JP7606992B2 - Working area prediction method, working area prediction system, and working area prediction program - Google Patents

Description

The present invention relates to a work area estimation method, a work area estimation system, and a work area estimation program, which can be suitably used, for example, to estimate a work area where agricultural work has been performed.

Patent Document 1 (JP Patent Publication No. 6946217) discloses a patent for a field registration device. This field registration device includes a position information storage processing unit and a field identification data generation unit. The position information storage processing unit acquires position information of multiple positioning points obtained by sampling positioning data at a predetermined interval using GNSS (Global Navigation Satellite System) by a positioning device mounted on a work vehicle while the work vehicle is moving. The field identification data generation unit performs convex hull processing on the point cloud of the multiple positioning points to identify the shape of the field.

The field registration device of Patent Document 1 is effective when there is a single work area. However, in cases where a work vehicle moves between multiple fields and performs agricultural work in these multiple fields, multiple positioning points will straddle the work areas where agricultural work was performed in the multiple fields and roads outside the work areas. In order to distinguish and identify the shapes of each of these multiple work areas, time-consuming tasks such as detecting positioning points outside the field while referring to map information and excluding them from the target of convex hull processing are required.

Patent No. 6946217

In view of the above circumstances, one of the objectives of the present disclosure is to provide a work area estimation method, a work area estimation system, and a work area estimation program for efficiently estimating the work area in which a work vehicle has performed agricultural work. Other issues and novel features will become apparent from the description of this specification and the accompanying drawings.

Below, the means for solving the problem are explained using the numbers used in (Mode for carrying out the invention). These numbers are added to clarify the correspondence between the description in (Claims) and (Mode for carrying out the invention). However, these numbers should not be used to interpret the technical scope of the invention described in (Claims).

According to one embodiment, the work area estimation method includes acquiring position information obtained by measuring the position at a plurality of positioning points (P) passed by a work vehicle (2) moving in a farm field (9) and calculating the distance between any two positioning points (P) whose positions are measured consecutively among the plurality of positioning points (P) and the time interval between the positioning times at the two positioning points (P) (S02). The work area estimation method further includes generating graph information of a connected graph in which any two positioning points (P) among the plurality of positioning points (P) are connected as a node (N) when the distance is shorter than a first threshold and the time interval is shorter than a second threshold (S03), and detecting a community that is a subgraph of the connected graph and has a higher density than another subgraph of the connected graph (S04). The work area estimation method further includes estimating (S05) a work area (81, 82, 83, 84) in which the work vehicle (2) performed agricultural work within the field (9) based on the community, and outputting (S06) work area information representing the work area (81, 82, 83, 84) to the outside.

According to one embodiment, the work area estimation system (1) includes an acquisition unit (421), a calculation unit (422), a generation unit (423), a detection unit (424), an estimation unit (425), and an output unit (426). The acquisition unit (421) acquires position information obtained by measuring the position at a plurality of positioning points (P) through which a work vehicle (2) performing agricultural work while moving in a farm field (9) passes. The calculation unit (422) calculates the distance between any two positioning points (P) whose positions are measured consecutively among the plurality of positioning points (P) and the time interval between the positioning times at the two positioning points (P). The generation unit (423) generates graph information of a connected graph in which any two positioning points (P) among the plurality of positioning points (P) are connected as nodes (N) when the distance is shorter than a first threshold value and the time interval is shorter than a second threshold value. The detection unit (424) detects communities, which are subgraphs of the connected graph and have a higher density than other subgraphs of the connected graph. The estimation unit (425) estimates work areas (81, 82, 83, 84) in which the work vehicle (2) performed farm work in the field (9) based on the communities. The output unit (426) outputs work area information representing the work areas (81, 82, 83, 84) to the outside.

According to one embodiment, the work area estimation program is a program for implementing a predetermined process by executing it. This process includes acquiring position information obtained by measuring the position at a plurality of positioning points (P) through which a work vehicle (2) that moves in a farm field (9) and performs farm work passes (S01), and calculating the distance between any two positioning points (P) that are consecutive in the order of position measurement among the plurality of positioning points (P) and the time interval between the positioning times at the two positioning points (P) (S02). This process further includes generating graph information of a connected graph in which any two positioning points (P) among the plurality of positioning points (P) are connected as a node (N) when the distance is shorter than a first threshold value and the time interval is shorter than a second threshold value (S03), and detecting a community that is a subgraph of the connected graph and has a higher density than another subgraph of the connected graph (S04). This process further includes estimating (S05) the work area (81, 82, 83, 84) in which the work vehicle (2) performed agricultural work within the field (9) based on the community, and outputting (S06) work area information representing the work area (81, 82, 83, 84) to the outside.

According to one embodiment, it is possible to efficiently estimate the work area in which a work vehicle performed agricultural work.

FIG. 1 is a diagram showing an example of a configuration of a working area estimation system according to an embodiment. FIG. 2 is a block circuit diagram showing an example of a configuration of a working area estimation device according to an embodiment. FIG. 3 is a flow chart showing an example of a configuration of a working domain estimation method according to an embodiment. FIG. 4A is a diagram for explaining generation of graph information of a connected graph according to one embodiment. FIG. 4B is a diagram for explaining generation of graph information of a connected graph according to one embodiment. FIG. 5A is a diagram for explaining generation of graph information of a connected graph according to one embodiment. FIG. 5B is a diagram for explaining generation of graph information of a connected graph according to one embodiment. FIG. 5C is a diagram for explaining generation of graph information of a connected graph according to one embodiment. FIG. 5D is a diagram for explaining generation of graph information of a connected graph according to one embodiment. FIG. 6 is a diagram for explaining community detection according to an embodiment. FIG. 7A is a diagram for explaining estimation of a working area according to one embodiment. FIG. 7B is a diagram for explaining estimation of a working area according to one embodiment. FIG. 7C is a diagram for explaining estimation of a working area according to one embodiment. FIG. 7D is a diagram for explaining estimation of a working area according to one embodiment. FIG. 7E is a diagram for explaining estimation of a working area according to one embodiment. FIG. 7F is a diagram for explaining estimation of a working area according to one embodiment.

The following describes, with reference to the attached drawings, the modes for implementing the working area estimation method, working area estimation system, and working area estimation program according to the present disclosure.

(Embodiment)
As shown in Fig. 1, a working area estimation system 1 according to one embodiment includes a working area estimation device 4. The working area estimation system 1 may further include an external terminal 5. The working area estimation device 4, the external terminal 5, and an on-board terminal 20 mounted on the work vehicle 2 may be communicatively connected via a network 3. The work vehicle 2 performs work such as agricultural work while moving within the fields 9A, 9B. The work vehicle 2 may further move between the fields 9A, 9B. Hereinafter, when the fields 9A, 9B are not distinguished from each other, they will be collectively referred to as the field 9.

As shown in FIG. 2, the working area estimation device 4 according to one embodiment may be configured as a so-called computer. That is, the working area estimation device 4 includes a bus 41, a calculation device 42, a storage device 43, a communication device 44, and an input/output device 45. The bus 41 is configured to connect the calculation device 42, the storage device 43, the communication device 44, and the input/output device 45 so that they can communicate with each other.

The calculation device 42 includes an acquisition unit 421, a calculation unit 422, a generation unit 423, a detection unit 424, an estimation unit 425, and an output unit 426. The storage device 43 includes a work area estimation program storage unit 431. The work area estimation program storage unit 431 stores the work area estimation program.

The calculation device 42 executes the working area estimation program to realize the processing of the acquisition unit 421, calculation unit 422, generation unit 423, detection unit 424, estimation unit 425, and output unit 426. In other words, the acquisition unit 421, calculation unit 422, generation unit 423, detection unit 424, estimation unit 425, and output unit 426 are virtual functional blocks that perform processing realized by the calculation device 42 and storage device 43 working together. The processing of these functional blocks will be described later.

The working area estimation program may be read from an external recording medium 430 and stored in the working area estimation program storage unit 431. The recording medium 430 may be a non-transitory and tangible medium.

The communication device 44 communicates with external devices including the in-vehicle terminal 20 and/or the external terminal 5 by wireless communication and/or wired communication via the network 3, for example under the control of the acquisition unit 421 or the output unit 426. The working area estimation program may be received from the outside via the network 3 by the communication device 44 and stored in the working area estimation program storage unit 431.

The input/output device 45 outputs information to the user and accepts operations input by the user. As an example, the input/output device 45 includes a display device that outputs images, a speaker that outputs sound, buttons that accept press operations, a microphone that accepts sound input, a touch panel that accepts touch operations and outputs images, etc.

The in-vehicle terminal 20 includes a positioning device and a communication device, and is configured to transmit position information representing a position measured by the positioning device using, for example, a Global Navigation Satellite System (GNSS) to the working area estimation device 4 via the network 3. The in-vehicle terminal 20 may be configured as a computer in which a calculation device executes a program to realize processing.

The external terminal 5 includes a communication device and a display device, and is configured to output information received from the working area estimation device 4 via the network 3 on the display device. The external terminal 5 may be configured as a computer in which a calculation device executes a program to realize processing.

With reference to the flowchart in FIG. 3, an example of the configuration of a working area estimation method according to one embodiment will be described. In other words, with reference to the flowchart in FIG. 3, an example of the operation of a working area estimation system 1 according to one embodiment will be described. In other words, the flowchart in FIG. 3 shows an example of the configuration of a working area estimation program executed by a working area estimation device 4 according to one embodiment.

Before the processing of the flowchart in FIG. 3 starts, the positioning device of the on-board terminal 20 of the work vehicle 2 measures its position and acquires position information. The point at which the on-board terminal 20 acquires position information is called a positioning point. The positioning point is included in the working trajectory that the work vehicle 2 passed while performing farm work in the field 9 or the movement trajectory that the work vehicle 2 moved without performing farm work. The on-board terminal 20 acquires position information at each of a plurality of times and stores the position information in the storage device in association with positioning time information that indicates the positioning time at which the position information was acquired. The on-board terminal 20 may perform positioning at a predetermined cycle. Strictly speaking, the position information indicates the position of the positioning device at the corresponding positioning time, but if the positioning device is fixed to the work vehicle 2, it actually indicates the position of the work vehicle 2 at that positioning time. The on-board terminal 20 starts acquiring position information when the work vehicle 2 starts operating by key-on operation. Furthermore, when the work vehicle 2 ends its operation by turning off the key, the in-vehicle terminal 20 stops acquiring the position information and transmits the position information stored in the storage device to the work area estimation device 4 via the network 3. When the work area estimation device 4 receives the position information, it starts the process of the flowchart in FIG. 3.

When the processing of the flowchart in FIG. 3 starts, step S01 is executed. In step S01, the acquisition unit 421 of the work area estimation device 4 acquires position information of the positioning point. More specifically, the acquisition unit 421 controls the communication device 44 to receive the position information transmitted from the vehicle-mounted terminal 20 and stores it in the storage device 43. The operation of the vehicle-mounted terminal 20 to measure the position of the work vehicle 2 and acquire the position information may be executed in step S01 or before step S01.

The position information of each positioning point may include, for example, the latitude and longitude of the positioning point. Furthermore, the position information of each positioning point may be recorded in association with the order in which the positioning was performed, in addition to positioning time information indicating the time at which the positioning was performed. The order of positioning may be calculated from the positioning time information.

After step S01, step S02 is executed. In step S02, the calculation unit 422 of the working area estimation device 4 calculates the distance and the time interval between two consecutive positioning points. More specifically, the calculation unit 422 calculates the distance between any two consecutive positioning points among the multiple positioning points based on the position information of these two positioning points. The calculation unit 422 also calculates the time interval between two positioning times respectively associated with these two positioning points based on these positioning times.

When the vehicle-mounted terminal 20 performs positioning periodically at a predetermined cycle, the time interval may be considered known and calculation of the time interval may be omitted. However, even in this case, the time interval may be calculated based on the recorded positioning time, taking into consideration the possibility that a malfunction occurred in receiving the GNSS signal.

After step S02, step S03 is executed. In step S03, the generating unit 423 of the working area estimation device 4 generates graph information of a connected graph in which two of the multiple positioning points that satisfy a predetermined condition are connected. This condition includes a first condition related to a first threshold and a second condition related to a second threshold. More specifically, the generating unit 423 performs a process of connecting the two positioning points when the distance between the two positioning points is shorter than the first threshold and the time interval between the two positioning times respectively associated with the two positioning points is shorter than the second threshold. The generating unit 423 performs this process for each combination of any two positioning points included in the multiple positioning points, thereby generating graph information representing a connected graph in which some of the positioning points are connected. Here, the graph information is data that represents the connected graph in a format that can be handled by a computer. The graph information includes, for example, information representing each of the nodes and edges included in the connected graph and information representing the connection relationship between the nodes and edges.

Generation of graph information of a connected graph according to one embodiment will be described with reference to Figures 4A and 4B. The first connected graph shown in the example of Figure 4A includes nodes N10, N11, N12, N13, N14, and N15. Hereinafter, when nodes N10 to N15 are not distinguished from one another, they will be collectively referred to as node N. In the example of Figure 4A, these nodes N represent multiple positioning points where positioning was performed while the work vehicle 2 was moving. The position of each node N in Figure 4A represents the position of the corresponding positioning point, and the connection of these nodes N represents that the order of positioning at the corresponding two positioning points is consecutive.

Specifically, in the example of FIG. 4A, node N10 is connected to node N11, node N11 is connected to node N12, node N12 is connected to node N13, node N13 is connected to node N14, and node N14 is connected to node N15. These connections indicate that the positioning points corresponding to nodes N10, N11, N12, N13, N14, and N15 have been located in this order.

Boundary C in FIG. 4A is an arc that is part of a circle centered at node N10. In other words, nodes N11, N12, N14, and N15, which are inside the circle, are closer to node N10 than the radius of the circle. Conversely, node N13, which is outside the circle, is farther from node N10 than the radius of the circle.

Of the connections in FIG. 4A, the connections drawn with solid lines represent connections with positioning points that were located before a predetermined time has elapsed from the positioning time of node N10. Conversely, the connections drawn with dashed lines represent connections with positioning points that were located after a predetermined time has elapsed from the positioning time of node N10. In the example of FIG. 4A, the positions of nodes N11, N12, N13, and N14 are measured before a predetermined time has elapsed from the positioning time of node N10. The remaining node N15 is measured after a predetermined time has elapsed from the positioning time of node N10.

Focusing on node N10 in Figure 4A, nodes N11, N12, and N14 satisfy the condition that their distance from node N10 is shorter than the radius of the boundary line C circle, and their positioning time is shorter than a predetermined time from the positioning time of node N10, but the remaining nodes N13 and N15 do not. This is represented as in the second connected graph in Figure 4B. In the example of Figure 4B, node N10 of interest is connected to nodes N11, N12, and N14, which satisfy the above conditions, and is not connected to nodes N13 and N15, which do not satisfy the above conditions.

In the examples of Figures 4A and 4B, nodes N11, N12, and N14 that are connected to node N10 in the second connection graph of Figure 4B are drawn with solid lines. Conversely, nodes N13 and N15 that are not connected to node N10 in the second connection graph of Figure 4B are drawn with dashed lines.

Next, the generation of graph information of a connected graph according to one embodiment will be described in more detail with reference to Figures 5A, 5B, 5C, and 5D. In the example of Figure 5A, a field 9 and positioning points P0, P1, P2, P3, P4, P5, P6, P7, P8, P9, P10, P11, and P12 are depicted. Hereinafter, when the positioning points P0 to P12 are not distinguished from each other, they are collectively referred to as positioning points P. Note that in Figure 5A, the positioning point P, which does not actually have an area, is depicted as a circle centered on the positioning point P in order to make it easier to see. This also applies to Figures 5B, 5C, and 5D, which will be described later. The positioning point P0 is located outside the field 9, and the remaining positioning points P1 to P12 are located inside the field 9. The arrows connecting the positioning points P0 to P12 in this order indicate that the work vehicle 2 performed positioning while passing through the positioning points P0 to P12 in this order. In the example of FIG. 5A, positioning at each positioning point P is performed at a predetermined positioning cycle, and this positioning cycle is one minute. Also, in the example of FIG. 5A, the work vehicle 2 passes through positioning points P0 to P12 while moving at a constant speed, and the distance between two consecutive positioning points P in the positioning order is 5 meters. However, these specific numerical values are merely examples and do not limit the embodiment.

Figure 5B shows an example of a connected graph generated when positioning point P0 is first focused on. Here, boundary line C0 is a circle centered on positioning point P0. The radius of this boundary line C0 is a first threshold used to determine whether or not to connect a certain positioning point P to positioning point P0. In the example of Figure 5B, the first threshold is 11 meters. As described above, in one embodiment, the first condition included in the conditions for connecting the focused positioning point P0 to a certain positioning point P is that the distance from positioning point P0 to a certain positioning point P is shorter than the first threshold. In the example of Figure 5B, positioning points P1, P2, and P8 satisfy the first condition, and the remaining positioning points P3 to P7 and P9 to P12 do not satisfy the first condition. Although boundary line C0 overlaps with the circles representing positioning points P7 and P9, it does not reach the center points of these circles, so the distance from the positioning point P0 of interest to positioning point P7 or P9 is longer than the first threshold value, and therefore positioning points P7 and P9 are determined not to satisfy the first condition.

As described above, in one embodiment, the second condition included in the conditions for connecting the positioning point P0 of interest with a certain positioning point P is that the time interval from the positioning time at which the work vehicle 2 measures its position at the positioning point P0 to the positioning time at that positioning point P is shorter than the second threshold value. In the example of FIG. 5B, the second threshold value is 7 minutes and 30 seconds. In the example of FIG. 5B, the positioning points P1 to P7 satisfy the second condition, and the remaining positioning points P8 to P12 do not satisfy the second condition.

As described above, in one embodiment, the conditions for connecting a positioning point P0 of interest to a certain positioning point P include a first condition related to a first threshold and a second condition related to a second threshold. In the example of FIG. 5B, positioning points P1 and P2 satisfy the first condition and the second condition. In the example of FIG. 5B, the edges of the second connection graph generated in step S03 of the flowchart in FIG. 3 are indicated by thick solid lines. More specifically, the second connection graph in FIG. 5B includes an edge connecting positioning point P0 to positioning point P1, and an edge connecting positioning point P0 to positioning point P2.

Note that the remaining positioning points P3 to P12 do not satisfy the first condition and/or the second condition, and are therefore not included in the second connection graph in FIG. 5B. In FIG. 5B, the positioning points P3 to P12 that do not satisfy the first condition and/or the second condition are indicated by dashed lines.

Figure 5C shows an example of a third connected graph that is generated when first focusing on positioning point P0 and then on positioning point P1. The third connected graph shown in Figure 5C includes edges represented by thick solid lines and edges represented by thin dashed lines. The edges represented by thick solid lines represent connections between positioning point P1 and other positioning points P. The edges represented by thin dashed lines are edges of the second connected graph shown in Figure 5B that are not included in the third connected graph shown in Figure 5C.

In the example of FIG. 5C, the boundary line C1 is a circle centered on the positioning point P1. The radius of this boundary line C1 is the first threshold value related to the first condition. The first threshold value in the example of FIG. 5C is the first threshold value used in the process described with reference to FIG. 5B, and is 11 meters as an example. In the example of FIG. 5C, the positioning points P0, P2, P3, and P7 to P9 satisfy the first condition related to the first threshold value, and the remaining positioning points P4 to P6 and P10 to P12 do not satisfy the first condition. Note that although the boundary line C1 overlaps with the circles representing the positioning points P6 and P10, it does not reach the center points of these circles, so the distance from the positioning point P1 of interest to the positioning points P6 or P10 is longer than the first threshold value, and therefore the positioning points P6 and P10 are determined not to satisfy the first condition.

In the example of FIG. 5C, the second threshold is the second threshold used in the process described with reference to FIG. 5B, and is, for example, 7 minutes and 30 seconds. In the example of FIG. 5C, positioning points P0 and P2 to P8 satisfy the second condition, and the remaining positioning points P9 to P12 do not satisfy the second condition.

In the example of FIG. 5C, the positioning points P0, P2, P3, P7, and P8 satisfy the first and second conditions. In the example of FIG. 5C, the edges connecting the positioning point P1 to the other positioning points P in the third connection graph generated in step S03 of the flowchart in FIG. 3 are shown with thick solid lines. More specifically, the third connection graph in FIG. 5C includes edges connecting the positioning point P1 to the positioning points P0, P2, P3, P7, and P8, respectively. The third connection graph shown in FIG. 5C further includes an edge connecting the positioning point P0 to the positioning point P2, which is an edge not related to the positioning point P1, in the second connection graph shown in FIG. 5B.

Then, by focusing on each of the positioning points P2 to P12 in turn, a connection graph is generated by adding edges connecting the focused positioning point P to a positioning point P that satisfies a first condition related to the first threshold value and a second condition related to the second threshold value. Figure 5D shows an example of a connection graph generated as a result of combining the connection graphs generated by focusing on each of the positioning points P0 to P12.

After step S03 in the flowchart of FIG. 3, step S04 is executed. In step S04, the detection unit 424 of the working domain estimation device 4 detects a community from within the connected graph. A community is a subgraph of the connected graph in which nodes are more densely connected than in other parts. In one embodiment, a so-called network analysis technique is used to detect a subgraph of the connected graph in which connections are more dense.

With reference to Figure 6, a method of detecting communities from a connected graph using a technique known as network analysis will be described. The connected graph shown in the example of Figure 6 includes nodes N21, N22, N23, N24, N25, and N26, and edges E1, E2, E3, E4, E5, E6, and E7. Hereinafter, when there is no distinction between nodes N21 to N26, they will be referred to as nodes N. Furthermore, when there is no distinction between edges E1 to E7, they will be referred to as edges E.

In the example of FIG. 6, edge E1 connects nodes N21 and N22. Similarly, edge E2 connects nodes N22 and N23. Edge E3 connects nodes N21 and N23. Edge E4 connects nodes N23 and N24. Edge E5 connects nodes N24 and N25. Edge E6 connects nodes N24 and N26. Edge E7 connects nodes N25 and N26.

First, for each combination of any two nodes included in the connected graph, the shortest path for moving along the edges between these two nodes is calculated. In the example of FIG. 6, a method is known for determining the shortest path for moving from one node N included in the connected graph to another node N included in the same connected graph via an edge E included in the same connected graph. As an example, the shortest path for moving from node N21 to node N22 is edge E1. Similarly, the shortest path for moving from node N21 to node N26 is edges E3, E4, and E6. Such shortest paths are calculated for each combination of any two nodes N included in the connected graph shown in the example of FIG. 6.

Next, for each edge included in the connected graph, the number of times it appears on the calculated shortest path is calculated. In the example of Figure 6, edge E1 appears only once on the shortest path between nodes N21 and N22, so the number of times edge E1 appears on the shortest path is 1. Similarly, edge E7 appears on the shortest path is also 1. Furthermore, edge E2 appears a total of four times on the shortest paths between node N22 and each of nodes N23, N24, N25, and N26, so the number of times edge E2 appears on the shortest path is 4. Similarly, edges E3, E5, and E6 appear on the shortest path are also 4. Furthermore, edge E4 appears a total of nine times in the shortest paths between node N21 and each of nodes N24, N25, and N26, the shortest paths between node N22 and each of nodes N24, N25, and N26, and the shortest paths between node N23 and each of nodes N24, N25, and N26, so the number of times edge E4 appears in the shortest paths is nine.

Next, among the edges in the connected graph, the edge that appears the most number of times on the shortest path is detected. In the example of Figure 6, edge E that appears the most number of times on the shortest path is edge E4.

Next, the edge that appears most frequently on the shortest path is removed from the connected graph, thereby dividing the connected graph and detecting the communities of the nodes. In the example of FIG. 6, the edge E4 that appears most frequently on the shortest path is removed from the connected graph, thereby detecting a first community 8A including nodes N21, N22, and N23, and a second community 8B including nodes N24, N25, and N26.

If removing the edge that appears most frequently on the shortest path does not split the connected graph, the nodes' communities are found by continuing to remove edges in descending order of frequency of appearance on the shortest path until the connected graph is split.

After step S04 in the flowchart of FIG. 3, step S05 is executed. In step S05, the estimation unit 425 of the work area estimation device 4 estimates the work area in which the work vehicle 2 performed work, based on the community detected in step S04. More specifically, the estimation unit 425 estimates the position and shape of the work area corresponding to each community by calculating the convex hull or concave hull of the positioning points included in each detected community, based on the position information of these positioning points.

As an example, a case will be described where the position information shown in FIG. 7A has been acquired. In FIG. 7A, multiple dots represent multiple positioning points at which the position of the work vehicle 2 has been measured, and multiple straight lines represent the movement trajectory of the work vehicle 2 moving without working, and the working trajectory of the work vehicle 2 moving while performing agricultural work. Steps S02 to S04 of the flowchart in FIG. 3 are executed based on the position information in FIG. 7A, and the communities of the positioning points shown in FIG. 7B, FIG. 7C, FIG. 7D, and FIG. 7E are detected. Then, by calculating the convex hull of the positioning points included in each of the detected communities, the positions and shapes of the work areas 81, 82, 83, and 84 corresponding to each community are inferred, as shown in FIG. 7F.

After step S05 in the flowchart of FIG. 3, step S06 is executed. In step S06, the output unit 426 of the work area estimation device 4 outputs work area information representing the estimated work area. More specifically, the output unit 426 controls the input/output device 45 to output the work area information representing the position and shape of the estimated work area to the outside. As an example, a display device or touch panel included in the input/output device 45 may display the estimated shape of the work area so that the user can visually recognize it. At this time, the estimated shape of the work area may be displayed superimposed on map information of the area including the work area so that the user can intuitively understand it. Furthermore, the positioning point P included in the work area may be displayed superimposed on the estimated shape and map information.

After step S06, the processing of the flowchart in FIG. 3 ends.

As described above, according to one embodiment, the work area can be efficiently estimated by automatically excluding positioning points outside the work area based on position information measured by the work vehicle 2 at multiple positioning points.

(Variation 1)
In the above-mentioned first embodiment, a configuration example in which a so-called network analysis technique is used to detect a community by excluding edges that appear most frequently on the shortest path between nodes in step S04 of the flowchart in Fig. 3 has been described with reference to Fig. 6. As a variation of this configuration example, a configuration example in which a community is detected by excluding nodes in the connection graph generated in step S03 of the flowchart in Fig. 3, each of which has an edge count less than a predetermined threshold, will be described.

In the example of the connected graph shown in FIG. 5D, the total number of edges of the positioning point P0 as a node is 2, the number of edges of the positioning points P4, P9, and P12 as nodes is 4, the number of edges of the positioning point P1 as a node is 5, the number of edges of the positioning points P3, P5, P8, P10, and P11 as nodes is 6, the number of edges of the positioning point P2 as a node is 7, and the number of edges of the positioning points P6 and P7 as nodes is 9. Here, if the threshold value of the total number of edges of the node is, for example, 3, the positioning point P0 as a node outside the field 9 is distinguished from the positioning points P1 to P12 as nodes inside the field 9. This is due to the difference that when the work vehicle 2 moves outside the field 9 without performing farm work, there is a high probability that its movement trajectory will be a narrow single road, and when the work vehicle 2 moves inside the field 9 while performing farm work, there is a high probability that its operation trajectory will include multiple parallel and adjacent lines. By excluding positioning point P0 as a node whose total number of edges is less than the threshold from the connected graph, a community consisting of positioning points P1 to P12 as nodes remaining in the connected graph is detected.

The threshold for the total number of edges for a node may be set in advance to an appropriate value based on previously acquired location information.

(Variation 2)
In the above-described first embodiment, in step S06 of the flowchart in FIG. 3, the configuration in which the working area estimation device 4 displays and outputs the working area information representing the estimated working area has been described. As a modified example of this configuration, the working area information may be output by being displayed by the external terminal 5. In this case, the output unit 426 of the working area estimation device 4 controls the communication device 44 to transmit the working area information to the external terminal 5 via the network 3. The communication device of the external terminal 5 receives the working area information, and the display device of the external terminal 5 outputs the working area information to the outside. At this time, the estimated shape of the working area may be displayed superimposed on map information of the area including the working area so that the user can intuitively understand. Furthermore, the positioning point P included in the working area may be displayed superimposed on the estimated shape and map information.

The invention made by the inventor has been specifically described above based on the embodiment, but it goes without saying that the present invention is not limited to the embodiment and can be modified in various ways without departing from the gist of the invention. Furthermore, the features described in the embodiment can be freely combined as long as they are not technically inconsistent.

REFERENCE SIGNS LIST 1 Working area estimation system 2 Work vehicle 20 On-board terminal 3 Network 4 Working area estimation device 41 Bus 42 Calculation device 421 Acquisition unit 422 Calculation unit 423 Generation unit 424 Detection unit 425 Estimation unit 426 Output unit 43 Storage device 430 Recording medium 431 Working area estimation program storage unit 44 Communication device 45 Input/output device 5 External terminal 8A, 8B Community 81, 82, 83, 84 Working area 9, 9A, 9B Field C, C0, C1 Boundary line E1, E2, E3, E4, E5, E6, E7 Edge N10, N11, N12, N13, N14, N15 Node N21, N22, N23, N24, N25, N26 Node P0, P1, P2, P3, P4, P5, P6, P7, P8, P9, P10, P11, P12 Positioning points

Claims (7)

acquiring position information obtained by measuring the position at a plurality of positioning points passed by a work vehicle that moves within a farm field while performing agricultural work;
Calculating a distance between any two positioning points that are consecutive in the order of measuring the positions among the plurality of positioning points, and a time interval between the measurement times of the positioning points at the two positioning points;
generating graph information of a connected graph in which any two of the plurality of positioning points are connected as nodes when the distance between the two is shorter than a first threshold and the time interval between the two is shorter than a second threshold;
detecting communities, which are subgraphs of the connected graph that have a higher density than another subgraph of the connected graph;
estimating a work area in the field in which the work vehicle performed the agricultural work based on the community;
and outputting to the outside working area information representing the working area. 2. The working space estimation method according to claim 1,
The detecting comprises:
removing edges of the connected graph in descending order of frequency of appearance on a shortest path between any two of the plurality of positioning points until the subgraph is separated from the connected graph. 2. The working space estimation method according to claim 1,
The detecting comprises:
excluding nodes of the connected graph having a total number of edges below a third threshold. The working area estimation method according to any one of claims 1 to 3,
The connection graph includes, as nodes, positioning points that the work vehicle has passed through from when it starts to when it finishes its operation. The working space estimation method according to any one of claims 1 to 4,
The outputting step includes:
and displaying the estimated shape of the work area by superimposing it on map information. an acquisition unit that acquires position information obtained by measuring the position of a work vehicle that moves within a farm field while performing agricultural work at a plurality of positioning points through which the work vehicle passes;
a calculation unit that calculates a distance between any two positioning points that are consecutive in the order of measurement of the positions among the plurality of positioning points, and a time interval between the measurement times of the positioning points at the two positioning points;
a generation unit that generates graph information of a connected graph in which any two of the plurality of positioning points are connected as nodes when the distance between the two is shorter than a first threshold and the time interval between the two is shorter than a second threshold;
a detector for detecting communities, the subgraphs of the connected graph having a higher density than another subgraph of the connected graph;
an estimation unit that estimates a work area in the field in which the work vehicle performed the agricultural work based on the community;
and an output unit that outputs working area information representing the working area to an outside. A work area estimation program for implementing a predetermined process by executing the work area estimation program,
The process comprises:
acquiring position information obtained by measuring the position at a plurality of positioning points passed by a work vehicle that moves within a farm field while performing agricultural work;
Calculating a distance between any two positioning points that are consecutive in the order of measuring the positions among the plurality of positioning points, and a time interval between the measurement times of the positioning points at the two positioning points;
generating graph information of a connected graph in which any two of the plurality of positioning points are connected as nodes when the distance between the two is shorter than a first threshold and the time interval between the two is shorter than a second threshold;
detecting communities, which are subgraphs of the connected graph that have a higher density than another subgraph of the connected graph;
estimating a work area in the field in which the work vehicle performed the agricultural work based on the community;
and outputting to the outside work area information representing the work area.

Images