Work System

Abstract

To provide a work system capable of improving the accuracy of determining whether or not something is a terrain. [Solution] The work system 1 acquires three-dimensional point cloud data of the work site, clusters the point cloud according to the proximity of each point in the three-dimensional point cloud data, and determines whether the cluster obtained by clustering the three-dimensional point cloud data in a clustering processing unit 71 corresponds to the terrain, and determines whether the clustered cluster corresponds to the terrain based on whether at least a portion of the point cloud contained in the clustered cluster is contained within a predetermined terrain area A associated with the work machine 10.

Description

[Technical Field]

[0001] The present invention relates to a work system. [Background technology]

[0002] For example, Patent Document 1 describes classifying low frequency components of three-dimensional point cloud data obtained by three-dimensional measurement as point cloud data that constitutes a topography. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2002-74323 Summary of the Invention [Problem to be solved by the invention]

[0004] However, when the above-mentioned conventional method is applied to a work site where the topography changes in an undulating manner, there is a risk of erroneous detection of the topography.

[0005] Therefore, an object of the present invention is to provide a work system that can improve the accuracy of determining whether or not something is a terrain. [Means for solving the problem]

[0006] The work system has a three-dimensional point cloud data acquisition unit, a clustering processing unit, and a terrain cluster determination unit. The three-dimensional point cloud data acquisition unit acquires three-dimensional point cloud data of a work site. The clustering processing unit clusters the point cloud according to the proximity of each point in the three-dimensional point cloud data. The terrain cluster determination unit determines whether or not a cluster obtained by clustering the three-dimensional point cloud data by the clustering processing unit corresponds to terrain. The terrain cluster determination unit determines whether or not the cluster corresponds to terrain based on whether at least a portion of the point cloud contained in the cluster is included in a terrain area that is a predetermined area associated with a work machine. [Effects of the Invention]

[0007] The above-described work system can improve the accuracy of determining whether or not something is terrain. [Brief explanation of the drawings]

[0008] [Figure 1] FIG. 2 is a side view of the work machine 10 and other components of the work system 1. [Figure 2] FIG. 2 is a block diagram of the work system 1 shown in FIG. [Figure 3] 3 is a flowchart of the processing of the controller 70 and the like shown in FIG. 2. [Figure 4A] 3 is a side view of a point cloud detected by a three-dimensional point cloud data acquisition unit 43 shown in FIG. 2. FIG. [Figure 4B] 3 is a top view of a point cloud detected by a three-dimensional point cloud data acquisition unit 43 shown in FIG. 2. FIG. [Figure 5A] FIG. 10 is a side view of clusters clustered based on point clouds. [Figure 5B] FIG. 10 is a top view of clusters clustered based on point clouds. [Figure 6A] FIG. 10 is an example of a side view of a cluster when there is a trench T5 crossing the terrain. [Figure 6B] This is an example of a top view of a cluster when there is a trench T5 crossing the terrain. [Figure 7] This is an example of a top view of a cluster when there is a trench T5 running across the terrain. [Figure 8] 10 is an example of a side view of a cluster when a structure T7 or a ground T1 has a slope T1a. [Figure 9] 10 is a modified side view showing an example of the arrangement of the three-dimensional point cloud data acquisition unit 43. FIG. [Figure 10] 2 is a modified example of the block diagram of the work system 1 shown in FIG. [Figure 11] 2 is an example of a display unit 80 of the work system 1 shown in FIG. DETAILED DESCRIPTION OF THE INVENTION

[0009] A working system 1 including a working machine 10 (see FIG. 1) will be described with reference to the drawings.

[0010] The work system 1 is a system that improves the detection accuracy of point clouds corresponding to terrain from three-dimensional point cloud data acquired from a work site. The work system 1 is a system that clusters the three-dimensional point cloud data and determines whether the clusters correspond to the terrain. As shown in FIG. 1, the work system 1 includes a work machine 10, a three-dimensional point cloud data acquisition unit 43, a controller 70 (computer), and a display unit 80 (see FIG. 2). In this embodiment, the work machine 10 has a configuration that includes the three-dimensional point cloud data acquisition unit 43, the controller 70, and the display unit 80, but is not limited to this.

[0011] As shown in FIG. 1, the work machine 10 is a machine that performs work. The work machine 10 may be a construction machine that performs construction work, or a loading and unloading machine that performs loading and unloading work. The work machine 10 may be, for example, a shovel or a crane. The work machine 10 may be a bulldozer or a wheel loader. The following description will be given of the case where the work machine 10 is a shovel.

[0012] The work machine 10 may operate in response to operation by a worker (operator). For example, the work machine 10 may be operated by a worker in a cab 13c (described later), or may be remotely operated from outside the work machine 10. The work machine 10 may sometimes be operated by automatic control. The automatic control may be automatic operation or semi-automatic operation (machine control described later).

[0013] The work machine 10 comprises a machine body 10a, an attachment 15, a drive control unit 17 (see FIG. 2), and an actuator 30 (see FIG. 2). The machine body 10a is the main body portion of the work machine 10. The machine body 10a comprises a lower body 11 and an upper rotating body 13.

[0014] The lower body 11 rotatably supports the upper rotating body 13. The lower body 11 may be a lower traveling body that can travel on a traveling surface (such as the ground). When the lower body 11 is capable of traveling, the lower body 11 may be provided with crawlers or wheels.

[0015] The upper rotating body 13 is mounted on the lower main body 11 so as to be able to rotate. A boom 15a and the like are attached to the upper rotating body 13. The upper rotating body 13 is equipped with a cab 13c. The cab 13c is an area where a worker (operator) can operate the work machine 10. When the work machine 10 operates in response to operation by the worker, the work machine 10 may be operated by the worker in the cab 13c, or may be remotely operated from outside the work machine 10.

[0016] (direction) The direction in which the rotation axis of the upper rotating body 13 relative to the lower main body 11 extends is defined as the vertical direction Z. In the vertical direction Z, the side (facing) from the lower main body 11 toward the upper rotating body 13 is defined as the upper side Z1, and the opposite side is defined as the lower side Z2. The vertical direction Z may be vertical. The direction in which the rotation axis of the boom 15a (described below) relative to the upper rotating body 13 extends is defined as the lateral direction Y. The direction perpendicular to both the vertical direction Z and the lateral direction Y is defined as the front-rear direction X. When viewed from the vertical direction Z, the front-rear direction X is the direction in which the central axis of the attachment 15 extends in the longitudinal direction of the attachment 15 (the front-rear direction X of the attachment 15). In the front-rear direction X, the side from which the attachment 15 protrudes relative to the upper rotating body 13 is defined as the rear side X1, and the opposite side is defined as the front side X2.

[0017] The attachment 15 is the part that performs work. The attachment 15 is attached to the machine body 10a. For example, the attachment 15 includes a boom 15a, an arm 15b, and a tip attachment 15c. The boom 15a is attached to the upper rotating body 13 so as to be rotatable (able to raise and lower, and rotatable in the forward-backward direction X and the upward-downward direction Z). The arm 15b is attached to the boom 15a so as to be rotatable (able to rotate in the forward-backward direction X and the upward-downward direction Z).

[0018] The tip attachment 15c is provided at the tip of the attachment 15. The tip attachment 15c is rotatably attached to the arm 15b (rotatable in the forward / backward direction X and the upward / downward direction Z). The tip attachment 15c may be a bucket capable of scooping and digging a work object. The tip attachment 15c may be equipped with a device for clamping a work object (grapple, nibbler, rotating fork, etc.), a device for crushing a work object (breaker, etc.), or a magnet for attracting a metal work object.

[0019] The work object is an object that is the target of work by the work machine 10. The work object may be soil or sand, rock, a magnetic material (metal, etc.), resin, waste, wood (logs, etc.), or a structure (block, etc.). If the work object is soil or sand, the work object may be in soil, granules, chips, powder, etc.

[0020] The drive control unit 17 (see Figure 2) controls the actuator 30. The drive control unit 17 may include a hydraulic circuit that controls a hydraulic actuator that is operated by hydraulic pressure. The drive control unit 17 may include an electric circuit that controls an electric actuator that is operated by electricity. The drive control unit 17 controls the travel of the lower body 11. The drive control unit 17 controls a swing motor 33 that swings the upper swing body 13 relative to the lower body 11. The drive control unit 17 controls a boom cylinder 35a that rotates (raises and lowers) the boom 15a relative to the upper swing body 13. The drive control unit 17 controls an arm cylinder 35b that rotates the arm 15b relative to the boom 15a. The drive control unit 17 controls a tip attachment cylinder 35c that rotates the tip attachment 15c relative to the arm 15b.

[0021] The actuator 30 is a device that moves the work machine 10. The actuator 30 may be equipped with a hydraulic actuator that is operated by hydraulic pressure, or an electric actuator that is operated by electricity. The actuator 30 may be equipped with a motor that drives rotation, or may be equipped with a cylinder that drives extension and retraction (telescopic cylinder). The actuator 30 may be equipped with a travel motor that drives the work machine 10. The actuator 30 may be equipped with a swing motor 33 that rotates the upper swing body 13 relative to the lower body 11. The actuator 30 may be equipped with a cylinder that moves the attachment 15.

[0022] The detection unit 40 detects various conditions. Part or all of the detection unit 40 may be mounted on the work machine 10, or may be located external to the work machine 10. The same applies to the input unit 35, controller 70, and display unit 80, which will be described later, that may be mounted on the work machine 10 or located external to the work machine 10. As shown in FIG. 2 , the detection unit 40 includes a position detection unit 41, a three-dimensional point cloud data acquisition unit 43, a direction detection unit 45, and an attitude detection unit 47.

[0023] The position detection unit 41 detects the position of the object to be measured. The position detection unit 41 detects the position of a specific part of the work machine 10. For example, the position detection unit 41 may detect the position of a specific part of the upper rotating body 13, or may detect the position of a specific part of the attachment 15. The position detection unit 41 may detect the position using electromagnetic waves (light, radio waves, etc.). The position detection unit 41 may use a satellite positioning system, for example, a global navigation satellite system (GNSS). The position detection unit 41 may use a (terrestrial) transmitter and receiver without using a satellite, or may use reflection of light (e.g., laser light) (e.g., a total station, etc.). The position detection unit 41 may calculate the position of the object to be measured based on position information detected by multiple types of devices.

[0024] The three-dimensional point cloud data acquisition unit 43 acquires three-dimensional point cloud data indicating the distance to an object. The three-dimensional point cloud data acquisition unit 43 may be an imaging device. The three-dimensional point cloud data acquisition unit 43 may detect three-dimensional information of an object by irradiating the object with waves such as electromagnetic waves and detecting the reflected waves. The three-dimensional point cloud data acquisition unit 43 may include a TOF (Time Of Flight) sensor that detects the distance based on the time from when the waves are irradiated until the reflected waves return, or may include a sensor that detects the distance based on the frequency of the reflected waves. The three-dimensional point cloud data acquisition unit 43 may include a device that detects three-dimensional information using light (e.g., laser light), such as a LiDAR (Light Detection and Ranging). The three-dimensional point cloud data acquisition unit 43 may also include a device that detects three-dimensional information using radio waves (e.g., millimeter-wave radar).

[0025] Only one three-dimensional point cloud data acquisition unit 43 may be provided, or multiple three-dimensional point cloud data acquisition units 43 may be provided. Only one type (such as one system) of three-dimensional point cloud data acquisition unit 43 may be used, or multiple types of three-dimensional point cloud data acquisition units 43 may be combined. The three-dimensional point cloud data acquisition unit 43 may detect three-dimensional information of the imaged object based on a three-dimensional image (distance image) and a two-dimensional image. The three-dimensional point cloud data acquisition unit 43 may be mounted on the work machine 10. The mounting position of the three-dimensional point cloud data acquisition unit 43 on the work machine 10 is not limited, and may be, for example, above the cab 13c, or may be disposed on the side or underside of each component of the attachment 15. The three-dimensional point cloud data acquisition unit 43 may also be disposed outside the work machine 10 (for example, at the work site). If the three-dimensional point cloud data acquisition unit 43 is located outside the work machine 10, it may be possible to detect positions (such as areas shaded by the attachment 15) that cannot be detected when the three-dimensional point cloud data acquisition unit 43 is mounted only on the work machine 10. Furthermore, if the three-dimensional point cloud data acquisition unit 43 is located outside the work machine 10, the work system 1 of this embodiment can be applied even if the work machine 10 does not have a three-dimensional point cloud data acquisition unit 43.

[0026] The direction detection unit 45 detects the direction (orientation, posture) of the object to be measured. The direction detection unit 45 detects the direction of a specific part of the work machine 10. For example, the direction detection unit 45 may detect the direction of a specific part of the upper rotating body 13, or may detect the direction of a specific part of the attachment 15. The direction detection unit 45 may detect the orientation of the object to be measured using geomagnetism. The direction detection unit 45 may detect the direction of the object to be measured based on the positions of multiple parts of the object to be measured relative to the work site.

[0027] The attitude detection unit 47 may detect the position and orientation of the work machine 10 relative to the work site. The attitude detection unit 47 may detect the position and orientation of a reference position of the work machine 10 relative to the work site. The reference position of the work machine 10 is, for example, a specific position on the upper rotating body 13 or the lower main body 11. The reference position of the work machine 10 may be the attachment portion (boom foot) of the boom 15a to the upper rotating body 13, or a specific position on the rotation center axis of the upper rotating body 13 relative to the lower main body 11. The attitude detection unit 47 may detect the inclination of the work machine 10 with respect to the horizontal plane. The attitude detection unit 47 may detect information (angle, angular velocity, angular acceleration, etc.) about the rotation of the boom 15a relative to the upper rotating body 13. The attitude detection unit 47 may detect information (angle, angular velocity, angular acceleration, etc.) about the rotation of the boom 15a relative to the upper rotating body 13. The attitude detection unit 47 may detect information about the rotation of the arm 15b relative to the boom 15a. The attitude detection unit 47 may detect information on the rotation of the bucket relative to the arm 15b.

[0028] The input unit 60 is used to input information (input device). The input unit 60 is used to input information used for control in the controller 70. For example, the input unit 60 may be used to set a topographical area A (see FIGS. 6A and 6B). As will be described later, the topographical area A is used to determine whether a cluster obtained by clustering three-dimensional point cloud data corresponds to a topography.

[0029] The input unit 60 is operated by an operator and outputs a signal in accordance with the operation. The input unit 60 outputs information to the controller 70. The input unit 60 may be equipped with a touch panel, a mouse, a keyboard, or a switch. The input unit 60 may be provided on a tablet, a smartphone, or a personal computer. The input unit 60 may be provided on a client device or a server device. The input unit 60 may be provided on the work machine 10, and may be provided in the cab 13c, for example. The input unit 60 may be provided on a remote control device for remotely operating the work machine 10. The input unit 60 may be provided on an operation unit (e.g., an operation lever) provided on the cab 13c or the remote control device, or may be provided on the operation unit or a display (e.g., a cluster gauge, etc.).

[0030] The input unit 60 may have an operation unit. The operation unit is configured so that it can be operated by an operator. Operations for moving the work machine 10 are input to the operation unit. The operation unit may be provided in the driver's cab 13c, or may be provided in a remote control device for remotely operating the work machine 10. The operation unit may include a lever or a pedal.

[0031] The operation unit (not shown) outputs a command according to the operation. The operation unit may output a command according to the operation amount. The command output by the operation unit may be pilot hydraulic pressure or an electrical signal. The operation unit may include a hydraulic remote control valve or an angle sensor (e.g., a variable resistor). An operation (travel operation) for traveling the lower body 11 (see FIG. 1) may be input to the operation unit. An operation (swing operation) for rotating the upper rotating body 13 (see FIG. 1) relative to the lower body 11 may be input to the operation unit. An operation (attachment operation) for moving the attachment 15 (see FIG. 1) may be input to the operation unit. An operation (boom operation) for rotating the boom 15a (see FIG. 1) relative to the upper rotating body 13 may be input to the operation unit. An operation (arm operation) for rotating the arm 15b (see FIG. 1) relative to the boom 15a may be input to the operation unit. An operation (tip attachment operation) for rotating tip attachment 15d (see FIG. 1) relative to arm 15b may be input to the operation unit.

[0032] The controller 70 is a computer that inputs and outputs signals, performs calculations (processing), stores information, etc. The functions of the controller 70 are realized by executing a program stored in a storage unit 70b of the controller 70 on a calculation unit 70a. The controller 70 and other devices may be connected via wireless communication or wired communication. The components of the controller 70 may be connected to each other via wireless communication or wired communication. For example, communication is performed via a communication means such as a mobile phone line, an optical line, a wireless LAN (Local Area Network), or a wired LAN.

[0033] For example, information is input to the controller 70 from the detection unit 40 and the input unit 60. For example, the controller 70 outputs a command (signal) to the drive control unit 17 to operate the work machine 10. For example, the controller 70 outputs information to the display unit 80. The controller 70 may be mounted on the work machine 10 or may be located outside the work machine 10. The controller 70 may be distributed and located in multiple parts (a distributed system may be formed).

[0034] The controller 70 includes a calculation unit 70a and a storage unit 70b. The calculation unit 70a performs calculations (processes) on information. The storage unit 70b stores information. Focusing on the functions of the controller 70, the controller 70 includes a clustering processing unit 71, a terrain cluster determination unit 72, and a terrain point cloud extraction unit 73. Specific functions (processing) executed by the controller 70 will be described later.

[0035] The functions of the controller 70 may be realized by a client device (not shown) and a server device (not shown). Each of the client device and the server device is a computer. The input unit 60, the controller 70, and the display unit 80 may be provided in either the client device or the server device, or in both. For example, the memory unit 70b and the calculation unit 70a of the controller 70 may be provided in either the client device or the server device, or in both. Only one client device and one server device may be provided, or multiple client devices and one server device may be provided. The client device and the server device may be connected by wireless communication or by wired communication. For example, communication is performed by a communication means such as a mobile phone line, an optical fiber line, a wireless LAN (Local Area Network), or a wired LAN.

[0036] The display unit 80 is a device that displays information. The display unit 80 may include a display device (monitor). The display unit 80 displays information based on a signal output from the controller 70. The display unit 80 outputs light (such as a display). The display unit 80 may be provided on a tablet, a smartphone, or a personal computer. The display unit 80 may be provided in the cab 13a (see FIG. 1). The display unit 80 may be provided on a remote control device for remotely controlling the work machine 10 (see FIG. 1). The display unit 80 may include a projection device that projects onto an object such as the ground. The display unit 80 may include a light source. The display unit 80 may include a device that uses VR (Virtual Reality) technology (a VR device) or a device that uses AR (Augmented Reality) technology (an AR device). The display unit 80 may change at least one of the hue, density (transparency), brightness, and saturation of the light it outputs. The work system 1 (work machine 10) may have an output unit that outputs information based on a signal output from the controller 70. The output unit may output sound or vibration.

[0037] (Operation of work machine 10) As described above, the work machine 10 (see FIG. 1) may be operated by an operator in the cab 13a (see FIG. 1), may be remotely operated by an operator from outside the work machine 10 (by a remote control device), or may be automatically driven. The work machine 10 is a machine that utilizes information and communication technology (ICT) (e.g., ICT construction machinery). For example, the work machine 10 may be operated by an operator using the function of a machine guidance (MG) system. Specifically, a work plan is set in the controller 70. Then, guidance such as the position where work should be done is shown to the operator so that the work machine 10 can work according to the work plan. This guidance is output, for example, to an output unit provided in the cab 13a of the work machine 10 or an output unit provided in the remote control device. Then, the operator operates the work machine 10 according to the guidance. As a result, the work machine 10 moves according to the work plan.

[0038] Furthermore, for example, the work machine 10 (see FIG. 1) may be operated by a machine control (MC) system (semi-automatic operation). Specifically, a work plan is set in the controller 70. Then, for example, the worker operates only some of the elements of the attachment 15 (see FIG. 1) (for example, only the boom 15a (see FIG. 1)). At this time, the controller 70 automatically controls the elements that are not operated by the worker (for example, the arm 15b (see FIG. 1) and the tip attachment 15d (see FIG. 1)) so that the work machine 10 moves in accordance with the work plan. At this time, the controller 70 controls the operation of the work machine 10 based on information detected by the attitude detection unit 47 (the same applies to automatic operation). As a result, the work machine 10 moves in accordance with the work plan. Furthermore, for example, the work machine 10 may be operated by automatic operation. In this case, the controller 70 controls the operation of the work machine 10 so that the work machine 10 automatically moves in accordance with the work plan.

[0039] (process) Next, a flowchart of the processing executed in the work system 1 (work machine 10) will be described with reference to Figure 3. The work system 1 (mainly the controller 70) is configured to perform the following processing. A program stored in the controller 70 causes the controller 70 to execute the following operations. A work method in which the following operations are performed is implemented in the work system 1 (work machine 10). The work method causes the controller 70 (computer) to perform the following processing.

[0040] First, the controller 70 acquires three-dimensional point cloud data (S10). For example, in the example shown in FIG. 1, the imaging range 43a of the three-dimensional point cloud data acquisition unit 43 includes the ground T1, a work object T2 placed on the ground T1, a floating aircraft T3 (e.g., a drone), and a portion of the attachment 15. In the example shown in FIG. 1, as shown in FIGS. 4A and 4B, the controller 70 acquires three-dimensional point cloud data from the three-dimensional point cloud data acquisition unit 43. The three-dimensional point cloud data includes a point cloud P1 corresponding to the ground T1, a point cloud P2 corresponding to the work object T2, a point cloud P3 corresponding to the aircraft T3, and a point cloud P4 corresponding to the attachment 15. At this point, the three-dimensional point cloud data is not separated into point clouds P1 to P4, and also includes noise points (not shown) that are separated from the point clouds P1 to P4.

[0041] (About clustering processing) Then, the controller 70 (more specifically, the clustering processing unit 71) performs clustering processing on the acquired three-dimensional point cloud data (S20). The clustering processing unit 71 performs clustering (grouping) on ​​the three-dimensional point cloud data acquired by the three-dimensional point cloud data acquisition unit 43. The clustering processing unit 71 clusters the point cloud according to the proximity state of each point of the three-dimensional point cloud data.

[0042] For example, the clustering processing unit 71 uses a clustering method called DBSCAN (Density-Based Spatial Clustering of Applications with Noise). Specifically, the clustering processing unit 71 determines whether a number of points equal to or greater than a threshold exists within a specified radius for any point in the 3D point cloud data. If a number of points equal to or greater than the threshold exists within the specified radius, the clustering processing unit 71 determines the region of the specified radius as one cluster. The clustering processing unit 71 then repeats the above determination for each point within the region of the specified radius, continuing to expand the cluster region as long as the condition is satisfied. Furthermore, for example, the clustering processing unit 71 uses a clustering method called OPTICS (Ordering Points To Identify the Clustering Structure). Specifically, the clustering processing unit 71 does not predefine a threshold like DBSCAN. The clustering processing unit 71 determines a threshold based on the distribution trend of the entire acquired 3D point cloud data and determines clusters in the same way as DBSCAN.

[0043] In this way, when the controller 70 (more specifically, the clustering processing unit 71) acquires three-dimensional point cloud data such as those shown in FIGS. 4A and 4B, it clusters the data into the following clusters. As shown in FIGS. 5A and 5B, the controller 70 determines point clouds P1 and P2, in which points are close to each other, as one cluster C1. The controller 70 also determines point cloud P3, in which points are close to each other, as one cluster C2. The controller 70 also determines point cloud P4, in which points are close to each other, as one cluster C3.

[0044] (Determining whether a cluster corresponds to the terrain) Next, the controller 70 (more specifically, the terrain cluster determination unit 72) determines whether all of the clustered clusters (C1, C2, C3) correspond to the terrain (S30). The controller 70 determines whether any part of the cluster is included in terrain area A (see FIGS. 5A and 5B) for all of the clustered clusters. The controller 70 determines that all clusters included in terrain area A are terrain clusters that correspond to the terrain. In the example shown in FIGS. 5A and 5B, the controller 70 determines that cluster C1 is a terrain cluster that corresponds to the terrain. Note that the conditions for determining that a cluster is a terrain cluster may include whether the cluster is within the range of the reach of the attachment 15 of the work machine 10, or within a range that is this range plus a predetermined distance.

[0045] (Regarding terrain area A) The topographical area A is an area that is set in advance. For example, the topographical area A is set in advance before the determination of the topographical cluster. For example, the topographical area A is set to an area in the imaging range 43a of the three-dimensional point cloud data acquisition unit 43 where a point cloud that represents the topography is likely to be detected. For example, specifically, the topographical area A may be determined according to a specific position of the lower body 11 of the work machine 10 (for example, the underside of the lower body 11). The specific position of the lower body 11 of the work machine 10 may be acquired by the attitude detection unit 47. Furthermore, for example, specifically, the topographical area A may be determined according to the measured position of the ground surface T1. The position of the ground surface T1 may be acquired by the position detection unit 41.

[0046] In the example shown in FIGS. 5A and 5B, the position of the terrain area A is set to the rear side X1 of the work machine 10. The position of the terrain area A may be changed depending on the arrangement of the work machine 10. The position of the terrain area A may be set to a predetermined area of ​​the work site or the entire work site. The position of the terrain area A may be made inputtable by the input unit 60. The terrain area A is set to a dimension L1 from the ground surface T1 to the upper side Z1 and a dimension L2 from the ground surface T1 to the lower side Z2. The terrain area A is set to a dimension L3 in the fore-aft direction X. The terrain area A is set to a dimension L4 in the lateral direction Y. The dimensions L1 to L4 of the terrain area A may be changed depending on the shape of the work site where the work machine 10 is arranged. The dimensions L1 to L4 of the terrain area A may be set depending on the dimensions of the work machine 10. The dimensions L1 to L4 of the terrain area A may be arbitrarily changeable. The dimensions L1 to L4 of the terrain area A may be made inputtable by the input unit 60.

[0047] Next, the controller 70 (more specifically, the terrain point cloud extraction unit 73) extracts the point clouds belonging to the clusters of the terrain clusters as one point cloud corresponding to the terrain (S40), and ends this process. In the example shown in FIGS. 5A and 5B, the point cloud belonging to cluster C1 is extracted as the point cloud corresponding to the terrain. Although not shown, the controller 70 may use the extracted point cloud corresponding to the terrain as terrain data as is. Furthermore, the controller 70 may complement information between points constituting the point cloud for the extracted point cloud corresponding to the terrain, and calculate three-dimensional information that more closely resembles the terrain, and use this as terrain data.

[0048] (Variation 1) In the above example, the controller 70 determined that the cluster that is partially included in the terrain region A is a terrain cluster. On the other hand, the controller 70 may determine that, in addition to the cluster that is partially included in the terrain region A, other clusters that include points lower than the terrain cluster are terrain clusters. The controller 70 may also determine that other clusters that include points lower than the height of the point cloud (point cloud of the terrain cluster) that belongs to the cluster determined to correspond to the terrain correspond to the terrain. In this determination, the height of the highest point of the terrain cluster may be compared with the height of the lowest point of the other cluster.

[0049] As described above, in this modified example, the controller 70 performs a first terrain cluster determination for each clustered cluster to determine whether it is included in terrain region A. As a result of the first terrain cluster determination, the controller 70 determines that a cluster that is partially included in terrain region A is a terrain cluster. Furthermore, the controller 70 performs a second terrain cluster determination for each cluster excluding the terrain cluster to determine whether it includes a point below the height of the terrain cluster. As a result of the second terrain cluster determination, the controller 70 further determines that a cluster that includes a point below the height of the terrain cluster is a terrain cluster. The second terrain cluster determination may be performed repeatedly. In other words, a second terrain cluster determination may be performed based on the cluster determined to be a terrain cluster in the second terrain cluster determination.

[0050] 6A and 6B, a case will be described in which there is a groove T5 in the ground surface T1 that crosses the far side X1 of the work machine 10. The groove T5 has a slope T5a on the near side X2, a bottom surface T5b, and a slope T5c on the far side X1. It is assumed that the imaging range 43a of the three-dimensional point cloud data acquisition unit 43 does not include the slope T5a.

[0051] The controller 70 (more specifically, the clustering processing unit 71) clusters the point cloud (not shown) of the ground surface T1 located on the far side X1 of the work machine 10 into cluster C4. Furthermore, the controller 70 clusters the point cloud including the bottom surface T5b and the slope T5c into cluster C5. Then, the controller 70 (more specifically, the terrain cluster determination unit 72) determines that the cluster C4 included in the terrain area A is a terrain cluster. Furthermore, the controller 70 (more specifically, the terrain cluster determination unit 72) determines that the cluster C5 including points located at a lower height than the cluster C4 (such as the point cloud (not shown) of the bottom surface T5b and the point cloud (not shown) of the slope T5c) is a terrain cluster.

[0052] In this way, even if the three-dimensional point cloud data acquisition unit 43 obtains a terrain point cloud that cannot be recognized as a continuous cluster, the controller 70 can appropriately determine the terrain cluster. For example, as shown in FIG. 7, even if a ditch T6 running across the lateral direction Y of the work machine 10 exists, the controller 70 can appropriately determine the terrain cluster. Note that the left part of FIG. 7 shows a cross-sectional view of the ditch T6 as viewed from the rear side X1. For example, as shown in FIG. 8, if a structure T7 that is not terrain, such as a roof, exists, a cluster C6 based on the point cloud of the structure T7 is outside the terrain area A. Therefore, the non-terrain cluster C6 is appropriately excluded from the terrain cluster. Furthermore, even if a slope T1a exists on the ground T1 as shown in FIG. 8, the controller 70 can determine the cluster C7 of the ground T1 that includes the slope T1a as the terrain cluster. Furthermore, as shown in FIG. 9, even if the mounting position of the three-dimensional point cloud data acquisition unit 43 is changed, the controller 70 can, for example, exclude a cluster C8 of the aircraft T3 and appropriately determine a terrain cluster C9 as the terrain cluster.

[0053] (Variation 2) 10, in this modification, the controller 70 has, in addition to the configuration of the embodiment described above, a position information acquisition unit 74, an orientation determination unit 75, and a coordinate conversion unit 76. The controller 70 may also have a difference calculation unit 77. The three-dimensional point cloud data acquisition unit 43 is attached to the work machine 10.

[0054] The position information acquisition unit 74 acquires position information of the work machine 10 relative to the work site. For example, the position information acquisition unit 74 acquires the position information of the work machine 10 relative to the work site based on the positioning information detected by the position detection unit 41 and the attitude information of the work machine 10 detected by the attitude detection unit 47. For example, the orientation determination unit 75 acquires azimuth angle information, which is information about the azimuth angle of the three-dimensional point cloud data acquisition unit 43 when the three-dimensional point cloud data was acquired. For example, specifically, the orientation determination unit 75 acquires the azimuth angle of the three-dimensional point cloud data acquisition unit 43 based on the positioning information detected by the position detection unit 41, the attitude information of the work machine 10 detected by the attitude detection unit 47, and the imaging direction of the three-dimensional point cloud data acquisition unit 43.

[0055] The coordinate conversion unit 76 converts the coordinate system of the three-dimensional point cloud data into a coordinate system of a target construction surface based on the work site, based on the position information and information including the azimuth angle from the three-dimensional point cloud data acquisition unit 43. For example, the coordinate conversion unit 76 converts the point cloud of the terrain cluster obtained in the coordinate system of the work machine 10 into the coordinate system of the target construction surface. The target construction surface is stored, for example, in the memory unit 70b of the controller 70. The target construction surface is used, for example, for automatic driving and for guiding the worker (operator). The coordinate system of the target construction surface is, for example, a global coordinate system. The global coordinate system may be, for example, the World Geodetic System or a plane rectangular coordinate system. The World Geodetic System (geocentric Cartesian coordinate system) is a coordinate system in which the coordinate origin is located at the center of gravity of the Earth, the X axis is directed toward the intersection of the Greenwich meridian and the equator, the Y axis is directed toward 90 degrees east longitude, and the Z axis is directed toward the North Pole. The plane rectangular coordinate system is a coordinate system in which the axis coinciding with the meridian at the origin of the coordinate system is the X-axis, values ​​pointing due north are positive, the axis perpendicular to the X-axis at the origin of the coordinate system is the Y-axis, values ​​pointing due east are positive, and the axis perpendicular to the X-axis and Y-axis is the Z-axis. Note that when the three-dimensional point cloud data acquisition unit 43 is fixedly installed at a work site or the like, the three-dimensional point cloud data may be acquired in the same coordinate system as the coordinate system of the target construction surface.

[0056] Specifically, for example, the controller 70 detects the position of the upper rotating body 13 (work machine position information) in the global coordinate system using the position detection unit 41 (e.g., GNSS). Then, the controller 70 acquires the position (including the angle (the acquisition direction of the three-dimensional point cloud data)) (azimuth angle information) of the three-dimensional point cloud data acquisition unit 43 relative to the upper rotating body 13. When the three-dimensional point cloud data acquisition unit 43 is attached to the upper rotating body 13, pre-stored position information of the three-dimensional point cloud data acquisition unit 43 relative to the upper rotating body 13 may be used. When the three-dimensional point cloud data acquisition unit 43 is attached to the attachment 15 (boom 15a or arm 15b), the position of the three-dimensional point cloud data acquisition unit 43 relative to the upper rotating body 13 may be calculated using the attitude information of the attachment 15 detected by the attitude detection unit 47. Then, the controller 70 calculates the position of the three-dimensional point cloud data acquisition unit 43 in the global coordinate system from the position of the upper rotating body 13 in the global coordinate system and the position of the three-dimensional point cloud data acquisition unit 43 relative to the upper rotating body 13. Then, the controller 70 uses the position of the three-dimensional point cloud data acquisition unit 43 in the global coordinate system to convert the point cloud data in the coordinate system based on the three-dimensional point cloud data acquisition unit 43 into the global coordinate system.

[0057] The difference calculation unit 77 calculates the difference between the terrain data based on the point cloud corresponding to the extracted terrain and the target construction surface. The difference between the terrain data and the target construction surface may be, for example, the amount of soil required to reach the target construction surface when discharging the work object onto the current terrain, or the amount of remaining soil required to excavate to reach the target construction surface. As shown in FIG. 11 , for example, the display unit 80 displays a work machine image 80A showing the work machine 10, a terrain data image 80B, and a target construction surface image 80C showing the target construction surface of the work machine 10, all superimposed on one another. The terrain data image 80B is an image showing terrain data based on a point cloud belonging to a cluster determined to correspond to the terrain. The terrain data image 80B may be an image of a point cloud corresponding to the terrain. The terrain data image 80B may be an image of a terrain cluster to which the point cloud corresponding to the terrain belongs. The terrain data image 80B may be an image of three-dimensional information that more closely resembles the terrain, with information between points supplemented based on the point cloud corresponding to the terrain. The display unit 80 may also display the difference between the terrain data and the target construction surface. In the example shown in Figure 11, the display unit 80 displays a difference display section 80D that shows the amount of soil required to reach the target construction surface, the amount of remaining soil to be excavated to reach the target construction surface, and the remaining work time required to reach the target construction surface.

[0058] (Effects of the first invention) [Configuration 1] The effects of the work system 1 shown in Figure 2 are as follows. The work system 1 has a three-dimensional point cloud data acquisition unit 43, a clustering processing unit 71, and a terrain cluster determination unit 72. The three-dimensional point cloud data acquisition unit 43 acquires three-dimensional point cloud data of a work site. The clustering processing unit 71 clusters the point cloud according to the proximity of each point in the three-dimensional point cloud data. The terrain cluster determination unit 72 determines whether or not a cluster (for example, clusters C1 to C3 shown in Figure 6A) obtained by clustering the three-dimensional point cloud data by the clustering processing unit 71 corresponds to the terrain. The terrain cluster determination unit 72 determines whether or not the clustered cluster corresponds to the terrain based on whether at least a portion of the point cloud contained in the clustered cluster is included in terrain area A, which is a predetermined area associated with the work machine 10.

[0059] In the above [Configuration 1], whether a cluster is a terrain cluster is determined by determining whether the clustered cluster has a point cloud included in a predetermined terrain area A. As a result, if an area that is likely to be terrain is set as terrain area A, it becomes possible to determine that the point cloud of a cluster that is continuous with terrain area A is also terrain, and it becomes possible not to determine that a point cloud of an object that is in a space that is not terrain is terrain. As a result, it is possible to improve the accuracy of determining whether or not something is terrain.

[0060] (Effects of the second invention) [Configuration 2] The work system 1 further includes a topographical point cloud extraction unit 73 that extracts only point clouds that belong to clusters that are determined to correspond to the topography.

[0061] In the above [Configuration 2], only the point clouds related to the terrain can be extracted.

[0062] (Effect of the third invention) [Configuration 3] The terrain cluster determination unit 72 further determines that other clusters that include points lower than the height of the point cloud belonging to the cluster determined to correspond to the terrain also correspond to the terrain.

[0063] In the above [Configuration 3], a cluster having a point below the determined topographical cluster is determined to be the topographical cluster. Therefore, even if the clusters corresponding to the topographical features are discontinuous, the topographical cluster can be determined appropriately.

[0064] (Effect of the fourth invention) [Configuration 4] The three-dimensional point cloud data acquisition unit 43 is an imaging device attached to the work machine 10. The work system 1 further has a position information acquisition unit 74, an orientation determination unit 75, and a coordinate conversion unit 76. The position information acquisition unit 74 acquires position information of the work machine 10 relative to the work site. The orientation determination unit 75 acquires azimuth angle information, which is information on the azimuth angle when the three-dimensional point cloud data was acquired, based on the imaging direction of the three-dimensional point cloud data acquisition unit 43. The coordinate conversion unit 76 converts the coordinate system of the three-dimensional point cloud data into the coordinate system of the target construction surface based on the work site, based on the position information and azimuth angle information.

[0065] In the above [Configuration 4], the coordinate system of the three-dimensional point cloud data can be aligned with the coordinate system of the target construction surface, making it easier to use the three-dimensional point cloud data for construction work performed by the work machine 10.

[0066] (Effect of the fifth invention) [Configuration 5] The work system 1 has a display unit 80 that displays topographical data based on a point cloud belonging to a cluster determined to correspond to the topography, and a target construction surface.

[0067] In the above [Configuration 5], the relative position of the target construction surface with respect to the topographical data can be displayed.

[0068] (Effect of the sixth aspect of the invention) [Configuration 6] The work system 1 has a difference calculation unit 77 that calculates the difference between the topographical data and the target construction surface.

[0069] In the above [Configuration 6], the difference between the topographical data and the target construction surface can be quantified. For example, this can be used to calculate the remaining work volume and work time.

[0070] (Variation) The above-described embodiments may be modified in various ways. For example, various examples (including modified examples) of the above-described embodiments may be combined in various ways. For example, the connections of the components shown in FIG. 1 and the like may be changed. For example, the number of components (including modified examples) of the above-described embodiments may be changed, or some of the components may not be provided. For example, the arrangement of the components may be changed. For example, the inclusion relationships of the components may be changed in various ways. For example, a component described as a lower-level component included in a higher-level component may not be included in this higher-level component, but may be included in another component. For example, what is described as multiple different components or parts may be combined into a single component or part. For example, what is described as a single component or part may be provided as multiple different components or parts. For example, the order of the steps in the flowchart shown in FIG. 3 may be changed, or some of the steps may not be performed. For example, each component may have only some of its features (functions, arrangement, shape, operation, etc.). [Explanation of symbols]

[0071] 1: Work system 10: Work machinery 43: Three-dimensional point cloud data acquisition section 71: Clustering processing section 72: Terrain cluster determination section 73: Terrain point cloud extraction part 74: Location information acquisition section 75: Orientation determining section 76: Coordinate conversion section 77: Difference calculation part 80:Display section A: Terrain area

Claims

1. a three-dimensional point cloud data acquisition unit that acquires three-dimensional point cloud data of a work site; a clustering processing unit that clusters the point cloud according to a proximity state of each point of the three-dimensional point cloud data; a topography cluster determination unit that determines whether or not the clusters obtained by clustering the three-dimensional point cloud data in the clustering processing unit correspond to topography; and the terrain cluster determination unit determines whether the clustered cluster corresponds to the terrain based on whether at least a portion of the point cloud included in the clustered cluster is included in a terrain region that is a preset region associated with a work machine. Working system.

2. The work system according to claim 1, The method further includes a topography point cloud extraction unit that extracts only a point cloud that belongs to the cluster determined to correspond to the topography. Working system.

3. The work system according to claim 2, the terrain cluster determination unit further determines that other clusters including points lower than the height of the point cloud belonging to the cluster determined to correspond to the terrain also correspond to the terrain; Working system.

4. The work system according to claim 3, the three-dimensional point cloud data acquisition unit is an imaging device attached to the work machine, a position information acquisition unit that acquires position information of the work machine relative to the work site; an orientation determination unit that acquires azimuth angle information, which is information about an azimuth angle when the three-dimensional point cloud data is acquired, based on the imaging direction of the three-dimensional point cloud data acquisition unit; a coordinate conversion unit that converts a coordinate system of the three-dimensional point cloud data into a coordinate system of a target construction surface based on the work site, based on the position information and the azimuth angle information; further comprising Working system.

5. The work system according to claim 1, a display unit that displays topographical data based on a point cloud belonging to a cluster determined to correspond to the topography and a target construction surface; Working system.

6. The work system according to claim 5, A difference calculation unit is provided to calculate a difference between the topographical data and the target construction surface. Working system.

Images