Work site identification system and work site identification method

The worksite identification system uses transport vehicle data to automatically identify and update mine worksites through clustering and convex hull analysis, addressing the inefficiencies of manual interviews and site changes.

WO2026234876A1PCT designated stage Publication Date: 2026-11-12KOMATSU LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
PCT/JP2026/005352
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2025-05-09
Filing Date
2026-02-13
Publication Date
2026-11-12

AI Technical Summary

Technical Problem

Current methods for identifying worksites in mines rely on interviews with managers, which are outdated and require frequent updates due to changes in loading and dumping sites, and there is a need for a more automated and accurate system to identify these sites using transport vehicle data.

Method used

A worksite identification system and method that extracts data from transport vehicle position data to identify clusters of work positions as worksites using clustering algorithms and convex hulls, assigning IDs to these clusters, and updating them based on location data over time.

Benefits of technology

Automatically identifies and updates worksites in mines by analyzing vehicle movement data, providing accurate and dynamic information on loading and soil disposal areas, and calculating cargo transport volumes.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure JP2026005352_12112026_PF_FP_ABST
    Figure JP2026005352_12112026_PF_FP_ABST
Patent Text Reader

Abstract

This identification system for identifying a work site in a mine comprises an extraction unit and an identification unit. The extraction unit extracts data for multiple work locations from location data of multiple transport vehicles. The identification unit identifies, as a work site from among the multiple work locations, a cluster of work locations where the number of work locations in a specific range is a prescribed number or more.
Need to check novelty before this filing date? Find Prior Art

Description

Worksite identification system and worksite identification method

[0001] The present disclosure relates to a worksite identification system and a worksite identification method.

[0002] In mines, mine managers plan the number of units allocated to each of the resource mining location, the resource transport destination, and the stripped topsoil disposal site, and construction machines operate based on these plans.

[0003] For example, Patent Document 1 discloses identifying the route traveled by a mining machine from the time it moves from a dumping site to a loading site until it moves back to the dumping site.

[0004] Japanese Unexamined Patent Publication No. 2013-105278

[0005] However, since the number and scope of worksites such as loading sites and dumping sites are determined by each mine, currently the only means to obtain data on loading sites and dumping sites is through interviews with managers. In addition, loading sites and dumping sites are modified or moved as a result of expanded excavation or landfilling with topsoil, so the data obtained through interviews needs to be updated each time.

[0006] An object of the present disclosure is to provide a worksite identification system and identification method capable of identifying worksites in a mine from position data of transport vehicles.

[0007] A worksite identification system for identifying worksites in a mine according to one aspect of the present disclosure includes an extraction unit and an identification unit. The extraction unit extracts data of a plurality of work positions from position data of a plurality of transport vehicles. The identification unit identifies, for the plurality of work positions, a cluster of work positions in which the number of work positions within a specific range is equal to or greater than a predetermined number as a worksite.

[0008] A worksite identification method for identifying worksites in a mine according to another aspect of the present disclosure includes an extraction step and an identification step. In the extraction step, data of a plurality of work positions is extracted from position data of a plurality of transport vehicles. In the identification step, for the plurality of work positions, a cluster of work positions in which the number of work positions within a specific range is equal to or greater than a predetermined number is identified as a worksite.

[0009] This disclosure provides a work area identification system and a work area identification method that can identify work areas in a mine from the location data of transport vehicles.

[0010] This is a diagram showing the configuration of a transport vehicle management system in a mine. This is a block diagram showing a transport vehicle and a data server. This is a block diagram showing the functions of a data server and a specific system. This is a schematic diagram showing the movement of a transport vehicle between the excavation location and the loading location. This is a diagram showing the vehicle status data at a given time. This is a diagram to explain the extracted location data. This is a diagram to explain the extracted location data. This is a diagram to explain clustering. (a) This is a diagram showing the loading location. (b) This is a diagram showing the loading location and the loading travel location. (a) This is a diagram showing the loading site as a result of clustering the loading location and the loading travel location. (b) This is a diagram showing an example of a loading site database. (a) This is a diagram showing the convex hull of the locations within the identified cluster. (b) This is a diagram showing the loading site and its convex hull. (a) This is a diagram showing an example of a loading site database. (b) This is a diagram showing the location data of the cluster. (c) This is a diagram showing the setting of the cluster ID. (d) This is a diagram showing the loading site database with IDs assigned to the clusters and saved. This is a diagram of a loading site. (a) This is a diagram showing the convex hull of the locations within the identified cluster. (b) This is a diagram showing the loading site and its convex hull. (a) A diagram showing an example of a loading area database. (b) A diagram showing cluster location data. (c) A diagram showing the setting of cluster IDs. (d) A diagram showing a loading area database with IDs assigned to clusters and saved. A diagram of a loading area. A flowchart explaining the operation of a specific system. A flowchart explaining the operation of a specific system. A flowchart explaining the operation of a specific system. A flowchart explaining the operation of a specific system. A flowchart explaining the operation of a specific system. A flowchart explaining the operation of a specific system. A diagram explaining the operation of a specific system. A diagram illustrating variations of the extracted location data.

[0011] The following describes an embodiment of a transport vehicle management system in a mine with reference to the drawings. Figure 1 is a diagram showing the configuration of a transport vehicle management system 1 in a mine. As shown in Figure 1, the transport vehicle management system 1 manages multiple transport vehicles 2 operating in the mine. The management system 1 includes multiple transport vehicles 2, a specific system 3 for identifying work sites, and a data server 4.

[0012] Each transport vehicle 2 transmits its own movement data as a movement data signal S1 to the data server 4. The data server 4 stores the received movement data. The identification system 3 identifies the work site based on the location data contained in the multiple movement data stored in the data server 4. The work site includes at least one of a loading area for loading cargo and a soil removal area for soil removal.

[0013] Transport vehicle 2 is, for example, a dump truck. Transport vehicle 2 loads cargo at the loading area in the mine and discharges the soil at the soil discharge area. Each transport vehicle 2 includes a vehicle body 11 and a vessel 12. The vehicle body 11 includes multiple wheels 13. The vessel 12 is in a horizontal position when cargo is loaded by a hydraulic excavator or the like, and when the loaded cargo is discharged, the front part is raised and it becomes inclined.

[0014] Figure 2A is a block diagram of the transport vehicle 2 and the data server 4. Figure 2B is a block diagram of the data server 4 and the specific system 3. The transport vehicle 2 includes a vessel sensor 14, a load sensor 15, a travel sensor 16, a vehicle position sensor 17, a vehicle controller 18, and a communication device 19.

[0015] The vessel sensor 14 detects whether the vessel 12 is in an inclined state or a horizontal state. The vessel sensor 14 may also be a sensor that detects the extension and retraction of the hydraulic cylinder that inclins the vessel 12. The vessel sensor 14 may also be an angle sensor that detects the inclination of the vessel. The vessel sensor 14 outputs vessel state data to the vehicle controller 18 indicating whether the vessel is in an inclined state or a horizontal state.

[0016] The load sensor 15 detects the load caused by the cargo loaded on the vessel 12. The load sensor 15 may be, for example, a load cell. The load sensor 15 outputs the load data to the vehicle controller 18.

[0017] The driving sensor 16 detects distance-related data for calculating the distance traveled. The driving sensor 16 may, for example, detect the rotation speed of the wheels 13. The driving sensor 16 may, for example, detect the speed of the vehicle body 11. The driving sensor 16 outputs the distance-related data to the vehicle controller 18.

[0018] The vehicle position sensor 17 detects the position data of the transport vehicle 2. The vehicle position sensor 17 may include a GNSS (Global Navigation Satellite System) receiver such as a GPS (Global Positioning System). Alternatively, the vehicle position sensor 17 may include a receiver of another positioning system. The vehicle position sensor 17 may also include a distance measuring sensor such as a LiDar, or an image sensor such as a stereo camera. The vehicle position sensor 17 outputs the position data to the vehicle controller 18. The position data indicates the current position of the transport vehicle 2.

[0019] The vehicle controller 18 creates a movement data signal S1 based on vessel state data, load data, mileage-related data, and position data. The vehicle controller 18 includes a processor 18a such as a CPU and a storage device 18b. The processor 18a performs the process of creating the movement data signal S1 based on vessel state data, load data, mileage-related data, and position data. The storage device 18b includes memory such as RAM or ROM, and an auxiliary storage device such as an HDD (Hard Disk Drive) or SSD (SolID State Drive). The storage device 18b stores data and programs for creating the movement data signal S1.

[0020] The movement data signal S1 includes time data, position data at that time, load data at that time, and vehicle status data at that time. The vehicle status data includes status data for empty vehicle stopped, empty vehicle driving, loading in progress, loaded vehicle driving, loaded vehicle stopped, and soil discharge.

[0021] The vehicle controller 18 determines the state of the vehicle from the vessel state data, load data, mileage-related data, and position data. If the vehicle controller 18 detects from the vessel state data that the vessel 12 is in a horizontal state, from the load data that no load is loaded, and from the mileage-related data that the vehicle body 11 is stopped, it determines that the vehicle is in an empty-vehicle-stopped state. If the vehicle controller 18 detects from the vessel state data that the vessel 12 is in a horizontal state, from the load data that no load is loaded, and from the mileage-related data that the vehicle body 11 is moving, it determines that the vehicle is in an empty-vehicle-driving state. If the vehicle controller 18 detects from the vessel state data that the vessel 12 is in a horizontal state, from the load data that the load is increasing, and from the mileage-related data that the vehicle body 11 is stopped, it determines that the vehicle is in a loading state. The vehicle controller 18 determines that the vehicle is in a loaded and traveling state if it detects from the vessel state data that the vessel 12 is in a horizontal state, from the load data that a load is loaded, and from the mileage-related data that the vehicle body 11 is traveling. The vehicle controller 18 determines that the vehicle is in a loaded and stopped state if it detects from the vessel state data that the vessel 12 is in a horizontal state, from the load data that a load is loaded, and from the mileage-related data that the vehicle body 11 is stopped. The vehicle controller 18 determines that the vehicle is in a soil-discarding state if it detects from the vessel state data that the vessel 12 is in an inclined state and from the mileage-related data that the vehicle body 11 is stopped.

[0022] The movement data signal S1 of the transport vehicle 2 is a signal that defines one cycle as the transition from the soil-discarding state to the loading state and then back to the soil-discarding state.

[0023] Figures 3(a) and 3(b) are schematic diagrams showing the movement of the transport vehicle 2 between the soil removal position Dp and the loading position Lp. Figure 3(a) shows the transport vehicle 2 removing soil at soil removal position Dp, loading cargo at the loading position Lp, and then heading back to the same soil removal position Dp. Figure 3(b) shows the transport vehicle 2 removing soil at soil removal position Dp1, loading cargo at the loading position Lp, and then heading to a different soil removal position Dp2. Note that when it is necessary to distinguish between soil removal positions, they will be written as Dp1 and Dp2, and when it is not necessary to distinguish them, they will be written as Dp. In this way, the transport vehicle 2 moves between the soil removal position Dp and the loading position Lp.

[0024] Figure 4 shows the vehicle status data at different times. As shown in Figure 4, at time t0, the transport vehicle 2 has finished unloading soil and is in an empty, stopped state. Next, at time t1, the transport vehicle 2 starts moving and is in an empty, moving state. Next, at time t2, the transport vehicle 2 stops moving and is in an empty, stopped state. Next, at time t3, the transport vehicle 2 is loading cargo onto the vessel 12 and is in a loading state. Next, at time t4, the transport vehicle 2 starts moving and is in a loaded, moving state. Next, at time t5, the transport vehicle 2 stops moving and is in a loaded, stopped state. Next, at time t6, the transport vehicle 2 is in an unloading state. The movement data signal S1 includes the status data of the transport vehicle 2 at each of these times.

[0025] In this way, the movement data signal S1 for one cycle, from the soil removal state to the loading state and then back to the soil removal state, is transmitted from each transport vehicle 2 to the data server 4.

[0026] Since the movement data signal S1 also includes position data at each time point, the vehicle state data and position data are also associated. Specifically, the loading position Lp data of the transport vehicle 2 can be obtained from the position data in the loading state. The soil removal position Dp data of the transport vehicle 2 can be obtained from the position data in the soil removal state.

[0027] In other words, the movement data signal S1 for one cycle, as shown in Figures 3(a) and 3(b), represents the data from when the transport vehicle 2 moves from the starting position, which is the soil removal position Dp, to the loading position Lp, and from the loading position Lp to the ending position, which is the soil removal position Dp.

[0028] The communication device 19 transmits the mobile data signal S1 to the data server 4. The communication device 19 can communicate with the data server 4 via wireless communication. The communication may also be conducted via a network such as the Internet; the important thing is that a signal can be transmitted from the communication device 19 to the data server 4.

[0029] The data server 4 receives the movement data signal S1 and stores the movement data. As shown in Figure 2A, the data server 4 includes a processor 4a such as a CPU and a storage device 4b. The processor 4a processes multiple location data to be stored in the storage device 4b. The storage device 4b includes memory such as RAM or ROM, and auxiliary storage such as an HDD (Hard Disk Drive) or SSD (SolID State Drive). The storage device 4b stores a program for storing multiple movement data, as well as the multiple movement data. The data server 4 may be a physical server or a cloud server.

[0030] The identification system 3 shown in Figure 1 identifies the loading area where cargo is loaded and the soil removal area where soil is removed. The functions of the identification system 3 can be performed by, for example, a personal computer or a portable terminal. As a portable terminal, a portable device such as a tablet or smartphone can be used.

[0031] As shown in Figure 1, the specific system 3 includes a processor 3a such as a CPU and a storage device 3b. The processor 3a performs processing to identify the loading site and the soil disposal site. The storage device 3b includes memory such as RAM or ROM, and auxiliary storage such as an HDD (Hard Disk Drive) or SSD (SolID State Drive). The storage device 3b stores data and programs for identifying the loading site and the soil disposal site, as well as a loading site database and a soil disposal site database. The specific system 3 may have multiple processors and multiple storage devices. The specific system 3 can communicate with the data server 4. Communication may be wired or wireless, or via a network such as the Internet; in short, it is sufficient that signals can be sent and received between the specific system 3 and the data server 4. The specific system 3 extracts the necessary data from the data server 4.

[0032] The specific system 3 includes an extraction unit 21, a identification unit 22, a calculation unit 23, and a setting unit 24 (see Figure 1), as shown in Figure 2B. The specific system 3 performs the functions of the extraction unit 21, the identification unit 22, and the calculation unit 23 by having a processor 3a execute a program stored in a storage device 3b. The functions of the extraction unit 21, the identification unit 22, and the calculation unit 23 may be performed by a single processor, or all or part of them may be performed by separate processors. Figure 5 is a diagram illustrating the location data extracted from the movement data in Figure 3(a). Figure 6 is a diagram illustrating the location data extracted from the movement data in Figure 3(b).

[0033] The extraction unit 21 extracts loading position data, loading travel position data, soil discharge position data, and soil discharge travel position data for a set number of days n_days from the location data of the movement data stored in the data server 4. If the above n_days is set to 1 day, the extraction unit 21 extracts loading position data, loading travel position data, soil discharge position data, and soil discharge travel position data for one day.

[0034] The data of the loading position Lp is position data of the transport vehicle 2 in a loading state. The data of the loading travel position Ltp includes position data when the transport vehicle 2 heads for the loading position Lp (referred to as loading approach position data) and position data when the transport vehicle 2 moves away from the loading position Lp (referred to as loading separation position data).

[0035] The loading approach position data is position data that is within a predetermined distance L_db from the loading position Lp and satisfies the altitude change x of -L_b<x<+L_a when the transport vehicle 2 heads for the loading position Lp. The loading separation position data is position data that is within a predetermined distance L_df from the loading position Lp and satisfies the altitude change x of -L_b<x<+L_a when the transport vehicle 2 moves away from the loading position Lp. The predetermined distance L_db and the predetermined distance L_df are preferably set to distances assumed to be within the range of the loading site in each mine. The altitude change -L_b<x<+L_a is also preferably set to an altitude change assumed to be within the range of the loading site in each mine.

[0036] The data of the dumping position Dp is position data of the transport vehicle 2 in a dumping state. The data of the dumping travel position Dtp includes position data when the transport vehicle 2 approaches the dumping position Dp (referred to as dumping approach position data).

[0037] The dumping approach position data is position data that is within a predetermined distance D_db from the dumping position Dp and satisfies the altitude change x of -D_b<x<+D_a when the transport vehicle 2 approaches the dumping position Dp.

[0038] When the dumping position Dp1 as the start position and the dumping position Dp2 as the end position are different as shown in FIG. 3(b), as shown in FIG. 6, the dumping approach position data is position data that is within a predetermined distance D_db from the dumping position Dp and satisfies the altitude change x of -D_b<x<+D_a when the transport vehicle 2 approaches the dumping position P3. The predetermined distance D_db is preferably set to a distance assumed to be within the range of the loading site in each mine. The altitude change -D_b<x<+D_a is also preferably set to an altitude change assumed to be within the range of the loading site in each mine.

[0039] The identification unit 22 identifies the loading site based on the loading position Lp data and the loading travel position Ltp data. The identification unit 22 performs clustering of the loading position data and the loading travel position data. For clustering, DBSCAN can be used, for example. Figure 7 is a diagram illustrating clustering. The search radius eps and the minimum number of samples are set. If the number of samples within the search radius (an example of a specific range) is greater than or equal to the minimum number of samples, they are determined to be samples of the same cluster. For example, the multiple samples shown in Figure 7 are classified into three clusters C1, C2, and C3. A cluster is a collection or a point cloud.

[0040] Figure 8(a) shows the loading location Lp on a map. Figure 8(b) shows the loading location Lp and the loading travel location Ltp on a map. As shown in Figures 8(a) and 8(b), extracting the loading travel location Ltp also increases the number of samples.

[0041] The identification unit 22 performs clustering of loading positions Lp and loading travel positions Ltp as shown in Figure 8(b). The search radius L_eps and the minimum number of samples L_sample are set appropriately according to the size of the mine. Clusters of loading positions Lp and loading travel positions Ltp are generated in which loading positions Lp and loading travel positions Ltp with a minimum number of samples L_sample or more exist within the search radius L_eps of a predetermined loading position Lp or loading travel position Ltp (an example of a specific range).

[0042] Figure 9(a) shows the results of clustering loading locations and loading travel locations. For example, as shown in Figure 9(a), two clusters are identified based on the clustering results of loading location Lp and loading travel location Ltp. The identification unit 22 identifies each of the two clusters as a loading site. The identification unit 22 assigns a cluster ID and date to loading location Lp and loading travel location Ltp within the cluster and stores them in the loading site database.

[0043] FIG. 9(b) is a diagram showing an example of the loading site database 31. For example, as shown in FIG. 9(a), in the loading site database, the ID of the first cluster classified on July 4, 2024 is set to L_240704_0, and the latitudes and longitudes of the loading position Lp and the loading travel position Ltp within the cluster L_240704_0, as well as the date added to the database, are stored. In addition, in the loading site database 31, the ID of the second cluster classified on July 4, 2024 is set to L_240704_1, and the latitudes and longitudes of the loading position Lp and the loading travel position Ltp within the cluster L_240704_1, as well as the date of cluster classification, are stored as a database.

[0044] The identifying unit 22 identifies each of the cluster L_240704_0 and the cluster L_240704_1 as a loading site. The ID of the cluster stored in the loading site database can be referred to as the ID of the loading site.

[0045] Similar to the loading site, the identifying unit 22 performs clustering of the dumping position and the dumping travel position based on the search radius D_eps and the minimum number of samples D_sample set for the dumping site, and identifies the classified clusters as dumping sites.

[0046] Furthermore, when the area enclosing the positions within the cluster identified by clustering contains the positions of previously identified loading sites at a predetermined proportion or higher, the identifying unit 22 regards the identified cluster as the same loading site in combination with the previously identified loading site. For example, assume that on a date subsequent to the date on which the clustering shown in FIG. 9(a) was performed, the loading positions Lp and the loading travel positions Ltp are extracted and clustering is performed, and as a result, clusters L240705_2, L240705_3, and L240705_4 as shown in FIG. 10(a) are classified.

[0047] The identifying unit 22 creates a convex hull that includes all positions for each cluster. A convex hull is the smallest convex set that includes all positions belonging to all clusters. For example, in FIG. 10(a), a convex hull ch_2 of all positions belonging to cluster L_240705_2 is created, a convex hull ch_3 of all positions belonging to cluster L_240705_3 is created, and a convex hull ch_4 of all positions belonging to cluster L_240705_4 is created. The identifying unit 22 determines whether or not these convex hulls ch_2, ch_3, and ch_4 contain more positions associated with the loading site L_240704_0 than a predetermined ratio (ratio), and determines whether or not these convex hulls ch_2, ch_3, and ch_4 contain more positions associated with the loading site L_240704_1 than the predetermined ratio (ratio). FIG. 10(b) is a diagram illustrating the convex hulls ch_2, ch_3, and ch_4 for cluster L_240704_0 and cluster L_240704_1. As shown in FIG. 10(b), the convex hull ch_4 contains more positions associated with the loading site L_240704_0 than the predetermined ratio (ratio). The convex hull ch_2 contains more positions associated with the loading site L_240704_1 than the predetermined ratio (ratio). There is no loading site whose positions are contained in the convex hull ch_3 in an amount greater than the predetermined ratio (ratio).

[0048] Therefore, the identifying unit 22 merges cluster L240705_4 into loading site L_240704_0, and identifies them as a single loading site. The identifying unit 22 merges cluster L240705_2 into loading site L_240704_1, and identifies them as a single loading site. Further, the identifying unit 22 identifies cluster L240705_3 as a single loading site.

[0049] Figure 11(a) shows the loading area database 31 where loading areas L_240704_0 and L_240704_1 are stored. Figure 11(b) shows the location data of three clusters L_240705_2, L_240705_3, and L_240705_4. As shown in Figure 11(c), the ID of cluster L_240705_4 is set to L_240704_0, the ID of cluster L240705_2 is set to L_240704_1, and the ID of cluster L_240705_3 is set to L_240705_3, and these are stored in the loading area database 31 as shown in Figure 11(d).

[0050] As a result, as shown in Figure 12, the loading sites identified on July 4, 2024 (see Figure 9(a)) can be updated based on the location data extracted on July 5, 2024. That is, the identification unit 22 identifies two loading sites L_240704_0 and L_240704_1 on July 4, 2024, and can identify three additional loading sites L_240704_0, L_240704_1, and L_240705_3 on July 5, 2024. Similar to loading sites, the identification unit 22 also identifies excavation sites as the same excavation site if the area surrounding the location within the cluster identified by clustering contains a predetermined proportion or more of the locations of previously identified excavation sites.

[0051] The identification unit 22 identifies two or more loading sites whose associated locations overlap with the identified cluster by a predetermined ratio or more, and if so, the multiple loading sites that overlap with the identified cluster by a predetermined ratio or more are identified as a single loading site.

[0052] For example, consider a case where, after the date the clustering shown in Figure 9(a) was performed, the loading location Lp and loading travel location Ltp are extracted and clustering is performed, resulting in the identification of cluster L240705_2 as shown in Figure 13(a). The identification unit 22 creates a convex hull containing all locations for each cluster. For example, in Figure 13(a), a convex hull ch_4 is created for all locations belonging to cluster L240705_2. The identification unit 22 determines whether these convex hulls ch_4 contain more locations associated with loading site L_240704_0 than a predetermined ratio. Figure 13(b) shows the convex hulls ch_4 for loading site L_240704_0 and loading site L_240704_1. As shown in Figure 13(b), the convex hull ch_4 contains more than a predetermined ratio of locations belonging to loading site L_240704_0. Furthermore, positions belonging to loading area L_240704_1 are included in the convex hull ch_4 at a predetermined rate or higher.

[0053] Therefore, the identification unit 22 combines clusters L240705_2 and L_240704_1 with cluster L_240704_0 to identify them as a single loading area.

[0054] Figure 14(a) shows the loading area database 31 where loading areas L_240704_0 and L_240704_1 are stored. Figure 14(b) shows the location data of cluster L240705_2. As shown in Figure 14(c), the ID of cluster L240705_2 is set to L240704_0, and the ID of loading area L240705_1 is set to L240704_0, and these are stored in the loading area database 31 as shown in Figure 14(d).

[0055] As a result, as shown in Figure 15, the loading area identified on July 4, 2024 (see Figure 9(a)) can be updated based on the location data extracted on July 5, 2024. In other words, the identification unit 22 identifies two loading areas on July 4, 2024, and on July 5, 2024, it can identify a single loading area L240704_0 formed by combining all the loading areas.

[0056] The identification unit 22, similar to the loading area, also identifies the soil disposal area if there are two or more soil disposal areas whose associated locations overlap with the identified cluster by a predetermined ratio of n or more, and the multiple loading areas that overlap with the identified cluster by a predetermined ratio of n or more are identified as a single soil disposal area.

[0057] The identification unit 22 does not identify clusters with a usage count of lim_cyc or less as loading sites and deletes them without saving them in the loading site database. The identification unit 22 does not identify clusters with a usage count of lim_cyc or less as unloading sites and deletes them without saving them in the unloading site database. The identification unit 22 determines the usage count based on the location data of loading sites and unloading sites in the received movement data signal S1. The identification unit 22 calculates the number of loading sites and unloading sites used in the movement data signal S1 acquired over a predetermined number of days and deletes the location data of IDs with a usage count of lim_cyc or less from the loading site database and the unloading site database.

[0058] The specific unit 22 deletes location data that has been added to the loading site database or soil disposal site database for a predetermined number of days n_delday. For example, in Figure 10(b), location data for July 4, 2025 will be deleted from the loading site database after 7 days.

[0059] The identification unit 22 associates the identified loading area and soil disposal area to create a combination of loading area and soil disposal area. The identification unit 22 can identify the loading area where the transport vehicle 2 loads cargo and the soil disposal area where it disposes soil from the received movement data signal S1. For example, the IDs of the loading areas included in the loading area database are L_240704_0, L_240704_1, and L_240705_3, and the IDs included in the soil disposal area database are D_240704_0, D_240704_1, and D_240705_5. If the ID of the loading area in the movement data signal S1 is L_240704_1 and the ID of the soil disposal area is D_240705_5, the identification unit 22 creates a combination of loading area L_240704_1 and soil disposal area D_240705_5. Furthermore, the specific unit 22 creates a combination of loading area L_240704_1 and soil removal area D_240704_1 if the ID of the loading area in another movement data signal S1 is L_240704_1 and the ID of the soil removal area is D_240704_1.

[0060] The calculation unit 23 calculates the amount of cargo to be transported from the loading site to the soil disposal site for each specified combination of loading site and soil disposal site. The movement data signal S1 of each transport vehicle 2 includes load data from the loading site to the soil disposal site. Therefore, the amount of cargo transported from a predetermined loading site to a predetermined soil disposal site over a predetermined number of days can be calculated from the load amount from the loading site to the soil disposal site determined from the load data of each transport vehicle 2.

[0061] The setting unit 24 shown in Figure 1 sets the parameters for extracting data from the received movement data signal S1. The parameters set are n_days, L_db, L_df, -L_b, +L_a, D_df, -D_b, +D_a, L_eps, L_sample, D_eps, D_sample, ratio, and n_delday. The setting unit 24 includes, for example, a display device and an input device. A display can be used as the display device. A touch display, mouse, keyboard, or voice input can be used as the input device. The operator can set the above parameters using the input device while checking the display unit. The display unit may also display the calculation results from the calculation unit 23. If the specific system 3 stores map data, the specified loading site and soil disposal site may be displayed on the map data. For example, the specified loading site and soil disposal site may be enclosed by lines, etc. (see Figure 9(a)).

[0062] Next, we will describe the operation of the identification system 3 in this embodiment, as well as the method for identifying the workplace.

[0063] Figure 16 is a flowchart illustrating the operation of the specific system 3. As shown in Figure 16, in step S11, the extraction unit 21 of the specific system 3 extracts the necessary data from the location data contained in the movement data stored in the data server 4. Next, in step S12, the identification unit 22 of the specific system 3 identifies the loading site and the soil disposal site using machine learning based on the extracted data. Next, in step S13, the specific system 3 assigns IDs to the loading site and the soil disposal site. Next, in step S14, the IDs of the loading site and the soil disposal site are combined. Details of each step are described below.

[0064] Figure 17 is a flowchart showing the operation of step S11 in Figure 16. In step S21, n_days, L_db, L_df, -L_b, +L_a, D_df, -D_b, +D_a, ratio, L_eps, L_sample, D_eps, D_sample, n_ration, lim_cyc, and n_delday are set using the setting unit 24.

[0065] Next, in step S22, the extraction unit 21 extracts the necessary data from the location data of the movement data stored in the data server 4. The extraction unit 21 extracts the loading position Lp data, loading travel position Ltp data, soil discharge position Dp data, and soil discharge travel position Dtp data for the set number of days n_days from the location data of the movement data stored in the data server 4.

[0066] Figure 18 is a flowchart showing the operation of step S12 in Figure 16. In step S31, the identification unit 22 performs clustering from the loading position Lp data and the loading travel position Ltp data using machine learning (e.g., DBSCAN) based on the set parameters L_eps and L_sample. Next, in step S32, the identification unit 22 performs clustering from the soil discharge position Dp data and the soil discharge travel position Dtp data using machine learning (e.g., DBSCAN) based on the set parameters D_eps and D_sample.

[0067] Figure 19 is a flowchart showing the operation of step S13 in Figure 16. In step S41, the identification unit 22 inputs the loading site ID into the loading site database based on the clustering results. Next, in step S42, the identification unit 22 inputs the soil disposal site ID into the soil disposal site database based on the clustering results.

[0068] Figures 20A to 20C are flowcharts showing the operation of steps S41 and S42. Steps S41 and S42 follow the same flow, as shown in Figures 20A to 20C. Below, we will explain the flow in Figures 20A to 20C using a loading area as an example.

[0069] In step S51, the identification unit 22 determines whether the database is empty or not. If the identification unit 22 determines that the database is empty, it assigns a new ID to all clusters. For example, if the loading area database is empty and two clusters are generated as shown in Figure 9(a), the IDs (L_240704_0 and L_240704_1) are assigned to each cluster and they are identified as loading areas.

[0070] Next, in step S53, the identification unit 22 deletes clusters whose usage count is less than or equal to a predetermined number of times (lim_cyc).

[0071] Next, in step S54, the identification unit 22 stores the ID, latitude, longitude, and date and time of addition to the database for the location within the cluster in the loading area database.

[0072] Next, in step S55, the identification unit 22 deletes data from the database that has been added to the database for a predetermined number of days (del_day), and the control ends.

[0073] On the other hand, if it is determined in step S51 that the database is not empty, the control proceeds to step S56. In step S56, the identification unit 22 starts loop control of the number of clusters in the clustering result. For example, if data for loading sites with ID L_240704_0 and L_240704_1 is stored in the loading site database on July 4, and clustering is performed on July 5, and three clusters (see L240705_2, L240705_3, and L240705_4 shown in Figure 10(a)) are identified, the loop between step S56 and step S67 is performed three times.

[0074] Next, in step S57, the identification unit 22 creates a convex hull using the location data belonging to the cluster. For example, as shown in Figure 10(a), a convex hull ch_2 is created using the location data belonging to the cluster L_240704_2.

[0075] Next, in step S58, the identification unit 22 starts loop control for the number of IDs in the database. For example, as shown in Figure 9(b), since the loading area database stores location data for IDs L_240704_0 and L_240704_1, the loop from step S58 to step S61 is performed twice.

[0076] In step S59, the identification unit 22 determines whether the convex hull contains more positions associated with a predetermined ID than a predetermined ratio. For example, the identification unit 22 determines whether the convex hull ch_2 contains more positions associated with L_240704_0 than a predetermined ratio.

[0077] In step S59, if it is determined that the convex hull does not contain more than a predetermined ratio of positions associated with each ID, the control proceeds to step S61. If the loop for the number of IDs has not yet finished, the control returns to step S58. For example, as shown in Figure 10(b), the position associated with L240704_0 is not included in the convex hull ch_2, so the control proceeds to step S61 and returns to step S58.

[0078] Next, in step S59, the identification unit 22 determines whether the convex hull contains more positions associated with other IDs in the database than a predetermined ratio. If, in step S59, it determines that the convex hull contains more positions associated with an ID than a predetermined ratio, the control proceeds to step S60, and the ID is added to the temporary storage list. For example, in step S59, the identification unit 22 determines whether the convex hull ch_2 contains more positions associated with L_240704_1 than a predetermined ratio. As shown in Figure 10(b), the convex hull ch_2 contains more positions associated with ID (L_240704_1) than a predetermined ratio, so L_240704_1 is added to the temporary storage list.

[0079] Next, the control proceeds to step S61, and if it has completed the loop control for the number of IDs in the database, the control proceeds to step S62.

[0080] In step S62, the identification unit 22 determines whether the temporary storage list is empty or not. If it is determined in step S62 that the temporary storage list is not empty, in step S63, the identification unit 22 determines whether the number of IDs in the temporary storage list is 1 or not. If it is determined in step S63 that the number of IDs in the temporary storage list is 1, in step S64, the identification unit 22 assigns the IDs in the temporary storage list to the cluster.

[0081] For example, as described above, if only L_240704_1 is added to the temporary storage list, in step S62 it is determined that the temporary storage list is not empty, and in step S63 it is determined that there is one ID in the temporary storage list, so the control proceeds to step S64. In step S64, the ID of L_240704_1 is assigned to the location data of cluster L240705_2. As a result, as shown in Figure 12, cluster L240705_2 is merged with L_240704_1 and identified as a single loading area.

[0082] After step S64, control proceeds to step S67. If the loop for the number of clusters in the clustering result is completed in step S67, control proceeds to step S53.

[0083] If the loop for the number of clusters in the clustering result has not finished, control returns to step S56, and steps S56 to S67 are repeated.

[0084] In step S62, if the temporary storage list is determined to be empty, the process of assigning a new ID to the cluster in step S66 will be explained. For example, in step S57, a convex hull ch_3 is created for a position within L240705_3. Since the convex hull ch_3 does not contain more than a predetermined ratio of positions associated with L_240704_0 and L_240704_1, the ID is not added to the temporary storage list in step S60. Therefore, in step S62, the temporary storage list is determined to be empty, and control proceeds to step S66. In step S66, the cluster for which the convex hull ch_3 was created is assigned the new ID L240705_3, and control proceeds to step S67.

[0085] In step S63, if it is determined that the number of IDs in the temporary storage list is not 1, step S65 will explain how to assign the oldest ID in the temporary storage list to the cluster. For example, if on July 4th the data for loading sites with ID L_240704_0 and L_240704_1 is stored in the loading site database, and clustering is performed on July 5th, resulting in the creation of cluster L240705_2 as shown in Figure 13(a), then the convex hull ch_4 of the positions within cluster L240705_2 will contain more positions associated with loading site L_240704_0 than a predetermined ratio, and more positions associated with loading site L240704_1 than a predetermined ratio. Therefore, in step S60, the two IDs L_240704_0 and L_240704_1 are added to the temporary storage list, and in step S63, it is determined that the number of IDs in the temporary storage list is not 1. In this case, the oldest ID, L_240704_0, is assigned to all of the cluster L_240705_2, loading area L_240704_0, and loading area L_240704_1 that created the convex hull ch_4. As a result, if a cluster spanning multiple loading areas is generated, the entire cluster can be treated as a single loading area.

[0086] Once the loop for the number of clusters resulting from the clustering process is complete, control proceeds to step S53. Steps S53, S54, and S54 have been described above and will therefore not be explained here.

[0087] Next, the combination of the loading area and unloading area IDs in step S14 will be explained. For example, suppose the IDs included in the loading area database are L_240704_0, L_240704_1, L_240705_3, and the IDs included in the unloading area database are D_240704_0, D_240704_1, D_240705_5. The identification unit 22 combines the loading area IDs and unloading area IDs and assigns a name to the combination. Based on the movement data signal S1, the identification unit 22 combines the loading area ID and unloading area ID to which the transport vehicle 2 is moving. For example, the identification unit 22 assigns a name to LD_1 for the combination of L_240704_0 and D_240704_0.

[0088] In this way, the loading area where the transport vehicle 2 loads its cargo and the soil removal area where it removes soil can be identified. Furthermore, the combination of the loading area where the transport vehicle 2 loads its cargo and the soil removal area where it removes the loaded cargo can be identified.

[0089] Furthermore, the calculation unit 23 can calculate the amount of cargo transported for each combination of loading and unloading areas from the load data in the movement data. This makes it possible to verify discrepancies between planned and actual transport amounts. In this way, for example, by visualizing the production volume for each combination of loading and unloading areas, the discrepancy between the plan and the actual results becomes clear, and the accuracy of the next plan can be improved. In addition, by tracking the daily changes in fuel consumption for each combination of loading and unloading areas, it becomes easier to identify where the factors causing fuel consumption deterioration are located.

[0090] The loading and unloading area identification system 3 according to this embodiment, as described above, includes an extraction unit 21 and an identification unit 22. The extraction unit 21 extracts data for multiple loading locations from the location data of multiple transport vehicles 2. The identification unit 22 identifies a cluster of loading locations as a loading area if, with respect to the multiple loading locations, the number of loading locations within a predetermined range of a predetermined loading location (an example of an identification range) is greater than or equal to a predetermined number. The extraction unit 21 extracts data for multiple unloading locations from the location data of multiple transport vehicles 2. The identification unit 22 identifies a cluster of unloading locations as an unloading area if, with respect to the multiple unloading locations, the number of unloading locations within a predetermined range of a predetermined unloading location (an example of an identification range) is greater than or equal to a predetermined number. This makes it possible to identify loading and unloading areas from the movement data of multiple transport vehicles.

[0091] In the identification system 3 according to this embodiment, the extraction unit 21 extracts data of loading travel positions from the position data up to predetermined distances L_db and L_df from the loading position, and the identification unit 22 identifies a cluster of loading positions and loading travel positions as a loading site if the number of loading positions and loading travel positions within a predetermined range (an example of an identification range) of a predetermined loading position or loading travel position is greater than or equal to a predetermined number. As a result, as shown in Figure 9(a), even if the sample size is small with only loading positions and it is difficult to identify a loading site using machine learning clustering, adding loading travel positions increases the sample size as shown in Figure 9(b), making it easier to identify a loading site.

[0092] In the identification system 3 according to this embodiment, the extraction unit 21 extracts data of the soil-discarding travel position from the soil-discarding position to a predetermined distance D_db from the position data, and the identification unit 22 identifies clusters of soil-discarding positions and soil-discarding travel positions where the number of soil-discarding positions and soil-discarding travel positions within a predetermined range (an example of an identification range) of a predetermined soil-discarding position or soil-discarding travel position is a loading site. As a result, even if the sample size is small with only soil-discarding positions and it is difficult to identify a soil-discarding site using machine learning clustering, adding soil-discarding travel positions increases the sample size as shown in Figure 9(b), making it easier to identify a loading site.

[0093] In the specific system 3 according to this embodiment, the extraction unit 21 extracts data for loading travel positions where the altitude change from the loading position is less than a predetermined value. If the altitude change is large, it is considered to be on a road far from the loading position, so by excluding location data with large altitude changes, the loading site can be identified more accurately.

[0094] In the identification system 3 according to this embodiment, the extraction unit 21 extracts data of the soil-discarding travel position where the change in altitude from the soil-discarding position is less than a predetermined value. If the change in altitude is large, it is considered to be on a track far from the soil-discarding position, so by excluding location data with large changes in altitude, the soil-discarding site can be identified more accurately.

[0095] Although one embodiment of the present invention has been described above, the present invention is not limited to the above embodiment, and various modifications are possible without departing from the spirit of the invention.

[0096] In the embodiment described above, the loading travel position data includes both the period within a predetermined distance L_db from the loading position Lp when the transport vehicle 2 is heading toward the loading position Lp, and the period within a predetermined distance L_df from the loading position Lp when the transport vehicle 2 is moving away from the loading position Lp, but it is not necessary to include either one of them.

[0097] Alternatively, the loading or unloading position may be determined based on the load. The load can be detected by the load sensor 15. For example, the position where the load increases can be identified as the loading position. Also, the position where the load decreases can be identified as the unloading position.

[0098] Alternatively, the loading position or soil discharge position may be determined based on the state or operation of the vessel 12. The state or operation of the vessel 12 can be detected by the vessel sensor 14. For example, the position when the vessel 12 is tilted can be determined as the soil discharge position.

[0099] In the embodiment described above, the data for the soil removal travel position includes only data within a predetermined distance D_db from soil removal position Dp when the transport vehicle 2 approaches soil removal position Dp. However, as shown in Figures 21(a) and 21(b), for example, the data for the soil removal travel position Dtp may also include data within a predetermined distance D_df from soil removal position Dp when the transport vehicle 2 moves away from soil removal position P2. Figure 21(a) shows the case where the soil removal position from which the vehicle departs toward the loading position and the soil removal position to which it moves toward from the loading position are the same. Figure 21(b) shows the case where the soil removal position from which the vehicle departs toward the loading position and the soil removal position to which it moves toward from the loading position are different. In the case of different positions, it is preferable to specify the soil removal area to which the vehicle moves toward from the loading position from the viewpoint of transporting the loaded cargo. Therefore, it is preferable that the data for the soil removal travel position includes data within a predetermined distance D_df when moving away from soil removal position Dp2 and data within a predetermined distance D_db when approaching soil removal position Dp2.

[0100] Furthermore, the data for the soil discharge travel position Dpt may include only the data for the period within a predetermined distance D_df from the soil discharge position Dp when the transport vehicle 2 is moving away from the soil discharge position P2, instead of the data for the period within a predetermined distance D_db from the soil discharge position Dp when the transport vehicle 2 is approaching the soil discharge position P2.

[0101] In the above embodiment, the data for the loading travel position Ltp and the soil discharge travel position Dtp are given the condition that the change in altitude is within a predetermined range, but the condition of altitude change is not required.

[0102] In the embodiment described above, DBSCAN is used as an example of clustering, but it is not limited to this, and other clustering methods may be used.

[0103] In the above embodiment, the specific system 3 extracts the necessary data from the location data stored in the data server 4. However, the system may also acquire location data for the days set by the setting unit 24, along with the movement data, and extract the necessary data from the acquired location data.

[0104] In the above embodiment, the movement data transmitted from the transport vehicle 2 is stored in the data server 4, but this is not limited to that. For example, the data server 4 may not be provided, and a specific system 3 may store the movement data transmitted from the transport vehicle 2.

[0105] In the above embodiment, it was stated that the functions of the specific system 3 can be executed by a personal computer or portable terminal, but it is not limited to this. For example, the extraction unit 21, the identification unit 22, and the calculation unit 23 of the specific system 3 may be executed by a server. The server may be a cloud server or a physical server. The setting unit 24 may be a personal computer or portable terminal provided separately from the specific system. The setting unit 24 may communicate with the server.

[0106] In the above embodiment, the specific system 3 specifies both the loading area and the soil disposal area, but it may also specify only one of them.

[0107] This disclosure provides a work area identification system and a work area identification method that can identify work areas in a mine from the location data of transport vehicles.

[0108] 1: Management system, 2: Transport vehicle, 3: Identification system, 3a: Processor, 3b: Memory device, 11: Vehicle body, 12: Vessel, 13: Wheels, 14: Vessel sensor, 15: Load sensor, 16: Driving sensor, 17: Vehicle position sensor, 18: Vehicle controller, 18a: Processor, 18b: Memory device, 19: Communication device, 21: Extraction unit, 22: Identification unit, 23: Calculation unit, 24: Setting unit, 31: Loading area database

Claims

1. A work area identification system for identifying work areas in a mine, comprising: an extraction unit that extracts data for multiple work locations from the location data of multiple transport vehicles; and an identification unit that identifies a cluster of work locations as a work area, where the number of work locations within a specific range is greater than or equal to a predetermined number.

2. The work area includes a loading area for loading cargo, the work position includes a loading position, the extraction unit extracts data of loading travel positions from the position data up to a predetermined distance from the loading position, and the identification unit identifies a cluster of loading positions and loading travel positions within the specified range, where the number of loading positions and loading travel positions is greater than or equal to a predetermined number, as the loading area, according to claim 1.

3. The work area includes a soil disposal area, the work position includes a soil disposal position, the extraction unit extracts data of soil disposal travel positions from the soil disposal position to a predetermined distance from the soil disposal position from the position data, and the identification unit identifies a cluster of soil disposal positions and soil disposal travel positions in the specified range where the number of soil disposal positions and soil disposal travel positions is greater than or equal to a predetermined number as the soil disposal area, the work area identification system according to claim 1.

4. The work area identification system according to claim 2, wherein the extraction unit extracts data of the loading travel position in the direction approaching the loading position or in the direction away from the loading position from the position data.

5. The work area identification system according to claim 3, wherein the extraction unit extracts data of the soil discharge travel position in the direction approaching the soil discharge position or in the direction away from the soil discharge position from the position data.

6. The work area identification system according to claim 2, wherein the extraction unit extracts data of the loading travel position that is less than a predetermined change in altitude from the loading position.

7. The work area identification system according to claim 3, wherein the extraction unit extracts data of the soil discharge travel position that is less than a predetermined change in altitude from the soil discharge position.

8. The work area includes a loading area for loading cargo, the work position includes a loading position, and the identification unit, if the area surrounding the plurality of loading positions in the generated cluster includes a predetermined proportion or more of the loading positions in the previously identified loading area, then the generated cluster is made to be the same loading area as the previously identified loading area.

9. The work area includes a soil disposal area for soil disposal, the work position includes a soil disposal position, and the identification unit, if the area surrounding the plurality of soil disposal positions in the generated cluster includes a predetermined proportion or more of the soil disposal positions in the previously identified soil disposal area, then the generated cluster is made the same soil disposal area as the previously identified soil disposal area.

10. The work area identification system according to claim 1, wherein the work area includes a loading area for loading cargo and a soil removal area for soil removal, the work location includes a loading location and a soil removal location, and the system further comprises a calculation unit that associates the identified loading area and soil removal area and calculates the amount of cargo to be transported from the loading area to the soil removal area.

11. The work area includes a loading area for loading cargo and a soil removal area for soil removal, the work location includes a loading location and a soil removal location, and the identification unit deletes the data of the loading location within the loading area or the data of the soil removal location within the soil removal area after a predetermined number of days have elapsed, the work area identification system according to claim 1.

12. A method for identifying a work area in a mine, comprising: an extraction step of extracting data for multiple work locations from location data of multiple transport vehicles; and a identification step of identifying a cluster of work locations as a work area, where the number of work locations within a specified range is greater than or equal to a predetermined number.

13. The method for identifying a workplace according to claim 12, wherein the workplace includes a loading area for loading cargo, the work location includes a loading location, the extraction step extracts data of loading travel locations from the loading location to a predetermined distance from the loading location from the location data, and the identification step identifies a cluster of loading locations and loading travel locations in the identified range that has a predetermined number or more of the loading locations and loading travel locations.

14. The method for identifying a workplace according to claim 12, wherein the workplace includes a soil disposal area, the work position includes a soil disposal position, in the extraction step, data of soil disposal travel positions from the soil disposal position to a predetermined distance is extracted from the position data, and in the identification step, a cluster of soil disposal positions and soil disposal travel positions in the identified range having a predetermined number or more is identified as the soil disposal area.

15. The method for identifying a workplace according to claim 13, wherein in the extraction step, data of the loading travel position in the direction approaching the loading position or in the direction away from the loading position is extracted from the position data.

16. The method for identifying a workplace according to claim 14, wherein in the extraction step, data of the soil-discarding travel position in the direction approaching the soil-discarding position or in the direction away from the soil-discarding position is extracted from the position data.

17. The method for identifying a workplace according to claim 13, wherein in the extraction step, data of the loading travel position is extracted from the loading position if the change in altitude is less than a predetermined value.

18. The method for identifying a workplace according to claim 14, wherein the extraction step extracts data of the soil-discarding travel position that is less than a predetermined change in altitude from the soil-discarding position.

19. The method for identifying a workplace according to claim 12, wherein the workplace includes a loading area for loading cargo, the work location includes a loading location, and in the identification step, if the area surrounding the plurality of loading locations in the generated cluster includes a predetermined proportion or more of the loading locations in the previously identified loading area, the generated cluster, together with the previously identified loading area, is considered to be the same loading area.

20. The method for identifying a work area according to claim 12, wherein the work area includes a soil disposal area, the work location includes a soil disposal location, and in the identification step, if the area surrounding the plurality of soil disposal locations in the generated cluster includes a predetermined proportion or more of the soil disposal locations in the previously identified soil disposal area, the generated cluster is considered to be the same soil disposal area together with the previously identified soil disposal area.

21. A method for identifying a work area according to claim 12, further comprising: a work area including a loading area for loading cargo and a soil removal area for soil removal; a work location including a loading area and a soil removal area; and a calculation step of relating the identified loading area and soil removal area to calculate the amount of cargo to be transported from the loading area to the soil removal area.

22. The method for identifying a workplace according to claim 12, wherein the workplace includes a loading area for loading cargo and a soil removal area for soil removal, the work location includes a loading location and a soil removal location, and the method further comprises a deletion step of deleting data of the loading location within the loading area or data of the soil removal location within the soil removal area after a predetermined number of days have elapsed.