Unattended system and unattended method

Abstract

To solve the problem that a large amount of cost is incurred when an underwater mobile robot and personnel required for its operation are carried on a ship and operated by reciprocating for long distance every time when maintenance is performed, therefore, reduction in costs required for maintenance of an on-water wind power generator has been demanded.SOLUTION: An unmanned maintenance system 1 includes an underwater mobile robot 14 which can be autonomously moved in water by autonomous control and can maintain an underwater facility relating to an on-water wind power generation device 2. In this way, it is possible to reduce cost required for maintenance of the on-water wind power generator.SELECTED DRAWING: Figure 3

Description

[Technical Field]

[0001] FIELD OF THE INVENTION Embodiments of the present invention relate to unattended maintenance techniques. [Background technology]

[0002] The power transmission cables of floating wind turbines installed on the ocean are kept afloat by floats to prevent them from rubbing against the seabed and becoming damaged. However, if aquatic organisms such as barnacles attach to the floats, they will sink, requiring periodic removal of the fouling. Furthermore, to ensure their integrity, visual inspections are also required after the fouling is removed. Furthermore, the mooring lines that secure the floats to prevent them from drifting away also require periodic removal of fouling and visual inspections. When floating wind turbines are installed in the open ocean, work must be carried out in areas too deep for divers to access. Therefore, attempts have been made to use underwater robots such as remotely operated underwater vehicles (ROVs). However, sending the underwater robots and the personnel required to operate them back and forth on a vessel over long distances each time maintenance is performed is extremely costly. Therefore, there is a need to reduce the maintenance costs of floating wind turbines. [Prior art documents] [Non-patent literature]

[0003] [Non-Patent Document 1] "Cleaning brush and UT measurement gauge for Observer mini-ROV", Internet<URL:https: / / www.youtube.com / watch?v=6eJ6fxpdCK0> ,[Searched on October 21, 2021] [Non-patent document 2] "Clean up dirt with a brush! Industrial underwater drone (ROV) MOGOOL", Internet<URL:https: / / www.youtube.com / watch?v=RSdy5jibHJ0> ,[Searched on October 21, 2021] Summary of the Invention

Problems to be Solved by the Invention

[0004] The problem to be solved by the present invention is to provide an unmanned maintenance technology capable of reducing the cost of maintaining an offshore wind power generation device.

Means for Solving the Problems

[0005] The unmanned maintenance system according to an embodiment of the present invention includes an underwater mobile robot that can move underwater unmanned by autonomous control and can maintain underwater facilities related to an offshore wind power generation device. , the underwater mobile robot includes an underwater moving device for moving in water, a cleaning device for removing attachments adhering to the surface of the underwater facility, and a control device for performing the autonomous control. The control device includes a position and orientation estimation unit for estimating the position and orientation of the underwater mobile robot based on information obtained by a device mounted on the underwater mobile robot, and an object recognition unit for recognizing a target part to be maintained of the underwater facility. The position and orientation estimation unit creates an environmental map including information on the surrounding environment of the underwater mobile robot while estimating the position and orientation of the underwater mobile robot. The object recognition unit recognizes the target part based on the position and orientation of the underwater mobile robot and the environmental map. The device includes a camera or a laser sensor. The device extracts feature points of objects around the underwater mobile robot, which are information on the surrounding environment. The position and orientation estimation unit creates the environmental map including a three-dimensional model showing the shape of the target part based on the three-dimensional feature point group data obtained by aggregating the feature points. 。

Effects of the Invention

[0006] According to an embodiment of the present invention, an unmanned maintenance technology capable of reducing the cost of maintaining an offshore wind power generation device is provided.

Brief Description of the Drawings

[0007]

Figure 1

Figure 2

Figure 3

Figure 4

Embodiments for Carrying Out the Invention

[0008] Hereinafter, embodiments of the unmanned maintenance system and the unmanned maintenance method will be described in detail with reference to the drawings.

[0009] Reference numeral 1 in Fig. 1 represents the unmanned maintenance system of this embodiment. This unmanned maintenance system 1 is for performing maintenance on the offshore wind power generation device 2 installed on the ocean 3 without human intervention. In this embodiment, the term "maintenance" includes at least one of the meanings of cleaning and inspection. Further, the term "maintenance" may include various meanings such as various maintenance, tests, operation confirmation, repairs, recovery, removal, etc.

[0010] As an offshore facility, the offshore wind power generation device 2 includes a plurality of blades 5 that rotate about a central axis of a hub 4. When wind hits these blades 5, the hub 4 rotates about the central axis, and a generator (not shown) provided inside the nacelle 6 generates electricity. In this embodiment, an upwind type propeller type windmill that is a lift type windmill and a horizontal axis windmill is exemplified.

[0011] Inside the hub 4, a variable pitch mechanism (not shown) for changing the pitch angle of the blade 5 is provided. Inside the nacelle 6, a brake device (not shown) and the like are provided. Further, an azimuth changing mechanism (not shown) for changing the azimuth of the nacelle 6 and the like are also provided. Also, a speed increaser (not shown) may be provided. This nacelle 6 is provided at the upper part of a tower 7 standing on the ocean 3.

[0012] In addition, as underwater facilities, the offshore wind power generation device 2 includes a floating body 8 for floating the tower 7 on the ocean 3, a mooring cable 9 for mooring the floating body 8, and a power transmission cable 10 for sending the generated power to the land.

[0013] The mooring cable 9 is a huge metal chain that ties the floating body 8 to the seabed 11. A plurality of mooring cables 9 are provided for one floating body 8. The lower ends of these mooring cables 9 are fixed to the seabed 11, and the offshore wind power generation device 2 does not need to be carried away by ocean currents even when floating on the floating body 8 and is fixed at a fixed position. In this embodiment, a floating type offshore wind power generation device 2 is exemplified, but this embodiment may also be applied to a fixed type offshore wind power generation device 2 that is implanted on the seabed 11.

[0014] The unmanned maintenance system 1 includes an unmanned mother ship 12, an aerial mobile robot 13, and an underwater mobile robot 14. Here, the aerial mobile robot 13 and the underwater mobile robot 14 are the unmanned mobile robots of this embodiment. In the unmanned maintenance system 1 of this embodiment, both the aerial mobile robot 13 and the underwater mobile robot 14 are mounted on the unmanned mother ship 12, but at least the underwater mobile robot 14 may be mounted on the unmanned mother ship 12.

[0015] The aerial mobile robot 13 can move unmanned in the air by autonomous control or remote operation, and the underwater facilities of the offshore wind power generation device 2 are inspected using this aerial mobile robot 13.

[0016] The underwater mobile robot 14 can move unmanned in the water by autonomous control or remote operation, and the underwater facilities of the offshore wind power generation device 2 are inspected using this underwater mobile robot 14. The inspection of the underwater facilities includes, for example, visual inspection of the power transmission cable 10 and the mooring cable 9, acquisition of three-dimensional shape data of the power transmission cable 10 and the mooring cable 9, measurement of the wear amount on the surface of the power transmission cable 10 and the mooring cable 9, and the like.

[0017] In addition, the underwater mobile robot 14 also cleans the underwater facilities of the offshore wind power generation device 2. The unmanned mother ship 12 and the underwater mobile robot 14 are connected by a power supply cable 15. The unmanned mother ship 12 is provided with a power supply device (not shown) for supplying power to the underwater mobile robot 14. The underwater mobile robot 14 receives power supply from an external power supply device via the power supply cable 15.

[0018] The unmanned mother ship 12 can navigate unmanned on the ocean 3 by autonomous control or remote operation, can carry the aerial mobile robot 13 and the underwater mobile robot 14, and is a ship from which the aerial mobile robot 13 and the underwater mobile robot 14 can depart. In this embodiment, a plurality of aerial mobile robots 13 and a plurality of underwater mobile robots 14 are carried by the unmanned mother ship 12.

[0019] A large number of offshore wind power generation devices 2 are erected on the ocean 3, thereby constructing an offshore wind power generation site. The unmanned mother ship 12 performs maintenance on these offshore wind power generation devices 2. For example, the unmanned mother ship 12 patrols a plurality of offshore wind power generation devices 2 through autonomous control or remote operation. When the unmanned mother ship 12 approaches the offshore wind power generation device 2, the aerial mobile robot 13 and the underwater mobile robot 14 set off to perform maintenance on the offshore wind power generation device 2. The underwater mobile robot 14 mainly performs cleaning and inspection of the offshore wind power generation device 2.

[0020] Note that one unmanned mother ship 12 may perform maintenance on a plurality of offshore wind power generation devices 2, or a plurality of unmanned mother ships 12 may be dispatched to the offshore wind power generation site, and the plurality of unmanned mother ships 12 may perform maintenance on the plurality of offshore wind power generation devices 2.

[0021] In addition, data obtained by the aerial mobile robot 13 and the underwater mobile robot 14, for example, image data when photographing the offshore wind power generation device 2, or three-dimensional shape data when measuring the offshore wind power generation device 2, etc. are aggregated by the unmanned mother ship 12. Then, these data are sent from the unmanned mother ship 12 to a predetermined data center that manages information about the offshore wind power generation device 2.

[0022] The offshore wind power generation site is provided in the open sea far from the coast. The unmanned mother ship 12 can navigate from the port serving as the base to the offshore wind power generation site through autonomous control or remote operation. Note that the unmanned mother ship 12 returns to the port regularly for maintenance.

[0023] In this embodiment, since all maintenance of the offshore wind power generation device 2 can be performed unmanned, cost reduction and labor saving can be achieved. For example, by making the cleaning and inspection of the offshore wind power generation device 2 unmanned, cleaning and inspection can be performed 24 hours a day, 365 days a year. Furthermore, when the weather is bad such as typhoons or rough seas, or when the weather changes suddenly, the personnel involved in cleaning and inspection are not exposed to danger.

[0024] Note that the unmanned mother ship 12 can relay the communication between the aerial mobile robot 13 and the underwater mobile robot 14. For example, when the aerial mobile robot 13 or the underwater mobile robot 14 communicates with the headquarters at a remote location (ground station), the unmanned mother ship 12 relays the communication. In this way, the administrator (user) at the headquarters can remotely operate the aerial mobile robot 13 or the underwater mobile robot 14.

[0025] The unmanned mother ship 12 is provided with a ship moving device (not shown) for navigating the unmanned mother ship 12. This ship moving device includes a screw for obtaining propulsion force underwater, a motor for driving the screw, a rudder for determining the traveling direction, and the like.

[0026] The unmanned mother ship 12 detects its own position using a satellite positioning system, a radar, a camera, and the like. In addition, the unmanned mother ship 12 can detect the position of the offshore wind power generation device 2 using at least one of a radar, a laser, a sonar, a camera, and the like.

[0027] The unmanned mother ship 12 detects the position of the offshore wind power generation device 2, and measures the relative position between the offshore wind power generation device 2 and the aerial mobile robot 13 or the underwater mobile robot 14 with the unmanned mother ship 12 as a reference (absolute position). Then, based on this relative position, the position of the aerial mobile robot 13 or the underwater mobile robot 14 is controlled. In this way, the unmanned mother ship 12 can grasp the accurate positional relationship between the offshore wind power generation device 2 and the aerial mobile robot 13 or the underwater mobile robot 14. In addition, the administrator (user) can operate the aerial mobile robot 13 or the underwater mobile robot 14 in real time.

[0028] The unmanned mother ship 12 detects the position of the aerial mobile robot 13 using at least one of a radar, a laser, a camera, and the like. When the unmanned mother ship 12 detects the position of the aerial mobile robot 13, the unmanned mother ship 12 transmits information indicating the position to the aerial mobile robot 13. The aerial mobile robot 13 can accurately grasp its own position based on the information indicating the position sent from the unmanned mother ship 12.

[0029] The unmanned mother ship 12 is provided with a takeoff and landing port 16 (Fig. 1) where the aerial mobile robot 13 can take off and land. Charging can be performed on the aerial mobile robot 13 that has landed at this takeoff and landing port 16. By doing so, since the unmanned mother ship 12 can charge the aerial mobile robot 13, the aerial mobile robot 13 can be inspected over a long period of time.

[0030] The unmanned mother ship 12 detects the position of the underwater mobile robot 14 using at least one of a laser, sonar, camera, etc. When the unmanned mother ship 12 detects the position of the underwater mobile robot 14, it transmits information indicating the position to the underwater mobile robot 14. The underwater mobile robot 14 can accurately grasp its own position based on the information indicating the position sent from the unmanned mother ship 12.

[0031] The unmanned mother ship 12 is equipped with a loading and unloading device 17 for putting the underwater mobile robot 14 into the water or recovering it from the water. For example, the loading and unloading device 17 includes a crane for lifting the underwater mobile robot 14 and a winding device for winding up the power supply cable 15.

[0032] The unmanned mother ship 12 supplies power to the underwater mobile robot 14. For example, power can be supplied via a power supply cable 15 (Fig. 1) extending from the unmanned mother ship 12 to the underwater mobile robot 14. By doing so, since the underwater mobile robot 14 can receive power supply from the unmanned mother ship 12, the underwater mobile robot 14 can perform cleaning and inspection over a long period of time. Note that the power supply to the underwater mobile robot 14 includes a mode of charging a battery (not shown) mounted on the underwater mobile robot 14. Also, the power supply cable 15 includes a signal line for transmitting and receiving signals.

[0033] As shown in Fig. 2, the underwater mobile robot 14 includes a communication unit 50, an underwater mobile device 51, a swimming sensor 52, an underwater inspection device 53, a cleaning device 54, and a control device 55.

[0034] The communication unit 50 is used for communication with the unmanned mother ship 12. This communication is carried out by wire. Also, the communication unit 50 is used for relaying the unmanned mother ship 12 and communicating with a predetermined device (not shown) at a remote location (ground station). For example, the communication unit 50 is used to send and receive information regarding the cleaning of underwater facilities with the headquarters that manages the underwater mobile robot 14 on land. In this way, the administrator (user) at the headquarters can remotely control the underwater mobile robot 14. For example, the administrator can appropriately switch the underwater mobile robot 14, which normally operates by autonomous control, to remote control to clean the underwater facilities.

[0035] Furthermore, the communication unit 50 is also used for communication with other underwater mobile robots 14. Note that the communication between the underwater mobile robot 14 and the unmanned mother ship 12 may be performed wirelessly. For example, in this embodiment, various communication technologies such as satellite communication networks, mobile communication networks, and the Internet can be used. Also, when the underwater mobile robot 14 is underwater, wired communication may be performed, and when the underwater mobile robot 14 surfaces, wireless communication may be performed.

[0036] The underwater movement device 51 is a device for submerging the underwater mobile robot 14. For example, the underwater movement device 51 includes a screw for obtaining propulsion force underwater, a motor for driving the screw, a rudder for determining the traveling direction, and a ballast for controlling buoyancy.

[0037] The swimming sensor 52 is a device for obtaining information necessary for the submergence of the underwater mobile robot 14. For example, the swimming sensor 52 includes a device for obtaining information using at least one of a laser, a sonar, a camera, a motion sensor, etc. Note that the motion sensor is a 9-axis sensor that combines an inertial sensor (3-axis acceleration sensor and 3-axis angular velocity sensor) and a 3-axis geomagnetic sensor. Also, the laser, sonar, camera, and motion sensor are devices mounted on the underwater mobile robot 14. Note that by using a sonar as the swimming sensor 52, for example, necessary information can be obtained even if the water becomes turbid during the cleaning operation.

[0038] In addition, in this embodiment, the information (images) obtained by the camera includes still images and moving images. Further, in this embodiment, a technique of performing predetermined image processing and recognizing an object depicted in the image can be used. Also, as the information obtained using a laser, point cloud data indicating the three-dimensional shape of the surface of an object is included.

[0039] The point cloud data is acquired using a laser sensor such as an infrared sensor or LiDAR. For example, by projecting a laser onto an object and receiving the reflected light with a light receiving element, the distance to the object can be measured. Also, the laser sensor measures the distance to surrounding objects using the ToF (Time of Flight) method of converting the delay time of the received pulse with respect to the projected pulse into distance.

[0040] Note that the underwater mobile robot 14 may use a stereo camera to simultaneously photograph a predetermined object from a plurality of different directions, thereby acquiring information in the depth direction to the object.

[0041] The underwater inspection device 53 is a device necessary for inspecting the underwater facilities of the offshore wind power generation device 2. This underwater inspection device 53 is also used when cleaning the underwater facilities. For example, it includes a device that obtains information using at least one of a laser, sonar, camera, etc. This underwater inspection device 53 includes an inspection camera, a three-dimensional laser scanner, etc. The inspection camera photographs the appearance of the offshore wind power generation device 2. The three-dimensional laser scanner measures the three-dimensional shape of the surface of the offshore wind power generation device 2.

[0042] The cleaning device 54 is a device necessary for cleaning the underwater facilities of the offshore wind power generation device 2. In this embodiment, the cleaning device 54 removes the deposits adhering to the surface of the underwater facilities. For example, aquatic organisms such as barnacles adhere to the floating body 8, the mooring cable 9, the power transmission cable 10, etc. The underwater mobile robot 14 uses the cleaning device 54 to periodically clean the surfaces of these underwater facilities.

[0043] For example, as shown in FIG. 3, the underwater mobile robot 14 includes a manipulator arm 62 and a rotary brush 63 provided at the tip of the manipulator arm 62. These constitute the cleaning device 54 of the present embodiment. Note that the cleaning device 54 may perform cleaning using a water jet or the like.

[0044] The underwater mobile robot 14 performs a cleaning operation of removing deposits adhering to the surface of the power transmission cable 10 or the like using the rotary brush 63. In addition to barnacles, the deposits to be removed include seaweeds such as moss, eggs of aquatic organisms, artificial objects such as garbage, scale, oil, sludge, dust, and the like. The underwater mobile robot 14 performs these removal operations by autonomous control or remote control.

[0045] As shown in FIG. 2, the control device 55 comprehensively controls various devices mounted on the underwater mobile robot 14. The control device 55 includes a position information acquisition unit 56, a position and attitude estimation unit 57, an object recognition unit 58, a swimming path planning unit 59, an inspection autonomous control unit 60, and a cleaning autonomous control unit 61. These are realized by a program stored in a memory or HDD being executed by a CPU.

[0046] The control device 55 is capable of executing an autonomous control program. The underwater mobile robot 14 performs a cleaning operation of removing deposits adhering to the surface of underwater facilities at least by autonomous control. In this way, since the underwater mobile robot 14 can clean underwater facilities while moving underwater by autonomous control, it can continuously perform operations. For example, it can operate 24 hours a day, 365 days a year. Therefore, it is possible to remove barnacle larvae before they adhere to and grow on the surface of underwater facilities, and even if the power required for cleaning underwater facilities is small, they can be easily removed. That is, the working reaction force of cleaning is reduced.

[0047] In addition, the program for autonomous control of the underwater mobile robot 14 executed by the control device 55 can be rewritten based on the control information received from the outside. For example, an administrator (user) at the headquarters can rewrite the program for autonomous control by remote operation. In this way, every time the cleaning plan is changed, the program for autonomous control of the underwater mobile robot 14 can be rewritten to cope with it.

[0048] In this embodiment, a plurality of underwater mobile robots 14 (FIG. 1) are provided, and the operation of one of the underwater mobile robots 14 is controlled based on the information obtained by the other underwater mobile robot 14. In this way, the plurality of underwater mobile robots 14 can cooperate with each other to clean and inspect the offshore wind power generation device 2.

[0049] For example, one of the underwater mobile robots 14 uses a swimming sensor 52 or an underwater inspection device 53 (3D laser scanner) to detect the position of the other underwater mobile robot 14. Furthermore, the position of the offshore wind power generation device 2 is also detected. Since one of the underwater mobile robots 14 is at the position closest to the other underwater mobile robot 14 and the offshore wind power generation device 2, their accurate positions can be detected. The information indicating these positions is sent to the other underwater mobile robot 14. Then, the other underwater mobile robot 14 can grasp the accurate positional relationship with the offshore wind power generation device 2 and perform a submerged navigation.

[0050] Next, the inspection and cleaning process executed by the control device 55 of the underwater mobile robot 14 will be described with reference to the flowchart of FIG. 4. Refer to the above-mentioned drawings.

[0051] First, in step S1, the position information acquisition unit 56 (FIG. 2) executes a position information acquisition process. Here, the position information acquisition unit 56 acquires the position information regarding the target part from the database that has accumulated the position information regarding the target part to be cleaned and inspected of the underwater facility.

[0052] For example, a management computer (not shown) provided in a headquarters located at a remote location (a ground station) has a database that stores information regarding the underwater facilities of the offshore wind power generation device 2. This database stores position information regarding the floating body 8, the mooring cable 9, the power transmission cable 10, etc., which are the target parts of the present embodiment. Note that the position information includes information on the shapes of these objects. The position information acquisition unit 56 acquires the position information sent from the headquarters.

[0053] In the next step S2, the position and attitude estimation unit 57 (FIG. 2) executes a position and attitude estimation process. Here, the position and attitude estimation unit 57 estimates the position and attitude of the underwater mobile robot 14 based on information obtained by devices mounted on the underwater mobile robot 14, for example, the swimming sensor 52. By doing so, the estimation accuracy of the position and attitude of the underwater mobile robot 14 can be improved.

[0054] The position and attitude estimation unit 57 estimates the position and attitude of the underwater mobile robot 14 based on, for example, an image captured by a camera, the three-dimensional shape of surrounding objects acquired by a laser sensor, and the acceleration and angular velocity acquired by a motion sensor. The position and attitude estimation unit 57 creates an environmental map including information on the surrounding environment of the underwater mobile robot 14 simultaneously with the estimation of the position and attitude of the underwater mobile robot 14. Note that the environmental map includes a three-dimensional model showing the shape of the target part.

[0055] The position and attitude estimation unit 57 can extract feature points of objects around the underwater mobile robot 14 using the information acquired by the camera and the laser sensor. The set of these feature points is the three-dimensional feature point group data. Then, the position and attitude estimation unit 57 analyzes the image (video) captured by the camera and tracks the feature points of the object (for example, the sides or corners of a box-shaped object) in real time. Based on this three-dimensional feature point group data, three-dimensional information on the position and attitude of the underwater mobile robot 14 can be estimated. Furthermore, the position and attitude estimation unit 57 can create an environmental map based on this three-dimensional feature point group data.

[0056] In addition, the position and orientation estimation unit 57 detects three-dimensional feature point group data of objects around the underwater mobile robot 14 at predetermined time intervals, and calculates the displacement of the position and orientation of the underwater mobile robot 14 based on the displacement between the three-dimensional feature point group data before and after in time series.

[0057] In addition, the position and orientation estimation unit 57 obtains the swimming path of the underwater mobile robot 14 prior to the current position and current orientation from a series of self-positions and self-orientations obtained in time series.

[0058] That is, the position and orientation estimation unit 57 performs control to estimate the position and orientation of the underwater mobile robot 14 based on images continuously captured when the underwater mobile robot 14 moves from the reference position. In this way, since the swimming path of the underwater mobile robot 14 moving from the reference position can be estimated from the images, the estimation accuracy of the position and orientation of the underwater mobile robot 14 can be improved.

[0059] In this embodiment, the position and orientation estimation unit 57 estimates the position and orientation of the underwater mobile robot 14 based on the information acquired by the camera and the laser sensor, but other embodiments may also be used. For example, the position and orientation estimation unit 57 may estimate the position and orientation of the underwater mobile robot 14 based on the information acquired by an RGB camera, a fish-eye camera, a gyro sensor, or an infrared sensor.

[0060] The environmental map created by the underwater mobile robot 14 is transmitted to the management computer (not shown) at the headquarters. Then, the management computer stores the environmental map in the database.

[0061] In the next step S3, the target recognition unit 58 (Fig. 2) executes target recognition processing. Here, the target recognition unit 58 recognizes the position and shape of the target part to be cleaned of the underwater facility. For example, the target recognition unit 58 recognizes the position and shape of the target part based on the position information acquired by the position information acquisition unit 56. In this way, the target part can be recognized by referring to the already accumulated position information. Note that the target part is an underwater facility such as the floating body 8, the mooring cable 9, and the power transmission cable 10.

[0062] Furthermore, the target recognition unit 58 additionally or alternatively recognizes the position and shape of the target part based on the position and attitude of the underwater mobile robot 14 estimated by the position and attitude estimation unit 57 and the environmental map created by the position and attitude estimation unit 57. In this way, the recognition accuracy of the target part can be improved. For example, when the already accumulated position information of the target part does not match the current situation, the recognition can be performed according to the current state of the target part. In particular, the shapes and positions of the power transmission cable 10 and the mooring cable 9 change depending on the direction and strength of the ocean current. Therefore, the target recognition unit 58 recognizes the current target part and enables the underwater mobile robot 14 to operate according to the state of the current target part.

[0063] The target recognition processing includes not only the processing of recognizing the target part but also the processing of recognizing the position and shape of the power supply cable 15. In addition, the target recognition processing includes the processing of recognizing the position and shape of the attachment. Furthermore, the target recognition processing includes the processing of distinguishing whether the attachment is a natural object or an artificial object.

[0064] Note that when the attachment recognized by the target recognition unit 58 is an artificial object, the cleaning operation may be aborted and only inspection may be performed. For example, when the artificial object is a fishing net, the underwater mobile robot 14 may get entangled in the fishing net when approaching, and subsequent activities may become impossible. Therefore, when the attachment is an artificial object, the underwater mobile robot 14 is controlled not to approach.

[0065] The object recognition process of this embodiment includes image processing of an image in which an object appears. Further, it includes a shape identification process for identifying the shape of the object.

[0066] Further, the object recognition process may include a process using machine learning. For example, the control device 55 may include a computer equipped with artificial intelligence (AI) that performs machine learning. Further, the control device 55 may include a deep learning unit that extracts a specific pattern from a plurality of patterns based on deep learning.

[0067] In the analysis using the computer of this embodiment, analysis techniques based on the learning of artificial intelligence can be used. 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 can be used. Further, the forms of machine learning include forms such as clustering and deep learning.

[0068] The system of this embodiment includes a computer equipped with artificial intelligence that performs machine learning. For example, this system may be composed of one computer equipped with a neural network, or may be composed of a plurality of computers equipped with a neural network.

[0069] Here, a neural network is a mathematical model that expresses the characteristics of brain functions by simulation using a computer. For example, it shows a model in which artificial neurons (nodes) that form a network by synaptic connections change the synaptic connection strength by learning and acquire problem-solving ability. Further, the neural network acquires problem-solving ability by deep learning.

[0070] For example, a neural network is provided with a hidden layer having six layers. Each layer of this hidden layer is composed of 300 units. Also, by pre-training a multi-layer neural network using learning data, it is possible to automatically extract feature amounts in the patterns of changes in the state of a circuit or system. Note that in a multi-layer neural network, on a user interface, it is possible to set an arbitrary number of hidden layers, an arbitrary number of units, an arbitrary learning rate, an arbitrary number of learning times, and an arbitrary activation function.

[0071] Note that a reward function may be set for each type of information item to be learned, and deep reinforcement learning in which the information item with the highest value is extracted based on the reward function may be used for the neural network.

[0072] For example, a CNN (Convolution Neural Network) with proven performance in image recognition is used. In this CNN, the hidden layer is composed of a convolutional layer and a pooling layer. The convolutional layer obtains a feature map by performing a filter process on nodes close to each other in the previous layer. The pooling layer further reduces the feature map output from the convolutional layer to obtain a new feature map. At this time, by obtaining the maximum value of the pixels included in the region of interest in the feature map, it is possible to absorb any slight deviation in the position of the feature amount.

[0073] The convolutional layer extracts local features of the image, and the pooling layer performs a process of summarizing the local features. In these processes, the image is reduced while maintaining the features of the input image. That is, in a CNN, it is possible to greatly compress (abstract) the amount of information held by the image. Then, using the abstracted image image stored in the neural network, it is possible to recognize the input image and classify the image.

[0074] Note that deep learning includes various methods such as autoencoders, RNNs (Recurrent Neural Networks), LSTMs (Long Short-Term Memory), and GANs (Generative Adversarial Networks). These methods may be applied to the deep learning of this embodiment.

[0075] The information regarding the target part acquired by the underwater mobile robot 14 is transmitted to the management computer (not shown) at the headquarters. Then, the management computer stores the information regarding the target part in the database. If the position information of the target part already stored in the database is old, it can be updated according to the current state of the target part.

[0076] In the next step S4, the swimming route planning unit 59 (Fig. 2) executes a swimming route planning process. Here, the swimming route planning unit 59 creates a plan for the swimming route of the underwater mobile robot 14 based on the environmental map created by the position and orientation estimation unit 57. The underwater mobile robot 14 autonomously moves following the planned swimming route. In this way, a swimming route considering the current state of the target part can be created. Therefore, for example, even if the position of the target part changes due to ocean currents, an appropriate swimming route can be planned.

[0077] For example, the underwater mobile robot 14 moves following the power transmission cable 10 and the mooring cable 9 whose shapes change due to ocean currents. Therefore, inspections can be carried out even in the presence of ocean currents. Furthermore, the administrator can perform cleaning and inspections automatically without remote operation.

[0078] In addition, the swimming route planning unit 59 estimates the position of the power supply cable 15 that changes as the underwater mobile robot 14 moves, and plans a swimming route in which the power supply cable 15 does not get entangled in the target part. For example, when a plurality of candidate swimming routes are created in which the underwater mobile robot 14 swims around the power transmission cable 10, the underwater mobile robot 14 is prevented from selecting a swimming route that goes around the power transmission cable 10 one or more times. By doing so, the power supply cable 15 is prevented from getting entangled in the power transmission cable 10.

[0079] Note that the position and shape of the power supply cable 15 are continuously recognized by the target recognition unit 58. The underwater mobile robot 14 may be provided with devices such as a laser sensor and a camera on the rear side or the upper side in the traveling direction in order to constantly recognize the power supply cable 15.

[0080] In the next step S5, the cleaning autonomous control unit 61 (FIG. 2) executes the cleaning autonomous control process. Here, the cleaning autonomous control unit 61 controls the underwater moving device 51 and the cleaning device 54 based on the swimming route planned by the swimming route planning unit 59, and autonomously dives the underwater mobile robot 14 to clean the target part. When there is a sea current, the underwater mobile robot 14 receives the force of being carried by the sea current, but since it can always swim following the swimming route planned by this autonomous control, the underwater mobile robot 14 can perform the cleaning work without being carried by the sea current. Note that the data obtained by the cleaning is transmitted to the management computer (not shown) of the headquarters.

[0081] In the next step S6, the inspection autonomous control unit 60 (FIG. 2) executes the inspection autonomous control process. Here, the inspection autonomous control unit 60 controls the underwater moving device 51 and the underwater inspection device 53 based on the swimming route planned by the swimming route planning unit 59, and autonomously dives the underwater mobile robot 14 to inspect the target part. Even when there is a sea current, the underwater mobile robot 14 can perform the inspection work without being carried by the sea current. Note that the data obtained by the inspection is transmitted to the management computer (not shown) of the headquarters.

[0082] Then, the control device 55 finishes the inspection and cleaning process. The above steps are at least part of the processes included in the inspection and cleaning process, and other steps may also be included in the inspection and cleaning process. Note that this inspection and cleaning process is a process that is repeated at regular intervals. By repeating this process, the unmanned maintenance method is executed. Note that this process may be interrupted and executed while the control device 55 is executing other main processes.

[0083] In the flowchart of the present embodiment, an example is shown in which each step is executed in series. However, the sequence of each step is not necessarily fixed, and the sequence of some steps may be reversed. Also, some steps may be executed in parallel with other steps. For example, in the present embodiment, the inspection autonomous control process is executed after the cleaning autonomous control process, but the cleaning autonomous control process may be executed after the inspection autonomous control process.

[0084] In the present embodiment, by using the underwater mobile robot 14 for cleaning, it is possible to perform cleaning and inspection even in deep parts where divers cannot work. In particular, cleaning can be performed in areas with a water depth of 50 m or more.

[0085] Also, to the data transmitted from the underwater mobile robot 14 to the management computer (not shown) at the headquarters, information on the environmental map or the 3D model of the target part is associated with the information obtained by inspection or cleaning. For example, the 3D model information is associated with the image of the target part. Therefore, the inspection results and cleaning results can be recorded and managed together with the information regarding the target part.

[0086] In the present embodiment, an example is shown in which the unmanned maintenance system 1 performs cleaning and inspection of the offshore wind power generation device 2, but other modes may also be possible. For example, the unmanned maintenance system 1 may perform only inspection or only cleaning. Additionally or alternatively, the unmanned maintenance system 1 may perform various maintenance, tests, operation checks, repairs, recoveries, removals, etc. of the offshore wind power generation device 2.

[0087] In addition, in this embodiment, an aspect where the offshore wind power generation device 2 is provided on the ocean 3 is exemplified, but other aspects may also be possible. For example, the offshore wind power generation device 2 may be provided on a lake. That is, the term "offshore" includes the meanings of the sea and the lake.

[0088] In addition, in this embodiment, as the offshore wind power generation device 2, a lift-type windmill and an upwind type propeller-type windmill with a horizontal axis are exemplified, but other aspects may also be possible. For example, the offshore wind power generation device 2 may be a downwind type propeller-type windmill. Further, the offshore wind power generation device 2 may be a Darrieus type windmill, a Gyromill type windmill, or a vertical wing type windmill, which is a lift-type windmill and a vertical axis windmill. Also, the offshore wind power generation device 2 may be a Savonius type windmill, a paddle type windmill, a cross-flow type windmill, or an S-shaped rotor type windmill, which is a drag-type windmill and a vertical axis windmill. Further, the offshore wind power generation device 2 may be a Magnus type windmill, which is a lift-type windmill and a horizontal axis windmill or a vertical axis windmill.

[0089] The control device 55 of this embodiment has hardware resources such as a processor (control unit) and a memory (storage unit), and is configured as a computer in which information processing by software is realized using the hardware resources by the CPU executing various programs. Further, the unmanned maintenance method of this embodiment is realized by causing a computer to execute various programs.

[0090] In addition, in this embodiment, an aspect where the control device 55 automatically controls various devices is exemplified, but other aspects may also be possible. For example, the control device 55 may receive an input operation of an administrator (user) of the unmanned maintenance system 1 and control various devices. That is, the control device 55 may also be a remote operation device for controlling various devices by manual operation of the administrator.

[0091] The system of this embodiment includes a control device with highly integrated processors such as a dedicated chip, FPGA (Field Programmable Gate Array), GPU (Graphics Processing Unit), or CPU (Central Processing Unit), a storage device such as ROM (Read Only Memory) or RAM (Random Access Memory), an external storage device such as HDD (Hard Disk Drive) or SSD (Solid State Drive), a display device such as a display, an input device such as a mouse or keyboard, and a communication interface. This system can be realized with a hardware configuration using an ordinary computer.

[0092] Note that the program executed by the system of this embodiment is provided by being pre-embedded in a ROM or the like. Alternatively, this program may be stored in a non-transitory computer-readable storage medium such as a CD-ROM, CD-R, memory card, DVD, flexible disk (FD), etc. in an installable or executable file format and provided.

[0093] Also, the program executed by this system may be stored on a computer connected to a network such as the Internet and provided by being downloaded via the network. Further, this system can also be configured by connecting separate modules that independently exhibit the functions of the components to each other via a network or a dedicated line and combining them.

[0094] According to the embodiment described above, by providing an underwater mobile robot that can move unmanned underwater by autonomous control and can maintain underwater facilities related to an offshore wind power generation device, the cost for maintaining the offshore wind power generation device can be reduced.

[0095] Although some embodiments of the present invention have been described, these embodiments are presented by way of example and are not intended to limit the scope of the invention. These embodiments can be implemented in various other forms, and various omissions, replacements, changes, and combinations can be made without departing from the gist of the invention. These embodiments or their modifications are included in the scope and gist of the invention, as well as in the invention described in the claims and its equivalent scope.

Explanation of Signs

[0096] 1... unmanned maintenance system, 2... offshore wind power generation device, 3... ocean, 4... hub, 5... blade, 6... nacelle, 7... tower, 8... floating body, 9... mooring cable, 10... power transmission cable, 11... seabed, 12... unmanned mother ship, 13... aerial mobile robot, 14... underwater mobile robot, 15... power supply cable, 16... takeoff / landing port, 17... input / output device, 50... communication unit, 51... underwater mobile device, 52... swimming sensor, 53... underwater inspection device, 54... cleaning device, 55... control device, 56... position information acquisition unit, 57... position and attitude estimation unit, 58... target recognition unit, 59... swimming route planning unit, 60... inspection autonomous control unit, 61... cleaning autonomous control unit, 62... manipulator arm, 63... rotating brush.

Claims

1. An underwater mobile robot that can move underwater autonomously and can maintain underwater facilities related to an offshore wind power generation device, The underwater mobile robot includes: An underwater moving device for moving underwater, A cleaning device for removing attachments adhering to the surface of the underwater facility, A control device for performing the autonomous control, and The control device includes: A position and orientation estimation unit that estimates the position and orientation of the underwater mobile robot based on information obtained by a device mounted on the underwater mobile robot, An object recognition unit that recognizes an object part to be maintained of the underwater facility, and The position and orientation estimation unit creates an environmental map including information on the surrounding environment of the underwater mobile robot while estimating the position and orientation of the underwater mobile robot, The object recognition unit recognizes the object part based on the position and orientation of the underwater mobile robot, the environmental map, The device includes a camera or a laser sensor, The device extracts feature points of objects around the underwater mobile robot, which are information on the surrounding environment, The position and orientation estimation unit creates the environmental map including a 3D model showing the shape of the object part based on the 3D feature point group data obtained by aggregating the feature points, An unmanned maintenance system.

2. The control device acquires the position information from a database storing the position information regarding the object part, The object recognition unit recognizes the object part based on the position and orientation of the underwater mobile robot, the environmental map, and the position information, The unmanned maintenance system according to Claim 1.

3. The control device plans a swimming route of the underwater mobile robot based on the environmental map and performs the autonomous control based on the swimming route, The unmanned maintenance system according to Claim 1 or Claim 2.

4. A power supply cable for receiving power supply from the outside is connected to the underwater mobile robot, The control device estimates the position of the power supply cable that changes as the underwater mobile robot moves and plans the swimming route in which the power supply cable does not get entangled with the object part, The unmanned maintenance system according to Claim 3.

5. The underwater mobile robot includes a communication unit for transmitting and receiving information regarding the maintenance of the underwater facility to and from a device at a ground station, The unmanned maintenance system according to any one of Claims 1 to 4.

6. An underwater mobile robot capable of moving unmanned in water is a method for maintaining underwater facilities related to an offshore wind power generation device, wherein the underwater mobile robot, an underwater moving device for moving in water, a cleaning device for removing deposits adhering to the surface of the underwater facility, and a control device for performing the autonomous control, and is provided with, wherein the control device, a position and attitude estimation unit that estimates the position and attitude of the underwater mobile robot based on information obtained by a device mounted on the underwater mobile robot, and an object recognition unit that recognizes an object portion to be the subject of maintenance of the underwater facility, and is provided with, wherein the position and attitude estimation unit creates an environmental map including information on the surrounding environment of the underwater mobile robot while estimating the position and attitude of the underwater mobile robot, and the object recognition unit recognizes the object portion based on the position and attitude of the underwater mobile robot and the environmental map, wherein the device includes a camera or a laser sensor, the device extracts feature points of objects around the underwater mobile robot, which are information on the surrounding environment, and the position and attitude estimation unit creates the environmental map including a three-dimensional model showing the shape of the object portion based on three-dimensional feature point group data obtained by aggregating the feature points, an unmanned maintenance method.

Images