Setting the autonomous operating area of a work vehicle or other work machine

Abstract

Some embodiments may include a touchscreen or other user input interface for selecting a portion of the displayed map, and one or more processors coupled to the touchscreen or other user input interface. The one or more processors may be configured to set an autonomous operating area for the work machine based on a user selection from the displayed map. The work machine may monitor its current position in the autonomous operating area and stop operation of at least one actuator or send a new operating signal to the actuator to modify operation of at least one power drive device of the work machine. Other embodiments may be disclosed and / or claimed.

Description

[Technical Field]

[0001] The present disclosure relates to off-highway work vehicles and other work machines, and some embodiments relate to establishing an autonomous operating domain for an off-highway work vehicle or other work machine. [Background technology]

[0002] Off-highway work vehicles, or other implements capable of traversing steep or uneven terrain, include special-purpose vehicles such as tractors, lawn mowers, construction vehicles, agricultural vehicles, etc. These implements may have a transportation system, such as wheels, treads, walking devices, crawlers, etc., for transporting the implement from one location to another. Power-driven transportation systems may be driven by any power source, such as a combustion engine, an electric motor, etc., or a combination thereof.

[0003] In addition to transportation systems, these implements may include tools for performing work tasks such as residential, commercial, and industrial work, such as mowing, spraying, harvesting, planting, digging, mining, grading, etc. These tools, also known as implements, include the following: Passive implements such as tractor-pulled plows and trailers with non-motorized transport systems. Power-driven implements such as power-driven hitches for positioning plows, mowers, excavators, lawn edgers, etc.

[0004] Various components of these work machines (e.g., powered devices of a transport system and / or powered implements) may be configured to operate autonomously (e.g., fully autonomous or semi-autonomous). A robotic lawnmower is an example of a work machine that operates fully autonomously. A tractor with an automatic steering system coupled with a steering wheel (or steering column) is an example of a semi-autonomous work vehicle (because an operator can manually steer the vehicle using the steering wheel). [Brief explanation of the drawings]

[0005] [Figure 1] FIG. 1 is a schematic diagram of a system according to various embodiments including a work machine and one or more computing devices for establishing an autonomous operating area for the work machine. [Figure 2] 2 is a flowchart illustrating operations that may be performed by the system of FIG. 1 to generate one or more virtual boundaries of an autonomous operating area. [Figure 3] 2 is a flowchart illustrating operations that may be performed by the implement of FIG. 1 using one or more generated virtual boundaries. [Figure 4] 1 is a flowchart illustrating a method according to various embodiments that may be performed using any of the systems described herein to generate boundary proposals. [Figure 5] 1 is a flowchart illustrating operations of a computing device that may be performed to visualize a boundary to an operator, according to various embodiments. [Figure 6A] 6 illustrates an exemplary user interface state that may be displayed by one or more computing device processors for tapping (selecting) the boundary detection button in the operational method of FIG. 5. [Figure 6B] 6B illustrates a diagram that may be displayed by one or more computing device processors to visualize boundaries to a user, the diagram being a next state of the exemplary user interface of FIG. 6A following activation of the detect boundaries button. [Figure 7] 6 is a schematic diagram of a boundary calculation module that performs boundary calculations related to the operation of the calculation device of FIG. 5. [Figure 8] 8 illustrates a process that may be performed by the boundary operation module of FIG. 7. [Figure 9] 1 illustrates a process according to various embodiments that may be performed by any boundary calculation module described herein in an embodiment that uses a generic segmentation model to generate data that can be enriched by a user. [Figure 10A]10A-10C schematically illustrate one state showing different exemplary visualizations according to various embodiments that may be displayed by any computing device processor described herein to execute the process described with reference to FIG. [Figure 10B] 10 is a schematic diagram illustrating another state showing different exemplary visualizations according to various embodiments that may be displayed by any computing device processor described herein to perform the process described with reference to FIG. [Figure 10C] 10 is a schematic diagram illustrating another state showing different exemplary visualizations according to various embodiments that may be displayed by any computing device processor described herein to perform the process described with reference to FIG. DETAILED DESCRIPTION OF THE INVENTION

[0006] As used in this specification and claims, the singular forms "a," "an," and "the" include the plural unless the context clearly contradicts otherwise. Furthermore, the term "having," "including," or "comprises" means "comprises." Furthermore, the term "coupled" does not exclude the presence of intermediate components between the connected components. The systems, apparatus, and methods described herein are not to be construed as limiting in any sense. Instead, the present disclosure contemplates all novel and non-obvious features and aspects of the various disclosed embodiments, alone and in various combinations and subcombinations with each other. The term "or" means "and / or," not "exclusive or" (unless expressly indicated).

[0007] The disclosed systems, methods, and apparatus are not limited to any particular aspect or feature, or combination thereof, nor do they require that one or more particular advantages exist or problems be solved. While any theory of operation facilitates description, the disclosed systems, methods, and apparatus are not limited to such theory of operation. While some operations of the disclosed methods are described in a particular sequential order for convenience of description, it should be understood that aspects of this description encompass rearrangements of the order of operations unless a particular order is required by specific text described below. For example, operations described in a sequential order may in some cases be rearranged or performed simultaneously. Furthermore, for simplicity, the accompanying drawings may not show the various ways in which the disclosed systems, methods, and apparatuses can be used in combination with other systems, methods, and apparatuses.

[0008] Additionally, terms such as "produce" and "provide" may be used herein to describe the disclosed methods. These terms are high-level abstractions of the actual operations that are performed. The actual operations that correspond to these terms will vary depending on the particular implementation and will be readily apparent to those skilled in the art. In some instances, values, procedures, or devices are described using terms such as "lowest," "best," and "minimum." Such descriptions are intended to indicate that a selection may be made from among numerous functional options available, with the understanding that such a selection need not be better, less, or more preferable than other options.

[0009] Specific examples are described with reference to directions such as "above," "below," "upper," "lower," etc. These terms are used for convenience and do not imply any particular spatial orientation.

[0010] Some autonomous systems use physical boundaries to regulate the operation of implements. In such systems, operators may establish physical boundaries by placing markers or other physical devices around a field, lot, or other area. However, this requires the operator's effort to place and maintain the markers and other physical devices, not to mention the cost of the markers and other physical devices themselves.

[0011] Various embodiments described herein may logically define an autonomous operating area using one or more independent computing devices or a user interface integrated into the work machine. The logical definition may generate boundary information and transmit it to the work machine. The transmitted boundary information may include one or more virtual boundaries recognizable by the work machine, or the work machine may derive one or more virtual boundaries from the transmitted boundary information. In this manner, a logically defined autonomous operating area for the work machine may be quickly defined.

[0012] This logically defined autonomous operating space may be utilized to control one or more power-driven devices of a transport system or implement of a work machine. For example, a robotic lawnmower may be configured to disconnect power to the robotic lawnmower's transport system and / or its cutting implement when its current position does not match the logically defined autonomous operating space, e.g., when the robotic lawnmower enters within a threshold distance or, in some embodiments, crosses (exceeds) the boundary of the logically defined autonomous operating space. In another example, a tractor or other work vehicle may use a logically defined autonomous operating space to determine a path its automated steering system should follow for a work task (e.g., to plan a path through a field that completely covers the field and ensures that the entire field is sprayed, planted, and harvested). These are just a few examples of how work vehicles and other work machines may utilize boundaries to control the autonomous operation of their components.

[0013] Additionally, autonomous passenger vehicles (e.g., on-highway vehicles) may have systems optimized to allow users to easily input a destination before beginning a journey. For example, many systems simplify the process by allowing operators to select from previous destinations or confirm a destination from a connected device. It would be desirable for work vehicles to have a similarly simplified user interface for operators to configure the autonomous operating area before beginning work.

[0014] In some embodiments, the implement or one or more independent computing devices connected thereto may suggest boundaries in selected portions of the map. The suggested boundaries and / or selected portions of the map may be determined based on user selections or any other inputs that indicate an operating area for the implement to operate in.

[0015] It may be possible and practical to utilize physically and logically defined autonomous operating regions. For example, some boundary segments may be defined by placing markers or other physical devices, while other boundary segments may be defined by one or more virtual boundaries. In various embodiments, outer and inner boundaries (e.g., in a robotic grass-cutting embodiment, areas not covered by grass, such as bark chips, trees, patios, gardens, etc.) may be defined by placing markers or other physical devices.

[0016] FIG. 1 illustrates a system 100 according to various embodiments that includes a work machine 120 and one or more computing devices 105 for establishing an autonomous operating domain for the work machine 120. The area may be defined by one or more virtual boundaries of a field, lot, or other physical area. The one or more computing devices 105 may transmit boundary information 19 that includes the one or more virtual boundaries or may include untranslated data from which the implement can derive the one or more virtual boundaries.

[0017] The work machine 120 may include one or more power drives 40, which may be components of a power-driven transport system or work machine and / or power-driven implements of the work machine. The actuators 30 may drive the operation of the driven devices 40 under the control of the controller 21. These power drives 40 and actuators 30 may include any power drives of a power-driven transport system or power-driven implements now known or developed in the future.

[0018] The controller 21 may include one or more processor sets, which may be implemented using any now known or later developed circuitry, such as logic circuits, application specific processors, general purpose processors that execute instructions stored in memory, or the like, or any combination thereof.

[0019] In some embodiments, the actuator 30 may be part of an autopilot system similar to the autopilot system described in U.S. Patent No. 10,822,017 (which is incorporated herein by reference), or other now known or later developed autopilot systems. In some embodiments, one or more processor sets of the controller 21 may perform any of the functions of the precision guidance system (PGS) described in U.S. Patent No. 10,986,767 (which is incorporated herein by reference), such as the steering controller functions and / or processor functions described in that U.S. patent.

[0020] The work implement 120 may include a positioning system for determining the current position of the work implement 120. In the illustrated embodiment, the positioning system may include sensors such as an inertial measurement unit 26 and a Global Navigation Satellite System (GNSS) receiver 27, although in other embodiments, the positioning system may include other local or remote sensors or other devices, now known or later developed, for collecting raw information from which the current position of the work implement 120 can be derived. The controller 21 may utilize any now known or later developed positioning algorithm to determine the current position of the work implement 120.

[0021] The one or more computing devices 105 may include a smartphone carried by an operator or other now known or later developed portable or stationary computing device. In some embodiments, the one or more computing devices 105 may also include a cloud device, server, or other remote resource that may assist the mobile or stationary computing device in generating the transmitted information 19. The portable or stationary computing device may upload information to a server / cloud (e.g., via a long-range wireless connection in the case of a portable device) for remote computation, and the computed information (for further processing and / or download to the implement 120) may be transmitted back to the portable or stationary computing device or downloaded directly from the server / cloud to the implement 120 (e.g., the cloud / server may download the transmitted information 19 directly to the implement 120).

[0022] In the illustrated embodiment, one or more independent computing devices 105 are provided that store boundary information 19, including one or more virtual boundaries, or raw data from which one or more virtual boundaries can be derived. The one or more independent computing devices 105 may be connected to the implement 40 by a wired or wireless connection, or alternatively, may download information to a memory card or other memory readable by the implement 40. The one or more computing devices 105 may include one or more processors 11 and a touch screen 15 or other user input interface that allows an operator to make selections from content displayed by the one or more computing devices 105.

[0023] Although the illustrated embodiments describe and show one or more separate computing devices, some embodiments may not require a separate computing device. In these embodiments, a touchscreen or other input device 15 may be incorporated into the implement. In such examples, the computing device operations described herein may be performed by one or more processors in the implement 120. In these embodiments, the implement may include multiple processors, one of which may control the actuators 30 and another of which may perform any computing device operations described herein.

[0024] Referring again to the illustrated embodiment, the one or more computing device processors 11 can obtain at least one location information value. In some implementations, the at least one location information value may be an address of a target property (e.g., a property where one or more work tasks are to be performed) specified by an operator. In other implementations, the at least one location information value may be a detected location associated with the work implement 120, such as a current location determined by a positioning system, or may be a detected location of a local computing device of the one or more computing devices 105 (particularly if the local computing device is connected to the work implement 120 via a wired or short-range wireless connection). The at least one location information value may be a value entered by an operator or received from another device connected to the work implement or the one or more computing devices 105 (e.g., additional work implements that may be nearby, or other nearby resources).

[0025] Using at least one location value, one or more computing device processors 11 can retrieve a map 5. The map 5 can be retrieved from a remote platform, such as a web mapping platform (which can retrieve maps by specifying a property or GPS location). The retrieved map 5 can be aerial imagery (e.g., images captured using a drone, aircraft, satellite, or a combination thereof), a topographical map, a transportation map, or a combination thereof. In various embodiments, the map 5 can include high-resolution, map-quality imagery, such as an orthomosaic map imagery that may be provided by a web mapping platform. In other embodiments, an operator or server can input any map described herein into one or more computing devices 105.

[0026] One or more computing devices 105 may present (display) information based on the map 5 on a touchscreen 15 or other display device associated with the one or more computing devices 105 to obtain selections from the operator and establish an autonomous operating area for the work implement 120. In some embodiments, the displayed information may include a graphical user interface through which the user can graphically indicate user selections from the map 5. In one specific example, the operator may add graphical markings via the graphical user interface using the touchscreen 15 or other input device of the user input interface of the one or more computing devices 105 to indicate the selected boundary.

[0027] By the user selecting on the presented map, the work machine 120 autonomously determines the work task. Perimeter information can be specified to define an enclosed area in which to perform the task. In other embodiments, the perimeter information may include one or more surrounding areas rather than defining a completely enclosed area. In any case, the perimeter information can exclude areas without grass coverage, such as bark shavings, trees, patios, gardens, etc.

[0028] One or more computing devices 105 may convert information selected by the operator from the map into coordinate information that can be interpreted by controller 21 to identify one or more virtual boundaries. In these specific examples, transmitted information 19 may include the coordinate information. In other embodiments, controller 21 may perform such conversions (e.g., controller 21 may derive the coordinate information from transmitted boundary information 19).

[0029] 2 illustrates operations (processes) 200 that may be performed by system 100 of FIG. 1 to generate one or more virtual boundaries of an autonomous operating area. In block 201, system 100 may identify location information values that indicate the geographic area in which the implement is intended to operate. The location information values may be addresses, locations detected using a positioning system, or combinations thereof. In some embodiments, the user interface may be configured to provide hints based on the vehicle's current or past locations (hints may prioritize current or past locations in a user selection list).

[0030] In block 202, the system 100 may obtain a map corresponding to the determined location information value. In block 203, the system 100 may display a user interface over or alongside at least a portion of the map.

[0031] In some embodiments, system 100 may detect the proposed boundary and include the proposed boundary on the displayed user interface, although this is not required. In embodiments in which the proposed boundary is displayed, system 100 may detect the boundary using pixel analysis, by accessing records (such as property boundary records), or the like, or a combination thereof. The user interface may include interface components for the user to accept or modify the proposed boundary.

[0032] In block 204, the system 100 can identify a selection range from the user interface that includes at least one peripheral location information value. In block 205, the system 100 can convert the at least one peripheral location information value into coordinate information. In block 206, the system 100 can use the coordinate information to generate one or more virtual boundaries that define an autonomous operating area for the work machine.

[0033] Figure 3 illustrates operations 300 that may be performed by the implement 120 of Figure 1 utilizing the generated virtual boundary or boundaries. In block 301, the implement 120 may monitor (manage) the implement's current position relative to the generated virtual boundary or boundaries.

[0034] If the work machine 120 determines in diamond block (decision tree) 302 that the current position does not match the autonomous operation area, then in block 303 the work machine 120 can stop operation of at least one powered drive unit 40 of the transportation system or an implement of the work machine 120, or change the operation of at least one powered drive unit 40. Otherwise (if there is a match), the work machine can continue monitoring the next current position in block 301.

[0035] FIG. 4 illustrates a method 400 that may be performed using any of the systems described herein to generate a boundary representation, according to various embodiments. In block 401, A location value may be identified that indicates a geographic region in which the machine is intended to operate. In block 402, a map corresponding to the identified location value may be obtained.

[0036] A user interface may be displayed over or alongside at least a portion of the map at block 403. At block 404, a selection range indicating at least one surrounding location information value may be identified from input to the user interface.

[0037] In block 405, the at least one boundary value is converted into coordinate information or other information that can be interpreted by the work machine. In block 406, the coordinate information can be used to generate one or more virtual boundary values that define at least a portion of the work machine's autonomous operating area. The work machine can perform any work machine operation based on the autonomous operating area, similar to the operation of FIG. 3.

[0038] In various embodiments, the user interface for defining the autonomous operating region may allow a user to assign different subregions with different restrictions. These different subregions may be defined by a user by processing and manipulating aerial imagery. For example, one subregion may be defined as a preferred transportation region (e.g., a preferred driving region). In this subregion, the implement may travel at a speed equal to or greater than the maximum driving speed of the other subregions, but may not use implements. In another subregion, the implement may travel at a lower maximum speed and may use implements. A practical application may be an autonomous lawnmower that travels along a path (trail) rather than across a lawn. In another practical application, the lawnmower may only use implements on the lawn, but not when traveling along the path.

[0039] In various embodiments, the autonomous operating domain is generated solely based on image analysis of aerial imagery (e.g., collected by an airborne device). In other embodiments, the autonomous operating domain may be generated based on a combination of image analysis of aerial imagery (e.g., imagery collected using an in-situ device, which may or may not be an aerial device) and site data. In another example, it may be practical and feasible to use an aerial drone to collect aerial and other site imagery of a site.

[0040] Further embodiments U.S. Provisional Application No. 63 / 398,308, filed August 16, 2022, describes embodiments that may be similar in various respects to the embodiments described herein. Any feature described in the embodiments of U.S. Provisional Application No. 63 / 398,308 may be combined with any feature of the embodiments described herein. The contents of the above U.S. Provisional Application are incorporated herein by reference, and are also explicitly redundantly described below with respect to FIGS. 5-8. FIG. 5 illustrates operations of a computing device that may be performed to visualize a boundary to an operator in accordance with various embodiments. FIG. 6A illustrates an exemplary user interface state that may be displayed by one or more computing device processors for tapping (selecting) the boundary detection button in the method of FIG. 5. FIG. 6B illustrates a next state of the exemplary user interface of FIG. 6A, which may be displayed by one or more computing device processors to visualize a boundary to a user after activation of the boundary detection button. FIG. 7 illustrates operations of a computing device that may be performed to visualize a boundary to an operator in accordance with various embodiments. FIG. 8 illustrates a process that may be performed by the boundary calculation module of FIG. 7.

[0041] In vehicle / implementation automation, vehicles such as drones, robots, or other machines (e.g., lawn mowers in some embodiments) can optionally utilize boundaries, such as safety boundaries or operating boundaries. In these systems, operators can define boundary areas within which the vehicle can operate normally, based on Global Navigation Satellite System (GNSS) coordinates, such as GPS coordinates.

[0042] If an operator defines an operating area, the vehicle's on-board computing system may use sensors (such as a GPS receiver) to continuously compare the vehicle's current position to the defined operating area. If the on-board computing system detects (e.g., based on the continuous comparison) that the vehicle has deviated from the defined operating area, the on-board computing system may generate a fault or other exception (e.g., the vehicle may request human assistance, may stop or reduce drive power to selected components (such as mower blades), or a combination thereof).

[0043] Some known user interfaces for selecting GNNS coordinates may require the operator to have knowledge of the GNNS coordinate identities and / or may be time-intensive. What is needed is an improved user interface for defining an operating region for a vehicle, such as a drone or robot (e.g., in some embodiments, a lawnmower), without requiring knowledge of the GNNS coordinate identities and / or more quickly than with existing user interfaces.

[0044] Various embodiments may include one or more computing devices (e.g., a user's smartphone and / or a remote cloud / server) that can be used to perform the method illustrated in Figure 5. The user interface can display a map based on aerial imagery (e.g., an orthomosaic map) to the user on a display. The map may be selected by the user entering location information (e.g., an address) into the system or by detecting the location of a drone, robot, etc.

[0045] A user can select a portion of the map to control a drone, robot, etc. In one embodiment, this can be done by tapping on a touch interface. Referring to Figures 6A and 6B, to control a vehicle within a park, a user can tap (select) a portion of the map showing the park in the initial state of the graphical user interface, as shown in Figure 6A. The system can recognize the selected location of the park, automatically generate a depicted boundary, and advance the graphical user interface to the next state, as shown in Figure 6B.

[0046] The system can generate the illustrated boundaries using any boundary identification algorithm now known or developed in the future. In one specific example, the processor described herein may include a boundary calculation module for executing the boundary identification algorithm. Figure 7 illustrates a boundary calculation module for executing the exemplary boundary identification algorithm shown in Figure 8.

[0047] In various embodiments, the module can access publicly available property line data and identify the edges of the park using the publicly available property line data. In another example, the system can use pixel analysis to identify proposed boundaries, which can be advantageous for recognizing property features that may not be visible in publicly available property line data. For example, if the vehicle is a lawnmower, the system can generate boundaries by using image analysis to identify areas covered with grass. In this case, areas without grass coverage, such as bark shavings, trees, patios, and gardens, can be excluded.

[0048] The illustrated boundary line may be a suggested boundary line, and the user may accept the suggested boundary line as is, or may modify the boundary line for any reason. In one example, the user may intend to use part of the lot as a parking lot or a jungle gym. In various embodiments, the presentation boundary is adjustable by dragging, resizing, etc.

[0049] Once the user approves the boundary, the system can generate GNSS coordinates from the approved boundary, which can convert locations on an image-based map (e.g., an orthomosaic map) to GNSS coordinates (e.g., GPS coordinates) using any algorithm now known or later developed.

[0050] In various embodiments, the system can output coordinate information that the user can use to program the vehicle. In some embodiments, the system can transmit the coordinate information to the vehicle and automatically apply operating boundaries to the vehicle.

[0051] In various embodiments, the system can generate additional information (other than the operating boundaries) that can assist the vehicle in operating within the operating boundaries. For example, referring to a lawnmower operating in a park, trees can be identified by performing pixel analysis. The lawnmower has its own sensors to identify trees while operating, but information indicating the presence of trees identified by pixel analysis of the map can be fused with information generated by the lawnmower's sensors to steer the vehicle within the operating area. Optionally, in various embodiments, the information generated by pixel analysis can be fused with sensor information generated by the vehicle to steer the vehicle within the operating boundaries autonomously or semi-autonomously.

[0052] [Determining boundaries using a general-purpose segmentation model] In any of the embodiments described herein, the task of determining boundaries can be performed using a general-purpose segmentation model (e.g., a Segment Anything Model (SAM)). This model does not generate semantic meaning for each segmented region generated. Instead, it generates only relevant object boundaries from a large training set.

[0053] Given an aerial image and focus points, a general-purpose segmentation model (e.g., the Segment Anything Model, or SAM) generates a set of candidate segmentation boundaries at the pixel level. This model does not generate semantic meaning for each segmented region; it only generates relevant object boundaries from a large training set. However, this autonomously generated initial data can be enriched by generating and displaying one or more visualizations that represent it. A user can select from one or more visualizations to enrich the autonomously generated initial data and generate subsequent data from which to define autonomous operating regions.

[0054] Semantic or instance segmentation models require extensive training and labeling and often fail to identify what a user might be interested in selecting in a map. In contrast, general-purpose segmentation models can be used to autonomously generate data that users can enrich by identifying regions of interest, providing more reliable and useful segmentation boundaries. The meaning of what is segmented is assigned by the user, such as an action region, a spray region, or a stationary region.

[0055] In some embodiments, the rasterized aerial image and the focus point (x,y) (tapped to indicate the center of the region of interest) are sent to a generic segmentation model. The output of this model is a pixel-level segmentation model that constitutes a set of candidate boundary predictions for various regions. These predictions continue a secondary process of user selection in which the user continues to select different boundaries (until the boundary of interest is selected as one of the boundary predictions). This process can also be done by forming a box around the boundary of interest in the primary image and selecting to exclude or include individual boundaries that are added to a final selection mask.

[0056] Once the mask is complete and the user approves it, a perimeter and associated polygon are generated for further refinement. Use the polygon's pick points to move, add, or remove vertices to reshape the perimeter. Once the final selection is made, the final mask boundary points are saved.

[0057] This process can be completed multiple times for multiple types of regions that are desired to be verified. For example, a user may want to determine a safety boundary that indicates the extent of a field. A user may also use this method to determine a speed region around a row of trees, which may consist of a connected mask of the trees. A "safety buffer" can be generated from the determined boundary, with a reference offset.

[0058] It may also be possible to use 3D data captured by drones in a Structure From Motion (SFM) process from LiDAR or multi-view images. If the 3D data includes color data (e.g., RGB), the same general-purpose segmentation model can be used to determine boundaries and refine them with more detailed information, such as the position of tree trunks rather than the crowns that are only visible from overhead images.

[0059] FIG. 9 illustrates a process according to various embodiments that may be performed by any boundary calculation module described herein in an embodiment that uses a generic segmentation model to generate data that can be enriched by a user. FIGS. 10A, 10B, and 10C each illustrate a schematic diagram of different exemplary visualizations according to various embodiments that may be displayed by any computing device processor described herein to perform the process described with reference to FIG. 9. With reference to FIG. 10A, the visualization of a selected location / site may appear differently from the visualizations of other nearby sites / buildings. With reference to FIG. 10B, a user can tap on a selected visualization to display a boundary presentation. With reference to FIG. 10C, a user can edit or approve the presented boundary.

[0060] In any embodiment described herein, multiple autonomous operating areas may be generated for the same work machine. One of these autonomous operating areas may include boundaries that differ from those of other autonomous operating areas. Each autonomous operating area may correspond to a different "mission" for the work machine. Different missions may be generated for the same work machine for the same physical area, such as for different times of day, lighting conditions, or weather conditions, and / or saved for later use by the user. A user may define one of the multiple missions as a currently active mission, and the work machine may utilize the autonomous operating area for the currently active mission. In a simple example, a user may define a first mission having a first operating area or safety boundary for the work machine to use during a specific time period, from a second mission having a different second operating area or safety boundary (for use outside of the specific time period).

[0061] It should be recognized that of the many potential embodiments to which the principles of the disclosed technology may be applied, the illustrated embodiments are merely preferred examples and should not be construed as limiting the scope of the present disclosure.

Claims

1. An apparatus including a work machine that performs one or more work tasks in an autonomous operating area, the autonomous operating area comprises one or more virtual boundaries for a field, lot, or other division of physical area; The working machine is 1) transportation systems; 2) one or more implements for performing the one or more work tasks; 3) one or more actuators for operating the power drives of the transportation system or the one or more implements; 4) a position determination system for determining a current location within said section of a field, lot, or other physical area; The work machine further includes one or more processor sets that control the one or more actuators; the processor set uses the positioning system to monitor a current position of the work implement relative to the one or more virtual boundaries of the autonomous operation area, and in response to the current position no longer coinciding with the autonomous operation area, stops or modifies operation of at least one of the one or more powered drives; the device further having a touch screen or other user input interface for selecting portions of the displayed map; The one or more processors are coupled to the touchscreen or other user input interface and configured to set the autonomous operating area based on user selections from the displayed map.

2. the one or more processors are configured to generate autonomously generated initial data at a pixel level based on information from the acquired map; the initial data includes a set of segmentation boundaries; the one or more processors are configured to generate a user input interface to collect user-specified data and to enrich the autonomously generated initial data to generate subsequent data; The apparatus of claim 1 , wherein the autonomous operating domain is established based on enriched subsequent data.

3. the user interface includes one or more visualizations showing a portion of the acquired map with the set of segmentation boundaries displayed thereon; 3. The apparatus of claim 2, wherein the user interface is configured to allow a user to further refine the segmentation boundary by using the one or more visualizations to move, add, or remove perimeter boundaries of a working area for the work implement.

4. The apparatus of claim 2 , wherein the autonomously generated initial data is generated using a generic segmentation model operated by a cloud device, server, or other remote resource.

5. The apparatus of claim 1 , wherein the autonomous operating region includes one or more speed regions.

6. the touchscreen or other user input interface connected thereto is integrated with the implement; The one or more devices connected to the touchscreen or other user input interface The apparatus of claim 1 , wherein the processor comprises one or more processor sets of the one or more processors of the work machine.

7. 7. The apparatus of claim 6, wherein stopping operation or modifying operation of at least one of the one or more actuators is commanded by a processor of the processor set that is different from the one or more processors connected to the touchscreen or other user input interface.

8. the one or more processors connected to the touchscreen or other user input interface are different from one or more processors in the processor set; the apparatus further comprises one or more computing devices connected to the work machine by a wired or wireless connection; the one or more computing devices including the touchscreen or other user input interface and the one or more processors connected thereto; the one or more computing devices transmit information about the autonomous operating area to the one or more processor sets via a wired or wireless connection; 2. The apparatus of claim 1, wherein monitoring by the one or more processor sets is initiated after receipt of the transmitted information.

9. The apparatus of claim 8 , wherein the one or more computing devices include a smartphone or other handheld device.

10. the touchscreen or other user input interface and at least one of the one or more processors connected thereto are part of a stationary device; 9. The apparatus of claim 8, wherein at least one processor of the stationary device is configured to download information about the autonomous operating area to a portable device connected to the work machine via a wired or wireless connection.

11. The apparatus of claim 10 , wherein the portable device includes a memory card or other memory readable by the one or more processor sets.

12. the one or more processors coupled to the touchscreen or other user input device; obtaining a map based on the received at least one location information value; configured to display a user interface on a display device on or alongside at least a portion of the map; the display device includes the touchscreen or other user input device associated with the one or more computing devices; The apparatus of claim 1 , wherein the user selection is generated based on user input received via a user interface.

13. the user selection includes a selection graphically indicated via the user interface; The apparatus of claim 12 , wherein at least one virtual boundary of the one or more virtual boundaries is derived from the user selection.

14. the set of one or more processors, or a processor connected to the touchscreen or other user input device, configured to convert a user-specified location on the imagery-based map into coordinate information; The autonomous operation area includes the coordinate information or is derived from the coordinate information.

1. The device described in 1.

15. one or more processors coupled to the touchscreen or other user input device are further configured to display a user input interface alongside at least a portion of the map including the presentation boundary; the user input interface modifies the proposed boundary using the user input interface if the user does not approve of the proposed boundary; and The device according to claim 1 , wherein the autonomous operating area is generated based on information about a position input via the user input interface, or is determined based on a position of the work machine or a position corresponding to an operator.

16. 16. The device of claim 15, wherein the presentation boundary is draggable using the user input interface to modify the presentation boundary.

17. The one or more processors coupled to the touchscreen or other user input device may further: obtaining the map based on at least one location information value; and The apparatus of claim 15 , configured to perform pixel analysis on content on the map to identify the presentation boundary.

18. the obtained map includes first data; The one or more processors coupled to the touchscreen or other user input device may further: accessing second data different from the first data; configured to perform boundary identification using the second data; The apparatus of claim 15 , wherein the proposed boundary is based on the result of the boundary identification.

19. The one or more processor sets further store a plurality of autonomous operation areas for the same work machine; Each autonomous operating domain corresponds to a different mission from multiple different missions, the one or more processor sets identifying a currently active mission from the plurality of different missions; The apparatus of claim 1 , wherein the current position of the work machine is monitored based on one autonomous operating area of the plurality of autonomous operating areas corresponding to a currently active mission.

Images