Yard inspection robot and program

Abstract

To provide a technology to travel automatically along a target route with high precision in order to inspect conveyor belts that transport raw materials stored in a raw material yard.SOLUTION: A yard inspection robot (1) includes a recognition unit (12) that recognizes rails from an image that includes the rails as a subject, an estimation unit (13) that estimates deviation and angle between a traveling direction and a target route, and a calculation unit (14) that calculates angular velocity for automatic travel to follow the target route by referring to the estimated deviation and angle.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to a yard inspection robot and a program.

Background Art

[0002] There is known a technology for recognizing an object existing on a traveling path and automatically traveling based on the recognized result. For example, Patent Document 1 discloses a technology for guiding a vehicle along a position separated by a predetermined shift amount smaller than half of the width between both white lines when one of the white lines cannot be recognized.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] Here, as an example using the technology of automatic traveling, there is also a yard inspection robot that automatically travels along the side of a belt conveyor that transports ore or coal from a raw material yard such as an ore yard or a coal yard and inspects the belt conveyor. The yard inspection robot preferably travels along a target path between the belt conveyor and a rail laid in parallel with the belt conveyor, the rail being for moving a hoist for loading ore or coal onto the belt conveyor. In this case, since the space between the belt conveyor and the rail is narrow, a high position accuracy of, for example, about ±10 to 20 cm is required with respect to the target path.

[0005] However, in the technology described in Patent Document 1, it is not assumed to require a high positional accuracy of about ±10 to 20 cm. Also, in the technology described in Patent Document 1, although a white line is recognized, there may be a case where a part of the rail for moving the crane that transports raw materials to the belt conveyor cannot be recognized because, for example, it is covered with soil. On the other hand, in the technology described in Patent Document 1, a case where a part of the white line cannot be recognized is not assumed. Therefore, when a part of the white line cannot be recognized, the positional accuracy of the vehicle may decrease. Thus, there is a problem in terms of positional accuracy when applying the technology described in Patent Document 1 to the yard inspection robot.

[0006] One aspect of the present invention aims to realize a technology for automatically traveling with high accuracy on a target path in order to inspect a belt conveyor that transports raw materials stored in a raw material yard.

Means for Solving the Problems

[0007] In order to solve the above problems, a yard inspection robot according to one aspect of the present invention is a yard inspection robot for inspecting a belt conveyor that transports raw materials stored in a raw material yard, and includes a rail laid parallel to the belt conveyor, and a recognition unit that recognizes the rail from an image including, as a subject, a rail for moving a crane that transports raw materials to the belt conveyor; an estimation unit that estimates, at each of a plurality of different positions in the traveling direction of the yard inspection robot from the yard inspection robot, a deviation between the traveling direction and a target path based on the rail, a deviation in a direction orthogonal to the traveling direction, and an angle formed by the traveling direction and the target path, based on the image and the rail recognized by the recognition unit; and a calculation unit that refers to the deviation and the angle estimated by the estimation unit and calculates an angular velocity for automatically traveling so that the yard inspection robot follows the target path.

Effects of the Invention

[0008] According to one aspect of the present invention, a technology for automatically traveling with high precision on a target path can be realized in order to inspect a belt conveyor that transports raw materials stored in a raw material yard.

Brief Description of the Drawings

[0009]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Mode for Carrying Out the Invention

[0010] 〔Embodiment 1〕 Hereinafter, an embodiment of the present invention will be described in detail.

[0011] (Outline of Yard Inspection Robot 1) The outline of the yard inspection robot 1 according to the present embodiment will be described with reference to FIG. 2. FIG. 2 is a diagram showing the outline of the yard inspection robot 1.

[0012] The yard inspection robot 1 is an autonomous mobile robot that inspects the belt conveyor BC that transports the raw material MA stored in the raw material yard MY. In FIG. 2, the yard inspection robot 1 automatically travels between one of the two rails RA1 (or RA2) of the rails RA laid parallel to the belt conveyor BC, which is for moving the hoist LE that transports the raw material MA to the belt conveyor BC, and the belt conveyor BC so as to follow the target path TR.

[0013] In FIG. 2, the target path TR of the yard inspection robot 1 is between the rail RA1 and the belt conveyor BC, but the target path TR of the yard inspection robot 1 is not limited to this. The target path TR of the yard inspection robot 1 may be any path based on the belt conveyor BC. As an example, the target path TR of the yard inspection robot 1 may be a path along the belt conveyor BC. Therefore, for example, the target path TR of the yard inspection robot 1 may be between the rail RA1 and the raw material yard MY in FIG. 2, but it is preferably near the belt conveyor BC. For the sake of easy explanation, hereinafter, the case where the yard inspection robot 1 travels between the rail RA1 and the belt conveyor BC will be described as an example.

[0014] (Configuration of Yard Inspection Robot 1) The configuration of the yard inspection robot 1 will be described with reference to FIG. 1. FIG. 1 is a diagram showing the configuration of the yard inspection robot 1. As shown in FIG. 1, the yard inspection robot 1 includes a control unit 10, a storage unit 20, a camera 30, a drive unit 40, and a sensor unit 50.

[0015] The storage unit 20 stores data referred to by the control unit 10. As an example, the storage unit 20 includes an image PIC including the rail RA as a subject and information indicating the target path TR. Examples of the storage unit 20 include, but are not limited to, a flash memory, an HDD (Hard Disk Drive), an SSD (Solid State Drive), or a combination thereof.

[0016] The camera 30 includes the rail RA that serves as a reference for the target path TR as a subject, and captures an image of the front or the rear in the traveling direction of the yard inspection robot 1. The camera 30 may capture images at predetermined intervals, or may capture images according to an instruction from the control unit 10 described later. The camera 30 supplies the captured image PIC to the control unit 10.

[0017] The drive unit 40 drives the yard inspection robot 1. As an example, the drive unit 40 drives the yard inspection robot 1 at the speed and angular velocity based on the commands of the speed and angular velocity supplied from the control unit 10. Examples of the drive unit 40 include, but are not limited to, wheels and motors.

[0018] The sensor unit 50 detects information around the yard inspection robot 1 and supplies a sensor signal indicating the information to the control unit 10. Examples of the sensor unit 50 include, but are not limited to, a camera that captures an image including the belt conveyor BC as a subject, a thermometer that measures the temperature of the belt conveyor BC, and a microphone that detects the sound emitted from the belt conveyor BC.

[0019] (Control unit 10) The control unit 10 controls each component included in the yard inspection robot 1. Also, as shown in FIG. 1, the control unit 10 includes an acquisition unit 11, a recognition unit 12, an estimation unit 13, a calculation unit 14, an inspection unit 15, a setting unit 16, and a drive control unit 17.

[0020] The acquisition unit 11 acquires data. As an example, the acquisition unit 11 acquires the image PIC captured by the camera 30. As another example, the acquisition unit 11 acquires the sensor signal supplied by the sensor unit 50. The acquisition unit 11 stores the acquired data in the storage unit 20.

[0021] The recognition unit 12 is a rail RA laid parallel to the belt conveyor BC, and recognizes the rail RA from an image PIC including the rail RA as a subject for moving a hoist LE that transports the raw material MA to the belt conveyor BC. The recognition unit 12 supplies information regarding the recognized rail RA to the estimation unit 13. An example of the information regarding the rail RA is information indicating the position of the rail RA in the image PIC.

[0022] The method by which the recognition unit 12 recognizes the rail RA from the image PIC is not limited. As an example, the recognition unit 12 differentiates the image PIC in the horizontal direction, discriminates a straight line using a method such as the Hough transform from points where the luminance value changes by a threshold value or more, and recognizes it as the boundary between the rail RA and the ground.

[0023] Based on the image PIC and the rail RA recognized by the recognition unit 12, the estimation unit 13 estimates, at each of a plurality of different positions p (p1 to pn, where n is a natural number of 2 or more) in the traveling direction of the yard inspection robot 1 from the yard inspection robot 1, the deviation between the traveling direction of the yard inspection robot 1 and the target path TR based on the rail RA, which is the deviation in the direction orthogonal to the traveling direction. Further, the estimation unit 13 estimates the angle formed by the traveling direction and the target path TR based on the image PIC and the rail RA recognized by the recognition unit 12. The estimation unit 13 supplies information indicating the estimated deviation and angle to the calculation unit 14. The process by which the estimation unit 13 estimates the deviation and angle will be described later.

[0024] Based on the deviation and angle estimated by the estimation unit 13, the calculation unit 14 calculates the angular velocity for the yard inspection robot 1 to automatically travel so as to follow the target path TR. As an example, the calculation unit 14 calculates the angular velocity at which the deviation estimated by the estimation unit 13 becomes 0 and the angle becomes 0 degrees. The calculation unit 14 supplies information indicating the calculated angular velocity to the drive control unit 17.

[0025] The inspection unit 15 inspects whether there is any abnormality in the belt conveyor BC. As an example, the inspection unit 15 refers to the sensor signal supplied from the sensor unit 50 and inspects whether there is any abnormality in the belt conveyor BC. The inspection unit 15 stores the inspection result in the storage unit 20.

[0026] As an example, when an image including the belt conveyor BC as a subject is supplied from the sensor unit 50, the inspection unit 15 measures the height of the deposit of the raw material MA that has fallen from the belt conveyor BC. If the height of the deposit is higher than a predetermined height, since the driving of the belt conveyor BC may be hindered, the inspection unit 15 determines that an abnormality has occurred in the belt conveyor BC.

[0027] As another example, when a sensor signal indicating the temperature of the belt conveyor BC is supplied from the sensor unit 50, the inspection unit 15 determines whether the temperature of the belt conveyor BC is higher than a predetermined temperature. If the temperature of the belt conveyor BC is higher than a predetermined temperature, since the temperature may be increased due to the wear of the bearing of the belt conveyor BC, the inspection unit 15 determines that an abnormality has occurred in the belt conveyor BC.

[0028] As yet another example, when a sensor signal indicating the sound emitted from the belt conveyor BC is supplied from the sensor unit 50, the inspection unit 15 determines whether the frequency and volume of the sound emitted from the belt conveyor BC are higher than a predetermined frequency and volume. If the frequency and volume of the sound emitted from the belt conveyor BC are higher than a predetermined frequency and volume, since an abnormality may have occurred in the driving of the belt conveyor BC, the inspection unit 15 determines that an abnormality has occurred in the belt conveyor BC.

[0029] In this way, since the inspection unit 15 inspects whether there is no abnormality in the belt conveyor BC, the yard inspection robot 1 can detect an abnormality in the belt conveyor BC.

[0030] The setting unit 16 sets a target path TR based on the belt conveyor BC. As an example, the setting unit 16 sets a path parallel to the rail RA at a predetermined length therefrom as the target path TR. The setting unit 16 stores information indicating the set target path TR in the storage unit 20.

[0031] In this way, since the setting unit 16 sets the target path TR based on the belt conveyor BC, the yard inspection robot 1 can set a target path TR according to the situation such as the length between the belt conveyor BC and the rail RA.

[0032] The drive control unit 17 controls the drive unit 40. As an example, the drive control unit 17 issues a command to the drive unit 40 to drive at the angular velocity calculated by the calculation unit 14. As another example, the drive control unit 17 issues a command to the drive unit 40 to travel in the traveling direction at a predetermined speed.

[0033] (Example of the process of the estimation unit 13 estimating the deviation and the angle) An example of the process of the estimation unit 13 estimating the deviation and the angle will be described with reference to FIGS. 3 to 5. FIG. 3 is a diagram showing a state where the yard inspection robot 1 takes an image PIC. In this example, as shown in FIG. 2, a case will be described as an example where the yard inspection robot 1 automatically travels so as to follow the target path TR at a position d [m] away from the rail RA1 between the rail RA1 and the belt conveyor BC.

[0034] As shown in FIG. 3, let the height from the ground to the center of the lens of the camera 30 of the yard inspection robot 1 be H [m]. Also, let the vertical angle of view of the camera 30 be α [rad] and the horizontal angle of view be β [rad] (not shown in FIG. 3). Also, among the ground along the traveling direction from the center of the lens of the camera 30 and included in the angle of view of the camera 30, the closest point to the yard inspection robot 1 is point p0, and any points on the ground on the opposite side of the yard inspection robot 1 from point p0 are points p1 and p2. That is, a plurality of positions different from each other in the traveling direction of the yard inspection robot 1 from the yard inspection robot 1 are points p0, p1, and p2.

[0035] Also, as shown in FIG. 3, with respect to the lens of the camera 30 of the yard inspection robot 1, let the distance from the lens to point p0 be s0 [m], the distance from the lens to point p1 be s1 [m], and the distance from the lens to point p2 be s2 [m]. In this case, the estimation unit 13 calculates s0 using the following formula (1). s0 = H / tan(α / 2) ···(1) Also, as shown in FIG. 3, let the height (field of view height) at the viewing angle of point p1 be h(s1) [m], and the height at the viewing angle of point p2 be h(s2) [m]. Here, if the distance from the lens of the camera 30 to an arbitrary point pn is sn and the height at the viewing angle is h(sn), the estimation unit 13 calculates h(sn) using the following formula (2). h(sn) = 2sn · tan(α / 2) ···(2) An example of the image PIC taken by the camera 30 of the yard inspection robot 1 shown in FIG. 3 is shown in FIG. 4. FIG. 4 is an example (PIC1) of an image taken by the camera 30 of the yard inspection robot 1.

[0036] As shown in FIG. 4, let the width (field of view width) at the viewing angle of point p1 be w(s1) [m], and the width at the viewing angle of point p2 be w(s2) [m].

[0037] Here, if the width at the viewing angle of an arbitrary point pn is w(sn), the estimation unit 13 calculates w(sn) using the following formula (3). w(sn) = 2sn · tan(β / 2) ···(3) Here, as shown in FIG. 4, the deviation between the traveling direction DM of the yard inspection robot 1 and the target path TR, and the deviation in the direction orthogonal to the traveling direction DM is defined as deviation y. Also, let the angle formed by the traveling direction DM and the target path TR be angle θ.

[0038] First, the estimation unit 13 refers to the image PIC1 and calculates the distance d1 from the point p1 to the rail RA1 in the width direction of the point p1 using the following formula (4). Here, let the total number of pixels in the width direction of the image PIC1 be w_pixel, and the number of pixels in the width direction of the image PIC1 from the point p1 to the rail RA1 be d_pixel1 (the number of pixels corresponding to d in FIG. 3). d1 = {w(s1) / w_pixel} × d_pixel1 ···(4) Further, the estimation unit 13 calculates the distance d1 from the point p1 to the rail RA1 from the estimated values of the deviation y and the angle θ using the following formula (5). d1 = d + s1·sin(θ) + y ···(5) Similarly for the point p2, the estimation unit 13 calculates the number of pixels d_pixel2 from the point p2 to the rail RA1, and calculates the distance d2 from the point p2 to the rail RA1 from the estimated values of the deviation y and the angle θ using the following formulas (4a) and (6). d2 = {w(s2) / w_pixel} × d_pixel2 ···(4a) d2 = d + s2·sin(θ) + y ···(6) By substituting the value of d1 calculated by formula (4) into the left side of formula (5), and substituting the value of d2 calculated by formula (4a) into the left side of formula (6), and solving formulas (5) and (6) for the angle θ and the deviation y, the estimation unit 13 can calculate the angle θ and the deviation y using the following formulas (7a) and (7b). θ = sin -1 {(d2 - d1) / (s2 - s1)} ···(7a) y = (d1·s2 - d2·s1) / (s2 - s1) - d ···(7b) The estimation unit 13 may further calculate the distance dn from each of the plurality of points n to the rail RA1, and estimate the angle θ and the deviation y that minimize the evaluation value J calculated by the least squares method using the following formula (8).

[0039]

Equation

[0040] In this way, the estimation unit 13 estimates the deviation y and the angle θ between the traveling direction DM and the target path TR by the least squares method. Therefore, the estimation unit 13 can accurately estimate the deviation y and the angle θ between the traveling direction DM and the target path TR.

[0041] The estimation unit 13 supplies information indicating the estimated angle θ and deviation y to the calculation unit 14. The calculation unit 14 calculates the angular velocity at which the angle θ becomes 0 degrees and the deviation y becomes 0. As an example, the estimation unit 13 calculates the curvature at which the yard inspection robot 1 is located on the target path TR in front of the distance sj, and calculates the angular velocity from the curvature and the speed of the yard inspection robot 1. The calculation unit 14 supplies information indicating the calculated angular velocity to the drive control unit 17. The drive control unit 17 issues a command to the drive unit 40 so as to drive at the angular velocity calculated by the calculation unit 14.

[0042] After the drive unit 40 is driven based on the command from the drive control unit 17, an example of the image PIC captured by the camera 30 of the yard inspection robot 1 is shown in FIG. 5. FIG. 5 is another example (PIC2) of the image captured by the camera 30 of the yard inspection robot 1.

[0043] As shown in FIG. 5, when the angle θ becomes 0 degrees and the deviation y becomes 0, the traveling direction DM of the yard inspection robot 1 coincides with the target path TR. That is, when the distances d1 from the point p1 to the rail RA1 and d2 from the point p2 to the rail RA2 are calculated using the formulas (5) and (6), respectively, both the distances d1 and d2 are equal to the distance d.

[0044] (Flow of the process executed by the yard inspection robot 1) The flow of the process executed by the yard inspection robot 1 will be described with reference to FIG. 6. FIG. 6 is a flowchart showing the flow of the process executed by the yard inspection robot 1.

[0045] (Step S11) In step S11, the acquisition unit 11 acquires an image PIC including the rail RA as a subject, which is captured by the camera 30. The acquisition unit 11 stores the acquired image PIC in the storage unit 20.

[0046] (Step S12) In step S12, the recognition unit 12 recognizes the rail RA from the image PIC. The recognition unit 12 supplies information regarding the recognized rail RA to the estimation unit 13.

[0047] (Step S13) In step S13, based on the image PIC and the rail RA recognized by the recognition unit 12, the estimation unit 13 estimates, at each of a plurality of different positions p (p1 to pn, where n is a natural number of 2 or more) in the traveling direction of the yard inspection robot 1 from the yard inspection robot 1, the deviation between the traveling direction of the yard inspection robot 1 and the target path TR based on the rail RA, which is the deviation in the direction orthogonal to the traveling direction. Further, the estimation unit 13 estimates the angle formed by the traveling direction and the target path TR based on the image PIC and the rail RA recognized by the recognition unit 12. The estimation unit 13 supplies information indicating the estimated deviation and angle to the calculation unit 14.

[0048] Note that the process of the setting unit 16 setting the target path TR based on the belt conveyor BC may be executed before step S13.

[0049] (Step S14) In step S14, based on the deviation and angle estimated by the estimation unit 13, the calculation unit 14 calculates the angular velocity at which the yard inspection robot 1 automatically travels so as to follow the target path TR. The calculation unit 14 supplies information indicating the calculated angular velocity to the drive control unit 17.

[0050] (Step S15) In step S15, the drive control unit 17 drives the drive unit 40 at the angular velocity calculated by the calculation unit 14.

[0051] Note that the process of the inspection unit 15 inspecting whether an abnormality has occurred in the belt conveyor BC may be executed at an arbitrary timing, or may be executed after any one of the steps shown in FIG. 6 (for example, after step S15).

[0052] (Effect of the yard inspection robot 1) As described above, based on the image PIC captured by the camera 30, the yard inspection robot 1 estimates the deviation y between the target path TR based on the rail RA and the traveling direction of the yard inspection robot 1, and the angle θ formed by the target path TR and the traveling direction of the yard inspection robot 1 at each of a plurality of positions. Then, the yard inspection robot 1 refers to the estimated deviation y and angle θ, and calculates the angular velocity for automatic driving so that the yard inspection robot 1 follows the target path TR.

[0053] For example, when using GPS (Global Positioning System), the position accuracy is about ±1 m. On the other hand, the yard inspection robot 1 calculates the deviation y and the angle θ based on the image PIC captured by the camera 30 provided in the yard inspection robot 1 of the periphery of the yard inspection robot 1. Therefore, the yard inspection robot 1 can automatically travel with a high position accuracy of about ±10 to 20 cm.

[0054] Also, when using RTK (Real Time Kinematic), the position accuracy is about ±2 cm. However, when using RTK, if a correction signal cannot be received from the Internet, the position accuracy deteriorates. That is, when using RTK (Real Time Kinematic), the position accuracy depends on the communication situation. On the other hand, the yard inspection robot 1 calculates the deviation y and the angle θ based on the image PIC captured by the camera 30 provided in the yard inspection robot 1. Therefore, the yard inspection robot 1 can automatically travel with high position accuracy without depending on the communication situation for the position accuracy.

[0055] In addition, in the case of the rail RA installed outdoors, part of it may be covered with soil. Even in such a case, the yard inspection robot 1 estimates the deviation y between the target path TR based on the rail RA and the traveling direction of the yard inspection robot 1 at each of a plurality of positions. Therefore, even when part of the rail RA is covered with soil, the yard inspection robot 1 estimates the deviation y between the target path TR and the traveling direction of the yard inspection robot 1 at the part where the rail RA is not covered with soil, so that the deviation y can be estimated with high accuracy.

[0056] In addition, the yard inspection robot 1 automatically travels so as to follow the target path TR based on the rail RA recognized based on the image PIC. Therefore, the yard inspection robot 1 does not need to create a map of the moving range of the yard inspection robot 1 using, for example, SLAM (Simultaneous Localization and Mapping). Also, the yard inspection robot 1 does not need to use, for example, measurement values by LiDAR (Light Detection And Ranging) to check whether there is a deviation from the target path based on the created map. That is, since the yard inspection robot 1 automatically travels based on the image PIC, it can automatically travel with a simple configuration and high position accuracy.

[0057] (Modification Example 1) In addition to the configuration of FIG. 1, the yard inspection robot 1 may be provided with a configuration for acquiring information indicating the position of the yard inspection robot 1. As an example, the yard inspection robot 1 may be provided with a GNSS (Global Navigation Satellite System) antenna that receives GPS signals. In this case, the estimation unit 13 of the yard inspection robot 1 also refers to the received GPS signal and estimates the deviation y and the angle θ. With this configuration, since the estimation unit 13 refers to the GPS signal in addition to the image PIC, the accuracy of the deviation y and the angle θ can be increased compared to the case of estimating the deviation y and the angle θ from the image PIC.

[0058] (Modification Example 2) In addition to the configuration shown in FIG. 1, the yard inspection robot 1 may be provided with a configuration for correcting the vertical vibration of the yard inspection robot 1. As an example, the yard inspection robot 1 may be provided with a gimbal. As another example, the yard inspection robot 1 may be provided with an IMU (Inertial Measurement Unit), and may correct image blurring based on the measurement values of the IMU and the differences from the front and rear images. With this configuration, even when the yard inspection robot 1 automatically travels on a rough path, after correcting the vertical vibration, the deviation y and the angle θ are estimated, so that the accuracy of the deviation y and the angle θ can be improved.

[0059] In the present invention, the distance d [m] between the target path TR and the rail RA1 may be zero. In this case, the target path TR will coincide with the rail RA1, and the yard inspection robot 1 will travel on the rail RA1.

[0060] (Image example) The target path was set to 1 [m] from the rail, and the yard inspection robot 1 was made to travel automatically. At this time, the image PIC3 taken by the camera 30 is shown in the upper part of FIG. 7. FIG. 7 is a diagram showing an example of an image taken by the camera 30.

[0061] Based on the image PIC3, the estimation unit 13 estimated the deviation y = 0.43 [m] and the angle θ = 1.8 [deg] by the least squares method using Equation (8). The calculation unit 14 calculated the angular velocity such that the deviation y = 0 [m] and the angle θ = 0 [deg] based on the estimated deviation y and angle θ, and the drive control unit 17 drove the yard inspection robot 1. At this time, the image PIC4 taken by the camera 30 is shown in the lower part of FIG. 7. As shown in the image PIC4 of FIG. 7, the yard inspection robot 1 driven based on the estimated deviation y and angle θ automatically traveled so as to follow the target path.

[0062] 〔Example of implementation by software〕 The function of the yard inspection robot 1 (hereinafter referred to as the "device") is a program for causing a computer to function as the device, and can be realized by a program for causing a computer to function as each control block of the device (particularly each part included in the control unit 10).

[0063] In this case, the above device includes, as hardware for executing the above program, a computer having at least one control device (for example, a processor) and at least one storage device (for example, a memory). By executing the above program with this control device and storage device, each function described in the above embodiments is realized.

[0064] The above program may be recorded on one or more computer-readable recording media, which are not transient. This recording medium may or may not be provided in the above device. In the latter case, the above program may be supplied to the above device via any wired or wireless transmission medium.

[0065] Also, part or all of the functions of each of the above control blocks can also be realized by a logic circuit. For example, an integrated circuit in which a logic circuit functioning as each of the above control blocks is formed is also included in the scope of the present invention. In addition to this, for example, it is also possible to realize the functions of each of the above control blocks by a quantum computer.

[0066] The present invention is not limited to the above-described embodiments, and various modifications are possible within the scope shown in the claims. Embodiments obtained by appropriately combining the technical means disclosed in different embodiments are also included in the technical scope of the present invention.

[0067] [Summary] The yard inspection robot according to Aspect 1 of the present invention is a yard inspection robot that inspects a belt conveyor that transports raw materials stored in a raw material yard, and is a rail laid parallel to the belt conveyor, and an image including, as a subject, a rail for moving a hoist that transports raw materials to the belt conveyor. A recognition unit that recognizes the rail from the image, and at each of a plurality of different positions in the traveling direction of the yard inspection robot from the yard inspection robot, a deviation between the traveling direction and a target path based on the rail, a deviation in a direction orthogonal to the traveling direction, and an angle formed by the traveling direction and the target path are estimated based on the image and the rail recognized by the recognition unit. An estimation unit, and a calculation unit that refers to the deviation and the angle estimated by the estimation unit and calculates an angular velocity for the yard inspection robot to automatically travel so as to follow the target path.

[0068] The yard inspection robot according to Aspect 2 of the present invention includes, in the above Aspect 1, an inspection unit that inspects an abnormality of the belt conveyor.

[0069] In the yard inspection robot according to Aspect 3 of the present invention, the estimation unit in the above Aspect 1 or 2 estimates the deviation and the angle between the traveling direction and the target path by the least squares method.

[0070] The yard inspection robot according to Aspect 4 of the present invention includes, in any one of the above Aspects 1 to 3, a setting unit that sets, from the rail, a position having a predetermined length in a direction orthogonal to the rail as the target path.

[0071] The program according to Aspect 5 of the present invention causes a computer to function as the recognition unit, the estimation unit, the calculation unit, the inspection unit, and the setting unit in the yard inspection robot according to any one of the above Aspects 1 to 4.

Explanation of Signs

[0072] 1 Yard inspection robot 10 Control unit 11 Acquisition unit 12 Recognition Unit 13 Estimation Unit 14 Calculation Unit 15 Inspection Unit 16 Setting Unit 17 Drive Control Unit 20 Memory Unit 30 Camera 40 Drive Unit 50 Sensor Unit

Claims

1. A yard inspection robot for inspecting a belt conveyor that transports raw materials stored in a raw material yard, a rail laid parallel to the belt conveyor, the recognition unit recognizing the rail from an image including, as a subject, a rail for moving a hoist that transports raw materials to the belt conveyor, at each of a plurality of different positions in the traveling direction of the yard inspection robot from the yard inspection robot, a deviation between the traveling direction and a target path based on the rail, a deviation in a direction orthogonal to the traveling direction, and an angle formed by the traveling direction and the target path are estimated based on the image and the rail recognized by the recognition unit, a calculation unit that calculates an angular velocity for automatic traveling so that the yard inspection robot follows the target path with reference to the deviation and the angle estimated by the estimation unit, A yard inspection robot comprising:

2. The yard inspection robot according to claim 1, further comprising an inspection unit for inspecting an abnormality of the belt conveyor.

3. The yard inspection robot according to claim 1 or 2, wherein the estimation unit estimates the deviation and the angle between the traveling direction and the target path by a least squares method.

4. The yard inspection robot according to claim 1 or 2, further comprising a setting unit that sets, as the target path, a position at a predetermined length in a direction orthogonal to the rail from the rail.

5. A program for causing a computer to function as the recognition unit, the estimation unit, and the calculation unit in the yard inspection robot according to claim 1.

Images