Inspection system and inspection method

The inspection system dynamically sets prohibited areas based on equipment operation status, improving safety and efficiency by using sensors to adapt safety margins, reducing manual labor and expanding robot navigation.

JP7581114B2Active Publication Date: 2024-11-12KK TOSHIBA
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
JP2021065259
Authority / Receiving Office
JP · JP
Patent Type
Patents
Current Assignee / Owner
Filing Date
2021-04-07
Publication Date
2024-11-12
Estimated Expiration
2041-04-07

AI Technical Summary

Technical Problem

Existing inspection systems using autonomous mobile robots face challenges in accurately setting prohibited areas to avoid equipment contact, which can be overly restrictive or require manual adjustment based on equipment operation status, leading to inefficiencies and increased labor.

Method used

An inspection system that utilizes an environmental map with an area setting unit to dynamically set prohibited areas based on the operating status of equipment, using sensors to identify operation states and adjust safety margins accordingly, reducing manual intervention.

Benefits of technology

The system effectively sets appropriate prohibited areas, enhancing safety and efficiency by allowing robots to navigate more freely while minimizing human intervention and adapting to equipment operation changes.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 0007581114000001
    Figure 0007581114000001
  • Figure 0007581114000002
    Figure 0007581114000002
  • Figure 0007581114000003
    Figure 0007581114000003
Patent Text Reader

Abstract

To provide an inspection technique that is able to appropriately set an inhibition area for inhibiting entry of an inspection robot.SOLUTION: An inspection system 1, which is an environment map used when an inspection robot 2 for inspecting a target device 7 to be inspected moves, includes: a map setting unit 31 that sets an environment map including information indicating a position of the target device 7; an operating state identification unit 32 that identifies an operating state indicating whether the target device 7 is operated or not; and an area setting unit 33 that, based at least on the operating state, sets an inhibition area N for inhibiting the inspection robot 2 entering a periphery of the target device 7 in the environment map.SELECTED DRAWING: Figure 3
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] FIELD OF THE DISCLOSURE Embodiments of the present invention relate to inspection techniques. [Background technology]

[0002] Inspection work at power generation facilities and the like is performed daily by humans patrolling the facilities. However, there are problems such as aging inspectors and labor shortages due to a society with a declining birthrate and an aging population. Therefore, there is a demand to use robots to reduce the burden on inspectors. In recent years, autonomous movement technology by robots has improved significantly, practical autonomous movement has been realized, and the practical use of robots is progressing in various fields. It is expected that robots will also be used for inspection work at power generation facilities in the future. For example, there is a known technology to inspect a specific plant using an autonomous mobile robot such as a drone. When performing such inspections, it is necessary to prevent the autonomous mobile robot from coming into contact with important equipment due to a malfunction or external factors. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] International Publication No. 2019 / 176710 [Non-patent literature]

[0004] [Non-Patent Document 1] E. Marder-Eppstein, DV Lu, D. Hershberger, “costmap_2d” http: / / wiki.ros.org / costmap_2d Summary of the Invention [Problem to be solved by the invention]

[0005] An autonomous mobile robot holds map information of the moving area, and moves while estimating its own position on the map. However, there is a possibility that the accuracy of the movement may vary due to a trouble in the robot itself or an external factor. In addition, there is a possibility that the robot may move unintentionally. Therefore, a technique is known in which a safety margin (prohibited area) is set in an area near a device or an obstacle to prevent the robot from coming into contact with the device or the obstacle. However, if a safety margin is set uniformly for all devices or obstacles, the area through which the robot can pass may be narrowed, and it may become difficult for the robot to pass through. In addition, devices are not always in operation, and if the device is stopped, the risk of the robot coming into contact with the device is low. Therefore, for stopped devices, there are cases where it is not necessary to set a prohibited area similar to that for operating devices. However, it is time-consuming to manually set a prohibited area every time the device is started or stopped. Therefore, there is a demand for a technique for appropriately setting a prohibited area while ensuring safety and reducing the manual effort.

[0006] An embodiment of the present invention has been made in consideration of the above circumstances, and has an object to provide an inspection technique that can appropriately set a prohibited area into which an inspection robot is prohibited from entering. [Means for solving the problem]

[0007] The inspection system according to an embodiment of the present invention is an environmental map used by an inspection robot that inspects a target device to be inspected when the inspection system moves, and includes a map setting unit that sets the environmental map including information indicating the position of the target device, an operation status identification unit that identifies an operation status indicating whether the target device is operating, and an area setting unit that sets a prohibited area around the target device on the environmental map based on at least the operation status, the prohibited area prohibiting the inspection robot from entering the target device, a type registration unit which registers type information indicating the type of the inspection robot, the type being distinguished by the moving mechanism mounted thereon; and an attribute registration unit which registers attribute information indicating the influence that the target device has on the surroundings, the influence being different depending on the type of the inspection robot; Equipped with The region setting unit sets the prohibited region based on the type information and the attribute information. . Effect of the Invention

[0008] According to an embodiment of the present invention, an inspection technique is provided that can appropriately set a prohibited area into which an inspection robot is prohibited from entering. [Brief description of the drawings]

[0009] [Figure 1] FIG. 1 is a configuration diagram showing an inspection system according to a first embodiment. [Diagram 2] FIG. 2 is a block diagram showing a management server according to the first embodiment. [Diagram 3] FIG. 1 is a block diagram showing an inspection robot according to a first embodiment. [Figure 4] FIG. 4 is an explanatory diagram showing a target device management table according to the first embodiment. [Diagram 5] FIG. [Figure 6] FIG. 4 is a perspective view showing prohibited areas of a first distance and a second distance. [Figure 7] FIG. 11 is an image showing the surface of a control panel photographed by a camera. [Figure 8] An image showing a generator captured by a thermal camera. [Figure 9] 6 is a flowchart showing an inspection process performed by an inspection robot. [Figure 10] FIG. 11 is a configuration diagram showing an inspection system according to a second embodiment. [Figure 11] FIG. 11 is a block diagram showing a management server according to a second embodiment. [Figure 12] FIG. 11 is a block diagram showing an inspection robot according to a second embodiment. [Figure 13] FIG. 11 is an explanatory diagram showing an inspection robot management table according to the second embodiment. [Figure 14] FIG. 11 is an explanatory diagram showing a target device management table according to the second embodiment; DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0010] (First embodiment) Hereinafter, embodiments of an inspection system and an inspection method will be described in detail with reference to the drawings. First, an inspection system and an inspection method according to a first embodiment will be described with reference to FIGS.

[0011] 1, reference numeral 1 denotes an inspection system according to a first embodiment. This inspection system 1 includes an inspection robot 2, a management server 3, and a management terminal 4. These are connected to each other via a predetermined network 5 (communication line).

[0012] The inspection system 1 is used to inspect various facilities installed in a predetermined plant 6 such as a power plant. Examples of the predetermined plant 6 include a power plant, a chemical plant, and a factory. In an inspection area of ​​such a plant 6, a plurality of target devices 7 to be inspected are provided. Examples of the target devices 7 include a generator, a control panel, a cooler, a distribution board, and a transformer facility. These target devices 7 are inspected using an inspection robot 2 arranged in the inspection area.

[0013] The inspection robot 2 can move on the ground using, for example, wheels. The inspection robot 2 is an unmanned mobile robot that has a self-location estimation function and can move autonomously to inspect the target device 7. The inspection robot 2 can move automatically, and can also be manually moved by remote control by the manager M.

[0014] Under normal circumstances, the inspection robot 2 periodically patrols the inspection area of ​​the plant 6 to inspect the target devices 7. Here, each target device 7 is connected to a management server 3 via a network 5. The operating status of each target device 7 is managed by the management server 3. The inspection robot 2 is also connected to the management server 3 via the network 5. The operation of the inspection robot 2 is managed by the management server 3.

[0015] Moreover, if an abnormality occurs in the target device 7, the inspection robot 2 rushes to inspect the target device 7. Therefore, the management server 3 can quickly grasp the failure or signs of failure in the important target device 7, and can minimize the subsequent adverse effects.

[0016] The manager M in the management office 8, which is located in a remote location from the plant 6, can remotely operate the inspection robot 2 and inspect the target equipment 7 through this remote operation. The manager M can also grasp the status of the target equipment 7 (presence or absence of anomalies) through the management server 3.

[0017] The management office 8 is provided with a management terminal 4 operated by the manager M. This management terminal 4 is configured with a predetermined computer such as a desktop PC, a notebook PC, or a tablet PC. In this embodiment, a desktop PC is illustrated. The management terminal 4 is connected to a display 9 which the manager M visually confirms, and a remote control terminal 10 which the manager M uses when remotely controlling the inspection robot 2. The remote control terminal 10 is provided with a joystick which the manager M uses when operating the inspection robot 2.

[0018] In the input section of the management terminal 4, predetermined information is input in response to an operation by an administrator M who uses the management terminal 4. This input section includes input devices such as a mouse or a keyboard. That is, the predetermined information is input to the management terminal 4 in response to an operation of these input devices. Furthermore, the predetermined information is input to the inspection robot 2 via this management terminal 4.

[0019] The output unit of the management terminal 4 outputs predetermined information. For example, the output unit controls an image displayed on the display 9.

[0020] In this embodiment, the display 9 is exemplified as a device that displays an image, but other modes may be used. For example, information may be displayed using a head-mounted display or a projector. Furthermore, a printer that prints information on a paper medium may be used instead of the display 9. In other words, the targets controlled by the output unit may include a head-mounted display, a projector, or a printer.

[0021] The management terminal 4 is connected to the inspection robot 2 and the management server 3 via a network 5. When the manager M remotely controls the inspection robot 2, communication is established between the management terminal 4 and the inspection robot 2 for sending and receiving signals for remote control. The manager M can then manually control the inspection robot 2 using the remote control terminal 10. Note that communication between the management terminal 4 and the inspection robot 2 may be established via the management server 3, and the inspection robot 2 may be remotely controlled via this management server 3.

[0022] The predetermined network 5 is exemplified by the Internet. The network 5 may be a LAN (Local Area Network), a WAN (Wide Area Network), or a mobile (cell phone) communication network. The inspection robot 2 is connected to the network 5 by wireless communication. The target device 7 is connected to the network 5 by wireless communication or wired communication.

[0023] The inspection robot 2 can measure its own position and orientation. For example, the self-position can be measured using a known satellite positioning system such as a global positioning system (GPS). The inspection robot 2 can measure its own position and orientation using simultaneous localization and mapping (SLAM) or visual simultaneous localization and mapping (VSLAM). The inspection robot 2 may also measure its own position and orientation using structure from motion (SfM) or the like.

[0024] 5 is a plan view of the inspection area of ​​the plant 6, and is also a layout drawing 11 of the inspection area. This layout drawing 11 is stored as an environmental map in the inspection robot 2. Note that, in this embodiment, the layout drawing 11 is illustrated as a plan view of the inspection area, but the layout drawing 11 may be 3D data that records the inspection area three-dimensionally.

[0025] As shown in this layout drawing 11, a large number of target devices 7 are arranged in an inspection area of ​​a plant 6. In this inspection area, a permitted area E is set which permits the entry (passage) of the inspection robot 2. In addition, around each target device 7, a prohibited area N is set which prohibits the entry of the inspection robot 2. Note that the prohibited area N is a planar area (two-dimensional area) in the vicinity of the periphery of the target device 7.

[0026] When the inspection robot 2 is used to automatically inspect the target device 7, the manager M can use the management terminal 4 to arbitrarily set the location where the inspection robot 2 will arrive. For example, the manager M specifies the desired arrival location of the inspection robot 2 in the layout drawing 11. The inspection robot 2 automatically generates a travel route R from the current location to the arrival location. Then, the inspection robot 2 autonomously moves along this travel route R to the arrival location.

[0027] Next, the prohibited area N will be described. In this embodiment, the inspection robot 2 identifies the operation status of each target device 7. Then, the range of the prohibited area N is set according to the operation status. For example, a prohibited area N having a different range is set for an operating target device 7 that is estimated to have a large adverse effect on the plant 6 when the inspection robot 2 comes into contact with it, and a prohibited area N having a different range is set for a non-operating target device 7 that is estimated to have a small adverse effect when the inspection robot 2 comes into contact with it.

[0028] As shown in Fig. 6, it is assumed that there is a target device 7A that is in operation and a target device 7B that is not in operation. Here, a prohibited area N is set within a range of a first distance L1 from the surface of the target device 7A that is in operation. Also, a prohibited area N is set within a range of a second distance L2 from the surface of the target device 7B that is not in operation. In this case, the second distance L2 is set to be shorter than the first distance L1.

[0029] In other words, the prohibited area N of the operating target device 7A is set to be wider than the prohibited area N of the non-operating target device 7B. In this way, a wider prohibited area N can be set for the operating target device 7A that is estimated to be adversely affected if the inspection robot 2 comes into contact with it. This reduces the possibility that the inspection robot 2 will come into contact with it.

[0030] Furthermore, by narrowing the range of the prohibited area N for the non-operating target device 7B, it is possible to widen the permitted area E through which the inspection robot 2 can pass. Note that the prohibited area N for the non-operating target device 7B may be eliminated. In other words, the second distance L2 may be set to zero.

[0031] In this manner, in this embodiment, the prohibited area N can be appropriately set according to the operating status of the target device 7. Furthermore, since the prohibited area N is automatically set according to the operating status, it is possible to save the manager M the trouble of individually setting the prohibited area N for each target device 7. Furthermore, since the range of the prohibited area N is set according to the operating status, an appropriate permitted area E taking safety into consideration is set, and the inspection robot 2 can pass through the permitted area E.

[0032] In addition, the target equipment 7A in operation includes not only the target equipment 7A that was in operation before the inspection robot 2 began its patrol inspection, but also the target equipment 7A that is scheduled to start operation during the inspection robot 2's patrol inspection.

[0033] Next, the system configuration of the inspection system 1 will be described with reference to the block diagrams shown in FIGS.

[0034] 2, the management server 3 includes a storage unit 13, a communication unit 14, and a server control unit 15. The management server 3 is configured as a computer having hardware resources such as a CPU, a ROM, a RAM, and a HDD, and in which information processing by software is realized using the hardware resources by the CPU executing various programs. Furthermore, the inspection method of this embodiment is realized by causing the computer to execute various programs.

[0035] Each component of the management server 3 does not necessarily have to be provided in one computer. For example, one system may be realized using a plurality of computers connected to each other via the network 5.

[0036] The storage unit 13 stores various information required when inspecting the target device 7. For example, it stores images taken by the inspection robot 2 or data acquired by the inspection robot 2. The storage unit 13 may also include a database. A database is a collection of information stored in a memory or HDD and organized so as to be searchable or accumulateable.

[0037] The communication unit 14 communicates with the inspection robot 2 and the management terminal 4 via the network 5. In addition, the communication unit 14 can be used to communicate with the target device 7.

[0038] The server control unit 15 comprehensively controls the management server 3. The server control unit 15 includes a target device management unit 16 and a robot management unit 17. These are realized by the CPU executing programs stored in the memory or HDD.

[0039] The target equipment management unit 16 manages the target equipment 7 provided in the inspection area of ​​the plant 6. For example, it controls the operation of the target equipment 7, such as starting or stopping it. Furthermore, it can receive data including a detection signal detected by a sensor (not shown) of the target equipment 7, and record the state of the target equipment 7 based on this data.

[0040] The robot management unit 17 manages the inspection robot 2 provided in the inspection area of ​​the plant 6. For example, it sets an environmental map including information indicating the position of the target device 7. Note that the same environmental map as that set by the robot management unit 17 is set in the inspection robot 2.

[0041] As shown in FIG. 3, the inspection robot 2 includes a camera 20, a thermo camera 21, a microphone 22, a magnetic sensor 23, a three-dimensional measurement sensor 24, a motion sensor 25, a movement mechanism unit 26, a memory unit 27, a communication unit 28, and a robot control unit 30.

[0042] The camera 20, the thermo camera 21, the microphone 22, and the magnetic sensor 23 are mounted on the inspection robot 2 and are sensors used for inspecting the target device 7. These sensors are used to detect the temperature, vibrations, and the like of the target device 7. These sensors are then used to detect events occurring in the target device 7, and the presence or absence of an abnormality in the target device 7 is ascertained based on the detection results. Note that the inspection robot 2 may be equipped with sensors other than these sensors. For example, a thermometer, a vibration meter, a pressure gauge, and the like may be equipped.

[0043] The camera 20 uses visible light or infrared light to capture images of the target device 7 in the vicinity of the inspection robot 2. The camera 20 is configured as an RGB camera or an infrared camera. The images captured by the camera 20 are stored in the storage unit 27 and are also sent to the management server 3.

[0044] The thermo camera 21 captures an image showing the temperature distribution of the target device 7 in the vicinity of the inspection robot 2. The image captured by the thermo camera 21 is stored in the storage unit 27 and is also sent to the management server 3.

[0045] The microphone 22 collects sounds around the inspection robot 2. The sounds detected by the microphone 22 include sounds emitted from the target device 7 in the vicinity of the inspection robot 2. The sounds can be used to determine vibrations or abnormalities in the target device 7. For example, based on whether or not sound is emitted from the target device 7, it can be determined whether the target device 7 is in operation or not. The sounds detected by the microphone 22 are stored in the storage unit 27 and sent to the management server 3.

[0046] The magnetic sensor 23 detects magnetism around the inspection robot 2. The magnetism detected by this magnetic sensor 23 includes magnetism emitted from the target device 7. For example, if the target device 7 is a generator or an electric motor, it is possible to know whether it is operating or not based on the magnetism emitted. The magnetic value detected by the magnetic sensor 23 is stored in the storage unit 27 and is also sent to the management server 3.

[0047] The three-dimensional measuring sensor 24 is mounted on the inspection robot 2 and measures the three-dimensional shapes of objects around the inspection robot 2. For example, a depth sensor is used as the three-dimensional measuring sensor 24. Note that an infrared sensor or a laser sensor such as LiDAR may also be used as the three-dimensional measuring sensor 24.

[0048] The three-dimensional measuring sensor 24 can measure the distance from the inspection robot 2 to a surrounding object, for example, by projecting a laser onto the object and receiving the reflected light with a light receiving element. The shooting direction of the camera 20 and the measurement direction of the three-dimensional measuring sensor 24 are the same. Furthermore, the three-dimensional measuring sensor 24 measures the distance from the inspection robot 2 to the surrounding object using a ToF (Time of Flight) method that converts the delay time between the projected pulse and the received pulse into distance. By using the three-dimensional measuring sensor 24, it is possible to generate three-dimensional point cloud data that includes information on the distance from the inspection robot 2 to the surrounding object and its shape.

[0049] The motion sensor 25 is a nine-axis sensor that combines an inertial sensor (a three-axis acceleration sensor and a three-axis angular velocity sensor) and a three-axis geomagnetic sensor. This motion sensor 25 is mounted on the inspection robot 2 and detects the acceleration that occurs when the inspection robot 2 moves. This motion sensor 25 can also detect gravitational acceleration. Furthermore, the motion sensor 25 detects angular velocity that occurs when the attitude (body orientation) of the inspection robot 2 changes. The attitude of the inspection robot 2 can also be grasped by geomagnetism. The acceleration values ​​and angular velocity values ​​detected by the motion sensor 25 are input to the robot control unit 30. The inspection robot 2 can measure its own position and attitude using this motion sensor 25.

[0050] The movement mechanism unit 26 rotates the wheels required for traveling of the inspection robot 2. The movement mechanism unit 26 is composed of, for example, a motor that drives the wheels, or an actuator that changes the direction of the wheels.

[0051] The storage unit 27 stores various information necessary for inspecting the target device 7. For example, it stores images captured by the camera 20, data acquired by various sensors, and the like.

[0052] The storage unit 27 also stores target device information on the target device 7, movement plan information on the movement of the inspection robot 2, patrol inspection plan information on patrol inspection, etc. The movement plan information includes information on the movement speed and movement range of the inspection robot 2.

[0053] In this embodiment, the same information as that stored in the inspection robot 2 is also stored in the management server 3. The information stored in the inspection robot 2 and the information stored in the management server 3 are synchronized with each other. The manager M can also manage the inspection robot 2 via the management server 3.

[0054] The inspection robot 2 moves autonomously based on the information stored in the storage unit 27, and performs unmanned inspection of each target device 7 in the inspection area.

[0055] The communication unit 28 communicates with the management server 3 and the management terminal 4 via the network 5. The manager M can operate or manage the inspection robot 2 via the network 5. Therefore, the manager M does not need to visit the inspection area of ​​the plant 6, and labor and management costs can be reduced.

[0056] The robot control unit 30 comprehensively controls the inspection robot 2. The robot control unit 30 includes a map setting unit 31, an operating status identification unit 32, an area setting unit 33, a route generation unit 34, a setting change unit 35, and an autonomous movement control unit 36. These are realized by the CPU executing programs stored in the memory or HDD.

[0057] The map setting unit 31 sets an environmental map (map information) including information indicating the position of the target device 7. This environmental map is stored in the storage unit 27 and is used when the inspection robot 2 moves autonomously. For example, this environmental map is used to generate a movement path. In addition, an environmental map similar to the one possessed by the inspection robot 2 is stored in the management server 3.

[0058] The operating status identification unit 32 identifies an operating status indicating whether the target device 7 is operating or not, based on information acquired by the camera 20, the thermo camera 21, the microphone 22, or the magnetic sensor 23. In this way, the operating status of the target device 7 can be autonomously identified by various sensors mounted on the inspection robot 2.

[0059] The target device 7 has different characteristics when it is in operation and when it is not in operation. For example, the moving parts of the target device 7 are stopped when it is not in operation. In addition, there are target devices 7 that are in a live state when they are in operation and generate a magnetic field, but are in a non-live state when they are not in operation and do not generate a magnetic field. In addition, there are target devices 7 that become hot when they are in operation but become cold when they are not in operation.

[0060] For example, at the stopping position of the inspection robot 2, it is possible to identify whether the target device 7 in the vicinity is operating or not based on the difference in sound level between operating and non-operating. Also, at the stopping position of the inspection robot 2, it is possible to detect whether the high voltage is in a live or non-live state based on the value of the magnetic sensor 23. It is possible to identify whether the target device 7 in the vicinity is operating or not.

[0061] As shown in FIG. 7, the inspection robot 2 takes an image of the panel surface of a control panel 40 as a target device 7 with the camera 20. The panel surface of the control panel 40 is provided with a display unit 41, a meter 42, and an indicator light 43. The display unit 41 displays characters indicating whether the control panel 40 is in operation or stopped. The meter 42 displays a predetermined measurement value. The indicator light 43 emits a different light color when the control panel 40 is in operation and when it is stopped. The operation status identification unit 32 recognizes the state of the panel surface of the control panel 40 included in the image by using a predetermined image processing technique. For example, an image of the panel surface of the control panel 40 in operation is compared with an image of the panel surface of the control panel 40 in non-operation, and a difference between these images is detected. Then, it is identified whether the control panel 40 (target device 7) is in operation or not.

[0062] As shown in FIG. 8, the inspection robot 2 takes an image of a generator 44 as the target device 7 with the thermo camera 21. In the image of the thermo camera 21, a heat-generating part 45 that is hotter than other parts is displayed in a color different from the other parts. For example, when the generator 44 is operating, the heat-generating part 45 appears, and when the generator 44 is not operating, the heat-generating part 45 does not appear. The operating status identification unit 32 identifies whether the generator 44 (target device 7) is operating or not based on the presence or absence of the heat-generating part 45 included in the image or its temperature. Also, the image of the generator 44 in operation is compared with the image of the generator 44 not in operation, and the difference between these images is detected. Then, it identifies whether the generator 44 (target device 7) is operating or not.

[0063] The robot control unit 30 (operation status identification unit 32) of this embodiment may include a computer equipped with artificial intelligence (AI) that performs machine learning. The robot control unit 30 may also include a deep learning unit that extracts a specific pattern from multiple patterns based on deep learning. The operation status of the target device 7 may be identified by a classifier generated by machine learning.

[0064] The analysis using the computer of this embodiment can use an analysis technique based on learning of artificial intelligence. For example, a learning model generated by machine learning using a neural network, a learning model generated by other machine learning, a deep learning algorithm, a mathematical algorithm such as regression analysis, etc. In addition, the form of machine learning includes forms such as clustering and deep learning.

[0065] The system of the present embodiment includes a computer equipped with artificial intelligence that performs machine learning. For example, the system may be configured with one computer equipped with a neural network, or may be configured with multiple computers equipped with neural networks.

[0066] Here, a neural network is a mathematical model that represents the characteristics of brain functions through computer simulation. For example, this model shows that artificial neurons (nodes) that form a network through synaptic connections change the strength of synaptic connections through learning, and acquire problem-solving abilities. Furthermore, neural networks acquire problem-solving abilities through deep learning.

[0067] For example, a neural network is provided with an intermediate layer having six layers. Each intermediate layer is composed of 300 units. In addition, by having the multi-layered neural network learn in advance using learning data, it is possible to automatically extract features in the pattern of changes in the state of a circuit or system. In addition, the multi-layered neural network can be configured on the user interface with any number of intermediate layers, any number of units, any learning rate, any number of learning times, and any activation function.

[0068] In addition, deep reinforcement learning may be used in the neural network, in which a reward function is set for each information item to be learned, and the information item with the highest value is extracted based on the reward function.

[0069] For example, we use a Convolution Neural Network (CNN), which has a proven track record in image recognition. In this CNN, the intermediate layer is composed of a convolution layer and a pooling layer. The convolution layer obtains a feature map by applying filtering to nearby nodes in the previous layer. The pooling layer further reduces the feature map output from the convolution layer to create a new feature map. At this time, by obtaining the maximum value of the pixels contained in the region of interest in the feature map, it is possible to absorb slight deviations in the position of the feature amount.

[0070] The convolution layer extracts local features of an image, and the pooling layer performs processing to consolidate local features. These processes reduce the size of the image while maintaining the features of the input image. In other words, CNN can significantly compress (abstract) the amount of information contained in an image. Then, the input image can be recognized and classified using the abstracted image stored in the neural network.

[0071] Note that there are various methods for deep learning, such as autoencoder, recurrent neural network (RNN), long short-term memory (LSTM), generative adversarial network (GAN), etc. These methods may be applied to the deep learning of the present embodiment.

[0072] 4, a target equipment management table is stored in the storage unit 27 of the inspection robot 2. The operation status identification unit 32 registers information indicating the operation status of the identified target equipment 7 in this target equipment management table. This target equipment management table is also stored in the management server 3. The respective target equipment management tables are synchronized with each other.

[0073] The target device management table has an item of the operating status of the target device 7 and an item of the prohibited area set around the target device 7, in association with the target device ID.

[0074] In the item of the target device ID, a target device ID is registered, which is identification information capable of individually identifying each target device 7. A unique target device ID is assigned to each target device 7. These target device IDs are the primary keys of the target device management table.

[0075] For example, when the inspection robot 2 photographs the target device 7 with the camera 20, the position of the target device 7 can be acquired based on the environmental map and the self-position. The operation status identification unit 32 acquires the target device ID of the target device 7 based on the position of the target device 7.

[0076] The operation status field registers the current operation status of the target device 7 based on the operation status of the target device 7 identified during inspection by the inspection robot 2. That is, information indicating whether the target device 7 is in operation or not is registered.

[0077] In addition, in this operation status item, the operation status scheduled in advance for the time period when the inspection robot 2 will perform the inspection may be registered. The information registered in this operation status item may be changed as appropriate during the inspection by the inspection robot 2. Furthermore, the operation status of the target device 7 identified during the previous inspection may be registered. Then, in the next inspection, these registered operation statuses may be used.

[0078] In the prohibited area item, the range of the prohibited area N set around the target device 7 is registered. For example, a "first distance" is registered in association with the target device ID of a target device 7 that is in operation. Also, a "second distance" is registered in association with the target device ID of a target device 7 that is not in operation. Note that, depending on a predetermined condition, a "second distance" may be registered in association with the target device ID of a target device 7 that is in operation.

[0079] In addition, in the target device management table, information indicating the position (coordinates) of the target device 7 on the environmental map, information regarding the shape, size, and volume of the target device 7, etc. may be registered in association with the target device ID.

[0080] 3, the area setting unit 33 sets a prohibited area N around the target device 7 on the environmental map, where the entrance of the inspection robot 2 is prohibited, based on at least the operating status of the target device 7. This area setting unit 33 sets a prohibited area N around the target device 7 on the environmental map, where the entrance of the inspection robot 2 is prohibited, based on information registered in the prohibited area item of the target device management table.

[0081] For example, as shown in Fig. 6, the area setting unit 33 sets a prohibited area N within a range of a first distance L1 from the surface of an operating target device 7A. On the other hand, the prohibited area N is set within a range of a second distance L2 from the surface of a non-operating target device 7B. In this way, a wide prohibited area N can be set for the operating target device 7A that is estimated to be adversely affected when the inspection robot 2 comes into contact with it. In addition, by narrowing the range of the prohibited area N for the non-operating target device 7B, the permitted area E through which the inspection robot 2 can pass can be expanded.

[0082] The area setting unit 33 may set the prohibited area N based on the operation status scheduled in advance for the time period when the inspection robot 2 will perform inspection, or may set the prohibited area N based on the operation status changed during inspection by the inspection robot 2. In other words, the setting of the prohibited area N may be changed during inspection by the inspection robot 2.

[0083] The path generating unit 34 generates a movement path R of the inspection robot 2 when sequentially inspecting the target devices 7. This movement path R is set on the environmental map. Then, the inspection robot 2 moves according to this movement path R.

[0084] The movement route R may be generated by the management server 3. That is, the management server 3 may include a route generation unit 34. Then, the movement route R received from the management server 3 is set in the storage unit 27 of the inspection robot 2. The inspection robot 2 may move autonomously along this set movement route R.

[0085] The setting change unit 35 performs a process of resetting the range of the prohibited area N to an appropriate range. For example, when the prohibited area N' corresponding to the movement route R is set to a range of a first distance L1 from the target device 7', the setting change unit 35 performs a process of resetting the prohibited area N' to a range of a second distance L2 from the target device 7'.

[0086] For example, as shown in FIG. 5, when the movement route R is set, it is assumed that a prohibited area N' exists near the movement route R. When the target devices 7' corresponding to the prohibited areas N' arranged on either side of the movement route R are in operation, the prohibited area N' is set within a range of a first distance L1 from the target devices 7'. However, if such a prohibited area N' exists, the area of ​​the movement route R is narrowed, making it difficult for the inspection robot 2 to travel. Therefore, the setting change unit 35 performs a process of re-setting these prohibited areas N' to a range of a second distance L2 from the target devices 7'. In this way, the prohibited area N' is less likely to interfere with the movement route R of the inspection robot 2. Furthermore, it is possible to avoid a situation in which the area of ​​the movement route R is narrowed by the prohibited area N'.

[0087] The autonomous movement control unit 36 ​​controls the autonomous movement of the inspection robot 2. The autonomous movement control unit 36 ​​estimates the self-position of the inspection robot 2 and controls the movement of the inspection robot 2 by referring to an environmental map.

[0088] Next, the inspection process executed by the inspection robot 2 will be described with reference to the flowchart of Fig. 9. The above-mentioned drawings will be referred to as appropriate. The following steps are at least a part of the processes included in the inspection process, and other steps may be included in the inspection process.

[0089] First, in step S1, the map setting unit 31 of the robot control unit 30 executes an environmental map setting process. In this environmental map setting process, an environmental map including information indicating the position of the target device 7 is set. This environmental map is used when the inspection robot 2 moves. The same environmental map is also set in the management server 3.

[0090] In the next step S2, the path generating unit 34 of the robot control unit 30 executes a path generating process. In this path generating process, a movement path R of the inspection robot 2 is generated. This movement path R is set on the environmental map. Then, when the inspection by the inspection robot 2 starts, the inspection robot 2 moves according to this movement path R. During the inspection, the inspection robot 2 acquires various pieces of information using the camera 20, the thermal camera 21, the microphone 22, or the magnetic sensor 23.

[0091] In the next step S3, the operating status identification unit 32 of the robot control unit 30 executes an operating status identification process. In this operating status identification process, the operating status indicating whether the target device 7 is operating or not is identified based on the information acquired by the camera 20, the thermal camera 21, the microphone 22, or the magnetic sensor 23. In addition, the operating status identification unit 32 registers information indicating the identified operating status of the target device 7 in the target device management table (FIG. 4).

[0092] In the next step S4, the area setting unit 33 of the robot control unit 30 executes an area setting process. In this area setting process, a prohibited area N (FIG. 5) that prohibits the inspection robot 2 from entering is set around the target equipment 7 on the environmental map based on the operation status of the target equipment 7 registered in the target equipment management table (FIG. 4) and the range of the prohibited area N. Note that the prohibited area N may be set according to the planned operation status of the target equipment 7 before the inspection by the inspection robot 2. Then, the range of the prohibited area N may be set based on the operation status of the target equipment 7 identified during the inspection by the inspection robot 2.

[0093] In the next step S5, the setting change unit 35 of the robot control unit 30 executes a setting change process. In this setting change process, when the prohibited area N' corresponding to the movement route R is set within a range of a first distance L1 from the target device 7', a process of resetting the prohibited area N' to a range of a second distance L2 from the target device 7' is performed. This setting change process is performed during inspection by the inspection robot 2. Note that, if the existence of the prohibited area N' corresponding to the movement route R is predicted before the start of inspection by the inspection robot 2, this setting change process may be performed before the start of inspection by the inspection robot 2, and the range of the prohibited area N' may be reconfigured to an appropriate range.

[0094] Then, the inspection robot 2 ends the inspection process. Note that this inspection process is a process that is repeated at regular time intervals. By repeating this process, the inspection method is executed by the inspection robot 2. Note that this process may be executed by interrupting the execution of another main process by the inspection robot 2.

[0095] The risk of the inspection robot 2 coming into contact with something changes depending on whether it is in operation or not. In this embodiment, the prohibited area N is set in accordance with such fluctuations in risk, making it possible to expand the area through which the inspection robot 2 can pass. This makes it possible to inspect or patrol many locations.

[0096] In the first embodiment, the operation status of the target device 7 is determined using various sensors mounted on the inspection robot 2, and an appropriate prohibited area N is set according to the operation status. This saves the manager M the trouble of setting the prohibited area N according to the operation status of the target device 7, and allows the permitted area E through which the inspection robot 2 can pass to be expanded while taking safety into consideration.

[0097] Second embodiment Next, an inspection system 1A and an inspection method according to a second embodiment will be described with reference to Fig. 10 to Fig. 14. Note that the same components as those shown in the above-described embodiment will be denoted by the same reference numerals and duplicated descriptions will be omitted.

[0098] As shown in FIG. 10, the inspection system 1A of the second embodiment is provided with a flying inspection robot 2A in addition to a mobile inspection robot 2. The flying inspection robot 2A is configured with a drone equipped with a flying mechanism such as a propeller. The inspection robot 2A moves through the air to inspect the target device 7. In this case, the movement path R (FIG. 5) becomes the flight path of the inspection robot 2A. Also, the prohibited area N (FIG. 6) becomes the three-dimensional space in the vicinity of the periphery of the target device 7.

[0099] In the inspection system 1A, inspection is performed using a mixture of a mobile inspection robot 2 and a flying inspection robot 2A. An appropriate prohibited area N is set for each of the inspection robots 2, 2A. In other words, the prohibited area N set for one target device 7 differs depending on the inspection robot 2, 2A.

[0100] In the second embodiment, the management server 3 transmits to the inspection robots 2, 2A management information indicating the operation status of each target device 7. The inspection robots 2, 2A identify the operation status of the target device 7 based on the received management information and set a prohibited area N.

[0101] 11, the server control unit 15 of the management server 3 includes a type registration unit 50 and an attribute registration unit 51 in addition to the configuration of the first embodiment described above. These are realized by the CPU executing a program stored in the memory or HDD.

[0102] The type registration unit 50 registers type information indicating the type of the inspection robot 2, 2A, which is distinguished by the moving mechanism mounted thereon.

[0103] The attribute registration unit 51 registers attribute information indicating the attributes of the target device 7 that are affected differently depending on the type of the inspection robot 2, 2A, and indicating the influence that the target device 7 has on the surroundings.

[0104] 13, an inspection robot management table is stored in the storage unit 13 of the management server 3. The inspection robot management table has an item for the type of the inspection robot 2, 2A in association with the robot ID. A type registration unit 50 registers type information in the inspection robot management table.

[0105] In the robot ID field, a robot ID is registered as identification information that can individually identify each of the inspection robots 2 and 2A. A unique robot ID is assigned to each of the inspection robots 2 and 2A. These robot IDs are the primary keys of the inspection robot management table.

[0106] 12, the robot control unit 30 of the inspection robot 2, 2A includes a situation information acquisition unit 52 in addition to the configuration of the first embodiment described above. This is realized by the CPU executing a program stored in the memory or HDD.

[0107] The status information acquisition unit 52 acquires management information indicating the operating status of the target device 7 sent from the management server 3. This management information includes information indicating whether the target device 7 is in operation or not in operation in association with the target device ID of each target device 7.

[0108] The operating status identification unit 32 of the second embodiment identifies the operating status of the target device 7 based on the management information. In this way, the operating status of the target device 7 can be identified based on the management information sent from the management server 3, thereby improving the accuracy of identifying the operating status. Then, the operating status identification unit 32 registers information indicating the identified operating status of the target device 7 in the target device management table (FIG. 14).

[0109] As shown in FIG. 14, a target device management table is stored in the storage unit 27 of the inspection robot 2, 2A.

[0110] The target device management table of the second embodiment has, in association with a target device ID, an item of the operating status of the target device 7, an item of a detailed status of the target device 7, an item of the operation start time of the target device 7, an item of a scheduled inspection time, an item of a prohibited area set around the target device 7, and an item of attributes of the target device 7. An attribute registration unit 51 registers attribute information in the target device management table.

[0111] In the detailed status item, detailed information about the target device 7 is registered. Not only the current operating status but also the planned operating status is registered. For example, even if the device is currently stopped, if it is scheduled to be operated during the inspection of the inspection robot 2, 2A, or if it is being prepared for operation, that fact is registered.

[0112] The operation start time field is registered with the operation start time of the target device 7 that is currently out of operation.

[0113] The item of scheduled inspection time is registered with the scheduled inspection time of the inspection robot 2, 2A. For example, the time period during which the inspection robot 2, 2A performs patrol inspection is registered.

[0114] Even if the target device 7 is currently out of operation, it may start operating during the inspection by the inspection robot 2, 2A. In this case, if the prohibited area N of the target device 7 is set within the range of the second distance L2 corresponding to the out-of-operation state, there is a concern that the inspection robot 2, 2A may approach or come into contact with the target device 7.

[0115] Therefore, the area setting unit 33 refers to the operation start time and the scheduled inspection time, and if the target equipment 7 is currently out of operation but is scheduled to be operated during the inspection by the inspection robot 2, 2A, registers the "first distance" in association with the target equipment ID of the target equipment 7. In this way, even if the operation status changes during the inspection by the inspection robot 2, 2A, the prohibited area N can be appropriately set. And, the risk of the inspection robot 2, 2A approaching or coming into contact with the target equipment 7 can be reduced.

[0116] Furthermore, even if the target equipment 7 is currently out of operation, if there is no risk of it being operated during the inspection by the inspection robot 2, 2A, that is, if there is sufficient time until the start time of operation of the target equipment 7, there is no need to change the setting of the prohibited area N.

[0117] In the attribute field, attribute information indicating the attributes of the target device 7 is registered. For example, if the target device 7 is equipped with a "ventilation fan," that fact is registered. Also, if the target device 7 generates a "strong magnetic field" during operation, that fact is registered.

[0118] The inspection robots 2, 2A have different performances depending on their movement mechanisms. For example, the inspection robots 2, 2A are affected by the surrounding environment differently depending on their types. The area setting unit 33 of the second embodiment sets the prohibited area N based on the type information and attribute information.

[0119] For example, in the case of a target device 7 equipped with a ventilation fan, wind is generated from the ventilation fan during operation. Therefore, if the flying inspection robot 2A passes near the target device 7, it may be affected by the wind. In particular, the movement accuracy of the flying inspection robot 2A is affected by the surrounding environment. In the vicinity of the target device 7, a high level of control is required to hover and adjust its posture. Therefore, the prohibited area N of the target device 7 is set to a wide range (first distance L1). On the other hand, in the case of a traveling inspection robot 2, since it is less susceptible to the wind generated from the ventilation fan, the prohibited area N of the target device 7 may be set to a narrow range (second distance L2).

[0120] Furthermore, in the case of target equipment 7 that generates a strong magnetic field using a high voltage during operation, there is a risk that the flying inspection robot 2A may be affected. In this case as well, the prohibited area N of the target equipment 7 is set to a wide range (first distance L1). On the other hand, in the case of a traveling inspection robot 2, since it is less susceptible to the influence of a strong magnetic field, the prohibited area N of the target equipment 7 may be set to a narrow range (second distance L2). In this way, different prohibited areas N can be set depending on the type of inspection robot 2, 2A.

[0121] The setting of the prohibited area N of the target device 7 may be the same regardless of the type of the inspection robot 2, 2A.

[0122] The inspection system and inspection method have been described based on the first and second embodiments, but the configuration applied in any one embodiment may be applied to the other embodiment, and the configurations applied in each embodiment may be combined.

[0123] For example, the operation status identification unit 32 may identify the operation status of some of the multiple target devices 7 based on information acquired by a sensor provided on the inspection robot 2, and identify the operation status of the other target devices 7 based on management information held by the management server 3.

[0124] In the flowcharts of the above-described embodiment, the steps are executed in series, but the order of the steps is not necessarily fixed, and some steps may be executed in reverse. Also, some steps may be executed in parallel with other steps.

[0125] The system of the above embodiment includes a control device with a highly integrated processor such as a dedicated chip, a Field Programmable Gate Array (FPGA), a Graphics Processing Unit (GPU), or a Central Processing Unit (CPU), a storage device such as a Read Only Memory (ROM) or a Random Access Memory (RAM), an external storage device such as a Hard Disk Drive (HDD) or a Solid State Drive (SSD), a display device such as a display, an input device such as a mouse or a keyboard, and a communication interface. This system can be realized with a hardware configuration using a normal computer.

[0126] The programs executed by the systems of the above-described embodiments are provided in advance in a ROM, etc. Alternatively, the programs may be provided in the form of installable or executable files stored in a non-transitory computer-readable storage medium such as a CD-ROM, CD-R, memory card, DVD, or flexible disk (FD).

[0127] The programs executed by this system may be stored on a computer connected to a network such as the Internet and provided by downloading them via the network. This system may also be configured by combining separate modules that independently perform the functions of the components and connect them to each other via a network or dedicated lines.

[0128] According to at least one of the embodiments described above, by providing an area setting unit that sets a prohibited area around the target equipment on the environmental map to prohibit the entry of the inspection robot based at least on the operating status, the prohibited area to prohibit the entry of the inspection robot can be appropriately set.

[0129] Although some embodiments of the present invention have been described, these embodiments are presented as examples and are not intended to limit the scope of the invention. These embodiments can be implemented in various other forms, and various omissions, substitutions, modifications, and combinations can be made without departing from the spirit of the invention. These embodiments or modifications thereof are within the scope of the invention and its equivalents as described in the claims, as well as within the scope and spirit of the invention. [Explanation of symbols]

[0130] 1 (1A)...inspection system, 2 (2A)...inspection robot, 3...management server, 4...management terminal, 5...network, 6...plant, 7 (7A, 7B, 7')...target equipment, 8...management office, 9...display, 10...remote control terminal, 11...layout drawing, 13...memory unit, 14...communication unit, 15...server control unit, 16...target equipment management unit, 17...robot management unit, 20...camera, 21...thermo camera, 22...microphone, 23...magnetic sensor, 24...3D measurement sensor, 25...motion sensor, 26...movement mechanism unit, 27...memory unit, 28...communication unit, 30...robot control unit, 31...map setting unit, 32...operation status identification unit, 33...area setting unit, 34...route generation unit, 35...setting change unit, 36...autonomous movement control unit, 40...control panel, 41...display unit, 42...meter, 43...indicator light, 44...generator, 45...heat generation unit, 50...type registration unit, 51...attribute registration unit, 52...situation information acquisition unit, E...permitted area, L1...first distance, L2...second distance, M...administrator, N(N')...prohibited area, R...movement route.

Claims

1. a map setting unit that sets an environmental map used by an inspection robot that inspects a target device to be inspected when the inspection robot moves, the environmental map including information indicating a position of the target device; an operating status identification unit that identifies an operating status indicating whether the target device is operating; an area setting unit that sets a prohibited area around the target device on the environmental map, the prohibited area prohibiting the inspection robot from entering, based on at least the operation status; A type registration unit that registers type information indicating the type of the inspection robot, the type information being distinguished by the moving mechanism mounted thereon; an attribute registration unit that registers attribute information indicating an influence that the target device has on its surroundings, the influence being different depending on the type of the inspection robot; Equipped with the region setting unit sets the prohibited region based on the type information and the attribute information. Inspection system.

2. The operation status identification unit identifies the operation status of the target equipment based on information acquired by a sensor provided in the inspection robot. The inspection system of claim 1 .

3. a management server for managing the operation status of the target device; The operation status identification unit identifies the operation status of the target device based on management information indicating the operation status of the target device held by the management server. The inspection system according to claim 1 or 2.

4. the region setting unit sets the prohibited region within a range of a first distance from the target device that is in operation, and sets the prohibited region within a range of a second distance from the target device that is not in operation, the second distance being shorter than the first distance; The inspection system according to any one of claims 1 to 3.

5. A path generating unit that generates a movement path of the inspection robot; a setting change unit that, when the prohibited area corresponding to the movement route is set within the range of the first distance from the target device, resets the prohibited area to the range of the second distance from the target device; Equipped with The inspection system according to claim 4.

6. A step in which a map setting unit sets an environmental map used when an inspection robot that inspects a target device to be inspected moves, the environmental map including information indicating a position of the target device; An operation status identification unit identifies an operation status indicating whether the target device is operating; a region setting unit setting a prohibited region around the target device on the environmental map, the prohibited region prohibiting entry of the inspection robot, based on at least the operation status; A step of registering type information indicating the type of the inspection robot, the type information being distinguished by a moving mechanism mounted thereon, by a type registration unit; An attribute registration unit registers attribute information indicating an influence that the target device has on its surroundings, the influence being different depending on the type of the inspection robot; Including, the region setting unit sets the prohibited region based on the type information and the attribute information. Inspection method.

Citation Information

Patent Citations

  • Robot control device, robot control method, and robot control system

    JP2020166352A

  • Unmanned mobile unit and patrolling check system

    WO2019176710A1

  • Driving route generating device, driving route generating method, driving route generating program, and drone

    WO2020071305A1