Method and apparatus for generating movement paths for multiple drones for inspecting industrial structures

Abstract

The present invention relates to a method and device for generating a movement path of a plurality of drones for inspecting an industrial structure. The method and device for generating a movement path of a plurality of drones for inspecting an industrial structure according to an embodiment of the present invention can identify a path that a drone can move and a path that the drone cannot move at a plurality of inspection points based on industrial structure information, and can derive an optimal movement path by optimizing a vehicle path problem model using a metaheuristic solution method so as to maximize the operational efficiency of the plurality of drones. In addition, the vehicle path problem model is further solved using an exact solution method, and a solution method can be determined from the exact solution method or the metaheuristic solution method based on the number of inspection points. This makes it possible to flexibly respond to on-site conditions and deliver accurate results.

Description

Technical Field

[0001] The present invention relates to a method for generating movement paths of a plurality of drones for inspecting industrial structures. Specifically, the present invention relates to a method for generating movement paths of a plurality of drones for inspecting industrial structures, which sets an optimal movement path for a drone to move to a plurality of preset inspection points according to the structure, outer shape, and characteristics of the industrial structure.

Background Art

[0002] The content described in this part only provides background information for this embodiment and does not constitute the prior art. Recently, along with the Fourth Industrial Revolution, the technology of using drones for photography has been commercialized and incorporated into many technologies. In particular, attempts to apply drone technology to industrial sites and the development of related technologies are underway. Conventionally, the method of checking the state of industrial structures has been carried out by photographing the main parts of industrial structures using a telescopic camera or by a person directly approaching the parts for inspection. Such a method using a telescopic camera has the drawback that it is difficult to confirm accurate results, and the method of direct inspection by a person requires a large number of personnel and multiple devices along with the risk of accidents, and has the drawbacks of taking a lot of time and cost. Therefore, there is a need for a technology that utilizes drones for inspecting industrial structures, and there is a situation where a general method and device for generating movement paths of a plurality of drones applicable to industrial structures of various sizes and shapes are required. In addition, when inspecting industrial structures using drones, it is desirable to perform the inspection using a plurality of drones in consideration of the size and structure of the industrial structure. Therefore, there is also a situation where there is a need for a method and device for generating an optimal movement path of drones that takes into account the battery capacity and cost of drones and does not cause collisions between a plurality of drones.

Summary of the Invention

[0003] The object of the present invention is to provide a method and apparatus for generating movement paths for multiple drones used in inspecting industrial structures, which identifies possible and impossible movement paths for multiple drones at multiple inspection points based on industrial structure information, optimizes a vehicle path problem model using a metaheuristic solution method to maximize the operational efficiency of multiple drones, and derives an optimal movement path. Furthermore, an objective of the present invention is to further solve the vehicle path problem model using an exact solution method, and to determine the solution method from either an exact solution method or a metaheuristic solution method based on the number of inspection points. This provides a method and apparatus for generating movement paths for multiple drones for inspecting industrial structures, which can flexibly respond to on-site conditions and transmit accurate results. The objects of the present invention are not limited to those mentioned above, and other objects and advantages of the present invention not mentioned will be understood from the description below and will be understood more clearly from the embodiments of the present invention. Furthermore, it will be readily apparent that the objects and advantages of the present invention can be achieved by the means and combinations set forth in the claims. [Means for solving the problem]

[0004] A route generation device for generating multiple drone movement paths for inspecting industrial structures according to several embodiments of the present invention to solve the aforementioned problem includes: a data transmission / reception unit that receives industrial structure information and multiple inspection points of the industrial structure; a route identification unit that, considering the industrial structure information, identifies routes in which the drone can move and routes in which the drone cannot move at the multiple inspection points and defines the connection relationships of the inspection points corresponding to the routes in which the drone can move; a graph completion unit that defines virtual routes between inspection points for which the connection relationships are not defined and completes the relationships of the multiple inspection points; a problem solving unit that solves a vehicle path problem model configured based on the multiple inspection points and outputs a solution; and a route generation unit that generates multiple drone movement paths based on the output solution. The problem-solving unit solves the vehicle path problem model through a metaheuristic solution method configured to search for a solution based on weights corresponding to the connection relationships of the multiple inspection points, evaluate the degree of fit by performing a suitability evaluation of the searched solution based on the operational efficiency for the multiple drones, terminate problem solving if the evaluated degree of fit satisfies the algorithm termination condition, and repeat problem solving by changing the weights if the evaluated degree of fit does not satisfy the algorithm termination condition, and the problem-solving unit terminates problem solving and outputs the searched solution after performing problem solving a predetermined number of times. Furthermore, the metaheuristic solution method is an ant colony system (ACS), and the problem-solving unit can further utilize pheromone information accumulated in existing pathways as positive feedback to search for a solution. Furthermore, the problem-solving unit is configured to solve the vehicle path problem model using an exact solution method. The problem-solving unit determines the solution method for the vehicle path problem model from the metaheuristic solution method or the exact solution method depending on the number of inspection points of the industrial structure. If the number of inspection points of the industrial structure exceeds a preset threshold of reference inspection points, the problem-solving unit can solve the vehicle path problem model using the metaheuristic solution method. If the number of inspection points of the industrial structure is less than or equal to the preset threshold of reference inspection points, the problem-solving unit can solve the vehicle path problem model using the exact solution method.

[0005] Furthermore, the vehicle path problem model can be configured to take into account the battery capacities of the multiple drones, and to treat inspection points that overlap between the sub-travel paths exclusively as starting points (depots). Furthermore, the graph completion unit can identify pairs of inspection points for which no connection relationship has been formed by searching for paths corresponding to elements that are 0 among the elements of the adjacency matrix, which shows the connection relationships of the plurality of inspection points, excluding the diagonal elements. Furthermore, in the graph completion unit, the virtual path can be defined as a detour path that connects pairs of inspection points for which the connection relationship has not been formed through at least one inspection point. Furthermore, the route identification unit generates a three-dimensional model of the industrial structure based on the industrial structure information, and uses the three-dimensional model of the industrial structure to identify routes along which the drone can move and routes along which the drone cannot move at the multiple inspection points. A method for generating movement paths for multiple drones for inspecting industrial structures according to some embodiments of the present invention includes the steps of: receiving industrial structure information and multiple inspection points of the industrial structure; identifying possible and impossible movement paths for the drones at the multiple inspection points, taking into account the industrial structure information; defining the connection relationships of the inspection points corresponding to the possible movement paths for the drones; defining virtual paths between inspection points for which the connection relationships are not defined to complete the relationships of the multiple inspection points; solving a vehicle path problem model configured based on the multiple inspection points and outputting a solution; and generating the movement paths based on the outputted solution, wherein the movement paths are for multiple drones The step of outputting the solution includes searching for a solution based on weights corresponding to the connectivity of the plurality of inspection points, evaluating the degree of fit by performing a suitability evaluation of the searched solution based on the operational efficiency for the plurality of drones, terminating the problem solving if the evaluated degree of fit satisfies the algorithm termination condition, and repeating the problem solving by changing the weights if the evaluated degree of fit does not satisfy the algorithm termination condition, and the step of outputting the solution terminates the problem solving and outputs the searched solution after performing the problem solving a predetermined number of times.

[0006] An industrial structure inspection system according to some embodiments of the present invention comprises a plurality of drones for inspecting an industrial structure, a plurality of control devices corresponding to each of the plurality of drones, and a route generation device for generating movement paths for the plurality of drones, wherein the movement path includes a sub-movement path corresponding to each of the plurality of drones, and the route generation device provides the sub-movement path to at least one of the corresponding drones and control devices, the route generation device includes a data transmission / reception unit for receiving industrial structure information and a plurality of inspection points of the industrial structure, a route identification unit that, considering the industrial structure information, identifies paths where the drones can move and paths where the drones cannot move at the plurality of inspection points and defines the connection relationships of the inspection points corresponding to the paths where the drones can move, a graph completion unit that defines virtual paths between inspection points where the connection relationships are not defined and completes the relationships of the plurality of inspection points, and the plurality of inspection points The system includes a problem-solving unit that solves a vehicle path problem model configured in a specific way and outputs a solution, and a path generation unit that generates movement paths for multiple drones based on the output solution, wherein the movement paths for the multiple drones include sub-movement paths corresponding to each of the multiple drones. The problem-solving unit searches for a solution based on weights corresponding to the connection relationships of the multiple inspection points, evaluates the degree of fit by performing a suitability evaluation of the searched solution based on the operational efficiency for the multiple drones, and solves the vehicle path problem model through a metaheuristic solution method configured to repeat problem solving by changing the weights if the evaluated degree of fit satisfies the algorithm termination condition, and if the evaluated degree of fit does not satisfy the algorithm termination condition. The problem-solving unit then terminates problem solving and outputs the searched solution after performing problem solving a predetermined number of times. Furthermore, the drone is configured to fly autonomously according to the travel path or according to the control signals of the control device, and the drone is further configured to generate position information and provide the generated position information to the control device, and the control device can display the travel path and the position information together on a display for the user to see. [Effects of the Invention]

[0007] The method and apparatus for generating travel paths for inspecting industrial structures according to embodiments of the present invention can identify paths that a drone can travel and paths that it cannot travel at multiple inspection points based on industrial structure information, and derive an optimal travel path by optimizing a vehicle path problem model using a metaheuristic solution method to maximize the operational efficiency of multiple drones. Furthermore, the route generation device and method for generating travel routes for industrial structure inspections are configured to further solve a vehicle route problem model using an exact solution method, and the solution method can be determined from either an exact solution method or a metaheuristic solution method based on the number of inspection points. Therefore, accurate results can be transmitted while flexibly responding to on-site conditions. In addition to the above, the specific effects of the present invention will be described together with the following specific details for carrying out the invention. [Brief explanation of the drawing]

[0008] [Figure 1] Figure 1 is a conceptual diagram illustrating an inspection system for industrial structures according to several embodiments of the present invention. [Figure 2] Figure 2 is a block diagram illustrating the configuration of the control device in detail. [Figure 3] Figure 3 is a block diagram illustrating the drone's configuration in detail. [Figure 4] Figure 4 is a block diagram illustrating the configuration of the route generation device in detail. [Figure 5] Figure 5 is an illustrative diagram illustrating the process of generating a travel path in a path generation device. [Figure 6] Figure 6 is an exemplary image visualizing a connected graph. [Figure 7] Figure 7 is an illustrative image illustrating the process of completing a connected graph. [Figure 8] Figure 8 is an illustrative image illustrating the process of completing a connected graph. [Figure 9] Figure 9 is an illustrative image illustrating the process of completing a connected graph. [Figure 10] Figure 10 is an illustrative image showing the generated movement path. [Figure 11] Figure 11 is a flowchart of a method for generating movement paths for multiple drones for inspecting industrial structures according to several embodiments of the present invention. [Figure 12] Figure 12 is a flowchart illustrating the detailed steps of stage (S140) in Figure 11. [Modes for carrying out the invention]

[0009] The terms or words used herein and in the claims shall not be construed to be limited to their general or dictionary meanings. In accordance with the principle that an inventor may define the concept of a term or word to best describe their invention, they should be interpreted as meanings and concepts consistent with the technical idea of ​​the present invention. Furthermore, it should be understood that the embodiments and configurations shown in the drawings herein represent only one embodiment of the present invention and do not represent the entire technical idea of ​​the present invention; therefore, various equivalents, variations, and applicable examples may exist that can replace them at the time of filing. The terms first, second, A, B, etc., as used herein and in the claims, may be used to describe various components, but such components shall not be limited by such terms. The terms are used solely for the purpose of distinguishing one component from another. For example, without departing from the scope of the present invention, the first component may be named the second component, and similarly the second component may be named the first component. The terms "and / or" include a combination of multiple related items or any of multiple related items. The terms used in this specification and the claims are merely used to describe specific embodiments and are not intended to limit the present invention. Singular expressions include plural expressions unless the context clearly indicates otherwise. It should be understood that terms such as "comprising" or "having" in this application do not preclude the presence or addition of features, numbers, steps, operations, components, parts, or combinations thereof described in the specification. Unless otherwise defined, all terms used in this specification, including technical or scientific terms, have the same meaning as commonly understood by those of ordinary skill in the technical field to which the present invention pertains.

[0010] Terms defined as in a commonly used dictionary should be interpreted as having a meaning consistent with the meaning in the context of the related art and should not be interpreted in an ideal or overly formal sense unless clearly defined in this application. Furthermore, each configuration, process, step, or method included in each embodiment of the present invention can be shared within a range that is technically non - contradictory to each other. Hereinafter, referring to FIGS. 1 to 12, an inspection system and method for industrial structures according to some embodiments of the present invention will be described in detail. FIG. 1 is a conceptual diagram for explaining an inspection system for industrial structures according to some embodiments of the present invention, FIG. 2 is a block diagram for explaining the configuration of a control device in detail, and FIG. 3 is a block diagram for explaining the configuration of a drone in detail. Referring to FIG. 1, an inspection system (10) for industrial structures according to some embodiments of the present invention can include a route generation device (100), a plurality of control devices (200), and a plurality of drones (300). The inspection system (10) for industrial structures can be a system that performs inspections on industrial structures using drones (300). In an embodiment, the industrial structure can mean a large-scale facility that requires a lot of personnel and costs for inspection or is dangerous for humans to directly perform inspections. The industrial structure may be a factory or a building constructed for the purpose of corporate operation, but is not limited thereto, and a large passenger aircraft illustratively shown in FIG. 1 may also be included in such industrial structures.

[0011] The route generation device (100), the plurality of control devices (200), and the plurality of drones (300) can exchange data via a network. In this case, the network can include a network based on wired Internet technology, wireless Internet technology, and short-range communication technology. The wired Internet technology can include, for example, at least one of a local area network (LAN) and a wide area network (WAN). Wireless internet technologies may include, for example, at least one of the following: Wireless LAN (WLAN), DLNA (Digital Living Network Alliance), Wibro (Wireless Broadband), WiMAX (World Interoperability for Microwave Access), HSDPA (High Speed ​​Downlink Packet Access), HSUPA (High Speed ​​Uplink Packet Access), IEEE 802.16, LTE (Long Term Evolution), LTE-A (Long Term Evolution-Advanced), Wireless Mobile Broadband Service (WMBS), and 5G NR (New Radio) technologies. However, this embodiment is not limited to these. Short-range communication technologies may include, for example, at least one of the following: Bluetooth, RFID (Radio Frequency Identification), Infrared Data Association (IrDA), UWB (Ultra-Wideband), ZigBee, Near Field Communication (NFC), Ultra Sound Communication (USC), Visible Light Communication (VLC), Wi-Fi, Wi-Fi Direct, and 5G NR (New Radio). However, this embodiment is not limited to these.

[0012] A drone (300), a control device (200), and a route generator (100) communicating over a network can comply with technical standards and standard communication schemes for mobile communications. For example, the standard communication scheme may include at least one of GSM (Global System for Mobile communication), CDMA (Code Division Multi Access), CDMA2000 (Code Division Multi Access 2000), EV-DO (Enhanced Voice-Data Optimized or Enhanced Voice-Data Only), WCDMA (Wideband CDMA), HSDPA (High Speed ​​Downlink Packet Access), HSUPA (High Speed ​​Uplink Packet Access), LTE (Long Term Evolution), LTEA (Long Term Evolution-Advanced), and 5G NR (New Radio). However, this embodiment is not limited thereto. For example, each drone (300) can be connected to its corresponding control unit (200) using short-range communication technology, and each control unit (200) and route generator (100) can be connected using wired or wireless internet technology. However, it is not limited to this. In this embodiment, the path generation device (100) can receive a plurality of inspection points for inspecting an industrial structure. The path generation device (100) can generate a travel path for a plurality of drones to move between the generated plurality of inspection points. The process of generating the travel path by the path generation device (100) will be described later. The travel path generated by the path generation device (100) may include a plurality of sub-travel paths divided according to the number of drones, and each drone (300) and its corresponding control device (200) may be provided with one of the plurality of sub-travel paths.

[0013] The control device (200) may be a user device for controlling the corresponding drone (300). Referring to Figure 2, the control device (200) may include a control unit (210), a display unit (220), a communication unit (230), a processor (240), and memory (250). The control unit (210) can receive user input. Exemplarily, the control unit (210) may include, but is not limited to, a joystick for controlling the drone's direction and at least one button for inputting user commands. In response to user input via the control unit (210), the processor (240) can generate user control signals to control the direction of movement of the drone (300) or to control the operation of components included in the drone (300). The memory (250) is configured to store data necessary for data processing by the processor (240) or data generated by the data processing of the processor (240). The communication unit (230) can be configured to exchange data with at least one of the route generator (100) and the drone (300) via the network. The processor (240) transmits the generated control signals to the drone (300) via the communication unit (230) so that the user can manually fly the drone (300). Furthermore, the processor (240) can display the sub-movement path received from the path generation device (100) on the display unit (220), and assist the user in manually operating the drone (300) by referring to the sub-movement path. In other words, the user can operate the drone (300) by referring to the sub-movement path displayed on the display unit (220), and can operate the drone (300) to follow the sub-movement path. Referring to Figure 3, the drone (300) may include a main body for flight (310), sensors (320), a communication unit (330), a processor (340), and memory (350). The main body (310) is configured to levitate the drone (300) and move it in one direction, taking into account the lift, gravity, thrust, and drag acting on the drone (300). The main body (310) may include multiple rotors configured to overcome gravity, generate lift to levitate the aircraft, and generate thrust to overcome air resistance. The main body (310) can move in one direction or change direction depending on the rotation speed of the rotors.

[0014] The sensor (320) can detect the structure of an industrial structure and generate sensing information. The sensor (320) may include an image sensor, and the sensing information may be image information generated by capturing the outline of the industrial structure. The user can check for cracks and damage to the outline of the industrial structure through the image information. Furthermore, the sensor (320) may include a LiDAR sensor, and the sensing information detected via the LiDAR sensor can be used to maintain the distance between the drone (300) and the industrial structure. In addition, the sensor (320) may be configured to include GPS, and the sensing information of the drone (300) may include location information. The communication unit (330) can be configured to exchange data with a corresponding control device (200) or route generator (100) via a network. In response to control signals provided from the control device (200) via the communication unit (330), the processor (340) controls the main unit (310) and sensors (320) to perform inspections of industrial structures using a drone (300). The memory (350) is configured to store data necessary for the data processing process by the processor (340) or data generated by the data processing by the processor (340). Here, the drone (300) can be manually flown under the control of a corresponding control device (200), but is not limited to this. The drone (300) can be configured to fly automatically along a sub-travel path generated by a path generation device (100). That is, the processor (340) can control the main body (310) and the sensors (320) respectively so that the drone (300) inspects industrial structures while flying automatically following the sub-travel path. In this embodiment, among the sensing information generated by the drone (300), image information and position information can be provided in real time or near real time by the corresponding control device (200). The processor (240) of the control device (200) can display the image information on the display unit (220) to help the user check the status of the industrial structure in real time or near real time. The processor (240) of the control device (200) can also display the position information together with the sub-movement path displayed on the display unit (220) to help the user check the real-time position of the drone (300) and control the drone (300). In manual operation mode, the user can control the drone (300) to follow the sub-movement path, taking into account the drone's real-time position. In automatic operation mode, the user can monitor the autonomous flight state of the drone (300) by comparing the drone's position with the sub-movement path. In an exemplary embodiment, if the user confirms that the drone (300) has deviated from the sub-path, the user can change the operating state of the drone (300) to manual operation mode and manually operate the drone (300). The route generation device (100) can generate a travel route for inspecting an industrial structure based on multiple inspection points of the industrial structure. The process of generating a travel route performed by the route generation device will be described below with reference to Figures 4 to 10.

[0015] Figure 4 is a block diagram illustrating the configuration of the path generation device in detail. Figure 5 is an illustrative diagram illustrating the process of generating travel paths in the path generation device. Figure 6 is an illustrative image visualizing a connected graph. Figures 7 to 9 are illustrative images illustrating the process of completing the connected graph. Figure 10 is an illustrative image showing the generated travel path. Drone inspections of industrial structures can be performed by either having the drone visit specific inspection points or by having the drone move along a path connecting the inspection points and inspecting the configuration along that path. Therefore, by defining multiple inspection points based on the structure of the industrial structure and determining the optimal order in which the drone visits these multiple inspection points, an optimal travel path can be generated. The problem of multiple drones (300) determining the order in which to visit multiple inspection points is an NP-hard problem, and it is a vehicle routing problem with constraints on vehicle capacity, taking into account the battery capacity limitations of the drones (300). The route generation device (100) according to an embodiment of the present invention can construct a vehicle routing problem model based on multiple inspection points of an industrial structure, solve the constructed vehicle routing problem model using a metaheuristic or exact method to derive an optimal solution, and generate movement paths for multiple drones based on the derived solution. Specifically, referring to Figure 4, the route generation device (100) includes a data transmission / reception unit (110), a route identification unit (120), a graph completion unit (130), a problem solving unit (140), and a route generation unit (150). In Figure 4, the data transmission / reception unit (110), the route identification unit (120), the graph completion unit (130), the problem solving unit (140), and the route generation unit (150) are shown as separate blocks distinct from each other. This is because the components constituting the route generation device (100) are functionally divided according to the operations performed by each component. Therefore, depending on the embodiment, some or all of the above components can be integrated into the same single device. Furthermore, Figure 5 shows the main processes for generating a travel path in the path generation device (100) and the relationships between each component, matching them together. The following explanation will be given with reference to Figures 4 and 5.

[0016] The data transmission unit (110) can receive data from multiple inspection points of the industrial structure to be inspected. The multiple inspection points of the industrial structure may be in a predefined state, taking into account the characteristics and specifications of the industrial structure, the inspection distance of the drone, etc. The multiple inspection points can be defined as locations corresponding to specific components of the industrial structure that require detailed inspection, but are not limited to this, and can also be defined as locations for generating a path to pass through specific components that require inspection. The drone (300) can inspect the main parts of the industrial structure while remaining in a fixed position for a certain period of time at each inspection point, and can perform inspections of the main parts of the industrial structure while moving between inspection points. Furthermore, the data transmission / reception unit (110) can also receive industrial structure information, including information on the characteristics and specifications of the industrial structure. In this embodiment, the multiple inspection points of the industrial structure and the industrial structure information may be, but are not limited to, data received from the industrial structure's control server (not shown). Furthermore, the data transmission / reception unit (110) can be configured to transmit the movement path generated through the process described later to at least one of the multiple drones (300) and multiple control devices (200). The route identification unit (120) can identify traversable and traversable routes at multiple inspection points, taking into account industrial structure information. In theory, multiple inspection points can be connected to form a path. However, considering the actual structure of industrial structures and the flight characteristics of drones (300) which are not adept at navigating curves, there may be points where movement is difficult or practically impossible. In other words, due to the geometric limitations of the industrial structure, it is possible that not all multiple inspection points are connected. The route identification unit (120) can identify routes where drone movement is impossible, taking into account industrial structure information. The route identification unit (120) identifies routes where drone movement is impossible, taking into account industrial structure information, and does not need to configure the connection relationships of inspection points that correspond to routes where movement is impossible. The route identification unit (120) can define routes of multiple inspection points as routes where drone movement is possible, rather than routes where drone movement is physically impossible.

[0017] The route identification unit (120) can determine the route based on a 3D model realized based on industrial structure information. In the embodiment, the route identification unit (120) can realize a 3D model of an industrial structure based on industrial structure information, and can reflect and display multiple determined inspection points together with the 3D model of the industrial structure. The route identification unit (120) uses the multiple inspection points displayed together with the 3D model of the industrial structure to identify routes that the drone (300) can travel and routes that the drone (300) cannot travel. The route identification unit (120) can determine a traversable route between multiple inspection points and determine a weight for the determined route. Here, the weight may be, but is not limited to, the travel cost applied in proportion to the distance between inspection points or the travel time between inspection points. Since the inspection points are not connected for routes that cannot be traversed, the route identification unit (120) does not need to set a weight for routes that cannot be traversed. The route identification unit (120) can connect all traversable routes between multiple inspection points and represent the connection relationships between the multiple inspection points in a connected graph. As shown in Figure 6, the connection relationships can be communicated to the user as visualized information in the connected graph. In the case of a connected graph constructed via a path identification unit (120), the state in which only traversable paths are connected corresponds to an incomplete graph. Referring to Figure 7, it can be seen that in the connected graph of nodes 1 to 5, no connection relationships are formed between node 1 and node 2, node 1 and node 3, node 2 and node 4, node 3 and node 5, and node 4 and node 5. In the adjacency matrix representing the relationships between nodes 1 to 5, it can be seen that the path in question is represented by 0, and paths with a connection relationship are represented by 1. In other words, pairs of inspection points that are determined to be traversable paths and for which no connection relationship is formed can be represented by 0 in the adjacency matrix. A complete graph is a graph in which all nodes (inspection points) are connected. Referring to Figure 8, we can see that nodes 1 through 5 are in a state where all connections have been formed. That is, in the adjacency matrix showing the relationships between nodes 1 through 5, only the diagonal matrix is ​​represented by 0. In the embodiment, the vehicle path problem model (VRP) can only be solved using a complete graph. That is, the connection relationships of the multiple inspection points linked via the path identification unit (120) are in an incomplete connection state, and completion is required to solve the vehicle path problem model (VRP). The graph completion unit (130) can generate a virtual path for completing the connectivity relationships between multiple inspection points. Here, completion can mean forming all connectivity relationships between multiple inspection points.

[0018] First, the graph completion unit (130) can generate virtual paths for pairs of inspection points that are not connected to each other in a connected graph, so that a connection relationship is defined. The graph completion unit (130) can use Dijkstra's algorithm to search for paths corresponding to the zero elements among the elements of the adjacency matrix, which shows the connection relationships of multiple inspection points, excluding the diagonal matrix. The graph completion unit (130) searches for the zero elements among the elements excluding the diagonal matrix and can define virtual paths for the connection relationships of the corresponding pairs of inspection points. Here, a virtual path can mean a detour path connected through at least one checkpoint. The graph completion unit (130) can set detour paths for pairs of checkpoints that are not connected. Referring to Figure 9, the path set between node 1 and node 2 (dotted line) is an implicit virtual path, and the path actually applied can be defined as a detour path passing through other pairs of checkpoints. This allows multiple checkpoints to be transformed into a completed state that can constitute a complete graph. That is, the adjacency matrix of the complete graph of multiple checkpoints can be a state where all elements except the diagonal matrix are non-zero. In such a completed state, a solution for multiple checkpoints is possible. Also, a detour path is a path connected via at least one checkpoint, and it may be given a higher weight than a state where pairs of checkpoints are directly connected in a connected graph. That is, a high detour cost is applied and may be naturally excluded in the process of deriving the optimal path described later. In another embodiment, the graph completion unit (130) can identify unconnected inspection points and define a virtual path by assigning arbitrary weights to the identified inspection points. The arbitrary weights correspond to values ​​higher than the weights based on the distance and travel time of pairs of inspection points, resulting in a higher detour cost, which is naturally eliminated in the process of deriving the optimal path described later. The problem-solving unit (140) can solve the problem by constructing a vehicle path model to find the optimal path between multiple fully connected inspection points, i.e., inspection points represented by a complete graph. The problem-solving unit (140) can solve the optimal path by constructing a vehicle path problem model (CVRP) with constraints on vehicle capacity that take into account the battery capacity limitations of the drone (300). Furthermore, the vehicle path problem model can be formalized by adding conditions to consider multiple drones and the battery capacity (usage time) of the drones, so that inspection points overlapping between sub-travel paths correspond exclusively to the starting point (depot). Therefore, multiple drones corresponding to the number of sub-travel paths can be operated simultaneously, enabling inspection of industrial structures. In the embodiment, the vehicle path problem model that optimizes the objective function while satisfying the following constraints can be constructed by a linear integer programming method as shown in Equation 1 below.

[0019]

number

[0020] Constraint (5) explicitly includes a capacity constraint, allowing us to confirm that the total flight time of the drone is less than or equal to the flight time corresponding to the drone's battery capacity. This allows us to set the optimal route considering the drone's battery capacity. The sub-tour prevention constraint (6) ensures that the path always starts from the starting point and ends at the starting point. The remaining essential constraint (7) specifies the definition domain of the variables, and the binary decision variables can be either 0 or 1. The problem-solving unit (140) can solve the formalized vehicle path problem model (CVRP) as described above using either an exact solution or a metaheuristic solution. In the embodiment, the problem-solving unit (140) can determine the solution method according to the number of inspection points generated considering the industrial structure information. The problem-solving unit (140) may include preset threshold values ​​for reference inspection points according to the type and size of the industrial structure. The problem-solving unit (140) can determine a method for solving the Vehicle Path Problem Model (CVRP) according to a threshold for the reference inspection points. If the number of inspection points exceeds the threshold, accurate calculations may consume a lot of time and resources, making it difficult to respond immediately to on-site requirements. To produce efficient results, it is necessary to appropriately select a solution method. The problem-solving unit (140) can solve the problem using an exact solution method if the number of inspection points is below a threshold. In the embodiment, the problem-solving unit (140) can solve the vehicle path problem model via the branch and bound method. The branch and bound method systematically enumerates all candidate solutions, estimates upper and lower limits for the values ​​to be optimized, and eliminates solutions that are determined to be impossible. Since solutions derived from the eliminated solutions are not considered, it is possible to reduce unnecessary time consumption and derive the optimal solution. Here, if the number of inspection points (nodes) increases, the difference in computation time between the two methods may increase exponentially. The problem-solving unit (140) can use a metaheuristic solution if the number of inspection points exceeds a threshold. In the case of industrial structures, a large number of inspection points may be derived due to the complexity and size of the structure, and for such industrial structures, problem-solving using a metaheuristic solution may be a more appropriate method.

[0021] Referring to Figure 5, the problem-solving unit (140) can search for a solution based on weights corresponding to the interconnection relationships of multiple inspection points. The problem-solving unit (140) can evaluate the degree of suitability of the searched solution by performing a suitability evaluation based on the operational efficiency of multiple drones. Specifically, the problem-solving unit (140) can check whether the inspection time for each drone is maximized and perform a suitability evaluation so that an efficient movement path can be derived with a small number of drones. The problem-solving unit (140) in the embodiment can output an optimal solution via a suitability determination algorithm min(max(inspection time for each drone)) that maximizes the inspection time for each drone and minimizes the number of drones. This makes it possible to generate movement paths for multiple drones that make maximum use of multiple inspection points with a minimum number of drones. The problem-solving unit (140) terminates problem-solving if the evaluated fitness satisfies the algorithm termination condition, and if the evaluated fitness does not satisfy the algorithm termination condition, it changes the graph weights and repeats problem-solving. The problem-solving unit (140) terminates problem-solving and outputs the results when the evaluated fitness satisfies the algorithm termination condition or when it has performed problem-solving for a predetermined number of iterations. In the embodiment, the problem-solving unit (140) can derive the optimal solution through an ant colony system (ACS) within a metaheuristic solution method. The ant colony system is a metaheuristic solution method that mimics the ability of real ants to find the shortest path from food to their home. Specifically, the ant colony system (ACS) is a system that applies the principle of heuristic search to ants called agents secreting pheromones along each path as they move toward their destination, and then the agents passing through that path use the pheromone information accumulated in that path to select the next path. The problem-solving unit (140) can further utilize the pheromone information accumulated in existing paths as positive feedback information to search for a solution. That is, the problem-solving unit (140) can perform problem solving (complete the search path) by further utilizing positive feedback (pheromone information) in addition to weights corresponding to the connection relationships of multiple inspection points, thereby outputting an even more optimized path.

[0022] The path generation unit (150) can generate a travel path based on the optimal solution derived from the problem solving unit (140) and the complete graph. The travel path may include sub-travel paths in which multiple drones (300) each move to multiple inspection points. The number of drones (300) can be the minimum possible, and the sub-travel paths can be configured so that each drone has as much flight time as possible. Nodes overlapping between sub-travel paths may have only one starting point (depot). Therefore, multiple drones, corresponding to the number of sub-travel paths, are operated simultaneously to perform inspections of industrial structures. Furthermore, in the embodiment, the problem-solving unit (140) can display the generated movement path together with the 3D model of the industrial structure. Figure 10 illustrates the schematic 3D model of the industrial structure and the generated movement path together, showing that four sub-movement paths constitute the movement path. That is, each of the four drones can perform inspections of the industrial structure along its corresponding sub-movement path. Furthermore, in the embodiment, the movement path may further include speed information corresponding to the inspection points. This allows the drone (300) to move sequentially to multiple inspection points in the order corresponding to the movement path, but to be controlled to move along each path according to a set speed. The path generation device for generating movement paths for multiple drones for inspecting industrial structures according to an embodiment of the present invention can identify paths that the drones can and cannot move along at multiple inspection points based on industrial structure information, and derive the optimal movement path by optimizing the vehicle path problem model using a metaheuristic solution method to maximize the operational efficiency of the multiple drones. Furthermore, the route generation device for generating travel routes for inspecting industrial structures is configured to further solve the vehicle route problem model using an exact solution method. Based on the number of inspection points, it can determine the solution method from either an exact solution method or a metaheuristic solution method. Therefore, it can communicate accurate results while flexibly responding to on-site conditions. Hereinafter, with reference to Figures 11 to 12, a method for generating movement paths for multiple drones for inspecting industrial structures according to several embodiments of the present invention will be described. The method for generating movement paths for multiple drones for inspecting industrial structures according to the embodiments is performed using the path generation device shown in Figures 1 to 10, and parts that overlap with the embodiments described above will be omitted or simplified.

[0023] Figure 11 is a flowchart of a method for generating movement paths for multiple drones for inspecting industrial structures according to some embodiments of the present invention. Figure 12 is a flowchart illustrating the detailed steps of step (S140) in Figure 11. Referring to Figure 11, the method for generating movement paths for multiple drones for inspecting industrial structures according to the embodiment is: The process includes the steps of: receiving industrial structure information and multiple inspection points of the industrial structure (S100); identifying possible and impossible routes for the drone to travel on at the multiple inspection points, taking into consideration the industrial structure information (S110); defining the connection relationships of the inspection points corresponding to the possible routes for the drone (S120); defining virtual routes between inspection points for which the connection relationships are not defined to complete the relationships between the multiple inspection points (S130); solving a vehicle path problem model constructed based on the multiple inspection points and outputting a solution (S140); and generating the travel path based on the output solution (S150). Here, the aforementioned movement path may include sub-movement paths corresponding to each of the multiple drones. In the embodiment, the step of outputting the solution (S140) may include solving the vehicle path problem model through a metaheuristic solution method configured to search for a solution based on weights corresponding to the connectivity of the plurality of inspection points, evaluate the degree of fit by performing a suitability evaluation of the searched solution based on the operational efficiency for the plurality of drones, terminate problem solving if the evaluated degree of fit satisfies the algorithm termination condition, and repeat problem solving by changing the weights if the evaluated degree of fit does not satisfy the algorithm termination condition. In the embodiment, the step of outputting the solution (S140) can be completed and the retrieved solution output once the problem solving process has been performed a predetermined number of times. In the embodiment, the metaheuristic solution method is an ant colony system (ACS), and the step of outputting the solution (S140) may include further utilizing pheromone information accumulated in the existing pathway as positive feedback to search for the solution.

[0024] In the embodiment, the step of outputting the solution (S140) can be further configured to solve the vehicle path problem model using an exact method. In the embodiment, the step of outputting the solution (S140) may include determining the solution method for the vehicle path problem model from among the metaheuristic solution method or the exact solution method, depending on the number of inspection points of the industrial structure. Referring to Figure 12, the step of outputting the solution (S140) may include the steps of: comparing the number of inspection points of the industrial structure with a preset threshold for reference inspection points (S142); solving the vehicle path problem model using a metaheuristic solution method if the number of inspection points of the industrial structure exceeds the preset threshold for reference inspection points (S144); and solving the vehicle path problem model using the exact solution method if the number of inspection points of the industrial structure is less than or equal to the preset threshold for reference inspection points (S146). In the embodiment, the vehicle path problem model can be configured to take into account the battery capacities of the multiple drones and to treat inspection points that overlap between the sub-travel paths exclusively as starting points (depots). In the embodiment, the step of completing the relationships between multiple inspection points (S130) may include searching for paths corresponding to elements that are 0 among the elements of the adjacency matrix showing the connectivity relationships between the multiple inspection points, excluding the diagonal matrix, to identify pairs of inspection points for which connectivity relationships have not been formed. In the embodiment, the virtual path can be defined as a detour path that connects pairs of inspection points that do not have the aforementioned connection relationship through at least one inspection point. In the embodiment, the step (S110) of identifying possible and impossible routes for the drone at the multiple inspection points, taking into account the industrial structure information, involves generating a 3D model of the industrial structure based on the industrial structure information, and using the 3D model of the industrial structure to identify possible and impossible routes for the drone at the multiple inspection points. The method for generating movement paths for multiple drones for industrial structure inspection according to the embodiment can also be embodied in the form of a computer-readable medium that stores computer-executable instructions and data. In this case, the instructions and data can be stored in the form of program code, which, when executed by a processor, can generate a predetermined program module and perform a predetermined operation. Furthermore, the computer-readable medium can be any available medium accessible by a computer, including all volatile and non-volatile media, and isolated and non-isolated media. Furthermore, the computer-readable medium can be a computer recording medium, which can include all volatile and non-volatile, isolated and non-isolated media embodied in any method or technique for storing information such as computer-readable instructions, data structures, program modules, or other data. For example, a computer recording medium may be a magnetic storage medium such as HDDs and SSDs, an optical recording medium such as CDs, DVDs and Blu-ray discs, or memory contained in a server accessible via a network.

[0025] Furthermore, the method for generating movement paths for multiple drones for inspecting industrial structures according to the embodiment can be embodied in a computer program (or computer program product) that includes instructions executable by a computer. The computer program includes programmable machine instructions processed by a processor and can be embodied in a high-level programming language, object-oriented programming language, assembly language, or machine language, etc. The computer program is also recorded on a tangible computer-readable recording medium (e.g., memory, hard disk, magnetic / optical medium, or SSD (Solid-State Drive)). Therefore, the method for generating movement paths for multiple drones for inspecting industrial structures according to the embodiment can be realized by executing the computer program described above on a computing device. The computing device may include a processor, memory, storage device, a high-speed interface connected to the memory and high-speed expansion ports, and at least some of the low-speed interfaces connected to the low-speed bus and storage device. Each of these components is connected to one another using various buses and is mounted on a common motherboard or in any other suitable manner. Here, the processor can process instructions within the computing device, such as instructions stored in memory or storage devices to display graphic information for providing a GUI (Graphical User Interface) on external input and output devices, such as a display connected to a high-speed interface. In another embodiment, a large number of processors and / or a large number of buses can be appropriately used with a large number of memories and memory configurations. Furthermore, the processor can be embodied in a chipset consisting of chips containing a large number of independent analog and / or digital processors. Memory also stores information within a computing device. In one example, memory may consist of volatile memory units or a collection thereof. In another example, memory may consist of non-volatile memory units or a collection thereof. Furthermore, memory may be other forms of computer-readable media, such as magnetic disks or optical disks. Furthermore, storage devices can provide computing devices with large-capacity storage space. Storage devices may be computer-readable media or configurations that include such media. For example, they may include devices within a Storage Area Network (SAN) or other configurations, and may be floppy disk drives, hard disk drives, optical disk drives, tape drives, flash memory, other similar semiconductor memory devices, or device arrays.

[0026] The above description is merely illustrative of the technical concept of this embodiment, and any person with ordinary skill in the art to which this embodiment belongs could make various modifications and variations without departing from the essential characteristics of this embodiment. Therefore, this embodiment is for illustrative purposes only, not to limit the technical concept of this embodiment, and the scope of the technical concept of this embodiment is not limited by such embodiment. The scope of protection of this embodiment should be interpreted in accordance with the following claims, and all technical concepts within an equivalent scope should be interpreted as being included in the scope of rights of this embodiment.

Claims

1. A path generation device for generating movement paths for multiple drones for inspecting industrial structures, A data transmission / reception unit that receives industrial structure information and multiple inspection points of industrial structures; A route identification unit that, taking into account the aforementioned industrial structure information, identifies routes in which the drone can and cannot move at a plurality of inspection points, and defines the connection relationships of inspection points corresponding to the routes in which the drone can move; A graph completion unit that defines virtual paths between inspection points where the aforementioned connection relationships are not defined, thereby completing the relationships between the multiple inspection points; A problem-solving unit that solves a vehicle route problem model configured based on the aforementioned multiple inspection points and outputs a solution; and The system includes a path generation unit that generates movement paths for multiple drones based on the output solution, The movement paths of the plurality of drones include a plurality of sub-movement paths on which each of the plurality of drones moves, and the plurality of sub-movement paths include a first sub-movement path for the movement path of the first drone and a second sub-movement path for the movement path of the second drone. The problem-solving unit solves the vehicle path problem model through a metaheuristic solution method configured to search for a solution based on weights corresponding to the connection relationships of the plurality of inspection points, evaluate the degree of fit by performing a suitability evaluation of the searched solution based on the operational efficiency for the plurality of drones, terminate the problem-solving if the evaluated degree of fit satisfies the algorithm termination condition, and repeat the problem-solving by changing the weights if the evaluated degree of fit does not satisfy the algorithm termination condition. The problem-solving unit terminates the problem-solving process and outputs the found solution when it has performed problem-solving a predetermined number of times. Route generation device.

2. The aforementioned metaheuristic solution is for an ant colony system (ACS). The problem-solving unit further utilizes the pheromone information accumulated in the existing pathway as positive feedback to search for a solution. The route generation device according to claim 1.

3. The problem-solving unit is further configured to solve the vehicle path problem model using an exact solution method. The problem-solving unit determines, from among the metaheuristic solution method and the exact solution method, a method for solving the vehicle path problem model according to the number of inspection points of the industrial structure. The problem-solving unit solves the vehicle path problem model using the metaheuristic solution method if the number of inspection points of the industrial structure exceeds a preset threshold of reference inspection points, and solves the vehicle path problem model using the exact solution method if the number of inspection points of the industrial structure is less than or equal to the preset threshold of reference inspection points. The route generation device according to claim 1.

4. The aforementioned vehicle route problem model is, Considering the battery capacities of the aforementioned multiple drones, the starting point included in the first sub-movement path and the starting point included in the second sub-movement path are the same, and the inspection points other than the starting points included in the first sub-movement path and the second sub-movement path are configured to be different from each other. The route generation device according to claim 1.

5. The aforementioned graph completion unit, In the adjacency matrix showing the connectivity relationships of the aforementioned multiple inspection points, paths corresponding to the elements that are 0 among the elements excluding the diagonal matrix are searched for to identify pairs of inspection points that do not have a connectivity relationship. The route generation device according to claim 1.

6. The aforementioned graph completion unit, The virtual path is defined as a detour path that connects pairs of inspection points that do not have the aforementioned connection relationship through at least one inspection point. The route generation device according to claim 5.

7. The route identification unit is, Based on the aforementioned industrial structure information, a three-dimensional model of the industrial structure is generated, and using the three-dimensional model of the industrial structure, paths along which the drone can move and paths along which the drone cannot move are identified at the multiple inspection points. The route generation device according to claim 1.

8. A method for generating movement paths for multiple drones for inspecting industrial structures, The stage of receiving industrial structure information and multiple inspection points of industrial structures; Taking into account the aforementioned industrial structure information, the step of identifying the paths along which the drone can and cannot move at the multiple inspection points; The step of defining the relationships between inspection points that correspond to the path on which the drone can move; A step of defining virtual paths between inspection points where the aforementioned connection relationships are not defined, thereby completing the relationships between the multiple inspection points; The step of solving a vehicle route problem model constructed based on the aforementioned multiple inspection points and outputting a solution; and The process includes the step of generating the travel path based on the output solution, The movement paths of the plurality of drones include a plurality of sub-movement paths on which each of the plurality of drones moves, and the plurality of sub-movement paths include a first sub-movement path for the movement path of the first drone and a second sub-movement path for the movement path of the second drone. The step of outputting the solution includes solving the vehicle path problem model through a metaheuristic solution method configured to search for a solution based on weights corresponding to the connectivity of the plurality of inspection points, evaluate the degree of fit by performing a suitability evaluation of the searched solution based on the operational efficiency for the plurality of drones, terminate problem solving if the evaluated degree of fit satisfies the algorithm termination condition, and repeat problem solving by changing the weights if the evaluated degree of fit does not satisfy the algorithm termination condition. The step of outputting the aforementioned solution involves ending the problem-solving process and outputting the found solution after the problem-solving process has been repeated a predetermined number of times. A method for generating movement paths for multiple drones for inspecting industrial structures.

9. Multiple drones for inspecting industrial structures; Multiple control devices corresponding to each of the aforementioned multiple drones; and An industrial structure inspection system comprising a path generation device that generates movement paths for a plurality of drones, wherein the movement path includes a sub-movement path corresponding to each of the plurality of drones, and the path generation device provides the sub-movement path to at least one of the corresponding drones and control devices, The aforementioned route generation device is A data transmission / reception unit that receives industrial structure information and multiple inspection points of industrial structures; A route identification unit that, taking into account the aforementioned industrial structure information, identifies routes in which the drone can and cannot move at a plurality of inspection points, and defines the connection relationships of inspection points corresponding to the routes in which the drone can move; A graph completion unit that defines virtual paths between inspection points where the aforementioned connection relationships are not defined, thereby completing the relationships between the multiple inspection points; A problem-solving unit that solves a vehicle route problem model configured based on the aforementioned multiple inspection points and outputs a solution; and The system includes a path generation unit that generates movement paths for multiple drones based on the output solution, The movement paths of the plurality of drones include a plurality of sub-movement paths on which each of the plurality of drones moves, and the plurality of sub-movement paths include a first sub-movement path for the movement path of the first drone and a second sub-movement path for the movement path of the second drone. The problem-solving unit solves the vehicle path problem model through a metaheuristic solution method configured to search for a solution based on weights corresponding to the connection relationships of the plurality of inspection points, evaluate the degree of fit by performing a suitability evaluation of the searched solution based on the operational efficiency for the plurality of drones, terminate the problem-solving if the evaluated degree of fit satisfies the algorithm termination condition, and repeat the problem-solving by changing the weights if the evaluated degree of fit does not satisfy the algorithm termination condition. The problem-solving unit terminates the problem-solving process and outputs the found solution when it has performed problem-solving a predetermined number of times. Inspection system for industrial structures.

10. The drone is configured to fly autonomously along the travel path or to fly in accordance with the control signals of the control device. The drone is further configured to generate location information and provide the generated location information to the control device. The control device displays the travel path and the location information together on a display and provides them to the user. The inspection system for industrial structures according to claim 9.

Images