JP7315159B2 - Structure inspection system - Google Patents

Description

The present invention relates to a structure inspection system.

Structures are subject to periodic inspections to maintain safety. For example, in a concrete structure, a hammering rod, a hammer, or the like is used to perform a hammering test to check for cracks, floats, and other abnormal portions inside the concrete (Patent Documents 1 and 2). Repair work is promptly carried out on structures where abnormal parts are found. Such periodic inspections maintain the safety of structures.

JP 2016-166819 A JP 2017-150883 A JP 2017-227632 A

However, there are many structures that require inspection due to aging of the structures, and it takes an enormous amount of time to inspect all areas of the structures that require inspection.

An inspection system of Patent Document 3 has been proposed as a technique for inspecting structures and the like. In the inspection system disclosed in Patent Document 3, a user interface device receives an input for designating an inspection point, and based on the input from the user interface device and its current position, the mobile robot autonomously flies and moves to the inspection point, and the mobile robot inspects the inspection point using inspection means such as percussion means. However, in the inspection system of Patent Literature 3, there is a possibility that designation of inspection points depends on the intuition and experience of the user. For this reason, the inspection system of Patent Literature 3 suffers from omissions and wastefulness in inspection, and there is a problem that the structure cannot be efficiently inspected.

SUMMARY OF THE INVENTION Accordingly, the present invention has been devised in view of the above-described problems, and an object of the present invention is to provide a structure inspection system capable of efficiently inspecting structures.

A structure inspection system according to the present invention comprises an imaging means, an input means for inputting an image of a structure imaged by the imaging means, a database in which three or more degrees of association between past images of a structure and the deformation status of the structure with respect to the past image of the structure are stored in advance, a determination means for determining a deformation status of a structure based on the image of the structure input via the input means by referring to the linkage levels stored in the database, and a determination means for determining the deformation status of the structure determined by the determination means. and an inspection means for inspecting the altered portion of the structure specified by the specifying means.hammering soundInspection tool and the above percussive bar typehammering sounda sound collecting tool for collecting sound emitted by the inspection tool, the flying robot having a movable part for moving the percussion rod type hammering inspection tool, the percussion rod typehammering soundThe inspection tool has a shaft whose rear end is connected to the movable portion, and a tapper provided at the tip of the shaft. The hammering rod type hammering inspection tool is characterized in that, while the flying robot stays at a predetermined position, the shaft is moved with the movable portion as a base point, and the structure is tapped by moving the tapper so as to stroke the structure.

According to the present invention configured as described above, it is possible to efficiently inspect a structure.

BRIEF DESCRIPTION OF THE DRAWINGS It is a schematic diagram which shows an example of the structure inspection system which concerns on 1st Embodiment of this invention. 1 is a schematic diagram showing an example of a flying robot according to a first embodiment of the present invention; FIG. 1 is a block diagram showing an example of a flying robot according to a first embodiment of the present invention; FIG. It is a block diagram showing an example of a terminal device concerning a 1st embodiment of the present invention. FIG. 3 is a schematic diagram showing an example of a database according to the first embodiment of the present invention; FIG. It is a flow chart which shows an example of operation of a structure inspection system concerning a 1st embodiment of the present invention. (a) is a diagram showing a state in which the flying robot according to the first embodiment of the present invention flies below a structure, and (b) is a diagram of the structure shown in (a) as seen from the direction of arrow D in the figure. It is a figure which shows an example of the deformation|transformation location of the structure displayed on the display part. It is a schematic diagram which shows an example of an inspection by the inspection apparatus which concerns on 1st Embodiment of this invention. It is a schematic diagram which shows an example of an inspection by the inspection apparatus which concerns on 1st Embodiment of this invention. It is a schematic diagram which shows the 1st modification of the database based on 1st Embodiment of this invention. FIG. 5 is a schematic diagram showing a first modified example of the flying robot according to the first embodiment of the present invention; It is a mimetic diagram showing a structure inspection system concerning a 2nd embodiment of the present invention.

Hereinafter, a structure inspection system to which the present invention is applied will be described in detail with reference to the drawings.

(First embodiment)
FIG. 1 is a schematic diagram showing an example of a structure inspection system 200 according to the first embodiment of the present invention.

A structure inspection system 200 includes a flying robot 100 , a terminal device 150 and a database 400 . An inspector A of the structure 300 operates the terminal device 150 to remotely control the flying robot 100 . Also, the terminal device 150 is connected to the database 400 by wire or wirelessly. Structure 300 is, for example, a concrete structure. The structure 300 is, for example, a bridge such as a floor slab bridge.

FIG. 2 is a schematic diagram showing an example of the flying robot 100 according to the first embodiment of the present invention. FIG. 3 is a block diagram showing an example of the flying robot 100 according to the first embodiment of the invention.

A drone is used as the flying robot 100, for example. A flying robot 100 is provided with a body 1 , a landing gear 2 , a rotor 11 , an inspection device 20 , an imaging device 23 , a control device 60 and a communication device 70 .

A landing gear 2 is provided on the airframe 1 . The landing gear 2 is attached to the lower portion of the airframe 1, for example. The landing gear 2 supports the flying robot 100 in the landing state on a ground plane, for example, the ground.

The rotor blades 11 are provided on the fuselage 1 . The rotor blades 11 give buoyancy to the airframe 1 to allow the airframe 1 to fly. A propeller, a rotor, or a duct fan, for example, is used as the rotary blade 11 .

The inspection device 20 is provided on the airframe 1 . The inspection device 20 inspects the deformed portion of the structure 300 . One example of the inspection device 20 is a hammering inspection device. The hammering sound inspection device includes, for example, a hammering sound inspection tool 21 and a sound collecting tool 22 that collects sounds emitted from the hammering sound inspection tool 21 .

One example of the hammering test tool 21 is, for example, a percussion stick type hammering test tool. The percussion stick type hammering test tool includes a shaft 21a, a movable part 21b connecting the shaft 21a to the machine body 1, a percussion element 21c provided at the tip of the shaft 21a, and an elastic mechanism 21d for vertically moving the shaft 21a. The movable portion 21b moves the shaft 21a with the end of the shaft 21a as a base point. As a result, the percussion stick type hammering inspection tool can percuss the structure by moving the percussor 21c at the tip of the shaft 21a. 21 d of elastic mechanisms contain a spring material, for example. When the spring material is compressed, the shaft 21a is folded toward the airframe 1 side. When the contraction of the spring material is released, the shaft 21a is lifted. The tapping element 21c makes a "sound" by moving it on the structure 300 as if stroking it. From this "sound", it is determined whether or not there is an abnormal portion such as a crack or float inside the structure 300 . The sound collection tool 22 picks up this "sound". Although a steel ball is shown as an example of the tapper 21c, any tapper can be used as the tapper 21c.

Another example of the hammering test tool 21 is, for example, a percussion-type hammering test tool. In the hammering type hammering test tool, a hammering head is used instead of the hammer 21c. The striking head emits a "sound" by hitting the deformed part. From this "sound", it is determined whether there is an abnormal part inside the structure 300 or not. Moreover, since it is possible to conduct a test such as a Charpy impact test, the strength of the structure 300 can also be examined.

The imaging device 23 is, for example, a camera capable of capturing both still images and moving images. The camera captures, for example, the vicinity of the contact point of the tapper 21c with the surface of the structure. Also, by controlling the direction of the camera, it is possible to capture images of the front, surroundings, and rear of the flying robot 100, for example. The imaging device 23 can transmit the captured image to the terminal device 150 via the communication device 70 . Furthermore, the camera can transmit image information of, for example, the front to the terminal device 150 via the communication device 70 and display it live on the display unit 161 of the terminal device 150 . For example, when the flying robot 100 moves to a location that is difficult for the inspector A to visually recognize, the inspector A can operate the flying robot 100 while watching the image on the monitor.

The imaging device 23 may further include a position information acquisition unit 23a that acquires position information and a distance sensor 23b that measures the distance to the subject. GPS (Global Positioning System), for example, is used for the position information acquisition unit 23a. For the position information acquisition unit 23a, for example, a device for measuring the absolute position of the imaging device 23 with reference to a predetermined origin is used. For example, an infrared sensor, an ultrasonic sensor, or the like is used as the distance sensor 23b.

A control device 60 is provided in the airframe 1 . The control device 60 controls each of the rotor blade 11 , the inspection device 20 , and the imaging device 23 .

A communication device 70 is provided in the body 1 . The communication device 70 connects the control device 60 and the terminal device 150 by wired communication or wireless communication. The communication device 70 can transmit information acquired by the inspection device 20 and the imaging device 23 to the terminal device 150 and the database 400 .

The terminal device 150 is, for example, an electronic device such as a personal computer (PC), but in addition to the PC, it may be embodied in any other electronic device such as a mobile phone, a smartphone, a tablet terminal, a wearable terminal, or the like.

FIG. 4 is a block diagram showing an example of the terminal device 150 according to the first embodiment of the invention. The terminal device 150 includes a control unit 154 for controlling the entire terminal device 150, an input unit 155 for inputting various control commands, a communication unit 156 for performing wired or wireless communication with the communication device 70 of the flying robot 100 and the database 400, a determination unit 157 for determining optimum detection algorithm information, a storage unit 158 represented by a hard disk or the like for storing a program for performing a search to be executed, and a determination unit 157. and an identification unit 159 for identifying a deformed portion of the structure from the determination result are connected to the internal bus 160, respectively. Furthermore, the internal bus 160 is connected to a display unit 161 as a monitor for actually displaying information.

The control section 154 is a so-called central control unit for controlling each component implemented in the terminal device 150 by transmitting control signals via the internal bus 160 . In addition, the control unit 154 transmits commands for various controls through the internal bus 160 according to inputs through the input unit 155 .

The input unit 155 receives an execution instruction for executing a program from the user. The input unit 155 notifies the control unit 154 when the execution command is input by the user. The control unit 154 that has received this notification executes a desired processing operation in cooperation with each component including the determination unit 157 . The input unit 155 may be connected to a keyboard, a touch panel, or the like, and various information may be input by a user operating the keyboard or the like. The input unit 155 receives various types of information such as an image of a structure and incidental information of the structure.

The determination unit 157 determines a deformed portion of the structure from the information input via the input unit 155 . The determination unit 157 reads out various information stored in the storage unit 158 as necessary information in executing the determination operation. The determination unit 157 may be controlled by artificial intelligence. This artificial intelligence may be based on any known artificial intelligence technology.

When storage unit 158 comprises a hard disk, predetermined information is written to each address based on control by control unit 154, and this information is read as necessary. The storage unit 158 also stores a program for executing the present invention. This program is read and executed by the control unit 154 .

The specifying unit 159 specifies a deformed portion of the structure from the determination result of the determining unit 157 .

The display unit 161 is composed of a graphic controller that creates a display image based on control by the control unit 154 . The display unit 161 is implemented by, for example, a liquid crystal display (LCD).

The database 400 is a database relating to conditions for determining deformation conditions of structures to be provided. This database 400 stores information sent via a public communication network or information input by a user of this system. The database 400 also transmits this accumulated information to the terminal device 150 based on a request from the terminal device 150 . Note that the database 400 may be provided in the flying robot 100 or may be provided in the terminal device 150 . Database 400 may be controlled by artificial intelligence. This artificial intelligence may be based on any known artificial intelligence technology.

FIG. 5 is a schematic diagram showing an example of the database 400 according to the first embodiment of the invention. As shown in FIG. 5, the database 400 stores in advance three or more levels of association between past structure images as reference input parameters and structure deformation conditions as output solutions for the past structure images. The degree of association indicates the degree to which the past structure image and the deformation state of the structure are linked, and is indicated in three or more stages such as percentage, 10 stages, or 5 stages, for example. This degree of association is an index that indicates the high possibility that an image of a structure is associated with the deformation state of the structure, and indicates the accuracy in determining the optimal deformation state of the structure from the structure image.

For example, in FIG. 5, the image J1 of the past structure shows a degree of relevance “80%” with the deformation state “crack” of the structure, and a degree “60%” with the state of deformation “discoloration” of the structure. In this case, "crack" indicates a stronger connection with the image J1 than "discoloration".

The degree of association may be configured by a model that can be updated through so-called machine learning, or may be configured by a neural network. At this time, the degree of association may be represented by hidden layers and weight variables. Further, the degree of association may be composed of a network on the premise that deep learning is performed.

(Example of operation of structure inspection system 200)
Next, an example of the operation of the structure inspection system 200 according to this embodiment will be described. FIG. 6 is a flow chart showing an example of the operation of the structure inspection system 200 according to this embodiment. FIG. 7(a) is a diagram showing a state in which the flying robot according to the first embodiment of the present invention flies below a structure, and FIG. 7(b) is a diagram of the structure shown in FIG.

As shown in FIG. 7, one example of a structure to be inspected is a bridge 350 . The bridge 350 is, for example, a floor slab bridge.

First, in imaging step S11, an image of the structure 300 is captured. The inspector A controls the terminal device 150 to fly the flying robot 100 and place it on the lower surface 350a of the bridge 350, as shown in FIG. 7(a). The imaging device 23 provided in the flying robot 100 images an imaging region R on the lower surface 350a of the bridge 350, as shown in FIG. 7(b). When the image is captured, the position information acquisition unit 23a may acquire the position of the imaging device 23, and the distance sensor 23b may measure the distance to the lower surface 350a of the bridge 350. FIG.

Next, an input step S12 inputs the image captured in the imaging step S11. The imaging device 23 transmits the captured image to the terminal device 150 . In the terminal device 150, the image captured by the imaging device 23 is input to the input unit 155 as an input parameter.

Next, in the degree-of-association acquisition step S13, three or more degrees of degree of association between the past structure image and the deterioration state of the structure with respect to the past structure image are acquired in advance. The database 400 stores three or more degrees of association between past structural images as reference input parameters and structural deformation states as output solutions.

According to the example of FIG. 5, in the case of the image J1 of the past structure, the degree of relevance is set to 80% for the structural deformation state “crack” and 60% for the structural deformation state “discoloration”. In the case of the image J2 of the past structure, the degree of correlation is set to 90% for the structural deformation state “crack”, 20% for the structural deformation state “discoloration”, and 50% for the structural deformation state “surface peeling”.

These degrees of association may be set based on the images of past structures and the state of deformation of the structure as the determination result, which are stored in advance in the database 400 . This degree of association may be configured by a so-called neural network. This degree of association indicates the accuracy of judging the actual state of deformation of the structure based on the images of the structure in the past. For example, for the image J1 of the past structure, "crack" with a degree of association of 80% is the most accurate judgment, and "discoloration" with a degree of association of 60% is the next most accurate judgment. Similarly, for the image J2 of the past structure, "crack" with a degree of correlation of 90% is the most accurate judgment, followed by "peeling of the surface" with a degree of association of 50%, and "discoloration" with a degree of association of 20%.

Next, a determination step S14 refers to the degree of association acquired in the degree-of-association acquisition step S13, and determines the deformation state of the structure based on the image input in the input step S12.

When determining the deformation state of this structure, the degree of association shown in FIG. 5 acquired in advance in the degree of association acquisition step S13 is referred to. For example, if the structure image input in the input step S12 is the past structure image J1 or similar to it, the above-described degree of association is referred to, and the “crack” with the highest degree of association is selected as the optimum solution. However, it is not essential to select the one with the highest degree of association as the optimum solution, and "discoloration" and "surface peeling", which have a low degree of association but still have the association itself, may be selected as the optimum solution. In addition, it is of course possible to select an output solution that is not connected by an arrow. That is, the selection of the deformation status of the structure is not limited to the case of selecting in order from the highest degree of association, but may be selected in order from the lowest degree of association depending on the case, or may be selected in any other order of priority.

In addition, when the structure image is partially similar to the past structure image J2 but also partially similar to the past structure image J3 and it is not known which one to assign, for example, it may be determined which reference input parameter (past structure image) to assign based on the feature amount on the image. At this time, it may be determined to which reference input parameter is to be assigned based on the feature amount on the image through deep learning.

In this way, after assigning the image of the structure as the input parameter to the image of the past structure as the reference input parameter, the deformation state of the structure as the output solution is selected based on the degree of association set for the reference input parameter.

Then, the display unit 161 is caused to display the deformation state of the structure as an output solution. Accordingly, by viewing the display unit 161, the user can immediately grasp the determination result of the deformation state of the structure.

As a result of determining the state of deformation in determination step S14, if there is deformation in the structure, the process proceeds to identification step S15. If the result of determining the deformation state in determination step S14 is that there is no deformation in the structure, the process returns to imaging step S11, and an area different from imaging area R may be imaged.

Next, in the specifying step S15, a deformed portion of the structure is specified based on the deformed state of the structure determined in the determining step S14. The specified deformed portion of the structure is stored in the terminal device 150 . The identified deformed portion of the structure may be displayed on the display unit 161 . As a result, the user can immediately grasp the deformed portion of the structure by visually recognizing the display unit 161 .

FIG. 8 shows an example of a deformed portion of a structure displayed on the display unit 161. As shown in FIG. For example, as shown in FIG. 8(a), the specifying unit 159 performs image processing such as marking on a portion of the structural image input by the input unit 155 where structural deformation has occurred, thereby specifying the structural deformation site S. Alternatively, as shown in FIG. 8B, the identifying unit 159 may identify the deformed portion S of the structure by performing image processing such as marking on all the locations on the image of the structure input by the input unit 155.

In this manner, the specifying unit 159 may specify a deformed portion of the structure based on the deterioration state of the structure determined by the determining unit 157 and the image of the structure input to the input unit 155 .

For example, the identifying unit 159 may identify the deformed portion of the structure by calculating the position where the structure is deformed from the position of the imaging device 23 that captured the image of the structure acquired by the position information acquiring unit 23a and the distance to the lower surface 350a of the bridge 350 measured by the distance sensor 23b.

In this way, the specifying unit 159 may specify a deformed portion of the structure based on the deterioration state of the structure determined by the determining unit 157, the position information acquiring unit 23a, and the distance sensor 23b.

Next, in an inspection step S16, the inspection device 20 inspects the deformed portion of the structure identified in the identification step S15.

The inspector A controls the terminal device 150 while viewing the deformed portion of the structure displayed on the display unit 161 to inspect the deformed portion of the structure using the inspection device 20 .

Alternatively, the inspector A may control the control device 60 so that the inspection device 20 autonomously inspects the deformed portion of the structure identified in the identification step S15 without controlling the terminal device 150 .

When performing the inspection, as shown in FIG. 9, for example, by moving the flying robot 100 in the direction of the arrow P in the figure, inspection by the inspection device 20, for example, a hammering inspection by a percussion stick type hammering inspection tool may be performed. As a result, the tapping tip 21c can move the lower surface 350a of the bridge 350 as if stroking it.

When conducting an inspection, as shown in FIG. 10, the flying robot 100 may be held at a predetermined position and the movable portion 21b may be moved in the direction of arrow Q in the figure to perform an inspection using the inspection device 20, for example, a hammering inspection using a percussion stick type hammering inspection tool. As a result, the tapping tip 21c can move the lower surface 350a of the bridge 350 as if stroking it.

The sound collecting tool 22 then collects the sound from the percussor 22c, completing the examination.

After the inspection is completed, the process returns to the imaging step S11 again, another imaging area different from the imaging area R is imaged, and the above-described procedure is repeated.

In the procedure described above, in the imaging step S11, an aspect has been described in which the imaging region R is imaged once and then the input step S12 is performed, but other embodiments may be employed. For example, in the imaging step S11, the entire area of the lower surface 350a of the bridge 350 may be divided into different imaging areas, and the input step S12 may be performed a plurality of times.

According to the present embodiment, a determination unit 157 that refers to the degree of association stored in the database 400 and determines the deformation state of the structure based on the image input via the input unit 155, an identification unit 159 that identifies the deformation location of the structure based on the deterioration state determined by the determination unit 157, and an inspection device 20 that performs an inspection of the deformation location identified by the identification unit 159. As a result, it is possible to further inspect the deformed portion of the structure for which the deformed state has been determined. Therefore, it is possible to inspect the structure with high accuracy.

Further, according to the present embodiment, a determining unit 157 that refers to the degree of association stored in the database 400 and determines the deformation state of the structure based on the image input via the input unit 155, a specifying unit 159 that specifies a deformed portion of the structure based on the deterioration state determined by the determining unit 157, and an inspection device 20 that inspects the deformed portion specified by the specifying unit 159. In other words, inspection by the inspection device 20 can be omitted for locations that have not been identified as altered locations. As a result, inspection omissions and waste can be eliminated when inspecting structures. Therefore, it is possible to efficiently inspect the structure.

In particular, according to this embodiment, it is possible to easily grasp the deformation state of a structure without requiring special skill. Furthermore, by constructing the above-mentioned degree of association with artificial intelligence, it is possible to further improve the discrimination accuracy by learning this.

Further, according to the present embodiment, it is characterized in that the determination result of the optimum deformation state of the structure is determined through the degree of association set to three or more stages. The association degree can be described, for example, by a numerical value from 0 to 100%, but is not limited to this, and may be configured in any stage as long as it can be described by numerical values of three or more stages.

By searching based on the degree of association represented by numerical values of three or more levels, it is possible to search and display the results in descending order of the degree of association in a situation where the determination results of the deformation status of a plurality of structures are selected. If the results can be displayed in descending order of degree of association to the user in this way, it is possible to prompt the user to preferentially select the determination result of the deformation state of the structure that is more likely. On the other hand, it is possible to display even the determination result of the deformation state of a structure with a low degree of association as a second opinion, and it is useful when the analysis cannot be performed well with the first opinion.

In addition, according to the present embodiment, it is possible to make a judgment without overlooking the determination result of the deformation state of a structure with an extremely low degree of correlation such as 1%. It is possible to alert the user that even the determination result of the deformation state of the structure with a very low degree of association is connected as a slight sign, and that it may be useful as the determination result of the deformation state of the structure once in dozens or hundreds of times.

Furthermore, according to this embodiment, there is an advantage that a search policy can be determined by setting a threshold value by performing a search based on three or more degrees of association. If the threshold value is lowered, even if the degree of association is 1%, it can be picked up without omission. On the other hand, if the threshold value is increased, it is possible to detect the optimum deformation status of the structure with a high probability. It is possible to decide which one to place emphasis on based on the way of thinking of the user side and the system side, and it is possible to increase the degree of freedom in selecting such points to place emphasis on.

Furthermore, in this embodiment, the degree of association described above may be updated. This update may reflect information provided via a public communication network such as the Internet, for example. When new knowledge about the relationship between the input parameter and the output solution (determination result of the deformation state of the structure) is discovered through site information and writing that can be obtained from the public communication network, the degree of association is increased or decreased according to the knowledge.

The degree of association may be updated manually or automatically by the system side or the user side based on the contents of research data, papers, conference presentations, newspaper articles, books, etc. by experts, other than based on information that can be obtained from public communication networks. Artificial intelligence may be used in these update processes.

According to this embodiment, the flying robot 100 is further provided with at least a percussive stick type hammering inspection tool. As a result, percussive inspection can be performed while the flying robot 100 is flying. Therefore, inspection time can be shortened.

According to the present embodiment, the flying robot 100 is further provided with at least a stick-type hammering inspection tool. The flying robot 100 has a movable part 21b for moving the stick-type hammering inspection tool, and the stick-type hammering inspection tool moves the movable part 21b while the flying robot 100 stays at a predetermined position to percuss a structure. As a result, the percussive inspection can be performed while the flying robot 100 is in a stable state. Therefore, the accuracy of inspection can be improved.

FIG. 11 is a schematic diagram showing a first modified example of the database according to the first embodiment of the present invention. The example of FIG. 11 shows an example in which three or more degrees of association are set between a combination of a past structure image and additional information on the structure, and the determination result of the deterioration state of the structure for the combination.

The incidental information of the structure includes installation information regarding the installation of the structure, section information regarding the section of the structure, and management information regarding maintenance of the structure. Incidentally, the incidental information of the past structures is the past information.

The installation information includes location information on the location where the structure is installed, such as mountainous areas, harbor areas, sea atmosphere, splash zone, tidal zone, seawater area, and submarine soil, environmental information such as temperature and humidity, and vehicle information such as vehicles traveling on the structure and the frequency of vehicle travel.

The cross-sectional information includes the cross-sectional dimensions of the structure, the thickness of the concrete cover, the amount of rebars installed, the moisture condition of the surface of the structure, and the like.

The management information includes the repair history of the structure, the period of service since the structure was constructed, and the like.

For example, in the database 400, a combination of a past structure image J1 and past installation information is represented as a node N1, and three or more degrees of association between the node N1 and the deterioration state of the structure with respect to the node N1 are stored. Similarly, in the database 400, a combination of a past structure image J1 and past cross-sectional information is represented as a node N2, and three or more degrees of association between the node N2 and the deterioration state of the structure with respect to the node N2 are stored. In the database 400, a combination of a past structure image J2, past installation information, and past management information is represented as a node N3, and three or more degrees of association between the node N3 and the deterioration state of the structure with respect to the node N3 are stored. In the database 400, a combination of a past structure image J3 and past cross-sectional information is represented as a node N4, and three or more degrees of association between the node N4 and the deterioration state of the structure with respect to the node N4 are stored.

For the node N1, the degree of association for “crack” is set at 80%, and the degree of association for “discoloration” is set at 60%. Similarly, the node N2 is set with a relevance degree of 70% for “crack” and a relevance degree of 30% for “surface peeling”. The node N3 is set with a relevance degree of 50% for “discoloration” and a relevance degree of 20% for “surface peeling”. The node N4 is set with a relevance degree of 80% for “discoloration” and a relevance degree of 40% for “surface peeling”.

The degree of association of such combinations is acquired in advance. Next, the determination unit 157 refers to such a degree of association, and determines which combination of the image of the structure and the incidental information of the structure newly input via the input unit 155 corresponds to the combination of the image of the past structure and the incidental information of the past structure arranged on the left side of the degree of association. If a new combination of the past structure image J1 and the past installation information is input via the input unit 155, it corresponds to the node N1. Therefore, "crack" with a degree of association of 80% for the node N1, "discoloration" with a degree of association of 60%, and the like are selected.

According to this embodiment, the database 400 stores three or more degrees of association between a combination of a past structure image and past incidental information of the structure, and the deterioration state of the structure with respect to the combination, and the incidental information includes at least one of installation information, section information, and management information. As a result, it is possible to determine the state of deformation of the structure based on the installation information, cross-sectional information, and management information. Therefore, it is possible to further improve the accuracy of determining the deformation state of the structure.

FIG. 12 shows a first modification of the flying robot 100 according to the first embodiment of the invention. The flying robot 100 further has at least three omnidirectional wheels 31 , a wheel drive mechanism 41 attached to the omnidirectional wheels 31 , and a wheel displacement mechanism 51 attached to the wheel drive mechanism 41 . 12, one of the three omnidirectional wheels 31 is omitted from illustration, and only two omnidirectional wheels 31 are shown.

The omnidirectional wheels 31 can move the body 1 in all directions on the surface of the structure 300 . The omnidirectional wheels 31 are driven by a wheel drive mechanism 41 . The wheel drive mechanism 41 can drive the omnidirectional wheels 31 independently.

The omnidirectional wheels 31 include base rollers 32 and free rollers 33 . The base roller 32 is rotationally driven by the wheel drive mechanism 41 .

The free roller 33 is attached to the base roller 32 so as to be freely rotatable in an axial direction orthogonal to the rotation axis direction of the base roller 32 .

The wheel displacement mechanism 51 is provided on the body 1 . The wheel displacement mechanism 51 can displace the positions of the omnidirectional wheels 31 independently of each other in the direction of the arrow y in the figure. For example, the wheel displacement mechanism 51 expands and contracts in the direction of the arrow y in the figure, and displaces the position of the omnidirectional wheel 31 in the direction of the arrow y in the figure. In the state where the wheel displacement mechanism 51 is contracted most, the position of the omnidirectional wheel 31 in the direction of the arrow y in the figure is higher than the position of the rotor 11 in the direction of the arrow y in the figure. As a result, the omnidirectional mobile wheel 31 can be pressed against the surface of the structure 300 without being interfered by the rotor blade 11 .

The control device 60 further controls each of the wheel drive mechanism 41 and the wheel displacement mechanism 51 .

Thus, the flying robot 100 has at least three omnidirectional wheels 31 capable of moving in all directions on the surface of the structure 300, a wheel drive mechanism 41 for driving the omnidirectional wheels 31, and a wheel displacement mechanism 51 for displacing the omnidirectional wheels 31.

When imaging, the inspector A presses each of the three omnidirectional wheels 31 against the lower surface 350 a of the bridge 350 using the thrust of the rotor blade 11 . At this time, the flying robot 100 can stabilize the body 1 . Therefore, the imaging by the imaging device 23 can be performed with high accuracy.

When inspecting, the inspector A presses each of the three omnidirectional wheels 31 against the lower surface 350 a of the bridge 350 using the thrust of the rotor blade 11 . At this time, the tapper 21 c contacts the lower surface 350 a of the bridge 350 . Next, the inspector A causes the flying robot 100 to travel on the lower surface 350 a of the bridge 350 using the driving force of the omnidirectional wheels 31 while controlling the omnidirectional wheels 31 using the control device 60 . As a result, the tapping element 21c moves so as to stroke the deformed portion on the lower surface 350a of the bridge 350. As shown in FIG.

As a result, the wheel displacement mechanism 51 can keep the height of the hammer 21c from the body 1 to the deformed portion in the direction of arrow y (=height of the hammer 21c) constant while the hammer 21c is moving through the deformed portion. Therefore, for example, compared to the case where the height in the arrow y direction in the figure from the machine body 1 to the deformed portion is difficult to stabilize, the inspection by the inspection device 20, for example, the hammering inspection tool 21 can be performed with higher accuracy.

(Second embodiment)
FIG. 13 is a schematic diagram showing an example of a structure inspection system 600 according to the second embodiment of the invention. A detailed description of the configuration similar to that of the first embodiment will be omitted below.

The structure inspection system 600 is used when inspecting the bridge 350 as the structure 300 . A structure inspection system 600 includes a suspension device 700 , a terminal device 150 and a database 400 . An inspector A of the structure 300 operates the terminal device 150 to remotely control the suspension device 700 .

The suspension device 700 includes a suspension member 701 suspended from a side portion such as a balustrade of the bridge 350, a support member 702 spaced downward from the lower surface 350a of the bridge 350 and provided on the suspension member 701, an inspection device 20, an imaging device 23, a control device 60, and a communication device 70.

A steel material, a rope, or the like is used for the suspension member 701, for example. The upper end of the suspension member 701 is fixed to the bridge 350 , and the support member 702 is fixed to the lower end of the suspension member 701 .

For example, a steel material or the like is used for the support member 702 . The support member 702 is provided with the inspection device 20 , the imaging device 23 , the control device 60 , and the communication device 70 . The support member 702 is shaped like a rod and extends substantially horizontally. Suspension members 701 are fixed to both ends of the support member 702 . The supporting member 702 may be provided with a plurality of imaging devices 23 . Also, the support member 702 may be provided with a plurality of inspection devices 20 .

This embodiment also includes a determination unit 157 that refers to the degree of association stored in the database 400 and determines the deformation state of the structure based on the image input via the input unit 155, an identification unit 159 that identifies the deformation location of the structure based on the deterioration state determined by the determination unit 157, and an inspection device 20 that inspects the deformation location identified by the identification unit 159. In other words, inspection by the inspection device 20 can be omitted for locations that have not been identified as altered locations. As a result, inspection omissions and waste can be eliminated when inspecting structures. Therefore, it is possible to efficiently inspect the structure.

200 : Structure inspection system 100 : Flying robot 1 : Airframe 2 : Landing gear 11 : Rotor 20 : Inspection device 21 : Hammering inspection tool 21a : Shaft 21b : Movable part 21c : Sounder 21d : Elastic mechanism 22 : Sound collecting tool 22c : Sounder 23 : Imaging device 23a : Position information acquisition unit 23b : Distance sensor 31 : Omnidirectional mobile wheel 32 : Base roller 33 : Free roller 41 : Wheel drive mechanism 51 : Wheel displacement mechanism 60 : Control device 70 : Communication device 150 : Terminal device 154 : Control unit 155 : Input unit 156 : Communication unit 157 : Discrimination unit 158 : Storage unit 159 : Identification unit 160 : Internal bus 161 : Display unit 300 : Structure 350 : Bridge 350a : Lower surface 400 : Database 600 : Structure inspection system 700 : Suspension device 701 : Suspension member 702 : Support member

Claims (4)

imaging means;
an input means for inputting an image of a structure captured by the imaging means;
a database in which past images of structures and degrees of association of three or more levels of relationships between the past images of structures and deformation states of the structures with respect to the past images of structures are stored in advance;
determining means for determining a deformation state of the structure based on the image of the structure input through the input means, with reference to the degree of association stored in the database;
an identifying means for identifying a deformed portion of the structure based on the deformed state of the structure determined by the determining means;
inspection means for inspecting the deformed portion of the structure specified by the specifying means;
with
The specifying means specifies a deformed portion of the structure by performing a marking process on the input image of the structure,
further comprising a flying robot provided with at least the inspection means;
The inspection means includes a percussion stick type hammering inspection tool for percussing the structure, and a sound collecting tool for collecting the sound emitted by the percussion stick type hammering inspection tool,
The flying robot has a movable part for moving the percussion stick type hammering inspection tool,
The above percussive stick type hammering test tool
a shaft having a rear end connected to the movable part;
a tapper provided at the tip of the shaft;
The structure inspection system is characterized in that the percussive stick type hammering inspection tool is configured to percuss a structure by moving the shaft with the movable part as a base point while the flying robot remains at a predetermined position and moving the percussor so as to stroke the structure. The input means further receives incidental information of the structure,
The database stores in advance three or more degrees of association between combinations of images of past structures and incidental information of past structures, and deformation conditions of the structures with respect to the combinations,
2. The structure inspection system according to claim 1, wherein the incidental information includes at least one of installation information related to installation of structures, cross-sectional information related to cross-sections of structures, and management information related to maintenance of structures. 3. The structure inspection system according to claim 1, wherein the percussive stick type hammering inspection tool percusses the structure as the flying robot moves. The structure inspection system according to any one of claims 1 to 3, wherein the flying robot has at least three omnidirectional wheels capable of moving in all directions on the surface of the structure, a wheel drive mechanism for driving the omnidirectional wheels, and a wheel displacement mechanism for displacing the positions of the omnidirectional wheels.

Images