Work area inference method, work area inference system, and work area inference program

Abstract

To efficiently infer a work area where a working device completed farm work, with low load and high accuracy.SOLUTION: A work area inference method includes: (S01) acquiring position information obtained by measuring a position at each of points which a working device (2) performing farm work while moving in a field (9) passes; (S03) extracting points satisfying a first extraction condition based on the order of position measurement, as first extracted measured points; (S05) generating a first link graph obtained by linking first extracted measured points satisfying a first link condition, as first link nodes; (S08) generating a first candidate polygon by inserting points satisfying a first insertion condition based on the first link graph, out of points except the first extracted measured points, to the first link graph as first insertion points; (S10) inferring a work area where the working device completed farm work in the field, on the basis of the first candidate polygon; and (S11) outputting work area inference information indicating the inferred work area to the outside.SELECTED DRAWING: Figure 4

Description

[Technical Field]

[0001] 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. [Background technology]

[0002] Patent Document 1 (JP 2023-102321 A) discloses a work area estimation method for efficiently estimating the work area where a work vehicle has performed agricultural work. This work area estimation method measures the positions of multiple points passed by the work vehicle, calculates the distance and the time interval between positioning times for any two positioning points where the measurements are consecutive, generates graph information for a connected graph in which two positioning points where the distance and time interval are shorter than a threshold are connected as nodes, detects communities, which are subgraphs of the connected graph that have a higher density than other subgraphs, and estimates the work area based on these communities.

[0003] In the work area estimation method of Patent Document 1, if the positioning sampling period is shortened to improve the accuracy of estimating the work area, the amount of positioning data may become too large, making it difficult to perform calculations for network analysis. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Publication No. 2023-102321 Summary of the Invention [Problem to be solved by the invention]

[0005] In view of the above circumstances, an object of the present disclosure is to provide a work area estimation method, a work area estimation system, and a work area estimation program for estimating with high accuracy and efficiency the work area in which a work implement has performed agricultural work, while realizing a relatively low load for network analysis calculations. Other objects and novel features will become apparent from the description of this specification and the accompanying drawings. [Means for solving the problem]

[0006] The following describes the means for solving the problems using the numbers and symbols used in the description of the invention. These numbers and symbols are added in parentheses for reference purposes to show an example of the correspondence between the claims and the description of the invention. Therefore, the claims should not be interpreted as being limited by the parenthetical descriptions.

[0007] According to one embodiment, a work area estimation method includes acquiring position information (S01) that measures the position of each of a plurality of points passed by a work implement (2) that moves within a field (9) while performing agricultural work. The work area estimation method further includes extracting, from the plurality of points included in the position information, points that satisfy a first extraction condition based on the order in which the positions were measured, as first extracted positioned points (S03). The work area estimation method further includes generating a first connection graph in which, from the plurality of first extracted positioned points, first extracted positioned points that satisfy a first connection condition are connected as first connection nodes (S05). The work area estimation method further includes inserting, into the first connection graph, a point that satisfies a first insertion condition based on the first connection graph, from a plurality of points excluding the plurality of first extracted positioned points, to generate a first candidate polygon (S08). The work area estimation method further includes estimating, based on the first candidate polygon, a work area in the field (9) where the work implement (2) performed agricultural work (S10). The work area estimation method further includes outputting work area estimation information indicating the estimated work area to the outside (S11).

[0008] According to one embodiment, the work area estimation system (1) includes a position information acquisition unit (421), an extraction unit (422), a connection graph generation unit (423), a candidate polygon generation unit (424), an estimation unit (425), and an output unit (426). The position information acquisition unit (421) acquires position information obtained by measuring the position of each of a plurality of points passed by a work implement (2) that moves within a farm field (9) and performs agricultural work. The extraction unit (422) extracts, from the plurality of points included in the position information, points that satisfy a first extraction condition based on the order in which the positions were measured, as first extracted positioned points. The connection graph generation unit (423) generates a first connected graph that connects, as first connected nodes, first extracted positioned points that satisfy the first connection condition among the plurality of first extracted positioned points. A candidate polygon generation unit (424) generates a first candidate polygon by inserting a point that satisfies a first insertion condition based on the first connected graph as a first insertion point into the first connected graph from among the multiple points excluding the multiple first extracted positioning points. An estimation unit (425) estimates a work area in the field (9) where the work implement (2) performed farm work based on the first candidate polygon. An output unit (426) outputs work area estimation information indicating the estimated work area to the outside.

[0009] According to one embodiment, the work area estimation program is a program for executing a predetermined process. This process includes acquiring position information (S01) that measures the position of each of a plurality of points passed by a work implement (2) that moves within a farm field (9) to perform agricultural work. This process further includes extracting, from the plurality of points included in the position information, points that satisfy a first extraction condition based on the order in which the positions were measured, as first extracted positioned points (S03). This process further includes generating a first connected graph in which, from the plurality of first extracted positioned points, first extracted positioned points that satisfy a first connection condition are connected as first connected nodes (S05). This process further includes inserting, into the first connected graph, a point that satisfies a first insertion condition based on the first connected graph, from a plurality of points excluding the plurality of first extracted positioned points, as a first insertion point, to generate a first candidate polygon (S08). This process further includes estimating (S10) a work area in the field 9 where the work implement 2 has performed farm work, based on the first candidate polygon. This process further includes outputting (S11) work area estimation information representing the estimated work area to the outside. [Effects of the Invention]

[0010] According to one embodiment, the work area in which the work implement has performed farm work can be estimated efficiently, with low load, and with high accuracy. [Brief explanation of the drawings]

[0011] [Figure 1] FIG. 1 is a diagram illustrating an example of the configuration of a working area estimation system according to an embodiment. [Figure 2] FIG. 2 is a block circuit diagram showing an example of the configuration of a working area estimation device according to an embodiment. [Figure 3] FIG. 3 is a block circuit diagram illustrating an example of the configuration of an external terminal according to an embodiment. [Figure 4] FIG. 4 is a flowchart showing an example of a configuration of a working area estimation method according to an embodiment. [Figure 5] FIG. 5 is a table illustrating an example of location information according to one embodiment. [Figure 6] FIG. 6 is a table showing an example of position information of extracted positioning points according to one embodiment. [Figure 7] FIG. 7 is a diagram illustrating generation of a connected graph according to an embodiment. [Figure 8] FIG. 8 is a diagram illustrating generation of a connected graph according to an embodiment. [Figure 9] FIG. 9 is a diagram illustrating generation of a connected graph according to an embodiment. [Figure 10] FIG. 10 is a diagram illustrating generation of a connection graph according to an embodiment. [Figure 11] FIG. 11 is a diagram illustrating generation of a connected graph according to an embodiment. [Figure 12] FIG. 12 is a diagram illustrating generation of a connection graph according to an embodiment. [Figure 13] FIG. 13 is a diagram illustrating community detection according to an embodiment. [Figure 14] FIG. 14 is a diagram illustrating the insertion of a point according to an embodiment. [Figure 15] FIG. 15 is a diagram illustrating the insertion of a point according to an embodiment. [Figure 16] FIG. 16 is a diagram illustrating the estimation of the working area according to one embodiment. [Figure 17] FIG. 17 is a diagram illustrating the estimation of the working area according to one embodiment. [Figure 18] FIG. 18 is a diagram illustrating the estimation of the working area according to one embodiment. [Figure 19] FIG. 19 is a diagram illustrating the estimation of the working area according to one embodiment. [Figure 20] FIG. 20 is a diagram illustrating the estimation of the working area according to one embodiment. [Figure 21] FIG. 21 is a diagram illustrating the estimation of the working area according to one embodiment. [Figure 22] FIG. 22 is a table showing an example of position information of extracted positioning points according to one embodiment. [Figure 23] FIG. 23 is a diagram illustrating the insertion of a point according to an embodiment. [Figure 24] FIG. 24 is a diagram for explaining the degree of coincidence between a candidate polygon and a bitmap image of the corresponding extracted positioning point according to an embodiment. [Figure 25] FIG. 25 is a diagram for explaining the degree of coincidence between a candidate polygon and a bitmap image of the corresponding extracted positioning point according to an embodiment. [Figure 26] FIG. 26 is a diagram for explaining the degree of coincidence between a candidate polygon and a bitmap image of the corresponding extracted positioning point according to an embodiment. [Figure 27] FIG. 27 is a diagram for explaining the degree of coincidence between a candidate polygon and a bitmap image of the corresponding extracted positioning point according to an embodiment. [Figure 28] FIG. 28 is a diagram for explaining the degree of coincidence between a candidate polygon and a bitmap image of the corresponding extracted positioning point according to an embodiment. [Figure 29] FIG. 29 is a diagram for explaining the degree of coincidence between a candidate polygon and a bitmap image of the corresponding extracted positioning point according to an embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0012] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS With reference to the accompanying drawings, embodiments for implementing a working area estimation method, a working area estimation system, and a working area estimation program according to the present disclosure will be described below.

[0013] (Embodiment) 1, a work area estimation system 1 according to one embodiment includes a work area estimation device 4. The work area estimation system 1 may further include some or all of an on-board terminal 20 mounted on a work apparatus 2 and an external terminal 5. The work area estimation device 4, the external terminal 5, and the on-board terminal 20 mounted on the work apparatus 2 may be connected to a network 3 via wired communication and / or wireless communication.

[0014] The on-board terminal 20 is mounted on a work device 2. The work device 2 moves on the ground and performs agricultural work such as tilling and harvesting on a field 9 and the crops grown in the field 9. The work device 2 may also move between fields 9A and 9B. Hereinafter, when there is no need to distinguish between fields 9A and 9B, these will be collectively referred to as field 9. As an example, the work device 2 may be a work device such as a tiller for tilling, a transplanter or rice transplanter for transplanting, or a combine for harvesting, or may be a tractor for pulling work machines for harvesting, tilling, fertilizing, etc.

[0015] The onboard terminal 20 generates position information that measures the position of each of a plurality of points that the work device 2 passes through while working in the field. Position measurement may be performed using a GNSS (Global Navigation Satellite System) or a quantum compass that does not require an external auxiliary system such as an artificial satellite. The points where the position is measured are called positioning points. The position information includes positioning time information that indicates the positioning time when the position was measured at each positioning point. The onboard terminal 20 transmits the movement information to the work area estimation device 4.

[0016] The work area estimation device 4 estimates the work area in which the work device 2 has performed work, based on the position information received from the on-board terminal 20. The work area estimation device 4 further generates and outputs work area estimation information that represents the results of estimating the range of the work area. As an example, the work area information may be transmitted to the external terminal 5 via the network 3.

[0017] The external terminal 5 may include a smartphone, tablet terminal, personal computer, etc. that has communication functions, display functions, and input functions. The external terminal 5 notifies the user of the content of the work area estimation information received from the work area estimation device 4 by displaying it, etc.

[0018] 2, the working space estimation device 4 according to one embodiment may be configured as a so-called computer. In the example of FIG. 2, the working space 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 may be configured to realize communication between the calculation device 42, the storage device 43, the communication device 44, and the input / output device 45.

[0019] The calculation device 42 executes a work area estimation program according to one embodiment to implement the processing of a position information acquisition unit 421, an extraction unit 422, a connected graph generation unit 423, a candidate polygon generation unit 424, an estimation unit 425, and an output unit 426. The position information acquisition unit 421, the extraction unit 422, the connected graph generation unit 423, the candidate polygon generation unit 424, the estimation unit 425, and the output unit 426 are virtual functional blocks that execute processes realized by the calculation device 42 and the storage device 43 working together. The position information acquisition unit 421 acquires, from the onboard terminal 20, position information generated when the work device 2 performs work in the field 9. The extraction unit 422 extracts, from multiple points included in the position information, points that satisfy predetermined extraction conditions as extracted positioning points. The connected graph generation unit 423 generates a connected graph in which extracted positioning points that satisfy predetermined connection conditions are connected as connecting nodes. The candidate polygon generation unit 424 generates a candidate polygon by inserting points, excluding the extracted positioning points, that satisfy insertion conditions based on the connection graph into the connection graph. The estimation unit 425 estimates the work area based on the candidate polygon. The output unit 26 outputs work area estimation information indicating the estimated work area to the outside.

[0020] The storage device 43 includes a program storage unit 431 and a data storage unit 432. The working area estimation program may be read from an external recording medium 430 and stored in the program storage unit 431. The recording medium 430 may be a non-transitory and tangible medium. The data storage unit 432 may store location information acquired from the on-board terminal 20.

[0021] The communication device 44 communicates with external devices including the on-board terminal 20 and / or the external terminal 5 by wireless communication and / or wired communication via the network 3. The working area estimation program may be received by the communication device 44 from the outside and stored in the program storage unit 431.

[0022] 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 keyboard and / or a mouse that accepts input, etc.

[0023] 3, the external terminal 5 may be configured as a computer including a bus 51, a calculation device 52, a storage device 53, a communication device 54, and an input / output device 55, similar to the working area estimation device 4. The input / output device 55 may include a touch panel that integrates a display device that outputs images with a touchpad that accepts input by touch operation. The external terminal 5 realizes predetermined processing by the calculation device 52 executing a program stored in the storage device 53. The program may be stored in the storage device 53 from the outside via a recording medium 530 or the communication device 54.

[0024] An example of the configuration of a working area estimation method according to an embodiment will be described with reference to the flowchart in Figure 4. The processing of the working area estimation method may be started when the working area estimation device 4 is started. At this time, the processing of the working area estimation method is realized by the calculation device 42 of the working area estimation device 4 executing a working area estimation program.

[0025] When the processing of the flowchart in Fig. 4 starts, step S01 is executed. In step S01, the location information acquisition unit 421 in Fig. 2 receives and acquires location information transmitted from the mounted terminal 20 in Fig. 1. The acquired location information may be stored in the data storage unit 432 in Fig. 2.

[0026] As shown in the table of Fig. 5, 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 a number indicating 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 in which the positioning was performed may be calculated from the positioning time information.

[0027] After step S01 in FIG. 4, step S02 is executed. In step S02, the extraction unit 422 in FIG. 2 determines extraction conditions for extracting locations that satisfy the extraction conditions from among the multiple locations included in the location information. As an example, the extraction conditions are satisfied when the remainder of dividing a number representing the order in which the location of the target location was measured by a predetermined integer is equal to a predetermined value. In this case, the integer is referred to as the extraction divisor, and the value is referred to as the selection number. In this example, the extraction unit 422 determines the extraction divisor and selection number as the extraction conditions. The extraction unit 422 sets an integer between 0 and the extraction divisor minus 1 as the selection number. As an example, a different integer may be set as the selection number each time steps S03 to S09, described below, are repeatedly executed. Here, a case will be described in which the extraction divisor is 6, the selection number is set to an initial value of 0 when steps S03 to S09 are first executed, and the selection number is incremented by 1 each time steps S03 to S09 are repeatedly executed. However, the above extraction conditions are merely an example and do not limit the present embodiment.

[0028] After step S02 in Fig. 4, step S03 is executed. In step S03, the extraction unit 422 in Fig. 2 extracts points that satisfy the extraction conditions determined in step S02 in Fig. 4 from among the multiple points included in the location information. In other words, the extraction unit 422 in Fig. 2 thins out points that do not satisfy the extraction conditions determined in step S02 in Fig. 4 from among the multiple points included in the location information, and the remaining points are called extracted positioning points. By doing so, the processing load of the following steps S04 to S06 can be significantly reduced.

[0029] As an example, when the extraction divisor is 6 and the selection number is 0, the data of the extracted positioning point among the location information shown in the table of Figure 5 is the data of the point where the remainder when the number is divided by 6 is 0, as shown in the table of Figure 6.

[0030] After step S03 in Fig. 4, step S04 is executed. In step S04, the connected graph generation unit 423 in Fig. 2 calculates the distance and time interval between any two extracted positioning points that are consecutive in the order of position measurement among the extracted positioning points. Note that whether the order is consecutive or not is determined by skipping over the numbers of the thinned out points. As an example, in the case of Fig. 6, it is determined that the numbers "0" and "6" are consecutive, and the numbers "0" and "12" are not consecutive.

[0031] 2 calculates the distance between any two extracted positioning points that are consecutive in the order of positioning among the plurality of extracted positioning points, based on the position information of these two extracted positioning points. Also, the connected graph generation unit 423 in FIG. 2 calculates the time interval between two positioning times associated with these two extracted positioning points, respectively, based on these positioning times.

[0032] After step S04 in FIG. 4, step S05 is executed. In step S05, the connection graph generation unit 423 in FIG. 2 generates a connection graph that connects two extracted positioning points that satisfy a predetermined connection condition among the plurality of extracted positioning points. The connection condition includes a first condition related to a distance threshold and a second condition related to a time interval threshold. More specifically, the connection graph generation unit 423 performs a process of connecting the two extracted positioning points when the distance between the two extracted positioning points is shorter than a predetermined threshold distance and the time interval between two positioning times associated with the two extracted positioning points is shorter than a predetermined threshold time. The connection graph generation unit 423 performs this process for each combination of any two extracted positioning points included in the plurality of extracted positioning points, thereby generating graph information representing a connection graph in which some of the extracted positioning points are connected as nodes. Here, the graph information is data representing the connection 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 relationships between the nodes and edges.

[0033] Generation of a connection graph according to one embodiment will be described with reference to Figures 7 and 8. The first connection graph shown in the example of Figure 7 includes nodes N10, N11, N12, N13, N14, and N15. Hereinafter, when there is no need to distinguish between nodes N10 to N15, these will be collectively referred to as node N. In the example of Figure 7, these nodes N represent a plurality of extracted positioning points extracted in step S03 of Figure 4. The position of each node N in Figure 7 represents the position of the corresponding extracted positioning point, and the connection of these nodes N represents that the order of positioning at the corresponding two extracted positioning points is consecutive.

[0034] 7, node N10 and node N11 are connected, node N11 and node N12 are connected, node N12 and node N13 are connected, node N13 and node N14 are connected, and node N14 and node N15 are connected. These connections indicate that the extracted positioning points corresponding to nodes N10, N11, N12, N13, N14, and N15 are positioned in this order.

[0035] Boundary line C in Figure 7 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 this circle, are closer to node N10 than the radius of this circle. Conversely, node N13, which is outside this circle, is farther from node N10 than the radius of this circle.

[0036] Among the connections in Fig. 7, connections drawn with solid lines represent connections with positioning points where the positions were measured before a predetermined time has elapsed since the positioning time of node N10. Conversely, connections drawn with dashed dotted lines represent connections with positioning points where the positions were measured after a predetermined time has elapsed since the positioning time of node N10. In the example of Fig. 7, the positions of nodes N11, N12, N13, and N14 are measured before a predetermined time has elapsed since the positioning time of node N10. The position of the remaining node N15 is measured after a predetermined time has elapsed since the positioning time of node N10.

[0037] Focusing on node N10 in Figure 7, nodes N11, N12, and N14 satisfy the conditions that their distance from node N10 is shorter than the radius of the circle of boundary line C, and that their positioning time is shorter than a predetermined time from the positioning time of node N10, while the remaining nodes N13 and N15 do not. This is represented as in the second connected graph in Figure 8. In the example of Figure 8, node N10 of interest is connected to nodes N11, N12, and N14, which satisfy the above conditions, but is not connected to nodes N13 and N15, which do not satisfy the above conditions.

[0038] 7 and 8, nodes N11, N12, and N14 connected to node N10 in the second connected graph of Fig. 8 are drawn with solid lines. Conversely, nodes N13 and N15 not connected to node N10 in the second connected graph of Fig. 8 are drawn with dashed lines.

[0039] Next, generation of graph information of a connected graph according to one embodiment will be described in more detail with reference to Figures 9, 10, 11, and 12. In the example of Figure 9, a field 9 and extracted positioning points P0, P1, P2, P3, P4, P5, P6, P7, P8, P9, P10, P11, and P12 are depicted. Hereinafter, when the extracted positioning points P0 to P12 are not distinguished from one another, they will be collectively referred to as extracted positioning point P. Note that in Figure 9, the extracted positioning point P, which does not actually have an area, is depicted as a circle centered at the extracted positioning point P to make it easier to see. This also applies to Figures 10, 11, and 12, which will be described later. The extracted positioning point P0 is located outside the field 9, and the remaining extracted positioning points P1 to P12 are located inside the field 9. The arrows connecting the extracted positioning points P0 to P12 in this order indicate that the work device 2 measured its position while passing through the extracted positioning points P0 to P12 in this order. In the example of FIG. 9, the positioning at each extracted positioning point P is performed at a predetermined positioning interval, which is one minute. Also, in the example of FIG. 9, the work device 2 passes through the extracted positioning points P0 to P12 while moving at a constant speed, and the distance between two consecutive extracted positioning points P is five meters. However, these specific numerical values ​​are merely an example and do not limit the embodiment.

[0040] FIG. 10 shows an example of a connection graph generated when attention is first focused on extracted positioning point P0. Here, boundary line C0 is a circle centered on positioning point P0. The radius of this boundary line C0 is a threshold distance used to determine whether or not to connect a certain positioning point P to positioning point P0. In the example of FIG. 10, the threshold distance is 11 meters. As described above, in one embodiment, the distance-related condition included in the condition for connecting the extracted positioning point P0 of interest with a certain extracted positioning point P is that the distance from the extracted positioning point P0 to the certain extracted positioning point P is shorter than the threshold distance. In the example of FIG. 10, extracted positioning points P1, P2, and P8 satisfy the distance-related condition, while the remaining positioning points P3 to P7 and P9 to P12 do not satisfy the distance-related condition. Although boundary line C0 overlaps with the circles representing extracted positioning points P7 and P9, it does not reach the center points of these circles, so the distance from the extracted positioning point P0 of interest to extracted positioning point P7 or P9 is longer than the threshold distance, and therefore extracted positioning points P7 and P9 are determined not to satisfy the distance conditions.

[0041] As described above, in one embodiment, the condition regarding the time interval included in the condition for linking the extracted positioning point P0 of interest with a certain extracted positioning point P is that the time interval from the positioning time at which the working device 2 measures the position at the extracted positioning point P0 to the positioning time at that positioning point P is shorter than the threshold time. In the example of FIG. 10, the threshold time is 7 minutes and 30 seconds. In the example of FIG. 10, the positioning points P1 to P7 satisfy the condition regarding the time interval, while the remaining positioning points P8 to P12 do not satisfy the condition regarding the time interval.

[0042] As described above, in one embodiment, the conditions for connecting the extracted positioning point P0 of interest with a certain extracted positioning point P include a condition related to a threshold distance and a condition related to a threshold time. In the example of FIG. 10, the extracted positioning points P1 and P2 satisfy both conditions. In the example of FIG. 10, the edges of the second connection graph generated in step S05 of FIG. 4 are indicated by thick solid lines. More specifically, the second connection graph of FIG. 10 includes an edge connecting the extracted positioning point P0 with the extracted positioning point P1, and an edge connecting the extracted positioning point P0 with the extracted positioning point P2.

[0043] The remaining extracted positioning points P3 to P12 do not satisfy the condition related to the threshold distance and / or the condition related to the threshold time, and therefore are not connected to the extracted positioning point P0 in the second connection graph in Fig. 10. In Fig. 10, the extracted positioning points P3 to P12 that do not satisfy the condition related to the threshold distance and / or the condition related to the threshold time are indicated by dashed lines.

[0044] 11 shows an example of a third connected graph generated when first focusing on positioning point P0 and then on positioning point P1. The third connected graph shown in FIG. 11 includes edges represented by thick solid lines and edges represented by thin dashed lines. The edges represented by thick solid lines represent connections between extracted positioning point P1 and other extracted positioning points P. The edges represented by thin dashed lines are edges in the second connected graph shown in FIG. 10 that are not included in the third connected graph shown in FIG.

[0045] In the example of FIG. 11, boundary line C1 is a circle centered on extracted positioning point P1. The radius of this boundary line C1 is the threshold distance used in the connection condition. The threshold distance in the example of FIG. 11 is the threshold distance used in the process described with reference to FIG. 10, and is 11 meters, for example. In the example of FIG. 11, extracted positioning points P0, P2, P3, and P7 to P9 satisfy the condition related to the threshold distance, while the remaining extracted positioning points P4 to P6 and P10 to P12 do not satisfy the condition related to the threshold distance. Note that although boundary line C1 overlaps with the circles representing extracted positioning points P6 and P10, it does not reach the center points of these circles. Therefore, the distance from the extracted positioning point P1 of interest to extracted positioning point P6 or P10 is longer than the threshold distance, and therefore extracted positioning points P6 and P10 are determined not to satisfy the condition related to the threshold distance.

[0046] In the example of Fig. 11, the threshold time is the threshold time used in the process described with reference to Fig. 10, and is, for example, 7 minutes 30 seconds. In the example of Fig. 11, the extracted positioning points P0, and P2 to P8 satisfy the condition regarding the threshold time, and the remaining extracted positioning points P9 to P12 do not satisfy the condition regarding the threshold time.

[0047] In the example of Fig. 11, the extracted positioning points P0, P2, P3, P7, and P8 satisfy the conditions related to the threshold distance and the conditions related to the threshold time. In the example of Fig. 11, in the third connection graph generated in step S05 of Fig. 4, edges connecting the extracted positioning point P1 to the other extracted positioning points P are indicated by thick solid lines. More specifically, the third connection graph of Fig. 11 includes edges connecting the extracted positioning point P1 to the extracted positioning points P0, P2, P3, P7, and P8, respectively. The third connection graph of Fig. 11 further includes an edge connecting the extracted positioning point P0 to the extracted positioning point P2, which is an edge unrelated to the extracted positioning point P1, in the second connection graph of Fig. 10.

[0048] Thereafter, the extracted positioning points P2 to P12 are sequentially focused on, and a connection graph is generated by adding edges connecting the extracted positioning point P of interest with the extracted positioning point P that satisfies the condition related to the threshold distance and the condition related to the threshold time. Fig. 12 shows an example of a connection graph generated as a result of combining the connection graphs generated by focusing on the extracted positioning points P0 to P12, respectively.

[0049] After step S05 in Fig. 4, step S06 is executed. In step S06, the candidate polygon generation unit 424 of the working area estimation device 4 detects a community from the connected graph. A community is a subgraph of the connected graph in which nodes are more densely connected than 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 denser.

[0050] A method for detecting communities from a connected graph using a so-called network analysis technique will be described with reference to Figure 13. The connected graph shown in the example of Figure 13 includes nodes N21, N22, N23, N24, N25, and N26, and edges E1, E2, E3, E4, E5, E6, and E7. Hereinafter, when nodes N21 to N26 are not distinguished, they will be referred to as node N. Furthermore, when edges E1 to E7 are not distinguished, they will be referred to as edge E.

[0051] In the example of FIG. 13, 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.

[0052] First, for each combination of any two nodes included in the connected graph, the shortest path along the edges between these two nodes is calculated. In the example of FIG. 13, a method is known for determining the shortest path 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 from node N21 to node N22 is edge E1. Similarly, the shortest path 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. 13.

[0053] Next, for each edge included in the connected graph, the number of times it appears on the calculated shortest route is calculated. In the example of Figure 13, edge E1 appears only once on the shortest route between node N21 and node N22, so the number of times edge E1 appears on the shortest route is 1. Similarly, edge E7 appears on the shortest route is also 1. Furthermore, edge E2 appears a total of four times on the shortest routes between node N22 and each of nodes N23, N24, N25, and N26, so the number of times edge E2 appears on the shortest route is 4. Similarly, the number of times edges E3, E5, and E6 appear on the shortest route is 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.

[0054] Next, the edge that appears the most number of times on the shortest path is detected from among the edges included in the connected graph. In the example of Figure 13, the edge E that appears the most number of times on the shortest path is edge E4.

[0055] Next, the edge that appears most frequently on the shortest path is removed from the connected graph, thereby dividing the connected graph and detecting communities of nodes. In the example of Figure 13, by removing edge E4 that appears most frequently on the shortest path from the connected graph, a first community 8A including nodes N21, N22, and N23 and a second community 8B including nodes N24, N25, and N26 are detected.

[0056] If the connected graph is not split even after removing the edge that appears most frequently on the shortest path, the node community is detected by continuing to remove edges in descending order of their frequency of appearing on the shortest path until the connected graph is split.

[0057] After step S06 in FIG. 4, step S07 is executed. In step S07, the candidate polygon generation unit 424 in FIG. 2 inserts, as nodes in the connection graph, points that satisfy an insertion condition based on the connection graph, among points excluding the extracted positioning points. Here, the insertion condition based on the connection graph focuses on two extracted positioning points that are consecutive in the order of positioning and that are included in the community detected from the connection graph in step S06 in FIG. 4, among the extracted positioning points included in the connection graph. The point that was thinned out in step S03 in FIG. 4 from between the two extracted positioning points is inserted as a node in the connection graph between the two extracted positioning points. As a result, the inserted points are connected in series between the two extracted positioning points in the order of positioning.

[0058] An example of the processing of step S07 in Fig. 4 will be described with reference to Fig. 14 and Fig. 15. In this example, each of points D0 to D24 shown in Fig. 14 is positioned in the order of the numbers included in the symbols. In step S03 in Fig. 4, of points D0 to D24 in Fig. 14, points D0, D6, D12, D18, and D24 indicated by solid lines are extracted as extracted positioning points. In step S06 in Fig. 4, a community is detected that includes extracted positioning points D0, D6, D18, and D24 in Fig. 14 but does not include extracted positioning point D12. In this community, the extracted positioning points D0, D6, D18, and D24 are connected to each other as nodes in a connected graph. 14, the candidate polygon generation unit 424 of FIG. 2 determines in step S07 of FIG. 4 that points D1 to D5 satisfy the insertion condition, while focusing on the extracted positioning points D0 and D6 of FIG. 14. As a result, points D1 to D5 are inserted as nodes of the connection graph between the extracted positioning points D0 and D6 in the order of the numbers included in their respective codes, and are connected. Similarly, the candidate polygon generation unit 424 of FIG. 2 determines that points D19 to D23 satisfy the insertion condition, while focusing on the extracted positioning points D18 and D24 of FIG. 14. As a result, points D19 to D23 are inserted as nodes of the connection graph between the extracted positioning points D18 and D24 in the order of the numbers included in their respective codes, and are connected. Conversely, extracted positioning point D12, points D7 to D11, and D13 to 17, which are not included in the community, are not inserted into the connection graph. As a result, the connection graph shown in Fig. 15 is generated. In Fig. 15, extracted positioning point D0, points D1 to D5, extracted positioning points D6 and D18, points D19 to D23, and extracted positioning point D24 are connected in series in the order of the numbers included in the codes.

[0059] Step S08 is executed after step S07 in FIG. 4. In step S08, the candidate polygon generation unit 424 in FIG. 2 generates a candidate polygon as a candidate for the work area where the work device 2 performed the work, based on the connection graph into which the points satisfying the insertion condition have been inserted in step S07 of the community detected in step S06 of FIG. 4. More specifically, the candidate polygon generation unit 424 in FIG. 2 calculates the convex hull or concave hull of the points corresponding to the nodes included in the connection graph obtained in step S07 of FIG. 4 based on positional information of these points, thereby estimating the position and shape of the work area corresponding to each community as a polygon. The candidate polygon generated in this manner approximates the actual work area with higher accuracy than a polygon generated based on the connection graph before the points were inserted, due to the insertion of the points satisfying the insertion condition into the connection graph in step S07 of FIG. 4.

[0060] As an example, a case will be described where the position information shown in FIG. 16 has been acquired. In FIG. 16, multiple dots represent multiple points where the position of the work device 2 has been measured, and multiple straight lines represent the movement trajectory of the work device 2 moving without performing work and the operating trajectory of the work device 2 moving while performing farm work. Steps S02 to S04 of FIG. 4 are executed based on the position information in FIG. 16, and communities of points shown in FIGS. 17, 18, 19, and 20 are detected. Then, by calculating the convex hull of the points included in each detected community, the position and shape of work areas 81, 82, 83, and 84 corresponding to each community are estimated, as shown in FIG. 21.

[0061] After step S08 in FIG. 4, step S09 is executed. In step S09, the extraction unit 422 in FIG. 2 determines whether the extracted positioning point extracted in the last executed step S03 in FIG. 4 is the last part of the points included in the location information. If the result of the determination is "YES", the process proceeds to step S10 in FIG. 4. If the result of the determination is "NO", the process returns to step S03 in FIG. 4, and steps S03 to S09 are repeated. Here, before the process returns to step S03 in FIG. 4, the extraction unit 422 in FIG. 2 sets the next part of the points included in the location information as the extracted positioning point.

[0062] In the example used in the explanation of step S02 in Fig. 4 above, when the current selection number is equal to the extraction divisor - 1, the result of the determination is "YES." When the current selection number is within the range from 0 to the extraction divisor - 2, the result of the determination is "NO." Here, the extraction unit 422 in Fig. 2 increments the selection number before the process returns to step S03 in Fig. 4.

[0063] As a more detailed example of the process of step S09 in FIG. 4, a case where the extraction divisor is 6 and the current selection number is 0 will be described. The result of the determination in step S09 is "NO," so the extraction unit 422 in FIG. 2 increments the selection number from 0 to 1, and the process returns to step S03. As a result, in step S03, which is executed for the second time, the extraction unit 422 in FIG. 2 extracts, as extracted positioning points, points whose remainder when the number is divided by the extraction divisor 6 is 1, as shown in the table in FIG. 22, from the location information shown in the table in FIG. 5. Thereafter, in step S07, which is executed for the second time, points that satisfy the insertion condition based on the connection graph shown in FIG. 23 are determined, instead of FIG. 14, which was executed for the first time. As shown in FIG. 23, when only the extracted positioning points D1 and D19 are included in the connection graph among the extracted positioning points D1, D7, D13, and D19 shown in the table in FIG. 22, no points satisfy the insertion condition based on the connection graph exist. In this case, the connection graph obtained as a result of the processing in step S07 in FIG. 4 is the same as the connection graph in FIG.

[0064] Step S08 in FIG. 4 is repeatedly executed a number of times equal to the extraction divisor, thereby generating a total number of candidate polygons equal to the extraction divisor. The multiple candidate polygons generated in this manner are each generated based on a different portion of the points included in the location information, and therefore may have different accuracy relative to the actual work area. The result of the determination in step S09 in FIG. 4 is then "YES," and the process proceeds to step S10. In step S10, the estimation unit 425 in FIG. 2 estimates the work area based on the multiple candidate polygons generated in step S08 in FIG. 4. More specifically, the estimation unit 425 in FIG. 2 selects, from the multiple candidate polygons, the candidate polygon that most closely matches the bitmap image of the corresponding extracted positioning point as the estimated work area.

[0065] An example of a method for calculating the degree of coincidence will be described. First, the estimation unit 425 in Fig. 2 generates, for each of a plurality of candidate polygons, a first bitmap image which is a collection of rectangles including extracted positioning points corresponding to the candidate polygons, out of a plurality of rectangles obtained by dividing a plane including the range over which the operating device 2 has moved into a vertical and horizontal mesh. Next, the estimation unit 425 generates, for each of a plurality of candidate polygons, a second bitmap image which is a collection of rectangles including the candidate polygons, out of a plurality of rectangles obtained by dividing a plane including the range over which the operating device 2 has moved into a vertical and horizontal mesh. At this time, the mesh division method is the same in the first bitmap image and the second bitmap image.

[0066] Next, the estimation unit 425 in FIG. 2 calculates, for each of the multiple candidate polygons, the ratio of the total number of rectangles included in both the corresponding first bitmap image and the second bitmap image divided by the total number of rectangles included in at least one of the first bitmap image or the second bitmap image as the degree of similarity between the first bitmap image and the second bitmap image.

[0067] A more detailed example of the processing of step S10 in Fig. 4 will be described with reference to Fig. 24, Fig. 25, Fig. 26, Fig. 27, Fig. 28, and Fig. 29. Fig. 24 shows an example of a first candidate polygon, and Fig. 25 shows an example of a bitmap image of extracted positioning points corresponding to the candidate polygon in Fig. 24. Fig. 26 shows an example of a first candidate polygon, and Fig. 27 shows an example of a bitmap image of extracted positioning points corresponding to the candidate polygon in Fig. 26. Fig. 28 shows an example of a first candidate polygon, and Fig. 29 shows an example of a bitmap image of extracted positioning points corresponding to the candidate polygon in Fig. 28. In step S08 of FIG. 4, the candidate polygons of FIG. 24, FIG. 26, and FIG. 28 are generated, and when the matching rate of FIG. 24 and FIG. 25 is approximately 75%, the matching rate of FIG. 26 and FIG. 27 is approximately 80%, and the matching rate of FIG. 28 and FIG. 29 is approximately 90%, the estimation unit 425 of FIG. 2 adopts the candidate polygon of FIG. 28, which has the highest matching rate with the bitmap image of the extracted positioning point, as the estimation result of the working area.

[0068] After step S10 in FIG. 4, step S11 is executed. In step S11, the output unit 426 in FIG. 2 outputs work area information representing the estimated work area. More specifically, the output unit 426 controls the communication device 44 to output the work area information representing the position and shape of the estimated work area to the outside. As an example, the output unit 426 transmits the work area estimation information to the external terminal 5, and the external terminal 5 may display the estimated shape of the work area represented by the received work area estimation information on a display device included in the input / output device 55 in FIG. 3. At this time, the external terminal 5 may superimpose the estimated shape of the work area on map information of the area including the work area to enable the user to intuitively understand. Furthermore, the positioning point P included in the work area may be superimposed on the estimated shape and map information.

[0069] After step S11, the process of the flowchart in FIG. 4 ends.

[0070] As described above, according to one embodiment, when the working device 2 estimates a work area based on position information measured at multiple points, a connection graph is generated based on partial position information after some of the points have been thinned out, and then candidate polygons for the work area are generated based on the connection graph with some of the points inserted. Furthermore, from among the multiple candidate polygons generated corresponding to multiple conditions for thinning out the points, the candidate polygon that has the highest degree of match with the bitmap image of the corresponding extracted positioning point is selected as the estimated result for the work area. As a result, the processing load can be reduced while the work area can be estimated with high accuracy.

[0071] (Addendum) The working area estimation method, working area estimation system 1, and working area estimation program described in each embodiment can be described as follows.

[0072] The working space estimation method according to the first aspect includes: acquiring position information obtained by measuring the position at each of a plurality of points passed by a work implement that performs agricultural work while moving within a farm field; extracting, from the plurality of points included in the position information, points that satisfy a first extraction condition based on the order in which the positions are measured, as first extracted positioning points; generating a first connection graph in which first extracted positioning points that satisfy a first connection condition are connected as first connection nodes among the plurality of first extracted positioning points; inserting a point that satisfies a first insertion condition based on the first connectivity graph as a first insertion point into the first connectivity graph among the plurality of points excluding the plurality of first extracted positioning points, to generate a first candidate polygon; estimating a work area in the field where the work implement performed the farm work based on the first candidate polygon; outputting work area estimation information indicating the estimated work area to the outside; Includes.

[0073] A working area estimation method according to a second aspect is the working area estimation method according to the first aspect, extracting, from the plurality of points included in the position information, points that satisfy a second extraction condition based on the order in which the positions are measured, as second extracted positioning points; generating a second connection graph in which second extracted positioning points that satisfy the first connection condition are connected as second connection nodes among the plurality of second extracted positioning points; inserting a point that satisfies a second insertion condition based on the second connectivity graph as a second insertion point into the second connectivity graph among the plurality of points excluding the plurality of second extracted positioning points, to generate a second candidate polygon; further comprising Inferring the working space comprises: and estimating a work area in the farm field where the work implement performed the farm work, further based on the second candidate polygon. Including, the first extraction condition is satisfied when a remainder obtained by dividing a number representing the order in which the position of the point was measured by a predetermined integer is equal to a first value; The second extraction condition is satisfied when the remainder obtained by dividing the number representing the order in which the position of the point was measured by a predetermined integer is equal to a second value different from the first value.

[0074] A working area estimation method according to a third aspect is the working area estimation method according to the second aspect, Inferring the working space comprises: generating a first figure representing an arrangement of the first extracted positioning points; calculating a first degree of match that represents a degree of match between the first figure and the first candidate polygon; generating a second figure representing an arrangement of the second extracted positioning points; calculating a second degree of match representing a degree of match between the second figure and the second candidate polygon; comparing the first degree of match with the second degree of match, and selecting the first candidate polygon as the working area when the first degree of match is greater, and selecting the second candidate polygon as the working area when the second degree of match is greater; Further includes:

[0075] A working area estimation method according to a fourth aspect is the working area estimation method according to the third aspect, Generating the first figure includes: generating, as the first figure, a bitmap image that is a set of rectangles that include the first extracted positioning points among a plurality of rectangles obtained by dividing the farm field into a predetermined mesh; Including, Calculating the first degree of coincidence includes: generating, as a third figure, a bitmap image that is a set of rectangles that include the first candidate polygon among the plurality of rectangles; calculating, as the first degree of coincidence, a ratio obtained by dividing the total number of rectangles included in both the first graphic and the third graphic among the plurality of rectangles by the total number of rectangles included in at least one of the first graphic or the third graphic; Including, generating the second figure generating, as the second figure, a second bitmap image that is a collection of rectangles that include the second extracted positioning points among the plurality of rectangles; Including, Calculating the second degree of coincidence includes: generating, as a fourth figure, a bitmap image that is a set of rectangles that include the second candidate polygon among the plurality of rectangles; calculating, as the second degree of coincidence, a ratio obtained by dividing the total number of rectangles included in both the second graphic and the fourth graphic among the plurality of rectangles by the total number of rectangles included in at least one of the second graphic or the fourth graphic; Includes.

[0076] A working area estimation method according to a fifth aspect is the working area estimation method according to any one of the first to fourth aspects, generating the first connected graph Calculating a distance between any two first extracted positioning points that are consecutive in the order of measurement of the positions among the plurality of first extracted positioning points, and a time interval between the positioning times at the two first extracted positioning points; Connecting any two first extracted positioning points among the plurality of first extracted positioning points as two first connected nodes when the distance between the two first extracted positioning points is shorter than a predetermined threshold distance and the time interval between the two first extracted positioning points is shorter than a predetermined threshold time; Including, generating the first candidate polygon detecting a first sub-community, which is a subgraph of the first connected graph, the subgraph having a higher density than another subgraph of the first connected graph; extracting, from the plurality of first extracted positioning points, first extracted positioning points included in the first partial community as first partial positioning points, extracting, from the plurality of points included in the location information, points that satisfy a first insertion condition based on the first partial positioning points as the first insertion points, and linking the first insertion points to nodes included in the first partial community; the first partial community inserted with the first insertion point as a node is set as the first candidate polygon; Includes.

[0077] A working area estimation method according to a sixth aspect is the working area estimation method according to the fifth aspect, Detecting the first sub-community includes: removing edges of the first connected graph in descending order of frequency of appearance on the shortest path between any two first extracted positioning points among the plurality of first extracted positioning points until the subgraph is separated from the first connected graph; Includes.

[0078] A working area estimation method according to a seventh aspect is the working area estimation method according to the fifth aspect, Detecting the first sub-community includes: excluding nodes in the first connected graph whose total number of edges is less than a predetermined threshold value; Includes.

[0079] A working area estimation method according to an eighth aspect is a working area estimation method according to any one of the fifth to seventh aspects, The first connected graph includes, as nodes, points that the working device passes through from the start to the end of its operation.

[0080] A working area estimation method according to a ninth aspect is the working area estimation method according to any one of the first to eighth aspects, The outputting step includes: Displaying the estimated shape of the work area superimposed on map information Includes.

[0081] A working space estimation system according to a tenth aspect includes: a position information acquisition unit that acquires position information obtained by measuring the position of a work device that moves within a farm field while performing agricultural work at each of a plurality of points through which the work device passes; an extraction unit that extracts, from the plurality of points included in the position information, points that satisfy a first extraction condition based on the order in which the positions are measured, as first extracted positioning points; a connection graph generation unit that generates a first connection graph in which first extracted positioning points that satisfy a first connection condition are connected as first connection nodes among the plurality of first extracted positioning points; a candidate polygon generating unit that generates a first candidate polygon by inserting a point that satisfies a first insertion condition based on the first connection graph as a first insertion point into the first connection graph among the plurality of points excluding the plurality of first extracted positioning points; an estimation unit that estimates a work area in the field where the work implement performed the farm work based on the first candidate polygon; an output unit that outputs work area estimation information indicating the estimated work area to the outside; Equipped with.

[0082] A working space estimation program according to an eleventh aspect includes: A work area estimation program for realizing a predetermined process by executing the program, The process comprises: acquiring position information obtained by measuring the position at each of a plurality of points passed by a work implement that performs agricultural work while moving within a farm field; extracting, from the plurality of points included in the position information, points that satisfy a first extraction condition based on the order in which the positions are measured, as first extracted positioning points; generating a first connection graph in which first extracted positioning points that satisfy a first connection condition are connected as first connection nodes among the plurality of first extracted positioning points; inserting a point that satisfies a first insertion condition based on the first connectivity graph as a first insertion point into the first connectivity graph among the plurality of points excluding the plurality of first extracted positioning points, to generate a first candidate polygon; estimating a work area in the field where the work implement performed the farm work based on the first candidate polygon; outputting work area estimation information indicating the estimated work area to the outside; Includes.

[0083] The invention made by the inventor has been specifically described above based on the embodiments, but it goes without saying that the present invention is not limited to the embodiments and can be modified in various ways without departing from the spirit of the invention. Furthermore, the features described in the embodiments can be freely combined within the scope of technical compatibility. [Explanation of symbols]

[0084] 1. Work Area Prediction System 2. Work equipment 20 Devices 3 Network 4 Work area estimation device 41 Bus 42 Arithmetic unit 421 Location information acquisition unit 422 Extraction part 423 Connected Graph Generation Unit 424 Candidate polygon generation unit 425 Guessing part 426 Output Section 43 Storage device 430 Recording Media 431 Program Memory Unit 432 Data storage unit 44 Communication equipment 45 Input / Output Devices 5 External terminals 51 Bus 52 Arithmetic unit 53 Recording Device 530 Storage medium 54 Communication equipment 55 Input / Output Devices 8A and 8B communities 81, 82, 83, 84 work area Fields 9, 9A, and 9B C, C0, C1 border D0, D1, D2, D3, D4, D5, D6, D7, D8, D9, D10, D11, D12, D13, D14, D15, D16, D17, D18, D19, D20, D21, D22, D23, D24 points, extracted positioning points E1, E2, E3, E4, E5, E6, E7 edges N10, N11, N12, N13, N14, and N15 nodes N21, N22, N23, N24, N25, and N26 nodes P0, P1, P2, P3, P4, P5, P6, P7, P8, P9, P10, P11, P12 Positioning points, extracted positioning points

Claims

1. acquiring position information obtained by measuring the position at each of a plurality of points passed by a work implement that performs agricultural work while moving within a farm field; extracting, from the plurality of points included in the position information, points that satisfy a first extraction condition based on the order in which the positions are measured, as first extracted positioning points; generating a first connection graph in which first extracted positioning points that satisfy a first connection condition are connected as first connection nodes among the plurality of first extracted positioning points; inserting a point that satisfies a first insertion condition based on the first connection graph as a first insertion point into the first connection graph from among the plurality of points excluding the plurality of first extracted positioning points, to generate a first candidate polygon; estimating a work area in the field where the work implement performed the farm work based on the first candidate polygon; outputting work area estimation information indicating the estimated work area to the outside; Contains Work area estimation method.

2. 2. The working area estimation method according to claim 1, extracting, from the plurality of points included in the position information, points that satisfy a second extraction condition based on the order in which the positions are measured, as second extracted positioning points; generating a second connection graph in which second extracted positioning points that satisfy the first connection condition are connected as second connection nodes among the plurality of second extracted positioning points; inserting a point that satisfies a second insertion condition based on the second connection graph as a second insertion point into the second connection graph from among the plurality of points excluding the plurality of second extracted positioning points to generate a second candidate polygon; further comprising Inferring the working space comprises: and estimating a work area in the farm field where the work implement performed the farm work, further based on the second candidate polygon. Including, the first extraction condition is satisfied when a remainder obtained by dividing a number representing the order in which the position of the point was measured by a predetermined integer is equal to a first value; The second extraction condition is satisfied when a remainder obtained by dividing a number representing the order in which the position of the point was measured by a predetermined integer is equal to a second value different from the first value. Work area estimation method.

3. 3. The working area estimation method according to claim 2, Inferring the working space comprises: generating a first figure representing an arrangement of the first extracted positioning points; calculating a first degree of match representing a degree of match between the first figure and the first candidate polygon; generating a second figure representing an arrangement of the second extracted positioning points; calculating a second degree of match representing a degree of match between the second figure and the second candidate polygon; comparing the first degree of match with the second degree of match, and selecting the first candidate polygon as the working area when the first degree of match is greater, and selecting the second candidate polygon as the working area when the second degree of match is greater; Also includes Work area estimation method.

4. 4. The working area estimation method according to claim 3, Generating the first figure includes: generating, as the first figure, a bitmap image that is a collection of rectangles that include the first extracted positioning points among a plurality of rectangles obtained by dividing the farm field into a predetermined mesh; Including, Calculating the first degree of coincidence includes: generating, as a third figure, a bitmap image that is a collection of rectangles that include the first candidate polygon among the plurality of rectangles; calculating, as the first degree of coincidence, a ratio obtained by dividing the total number of rectangles included in both the first graphic and the third graphic among the plurality of rectangles by the total number of rectangles included in at least one of the first graphic or the third graphic; Including, Generating the second figure includes: generating, as the second figure, a second bitmap image that is a collection of rectangles that include the second extracted positioning points among the plurality of rectangles; Including, Calculating the second degree of coincidence generating, as a fourth figure, a bitmap image that is a set of rectangles that include the second candidate polygon among the plurality of rectangles; calculating, as the second degree of coincidence, a ratio obtained by dividing the total number of rectangles included in both the second graphic and the fourth graphic among the plurality of rectangles by the total number of rectangles included in at least one of the second graphic and the fourth graphic; Contains Work area estimation method.

5. The working area estimation method according to any one of claims 1 to 4, generating the first connected graph Calculating a distance between any two first extracted positioning points that are consecutive in the order of measurement of the positions among the plurality of first extracted positioning points, and a time interval between the positioning times at the two first extracted positioning points; connecting any two first extracted positioning points among the plurality of first extracted positioning points as two first connected nodes when the distance between the two first extracted positioning points is shorter than a predetermined threshold distance and the time interval between the two first extracted positioning points is shorter than a predetermined threshold time; Including, Generating the first candidate polygon includes: detecting a first sub-community, which is a subgraph of the first connected graph, the subgraph having a higher density than another subgraph of the first connected graph; extracting, from the plurality of first extracted positioning points, first extracted positioning points included in the first partial community as first partial positioning points, extracting, from the plurality of points included in the location information, points that satisfy a first insertion condition based on the first partial positioning points as the first insertion points, and linking the first insertion points to nodes included in the first partial community; the first partial community inserted with the first insertion point as a node is set as the first candidate polygon; Contains Work area estimation method.

6. 6. The working area estimation method according to claim 5, Detecting the first sub-community includes: removing edges of the first connected graph in descending order of frequency of appearance on the shortest path between any two first extracted positioning points among the plurality of first extracted positioning points until the subgraph is separated from the first connected graph; Contains Work area estimation method.

7. 6. The working area estimation method according to claim 5, Detecting the first sub-community includes: excluding nodes of the first connected graph whose total number of edges is less than a predetermined threshold value; Contains Work area estimation method.

8. 6. The working area estimation method according to claim 5, The first connected graph includes, as nodes, points that the working device passes through from the start to the end of its operation. Work area estimation method.

9. 2. The working area estimation method according to claim 1, The outputting step includes: Displaying the estimated shape of the work area superimposed on map information Contains Work area estimation method.

10. a position information acquisition unit that acquires position information obtained by measuring the position of a work device that moves within a farm field while performing agricultural work at each of a plurality of points through which the work device passes; an extraction unit that extracts, from the plurality of points included in the position information, points that satisfy a first extraction condition based on the order in which the positions are measured, as first extracted positioning points; a connection graph generation unit that generates a first connection graph in which first extracted positioning points that satisfy a first connection condition are connected as first connection nodes among the plurality of first extracted positioning points; a candidate polygon generating unit that generates a first candidate polygon by inserting a point that satisfies a first insertion condition based on the first connection graph as a first insertion point into the first connection graph among the plurality of points excluding the plurality of first extracted positioning points; an estimation unit that estimates a work area in the field where the work implement performed the farm work based on the first candidate polygon; an output unit that outputs work area estimation information indicating the estimated work area to the outside; Equipped with Work area estimation system.

11. A work area estimation program for realizing a predetermined process by executing the program, The process comprises: acquiring position information obtained by measuring the position at each of a plurality of points passed by a work implement that performs agricultural work while moving within a farm field; extracting, from the plurality of points included in the position information, points that satisfy a first extraction condition based on the order in which the positions are measured, as first extracted positioning points; generating a first connection graph in which first extracted positioning points that satisfy a first connection condition are connected as first connection nodes among the plurality of first extracted positioning points; inserting a point that satisfies a first insertion condition based on the first connection graph as a first insertion point into the first connection graph from among the plurality of points excluding the plurality of first extracted positioning points, to generate a first candidate polygon; estimating a work area in the field where the work implement performed the farm work based on the first candidate polygon; outputting work area estimation information indicating the estimated work area to the outside; Contains Work area estimation program.

Images