching movement, e.g. the route datapoint 1-3 is at another pre-determined distance from the wall, and thereby resulting in a localization uncertainty in the direction perpendicular to the normal driving direction F, which in this case is associated with the x-dimension. Manners for simulating localization uncertainty reductions in dependence of a calculated distance and / or orientation from a boundary under predefined conditions in the computer-implemented method are further described in relation to Figs. 8A, 8B, 9A, 9B and 10.

[0084]

[0074] Figs. 8A and 8B represent a displacement in which the vehicle moves towards a boundary, i.e. the normal driving direction F of the vehicle during the displacement 3 is substantially perpendicular to the boundary-section 5 it moves towards and the final route datapoint of the displacement is at a distance dffrom the boundary-section 5 that equals a distance to the frontal edge of the vehicle outline 2. The vehicle’s normal driving direction F is at an angle (p with respect to the x- or y-direction of the computational representation and the boundary-section 5 is at an angle ip with respect to the y- or x-direction, which is perpendicular to the vehicles normal driving direction F. The depicted situation equals a situation in which the actual vehicle recalibrates its position based on sensing a slight frontal resistance, which recalibration is simulated using a frontal touch model in which the localization uncertainty may be reduced using:

[0085] Thus the localization uncertainty calculated through the drive sensor model of a displacement in the x-direction Cxtis reset to zero when the angle of the boundary-section ip is pointing in the y- direction (Fig. 8A), while the localization uncertainty in the y-direction Cytis maintained as is.

[0086] Similarly, the localization uncertainty of a displacement in the y-direction Cytis reset to zero when the angle of the boundary-section ip is pointing in the x-direction, while the localization uncertainty in the x-direction CxLis maintained as is. In situations where the boundary-section 5 is at an angle displacement has the vehicle abutting a boundary-section 5 at an angle with respect to the coordinates of the graphical representation (as depicted in Fig. 8B for example), the localization uncertainty as calculated through the drive sensor model is reduced rather than reset. In the present example, the localization uncertainty in that case is scaled using the absolute value of the tangent of the angle of the boundary-section ip, offset by 90° to account for the normal driving direction F being perpendicular to the direction of the boundary-section. The scaling accounts for the localization uncertainty perpendicular to the normal driving direction F, which for the actual vehicle in the actual agricultural operating area could cause the vehicle to encounter the angled boundary-section at a different absolute position (i.e. higher or lower than indicated in Fig. 8B).

[0075] Figures 8C, 9A and 9B represent a displacement in which the vehicle has a normal driving direction F in a direction substantially parallel to the boundary-section 5, wherein Figure 8C shows an exemplary visual representation of simulating vehicle localization using an algorithm representing sideways force sensing and Figs. 9A and 9B show an exemplary visual representation of simulating vehicle localization using an algorithm representing a distance sensor. The distance sensors as depicted in these examples correspond to the distance sensors as indicated for the exemplary vehicle schematically shown in Fig. 1 A, i.e. the vehicle has a sensor located on either side and having a sensing range 51 , 52 substantially perpendicular to the normal driving direction F. The sensing range 51 , 52, depends on the particular sensor type used in the actual vehicle, and may be included in the vehicle data. For example, ultrasonic sensors may have a range between a minimum value larger than zero, such as around 10 cm, and a maximum value no larger than 4 meters, such as around 2 meters. Two-dimensional sensors such as lidar may have a sensing range of up to 20 meters. These sensing ranges are preferably defined with respect to the datapoint 1 , more preferably with respect to a maximum sideways distance dsof the vehicle outline 2 with respect to the datapoint 1 .

[0087]

[0076] Figure 8C represents a displacement in which the vehicle is intended to follow a wall, i.e. where the normal driving direction F of the vehicle during the displacement 3 is parallel to the boundary 5 and the final route datapoint of the displacement is at maximum sideways distance dsfrom the boundary-section 5. The depicted displacement equals a situation in which the actual vehicle recalibrates its position based on sensing a slight sideways resistance or touch. This recalibration situation is simulated using a sideways touch model, which is applied when the computer-implemented method has calculated that a final route datapoint of a displacement has a boundary-section 5 oriented parallel to the normal driving direction F at a maximum sideways distance ds. The sideways touch model may include the following equations for reducing the localization uncertainty:

[0088] Thus the localization uncertainty calculated through the drive sensor model of a displacement in the y-direction CyLis reset to zero when the angle of the boundary-section ip is pointing in the x- direction, while the localization uncertainty in the x-direction CxLis maintained as is. Similarly, the calculated localization uncertainty in the x-direction CxLis reset to zero when the angle of the boundary-section ip is pointing in the y-direction, while the localization uncertainty in the y- direction CyLis maintained as is. In situations where the boundary-section 5 is at an angle with respect to the coordinates of the graphical representation, the localization uncertainty calculated through the drive sensor model is reduced rather than reset, in a similar manner (i.e. using the absolute value of the tangent of the angle of the boundary-section ip) and for similar reasons as indicated for Fig. 8B.

[0089]

[0077] Instead of having a boundary-section 5 detected at distance d, a boundary-section 5b that is detectable to the distance sensor may be present within the sensor’s sensing range 51 , 52, 51 ’, 52’, 51 ”. Figs. 9A, 9B and 9C schematically show an exemplary visual representation of simulating vehicle localization using an algorithm representing a distance sensor, with Fig. 9A showing a parallel wall sensing simulation, and Figs. 9B and 9C showing (the same) a corner sensing simulation. Figures 9A and 9B each show a vehicle model having two distance sensors oriented in opposite directions from one another, substantially perpendicular to the driving direction F. Fig. 9A shows a sensor range that is, or is simplified to, a linear sensing range 51 ’, 52’, while Fig. 9B represents a conical sensing range 51 ’, 52’. Figure 9B shows a vehicle model having a two-dimensional sensing range 51 ” that extends around the vehicle. To determine detectable boundary-sections 5b and endpoints 6 thereof in the computational representation, first a sensor detection area 61 , 62, 61 ’, 62’, 62” is determined for each sensor, along the length of the displacement. Preferably, the sensor detection areas 61 , 62, 6T, 62’, 62” are defined by projecting the sensing range 51 , 52, 5T, 52’, 51” from the end-position of the displacement backwards along the length of the displacement, i.e. to the start-position of the displacement. Next, detectable boundary-sections 5b that intersect with the sensor detection area, as well as any endpoints 6, are identified and the localization uncertainty is reduced accordingly. The identification of detectable boundary-sections 5b and endpoints 6 respectively correspond to the distance sensors of the actual vehicle sensing the presence of a detectable boundary-section, such as a wall or fence, and a corner thereof.

[0090]

[0078] The recalibration based on sensing the presence of a detectable boundary-section, such as a wall or fence, is simulated using a wall sensor model. If a detectable boundary-section 5b intersects with the sensor detection area 62 at least until the end-position and has a length component parallel to the normal driving direction F, such as in the exemplary situation depicted in Fig. 9A, a similar logic as set out for the wall-follow displacement depicted in Fig. 8C may be used. In this similar logic equations similar to those presented for the sideways touch model are used to reduce the localization uncertainty perpendicular to the detectable boundary-section 5b, with the limitation that a minimum sideways localization uncertainty value kwcMis maintained to account for sensor uncertainty. Thus if the value of CxLis decreased according to the rules described for Fig. 8C, in case of the boundary-section being in the sensor detection area, the localization uncertainty is reduced in accordance with: Cxt= max{kwaU, Cxt }, such that a localization uncertainty perpendicular to the normal driving direction F is reduced to no less than the minimum sideways localization uncertainty value kwaa. The same applies to the value of Cy(. The minimum sideways localization uncertainty value kwcMpreferably forms a part of the vehicle data used as input to the computer-implemented method and is vehicle type / design specific.

[0091]

[0079] If the detectable boundary-section 5b terminates before the sensor detection area 62, seen in the direction of travel F, the localization uncertainty perpendicular to the normal driving direction F increases again from thereon. The increase in localization uncertainty is then calculated using the previously described drive sensor model. Thus the computer-implemented method verifies an end position of a detectable boundary-section 5b within the sensor detection area 61 ,62, and determines a remaining length of the sensor detection area 61 , 62 from an endpoint 6 of a detectable-boundary-section 5b to the end-position of the displacement to determine a localization uncertainty reduction and calculate any increases in localization uncertainty that occur thereafter during the same displacement.

[0092]

[0080] The recalibration based on the actual vehicle sensing the presence of a corner of a detectable boundary-section is simulated using a corner sensor model. The corner sensor model is applied to reduce the localization uncertainty in the normal driving direction F when a detectable boundary-section 5b intersecting with the sensor detection area 62’, 61 ” has a detectable endpoint 6 within the sensor detection area 62’, 61 ”, such as in the exemplary situation depicted in Figs. 9B and 9C. The corner sensor model may apply the following rules to reduce the localization uncertainty in the normal driving direction F:

[0093] > (min{Cy , Cx / \tan(ip)\}, if ip kn yi~ I Cyi totherwise

[0094] If these rules indicate a decrease in the value of CxL, the localization uncertainty is reduced in accordance with: Cxt= max{kcorner, Cxt }, to ensure the localization uncertainty maintains a minimum parallel value accounting for sensor uncertainty. The same applies to the value of Cy;. The minimum parallel localization uncertainty value kcorneralso preferably forms a part of the vehicle data used as input to the computer-implemented method. Moreover, the minimum parallel localization uncertainty value kcornermay be equal to the minimum sideways localization uncertainty value kwall, such that both values may be included in the vehicle data as a single value. If the endpoint 6 within the sensor detection area 61 , 62 is located at a distance from the end-position when seen in the direction of travel F, the localization uncertainty in the direction of travel F increases again from thereon, in accordance with the previously used rules, such as described in reference to Fig. 7 above.

[0095]

[0081] Figure 10 schematically shows another exemplary visual representation of simulating vehicle localization using an algorithm representing a distance sensor, wherein multiple boundary-sections 5a intersect with the sensor detection area 62 that is determined for the displacement 3. The sensor detection area 62 as depicted is subdivided into four sections 31 , 32, 33, 34 along the direction of travel F, to aid further explanation. The first sensor detection area section 31 is intersected by a single detectable boundary-section 5b, providing a similar situation to the one depicted in Fig. 9A. The second sensor detection area section 32 is intersected by two detectable boundary-sections 5b at two different distances perpendicular to the normal driving direction F, a first one of the two being the remainder of the detectable boundary-section 5b intersecting the first sensor detection area section 31 and having an endpoint at the outer edge of the second sensor detection area section 32. The third sensor detection area section 33 is intersected by a single detectable boundary-section 5b, which is the remainder of the second boundary-section 5b intersecting the second sensor detection area section 32. Thus, the second and third sensor detection area sections 32, 33 are each demarked by endpoints 6 of the first and second detectable boundary-sections 5b. The fourth sensor detection area section 34 is not intersected by any detectable boundary-sections 5b. To account for such situations, i.e. when a single sensor detection area comprises of at least two different sections having no, one or more detectable boundary-sections 5b, the computer-implemented method comprises one or more steps to identify at least one applicable localization uncertainty reduction model to be used and any additional localization uncertainty to be calculated.

[0096]

[0082] The computer-implemented method may, for example, include a step identifying one or more endpoints 6 within the sensor detection area 62 and discretize the sensor detection area 62 in sections 31 , 32, 33, 34 along the direction of travel F at each identified end point position 6. For the displacement, intermitted datapoints may then be generated that also correspond to the endpoint positions 6 seen along the direction of travel, and a localization uncertainty may then be calculated for each datapoint (i.e. intermitted datapoint or end route datapoint of the displacement) along the displacement in accordance with the gain, reduction and resets described above in reference to Figs. 7 - 9C.

[0097]

[0083] In the example depicted in Fig. 10, all endpoints 6 are identified and used to discretize the sensor detection area 61 in sections 31 , 32, 33, 34. An exemplary visual representation of the localization uncertainty calculated at each detection area section edge is included in Fig. 10, having a rectangular shape with a width and length corresponding to an associated localization uncertainty in each direction, respectively in the sideways direction and the direction of travel F. Although resulting in a high accuracy of calculated localization uncertainty along the displacement, analysis at all endpoints 6 may result in unnecessary calculations being performed, when consecutive (intermitted) datapoints along the displacement have the same localization uncertainty result, such as the results indicated for sensor detection area sections 31 , 32 and 33 in the example of Fig. 10.

[0098]

[0084] Additionally or alternatively to discretizing the sensor detection area into sections, the computer-implemented method may contain a step applying rules that identify what detectable boundary-section 5b and / or endpoint 6 within the sensor detection area 61 , 62, 6T, 62’, 61 ” to use forwhich localization uncertainty model and how. Preferably, hereto the detectable boundarysection 5b and / or endpoint 6 that is closest to the route datapoint corresponding to the end of the displacement is used. In particular, the method may apply at least a part of the following rules: Herein, first all detectable boundary-sections 5b that intersect with the sensor detection area are identified as detected lines by comparing the set of detectable boundary-sections (e.g. detectable lines Ldetectabie) with each (e.g. X) of the sensor detection areas 61 , 62, 61 62’, 61 ”. The detected lines may then optionally be filtered on a predetermined minimum length lmin, which is a vehicle data entry chosen to correspond to a boundary length ignored by the actual vehicle when reducing or resetting the position in order to ignore animal legs. For boundary-sections having an angle p different from the direction of travel F of the displacement, a length in the direction of travel F is preferably calculated and considered. The calculated length in the direction of travel F may be referred to as the length component of the boundary-section. Finally, the detectable boundarysection 5b and / or endpoint 6 closest to the route datapoint corresponding to the end route datapoint of the displacement is used to determine the localization uncertainty reduction. For an identified closest detectable boundary-section 5b, the localization uncertainty is reduced according to the same method described in reference to Fig. 9A. For an identified closest endpoint 6, the localization uncertainty is reduced according to the same method described in reference to Fig. 9B. This means that, when using the example of Fig. 10, the second boundary-section 5b intersecting the second sensor detection area sections 32 and 33 is identified as closest detectable boundary-section 5b, which will be used to reduce the localization uncertainty perpendicular to the normal driving direction F to no less than the minimum sideways localization uncertainty value kwcM. Further, the endpoint 6 of the second detectable boundary section 5b, that has a position on the edge of sensor detection area sections 33 and 34, is identified as the closest endpoint 6, which will be used to reduce the localization uncertainty in the normal driving direction F to no less than the minimum parallel localization uncertainty value kcorner. Thus, in the present example of Fig. 10, the reduced localization uncertainty as visualized through the rectangle at the intersection of sensor detection area sections 33 and 34 would be calculated.

[0099]

[0085] For proper accuracy, the gain in localization uncertainty due to the closest endpoint 6 being at a non-zero distance defrom the end-position of the displacement 3 in the sensor detection area 62 should also be accounted for. This may be achieved by including a further rule that calculates the distance debetween the closest endpoint 6 and the end-position, followed by a calculation of the associated localization uncertainty gain. Alternatively, the sensor detection area 62 may first be discretized into two sections based on the closes endpoint 6 being at the non-zero distance defrom the end-position of the displacement, using the position of this closest endpoint6 to mark the intersection and include an intermitted datapoint for localization uncertainty calculations. In the exemplary case depicted in Fig. 10, the intermitted datapoint would then be positioned at the intersection between sensor detection area sections 33 and 34. After the step of discretizing the sensor detection area into sections, the aforementioned calculations and rules are applied to each detection area section independently. Thus, in the case of Fig. 10, a first set of calculations is performed for a section comprising sensor detection area sections 31 , 32 and 33, and a second set of calculations is performed for sensor detection area section 34.

[0086] To capture any localization uncertainty reduction for displacements comprising a bend, such as a turn displacement, additional route datapoints may be included at either end of the bend, and sensor detection areas 61 , 62 may be determined for each straight section after and before the bend independently, while a section containing the bend may be disregarded for any localization uncertainty reduction calculations.

[0100]

[0087] Fig. 11 schematically shows an embodiment for a module 40 to calculate a minimum required clearance for use in a route planning method such as the computer implemented method 100 for planning a route for an autonomous agricultural vehicle. The module 40 uses a sequence of at least two route datapoints 41 , a computational representation of the agricultural area 42 and vehicle data relating to localization characteristics 43 as input and provides a minimum vehicle clearance for each of the at least two route datapoints 47 as output. The module 40 comprises the previously described drive sensor model 44, as well as a distance sensor model 45 and an optional touch sensor model 46. The distance sensor model 45 may include both the previously described wall sensor model and corner sensor model, for using the module 40 to simulate routes for an actual vehicle having one or more sideways oriented distance sensors, such as the vehicle depicted in Fig. 1A, or one or more alternative or additional appropriately adjusted wall and / or corner sensor models to simulate the actual vehicle having alternatively oriented, number and / or type(s) of distance sensors. The optional touch sensor model 46 includes the frontal touch model and sideways touch model as described above, for using the module 40 to simulate routes for an actual vehicle adapted to purposefully interact with boundaries of the agricultural operating area, such as the vehicle depicted in Fig. 1 A. The module 40 is adapted to first calculate the localization uncertainty through the drive sensor module 44, i.e. using a gain factor and a distance between a current and a previous route datapoint. Then the module 40 verifies if either the distance sensor module 45 or the optional touch sensor module 46 is applicable to calculate any reductions for the localization uncertainty that is originally calculated with the drive sensor module 44. If the distance sensor module 45 is applied, a remaining distance within the sensor detection area may be calculated as described above, which may be returned to the drive sensor module 44 for calculating a further localization uncertainty gain prior to outputting the resulting minimum required clearance.

[0101]

[0088] Additional or alternative modules may be introduced for autonomous agricultural vehicles of a different design, operating in adjusted agricultural operating spaces. For example, the autonomous agricultural vehicle may be configured to sense RFID tags which are positioned at fixed locations in the agricultural operating space. The RFID tags may then be included in the computational representation of the agricultural operating space and accounted for in the simulations performed by the module similarly to end points, e.g. reducing a localization uncertainty in the normal driving direction F upon determining an RFID tag being within a sensing area from a particular route datapoint. In a further example, the autonomous agricultural vehicle may be configured to sense magnetic strips that are provided in the floor of the agricultural operating space and navigate based on the sensing of these strips. The strips may then be included in the computational representation of the agricultural operating space and accounted for in the simulations performed by the module similarly to visible boundary sections 5b, e.g. have the sensing range adjusted to the range with respect to the vehicle outline 2 within which the vehicle senses the strips and apply a wall sensor model for reducing the localization uncertainty.

[0102] The configuration enabling the vehicle to sense the RFID tags and / or strips may also be considered distance sensors.

[0103]

[0089] The present invention has been described above with reference to a number of exemplary embodiments as shown in the drawings. It will be clear to a person skilled in the art that the scope of the invention is not limited to these examples, nor to the explicitly mentioned alternatives, but that a number of further variations and modifications thereof are possible without departing from the scope of the invention as defined in the attached claims.

Claims

CLAIMS1 . An animal farm system comprising: an autonomous agricultural vehicle (11) adapted to perform an animal related action in an agricultural operating area, comprising at least one drive wheel (19a, 19b) under the control of a control device (14), arranged to control the at least one wheel to move the vehicle in accordance with control signals based on a route using a drive sensor system (15) and at least one distance sensor (16,17) to monitor a pose of the vehicle along the route with respect to a normal driving direction (F) of the vehicle (11); and a data processing system (20) comprising a processor (25) programmed to carry out a method for generating a route, the method for generating the route comprising: forming a computational representation of the agricultural operating area, comprising coordinates relating to boundaries (5; 5a, 5b) of the agricultural operating area;- receiving userinput defining at least one displacement for the autonomous agricultural vehicle in the agricultural operating area, forming a list of route datapoints (%t,yt, <Pt) corresponding to associated vehicle poses in the agricultural operating area along the at least one displacement; determining for the at least one displacement a risk of colliding with a boundary (5; 5a, 5b) of the agricultural operating area by simulating the at least one displacement of the vehicle in the agricultural operating space; and relaying the risk, wherein the step of determining a risk of colliding with a boundary (5; 5a, 5b) comprises: calculating a minimum required clearance (C(, Cy() between the vehicle and the boundary (5; 5b) at each route datapoint (x^yt.cpi) with respect to a preceding route datapoint (xi-^yi^. pi- , using an approximation of localization uncertainty gain caused by inaccuracy of the drive sensor system (15) to calculate the minimum required clearance.

2. Animal farm system according to claim 1 , wherein the autonomous agricultural vehicle (1 1) further comprises a vehicle communication device (18) and wherein the data processing system (20) is configured to communicate with the vehicle (11) through the vehicle communication device (18) for sending the generated route to the vehicle.

3. Animal farm system according to claim 1 or 2, wherein the calculated minimum required clearance is cumulative along the sequence of route datapoints (x^yi.cpt).

4. Animal farm system according to any one of the preceding claims, wherein the step of determining a risk of colliding with a boundary (5; 5a, 5b) further comprises calculating if a boundary (5; 5b) of the agricultural operating area is within a predetermined distance from the route datapoint (%t,yt, <Pt), the predetermined distance being based on a sensing range (51 , 52; 51 ’, 52’; 51 ”) of the at least one distance sensor (16, 17), and, if a boundary (5; 5b) is found to be within the predetermined distance, reducing the calculated minimum required clearance independence of a position and orientation of the boundary (i ) with respect to the vehicle pose indicated by the route datapoint x^y^cpi).

5. Animal farm system according to claim 4, wherein the reducing of the calculated minimum required clearance in dependence of a position and orientation of the boundary is capped at a minimum localization uncertainty value (fcwa;;; Corner) based on a sensor uncertainty of the distance sensor (16, 17).

6. Animal farm system according to claim 4 or 5, wherein the calculating if a boundary (5) of the agricultural operating area is within a predetermined distance from the route datapoint (x^yi. pi) comprises determining a sensor detection area (61 , 62, 61 ’, 62’, 61 ”) along the movement between each two subsequent route datapoints based on the sensing range (51 , 52; 51 ’, 52’; 51 ”), preferably by projecting the sensing range (51 , 52; 51 ’, 52’; 51 ”) from the endposition of the displacement backwards along the length of the displacement, most preferably by calculating the convex hull through for a projection of the sensing range (51 , 52; 51 ’, 52’; 51 ”) at both the route datapoint x^y^cpi) and the preceding route datapoint (xi-1,yi-1, (pi-1').

7. Animal farm system according to any one of claims 4 - 6, wherein the boundaries (5) of the agricultural operating area in the computational representation are included as a plurality of boundary-sections (5a, 5b), each extending between two endpoints (6) and comprising a type indication indicating if the boundary-section is a non-detectable boundary-section (5a) or a detectable boundary-section (5b) for the distance sensor (16, 17) and wherein the calculated minimum required clearance is only reduced if the found boundary is a detectable boundarysection (5b).

8. Animal farm system according to claim 7, wherein the method for generating the route further comprises optimizing the computational representation of the agricultural operating area by merging boundary-sections (5a, 5b) of the same type indication if endpoints (6) of two respective boundary-sections have the same coordinates and the two respective boundarysections are parallel to one another.

9. Animal farm system according to any one of claims 7 or 8, in combination with claim 6, wherein the calculating if a boundary is within the predetermined distance from the route datapoint (x^yi. pi) further comprises comparing the sensor detection area (61 , 62; 61 ’, 62’; 61 ”) with the computational representation of the agricultural operating area and identifying all intersecting detectable boundary-sections (5b) and any associated endpoints (6) thereof.

10. Animal farm system according to claim 9, wherein identified intersecting detectable boundary-sections (5b) having a length smaller than a predetermined minimum length are disregarded as identified intersecting boundary-sections (5b), the length preferably being calculated in a direction parallel to the normal driving direction (F).1 1. Animal farm system according to claim 9 or 10, wherein the identifying all intersecting detectable boundary-sections (5b) and any associated endpoints (6) thereof comprises identifying which detectable boundary-section (5b) and which endpoint (6) is closest to the route datapoint(%j,yj,<Pj), preferably when considered in a direction parallel to the normal driving direction (F), and wherein the reduction of the calculated minimum required clearance is determined based on the identified closest detectable boundary-section (5b) and closest endpoint (6).

12. Animal farm system according to any one of claims 9 - 11 , further comprising determining a distance (de) between the endpoint (6) closest to the route datapoint x^y^cpi) and the endpoint (6), preferably in a direction parallel to the normal driving direction (F), and calculating a further gain in minimum required clearance (C(, Cy() using the approximation of localization uncertainty gain caused by inaccuracy of the drive sensor system (15) and the determined distance.

13. Animal farm system according to any one of claims 9 - 12, wherein an identified detectable boundary-section (5b) results in the calculated minimum required clearance being reduced in a direction perpendicular to the normal driving direction (F) and wherein an identified endpoint (6) results in the calculated minimum required clearance being reduced in a direction parallel to the normal driving direction (F).

14. Animal farm system according to any one of the preceding claims, wherein the method for generating the route further comprises: forming a computational representation (1 , 2) of the autonomous agricultural vehicle (11) defining a vehicle outline (2) representing at least a maximum vehicle width (w) and a maximum vehicle length (I) with respect to the central point (1) and a normal driving direction (F) corresponding to an orientation of the vehicle (<p;) indicated by a respective route datapoint (w^i); and preferably wherein the step of determining a risk of colliding with a boundary (5; 5a, 5b) comprises projecting the central point (1) and / or the vehicle outline (2) onto each route datapoint (%j,yj,<Pj) with the normal driving direction (F) oriented in accordance with the associated vehicle pose and determining the presence of a boundary (5; 5a, 5b) on or inside the vehicle outline (2).

15. Animal farm system according to claim 14, wherein the vehicle outline (2) representing at least a maximum vehicle width (w) and a maximum vehicle length (I) with respect to the central point (1) and a normal driving direction (F) represents at least a maximum vehicle width (w) and a maximum vehicle length (I) increased by the minimum required clearance.

16. Animal farm system according to claim 14 or 15, wherein the vehicle outline is increased with the determined minimum required clearance before determining the presence of a boundary (5; 5a, 5b) on or inside the vehicle outline (2).

17. Animal farm system according to any one of the preceding claims, wherein the method for generating the route further comprises: obtaining a vehicle model corresponding to the autonomous agricultural vehicle (11), the vehicle model comprising a drive sensor model representing the localization uncertainty gain of the vehicle in at least one direction with respect to a driving direction (F) and a distance sensor model representing the sensing range (51 , 52; 51 ’, 52’, 51 ”) of the at least one distance sensor(16, 17) with respect to the central point (1) and the driving direction (F), and optionally the minimum localization uncertainty value (kwau', kcorner)-18. Animal farm system according to any one of the preceding claims, wherein the method for generating the route further comprises comparing the determined minimum required clearance to a predetermined maximum clearance value and relaying a localization risk if the minimum required clearance has a value equal to or greater than the predetermined maximum clearance value.

19. Animal farm system according to any one of the preceding claims, wherein the autonomous agricultural vehicle is one of a manure removal vehicle comprising a manure collection device arranged to collect manure while moving along the route or a feed handling vehicle comprising a feed pusher and / or a feed providing outlet.

20. Animal farm system according to any one of the preceding claims, wherein the step of relaying a risk comprises outputting a risk-notification signal and / or message to a user, the signal and / or message optionally including a suggested adjustment of the displacement to remedy the risk.21 . A computer program for use with an animal farm system according to any one of claims 1 - 20, comprising instructions which, when the program is executed by a computer, cause the computer to: form a computational representation of the agricultural operating area comprising coordinates relating to boundaries (5; 5a, 5b) of the agricultural operating area; form a list of route datapoints .xi,yi,<pi') corresponding to associated vehicle poses in the agricultural operating area along the at least one displacement based on received user-input defining at least one displacement for the autonomous agricultural vehicle in the agricultural operating area; determine for the at least one displacement a risk of colliding with a boundary (5) of the agricultural operating area by simulating the at least one displacement of the vehicle in the agricultural operating space; and relaying the risk, wherein the step of determining a risk of colliding with a boundary (5; 5a, 5b) comprises calculating a minimum required clearance (C(, Cy() between the vehicle and the boundary (5) at each route datapoint x^y^ pi) with respect to a preceding route datapoint (x^, yt_ , cpi-') , using an approximation of localization uncertainty gain caused by inaccuracy of the drive sensor system (15) to calculate the minimum required clearance.

22. Computer program according to claim 21 , wherein the step of determining a risk of colliding with a boundary (5; 5a, 5b) further comprises calculating if a boundary (5) of the agricultural operating area is within a predetermined distance from the route datapoint (x / .y / .ipi), the predetermined distance being based on a sensing range (51 , 52; 51 ’, 52’, 51 ”) of the at leastone distance sensor (16, 17), and reducing the calculated minimum required clearance in accordance with a set of predefined rules if a boundary (5; 5a, 5b) is found to be within the predetermined distance, the set of predefined rules accounting for a position and orientation of the boundary (p) with respect to the vehicle pose indicated by the route datapoint (x^yi.cpi).

23. A data processing device containing the computer program of claim 21 or 22.

23. A user interface for use with the computer program according to claim 21 or 22 or data processing device according to claim 23, adapted to visualise the computational representation of the agricultural operating area and a graphical representation of an autonomous agricultural vehicle therein, the user interface being adapted to receive user input defining at least one displacement of the autonomous agricultural vehicle within the agricultural operating area with respect to a start pose.

24. User interface according to claim 23, further adapted to visualise a risk of colliding with a boundary (5; 5a, 5b) at an associated route datapoint (%t,yt,<Pt) in the visualized computational representation.

25. User interface according to claim 23 or 24, further adapted to visualize a localization risk if the minimum required clearance associated with a route datapoint x^y^cpi) has a value equal to or greater than the predetermined maximum clearance value.

Images