Smart control tower system for operating wirelessly controllable vehicles in aerial, ground, and underwater environments

Abstract

A smart control tower system for operating mobiles capable of wireless control in aerial, terrestrial, and underwater environments is disclosed. The smart control tower system comprises: a plurality of mobiles, each comprising a sensor module that collects image data, location information, and environmental data, and a drive module that receives control commands from a main control tower or a sub-control tower below and moves autonomously; a main control tower that generates the control commands to control the plurality of mobiles and performs disaster response judgments by analyzing the collected image data, location information, and environmental data; and a sub-control tower that receives control over the plurality of mobiles from the main control tower via a wireless network to respond to disaster sites that are difficult for the main control tower to access, and controls the plurality of mobiles independently or in cooperation based on the collected image data, location information, and environmental data.

Description

Technology Field

[0001] One embodiment of the present invention relates to a smart control tower system, and more specifically, to a smart control tower system for operating a mobile body capable of wireless control in aerial, terrestrial, and underwater environments. Background Technology

[0002] Recently, various disaster situations have been occurring frequently due to factors such as climate change, urban densification, and the advancement of industrial facilities. Consequently, the speed and accuracy of on-site response are receiving attention as key elements of disaster response systems (hereinafter also referred to as 'smart control tower systems'). In particular, existing human-centered response methods show limitations in responding to high-risk areas that are difficult to physically access; as a result, research on disaster response technologies utilizing unmanned vehicles is actively underway.

[0003] Conventionally, technologies utilizing single unmanned aerial vehicles (UAVs) or ground vehicles have primarily been applied to collect video data at disaster sites or perform simple reconnaissance missions. While these single-vehicle-based systems have the advantages of being relatively simple to implement and having low initial costs, they have limitations when precise and repetitive inspections are required for large areas or complex disaster sites.

[0004] For example, inspections performed by a single drone may result in blind spots and are vulnerable to issues such as flight path optimization, battery efficiency, and communication failures. Furthermore, the control of these unmanned vehicles has largely been based on manual remote operation, which slows down on-site response speeds and acts as an obstacle to ensuring real-time capability.

[0005] Furthermore, while there have been attempts to utilize multiple unmanned vehicles, most were based on individual operation rather than collaboration between the vehicles. Consequently, inefficiencies in operational efficiency arose, such as route duplication, missed inspections, and unbalanced mission distribution among the vehicles. Additionally, since the control of each vehicle was designed to be concentrated in a single control center or central server, issues such as communication failures, control delays, and single points of failure could occur.

[0006] Meanwhile, video data and environmental information collected by mobiles at disaster sites were often monitored directly by humans or interpreted without being quantified. Consequently, the objectivity and reliability of on-site response judgments were compromised, and this also became a cause of response delays. In particular, relying on visual intuition without quantitative criteria to identify physical defects such as structural cracks, corrosion, and deformation made it difficult to guarantee consistency in judgment. These technical limitations also imposed constraints on precise damage analysis and the establishment of long-term disaster prevention plans.

[0007] Furthermore, in existing disaster response systems, significant limitations arise in real-time control or data collection when control towers or operations centers cannot access the center of the disaster site. This is emerging as a serious problem, particularly in areas that are geographically difficult to access or in high-risk zones where hazardous substances have spread. Consequently, existing centralized control methods alone have limitations in proactively responding to various disaster situations, leading to an increasing demand for more flexible and decentralized control structures.

[0008] In conclusion, conventional disaster response systems have limitations such as unmanned mobile systems based on single or individual operation, manual control methods, a focus on the qualitative interpretation of video and environmental data, and centralized control structures; consequently, there is a growing need for a new type of disaster response system to overcome these limitations. There is a demand for an advanced system structure that efficiently operates multiple wirelessly controllable mobile units, enables rapid and quantitative response decisions based on collected data, and includes field-oriented distributed control. Prior art literature

[0009] Korean Registered Patent No. 10-2126502 (Registered June 18, 2020) The problem to be solved

[0010] One embodiment of the present invention aims to provide a smart control tower system that solves the aforementioned problems by enabling the efficient operation of multiple mobile units capable of wireless control in aerial, land, and underwater environments, improving the accuracy and reliability of disaster response through collaborative inspection and quantitative defect analysis among the mobile units, and enabling real-time and stable disaster site response through an organic control distributed structure between main and sub-control towers. means of solving the problem

[0011] A smart control tower system according to one embodiment for achieving the aforementioned purpose comprises: a plurality of mobile bodies, each comprising a sensor module for collecting image data, location information, and environmental data, and a driving module for autonomously moving by receiving control commands from a main control tower or a sub-control tower below; a main control tower that generates the control commands for controlling the plurality of mobile bodies and performs disaster response judgments by analyzing the collected image data, location information, and environmental data; and a sub-control tower that receives control over the plurality of mobile bodies from the main control tower via a wireless network to respond to disaster sites that are difficult for the main control tower to access, and controls the plurality of mobile bodies independently or in cooperation based on the collected image data, location information, and environmental data.

[0012] Meanwhile, since the description of the technology disclosed in this specification is merely an example for structural or functional explanation, the scope of the disclosed technology should not be interpreted as being limited by the examples described in the text. That is, since the examples are subject to various modifications and may take various forms, the scope of the disclosed technology should be understood to include equivalents capable of realizing the technical concept. Furthermore, since the purposes or effects presented in the disclosed technology do not imply that a specific example must include all of them or only such effects, the scope of the disclosed technology should not be understood as being limited by them.

[0013] Furthermore, the meaning of the terms described in this invention should be understood as follows. Terms such as "first," "second," etc., are intended to distinguish one component from another, and the scope of rights shall not be limited by these terms. For example, the first component may be named the second component, and similarly, the second component may be named the first component.

[0014] Furthermore, when it is stated that one component is "connected" to another component, it should be understood that while it may be directly connected to that other component, there may also be other components in between. Conversely, when it is stated that one component is "directly connected" to another component, it should be understood that there are no other components in between. Meanwhile, other expressions describing the relationships between components, such as "between" and "between," or "adjacent to" and "directly adjacent to," should be interpreted in the same way.

[0015] A singular expression should be understood to include a plural expression unless the context clearly indicates otherwise, and terms such as "include" or "have" are intended to specify the existence of the set-up features, numbers, steps, actions, components, parts, or combinations thereof, and should be understood not to preclude the existence or addition of one or more other features, numbers, steps, actions, components, parts, or combinations thereof. Effects of the invention

[0016] As described above, one embodiment of the present invention provides a smart control tower system capable of efficiently integrating and controlling a plurality of wirelessly controlled mobile units operable in aerial, land, and underwater environments, thereby enabling a rapid and safe response to high-risk areas such as disaster sites.

[0017] In addition, one embodiment of the present invention can quantitatively analyze the damage state of a structure through Pix-based quantification and deep learning-based object recognition technology based on image data and environmental data using a moving object, thereby ensuring much higher reliability and objectivity compared to existing subjective and qualitative inspection methods.

[0018] In addition, one embodiment of the present invention can minimize blind spots and automatically identify and supplement areas missed during inspection through a collaboration-based inspection algorithm among multiple mobile bodies and a real-time information sharing function among mobile bodies, and can simultaneously secure autonomy and efficiency by reconstructing inspection routes according to the operating status of mobile bodies, such as battery level and available range.

[0019] In addition, one embodiment of the present invention enables the flexible distribution of control authority between a main control tower and a sub-control tower, thereby allowing for stable on-site control even in situations where central control is difficult due to network failures or geographical limitations, and enables flexible and dynamic response to multiple disaster situations through a function for reassigning the roles of mobile units according to mission priority.

[0020] Accordingly, one embodiment of the present invention provides the effect of overcoming the limitations of existing disaster response systems through the collaborative operation of a plurality of wirelessly controlled mobile bodies, quantitative analysis, and an organic control distributed structure, and enabling the implementation of a smart disaster management system capable of more reliable and real-time response. Brief explanation of the drawing

[0021] FIG. 1 is a diagram schematically illustrating the configuration of a smart control tower system according to one embodiment of the present invention. FIG. 2 is a diagram exemplarily illustrating a plurality of mobile body configurations of a smart control tower system according to one embodiment of the present invention. FIG. 3 is a diagram illustrating the configuration of a main control tower for setting quantitative evaluation and disaster safety based on image data and environmental data according to an embodiment of the present invention. FIG. 4 is a diagram further illustrating a plurality of mobile body configurations for performing a collaboration-based inspection algorithm according to an embodiment of the present invention. FIG. 5 is a diagram illustrating an additional configuration of a main control tower for automatically assigning or switching control authority with a sub-control tower according to an embodiment of the present invention. FIG. 6 is a diagram illustrating an additional configuration of a main control tower for generating a re-shooting command by comparing the measured values ​​of the camera's shooting angle, illumination level, and focal length at the time of shooting with reference values, respectively, according to an embodiment of the present invention, and dynamically reassigning the role of each moving body according to priority. Specific details for implementing the invention

[0022] The present invention will be described in more detail below with reference to preferred embodiments to which the invention belongs.

[0023] FIG. 1 is a diagram schematically illustrating the configuration of a smart control tower system according to one embodiment of the present invention, and FIG. 2 is a diagram exemplarily showing the configuration of a plurality of mobile bodies of a smart control tower system according to one embodiment of the present invention.

[0024] At this time, Fig. 2 is to be cited as an additional reference when explaining the multiple moving bodies of Fig. 1.

[0025] Referring to FIG. 1, a smart control tower system according to one embodiment includes a plurality of mobile bodies (100), a main control tower (200), and a sub-control tower (300) to operate mobile bodies capable of wireless control in aerial, land, and underwater environments. At this time, any two of the plurality of mobile bodies (100), the main control tower (200), and the sub-control tower (300) may be connected by a wireless network (400).

[0026] In one embodiment, a plurality of moving bodies (100) may be objects that can freely move to a desired location (point, area) according to control commands from a main control tower (200) or / and a sub-control tower (300) to be described later.

[0027] For example, the plurality of mobile bodies (100) are preferably unmanned aerial vehicles such as drones, unmanned ground vehicles, or underwater exploration vessels capable of autonomous navigation, but are not necessarily limited thereto. These plurality of mobile bodies (100) can be configured in various sizes and shapes depending on the purpose of operation, and can be equipped with cameras, communication modules, sensor modules, and drive modules to operate according to commands from a remote location.

[0028] For example, a plurality of mobile bodies (100) may include a sensor module (110) that collects image data, location information and environmental data, a driving module (120) that receives control commands from a main control tower (200) and / or a sub-control tower (300) and moves autonomously, and a communication module (130) that performs communication functions with the main control tower (200) and / or the sub-control tower (300).

[0029] Looking at each configuration, the sensor module (110) may include various types of sensors depending on the operating environment and mission purpose of the mobile body (100). For example, the sensor module (110) may include an optical camera, an infrared camera, or a thermal imaging camera for collecting visual information on the driving or flight path of the mobile body (100), and through these image sensors, the condition of structures at the disaster site, surrounding terrain, temperature distribution, etc., can be visually identified.

[0030] In addition, the sensor module (110) according to one embodiment may include a Global Positioning System (GPS) module or other satellite-based position tracking sensor for tracking the current position of the moving body (100) in real time, and furthermore, by additionally including an inertial measurement unit (IMU), a gyroscope, a geomagnetic sensor, etc., it may be possible to enable more precise position correction and direction detection.

[0031] In addition, the sensor module (110) may further include environmental sensors for collecting information such as temperature, humidity, atmospheric pressure, gas concentration, and radiation levels of the surrounding environment. These sensors are useful for obtaining advance information to effectively respond to various types of disasters, such as fire, toxic gas leaks, floods, and collapse risks, and the collected environmental information can be used as a basis for situational judgment and control command generation at the main control tower (200) and sub-control tower (300), as well as for the safe operation of the mobile body itself.

[0032] In this way, the sensor module (110) according to one embodiment is configured to collect complex data such as images, location and environmental information in real time, and by placing sensors in various directions such as the front, side, and bottom of the moving body (100), it is possible to obtain information from a wider range and various angles.

[0033] In one embodiment, the drive module (120) may include a propulsion device and a direction control device for autonomously driving, flying, or navigating the vehicle (100). For example, in the case of an unmanned aerial vehicle, the drive module (120) may include a motor for rotating a rotor blade, a battery pack for power supply, a gyro control device for attitude stabilization, a speed control device, etc., and in the case of an unmanned ground vehicle, it may include an electric motor for driving a wheel or caterpillar track, a steering control device for forward and backward movement and left and right rotation, etc. Additionally, in the case of an underwater vehicle, the drive module (120) may include a propulsion propeller, a rudder, a buoyancy control device, etc.

[0034] These driving modules (120) can automatically follow a path or perform actions such as stopping, turning, and ascending / descending according to control commands received from the main control tower (200) and / or the sub-control tower (300), and in some embodiments, they may be configured to avoid obstacles or move to a specific target location on their own through a built-in autonomous driving algorithm.

[0035] In one embodiment, the communication module (130) may be configured as a wireless communication device for the mobile body (100) to transmit and receive data bidirectionally with the main control tower (200) and the sub-control tower (300). Specifically, the communication module (130) may include LTE, 5G, wireless LAN (Wi-Fi), satellite communication, LoRa (Long Range), or a dedicated RF communication module, and a single or multiple communication technologies may be applied in combination depending on the terrain conditions of the disaster area, network availability, mission characteristics, etc.

[0036] For example, in urban areas, high-resolution video data can be transmitted in real time using a high-speed bandwidth 5G network, and in mountainous or marine areas where communication infrastructure is lacking, continuous transmission and reception of data can be enabled using satellite communication or low-power long-range wireless technology. In addition, the communication module (130) can maintain minimal command execution and data synchronization even when central control is temporarily interrupted through a device-to-device (D2D) communication function between towers or between mobile bodies.

[0037] Next, in one embodiment, the main control tower (200) is preferably a main vehicle that can move freely on land, excluding areas that are difficult to access, but is not necessarily limited thereto.

[0038] This main control tower (200) generates control commands to control multiple mobile bodies (100) and transmits them to the multiple mobile bodies (100), and, for example, can collect image data, location information, and environment data from the multiple mobile bodies (100) through wireless communication.

[0039] For example, the main control tower (200) can determine the current location and movement path of each mobile body (100) in real time, and then generate control commands including individual path planning commands, stop or avoidance commands, search direction change commands, etc. for each mobile body by considering terrain information of the area or a pre-designated mission area.

[0040] These control commands are transmitted to each mobile unit, for example, via a wireless network, and the transmitted commands can be dynamically modified or updated according to the real-time situation or priority of the disaster site. In addition, the main control tower (200) can divide inspection zones so that multiple mobile units do not overlap, and can also include spacing control commands to prevent communication congestion or collisions.

[0041] Accordingly, a main control tower (200) according to one embodiment can perform disaster response judgment by analyzing image data, location information, and environmental data collected from a plurality of mobile bodies (100).

[0042] For example, the main control tower (200) can comprehensively analyze video data, GPS-based location information, and environmental data such as ambient temperature, humidity, and hazardous gas concentration received in real time from multiple mobile bodies (100), and determine that there is a high risk of structural collapse at a specific location or that the spread of toxic substances is in progress in a specific area.

[0043] Such analysis can be performed by comparing changes in collected data along a time axis or against predefined thresholds; based on this, access restriction orders for the relevant area can be issued to mobile units, or internal tower warning systems can be activated to induce the evacuation of nearby personnel. Furthermore, the tower can classify disaster types and determine response priorities based on the analysis results, utilizing this data as foundational information for subsequent mission allocation or the generation of field reports.

[0044] Next, in one embodiment, the sub-control tower (300) is preferably a means of transportation manufactured in a special form that allows it to move to an area that is difficult for the main control tower (200) to access, but is not necessarily limited thereto. At least one such means of transportation may be provided, but the multiple meanings thereof are omitted for convenience.

[0045] For example, the sub-control tower (300) may be a specially designed vehicle capable of accessing a disaster site that is difficult for the main control tower (200) to access. In particular, the sub-control tower (300) may be composed of a small electric vehicle-based autonomous vehicle, thereby allowing it to approach the center of a disaster site where it is difficult for ordinary vehicles to enter due to narrow or rough terrain, and efficiently control multiple vehicles from the center of the site independently or in conjunction with the main control tower (200).

[0046] That is, the sub-control tower (300) may be configured in the form of a tractor-type multi-purpose vehicle or a special structure transport vehicle with enhanced explosion-proof and waterproof functions, and may be designed to operate even in harsh field conditions such as high temperature, high humidity, smoke, and water damage. To this end, it may be equipped with a durable body structure, wheels or a crawler system capable of omnidirectional movement, and various external sensors and communication equipment.

[0047] In addition, the sub-control tower (300) according to one embodiment may be configured to enable independent operation and mission execution at the site by equipping a data processing device, a high-performance communication module, a power supply device, and a cooling system inside the vehicle, and may be in a form capable of switching between manned and unmanned modes so that nearby personnel can board or operate remotely when necessary.

[0048] A sub-control tower (300) having such a form can, for example, receive control over a plurality of mobile bodies (100) from the main control tower (200) via a wireless network to respond to a disaster site that is difficult for the main control tower (200) to access, and can control the plurality of mobile bodies independently or in cooperation based on video data, location information, and environmental data collected from the plurality of mobile bodies (100).

[0049] More specifically, the sub-control tower (300) may include an internal control system and a real-time data processing device for directly controlling a plurality of mobile bodies (100) within the scope of authority after being granted temporary control authority over the disaster area through wireless communication with the main control tower (200).

[0050] For example, the sub-control tower (300) can analyze high-resolution real-time video data to predict signs of collapse of a specific structure or the diffusion path of harmful gases, and based on the results, individually transmit commands to adjacent mobile bodies to detour, stop, or relocate. In addition, cooperative cluster control can be performed by comparing and analyzing the visibility and communication status between multiple mobile bodies within the disaster site, efficiently dividing the inspection target area so that multiple mobile bodies do not repeatedly inspect the same point, and coordinating tasks between mobile bodies.

[0051] Furthermore, in areas where the communication infrastructure is unstable, the sub-control tower (300) can independently make decisions without being connected to the main control tower (200) and execute self-path resetting or emergency avoidance control by considering the current location of each mobile body, remaining battery, obstacle conditions, etc.

[0052] As another example, the sub-control tower (300) can identify a location where personnel requiring rescue are presumed to exist around a disaster site based on video data received in real time from multiple mobile bodies (100), select the mobile body closest to that location, and immediately transmit a command to verify the site.

[0053] In this process, the sub-control tower (300) analyzes the coordinates and direction information of each mobile body and selects only the mobile bodies that are easy to access, or assigns tasks preferentially to mobile bodies that have secured a geographically advantageous path, thereby enabling cooperative shared control without functional overlap between mobile bodies.

[0054] In addition, if it is determined that a single mobile unit is unable to respond immediately due to reasons such as communication failure or battery shortage, the sub-control tower (300) can automatically switch the mission to another nearby mobile unit to minimize the gap in mission execution of the entire system.

[0055] As another example, the sub-control tower (300) can integrate and analyze video data, location information, and environmental data collected in real time from multiple mobile bodies (100) to dynamically identify safe entry routes and high-risk areas at the disaster site.

[0056] For example, at a wildfire site, the sub-control tower (300) can identify the boundaries of smoke spread and flames through video data collected from moving objects, and can determine the coordinates and movement paths of the moving objects through location information. In addition, by considering multiple sensor data such as temperature, carbon monoxide concentration, oxygen concentration, wind direction and speed among environmental data, the current combustion spread speed and danger range can be predicted.

[0057] The sub-control tower (300), having comprehensively assessed this information, can individually transmit emergency evasion commands to multiple mobile bodies located in the combustion path as a priority, and can additionally assign auxiliary missions, such as searching for remaining human lives in the area or detecting rescue request signals, to mobile bodies located on the opposite side of the flame spread and in relatively safe positions. Additionally, in high-temperature areas, access can be restricted to specific mobile bodies equipped with heat-blocking equipment, and other mobile bodies can be controlled to collect video data while maintaining a certain distance.

[0058] In this way, the sub-control tower (300) can control multiple mobile bodies independently or cooperatively depending on the situation by integratively considering video, location, and environmental data, and can simultaneously ensure the safety and efficiency of each mobile body's mission execution.

[0059] FIG. 3 is a diagram illustrating the configuration of a main control tower for setting quantitative evaluation and disaster safety based on image data and environmental data according to an embodiment of the present invention.

[0060] At this time, the configuration of the main control tower shown in Fig. 3 can be applied in the same way to the configuration of the sub-control tower (300), but this will be omitted.

[0061] Referring to FIG. 3, a main control tower (200) according to one embodiment may include a communication module (210), an analysis module (220), and a control module (230).

[0062] In one embodiment, the communication module (210) is a component for performing bidirectional wireless communication between a plurality of mobile bodies (100) and a main control tower (200) or a sub-control tower (300), and can utilize various wireless communication technologies. For example, the communication module (210) can perform stable data transmission and reception with a plurality of mobile bodies (100) based on wireless communication protocols such as LTE (Long Term Evolution), 5G NR (New Radio), Wi-Fi (IEEE 802.11 series), Bluetooth, or satellite communication (SATCOM).

[0063] More specifically, the communication module (210) can utilize a low-latency, high-bandwidth communication channel based on 5G mobile communication to rapidly transmit large amounts of information such as video data and environmental data, and in disaster areas where network infrastructure is inadequate or in environments where the communication network is damaged, it can operate in parallel with ad-hoc communication technology based on a low-orbit satellite communication network such as Star-link or a Mesh network. This wireless communication method can be adaptively selected according to the location of the mobile body (100), and the communication module (210) can be configured to automatically switch to an appropriate communication path depending on the situation.

[0064] In addition, the communication module (210) is not limited to the function of simply receiving data, but can perform a bidirectional transmission relay role by receiving video data, location information, and environmental data collected from multiple mobile bodies in real time and transmitting them to the analysis module (220), while also transmitting control commands generated as a result of analysis and control back to the mobile bodies (100). In particular, when multiple mobile bodies are distributed and operated at different locations within a disaster site, the communication module (210) can evaluate the communication quality (RSSI, SNR, etc.) of each mobile body in real time and schedule data transmission and reception based on priority. For example, in an environment where communication blind spots frequently occur, such as in mountainous terrain, the communication module (210) can configure a temporary multi-hop communication path by setting a specific mobile body that has secured the highest signal strength as a relay node, thereby minimizing communication interruptions between the sub-control tower (300) and multiple mobile bodies.

[0065] In one embodiment, the analysis module (220) can quantify the surface of a structure through a Pix-based numerical algorithm based on image data and environment data stored in memory (not shown).

[0066] For example, the analysis module (220) can perform an algorithm to quantify the surface state by analyzing the texture, color change, reflectance, contour distortion, etc. of the surface of the structure from each pixel of the image data.

[0067] More specifically, the analysis module (220) can divide the stored image data into resolution units and extract surface characteristics of the structure based on the RGB or HSV values ​​of each pixel. For example, by comparing and analyzing pixel-unit color contrast, brightness difference, or degree of refraction of contours in repeated images of the same structure, physical changes such as surface roughness, attachment of foreign substances, and wear can be quantified. In particular, by utilizing computer vision-based techniques such as SIFT (Scale Invariant Feature Transform) and SURF (Speeded Up Robust Features) on local pixel groups extracted from the image, surface feature points of the structure and the rate of surface change over time can be numerically represented.

[0068] For example, in images of the concrete surface of a bridge pier, the roughness of the rough surface, repeated crack formation patterns, and brightness gradient can be analyzed using a Pix-based quantification algorithm, and based on this, a reliability index or damage index between 0.0 and 1.0 can be calculated. These results can be compared and analyzed with past inspection history and utilized to quantify the degree of deterioration or the rate of damage progression in the same area.

[0069] Additionally, Pix-based quantification is not limited to a single frame image but can be extended to a time-series analysis method utilizing image sequences collected from multiple time periods. In this case, the analysis module (220) can align continuous shooting data for the same location of the structure to infer the rate of change relative to a specific point in time (e.g., a decrease in reflectance of more than 20% compared to 3 months ago), and can transmit a warning signal to the control module (230) if an abnormal change is detected.

[0070] Here, the aforementioned Pix-based quantification algorithm can quantitatively analyze the width, depth, and area of ​​cracks on the surface of a structure based on image data, and can also evaluate the progression of structural defects by converting the quantitatively analyzed results into a 3D model and comparing it with previous inspection history.

[0071] For example, the analysis module (220) can perform a preprocessing process to quantitatively analyze defects such as cracks on the surface of a structure based on image data, first automatically segmenting the surface area of ​​the structure within the image and detecting crack candidate areas for each segmented area. At this time, candidate areas are selected based on pixel-level contrast, edge contours, and changes in curvature within the image, thereby minimizing false detection caused by external factors such as backgrounds or shadows.

[0072] Subsequently, the analysis module (220) utilizes depth information linked to the image to quantitatively calculate parameters such as depth, width, and area for the detected crack area. For example, based on distance data collected from a matching image obtained at multiple shooting points or from a LiDAR, stereo camera, etc., it calculates the mapping relationship between the defect location in the image and the actual distance. Through this, the size (in px units) on the simple 2D image is converted into an actual physical size (in mm or cm units), and the length, maximum and average width, and depth distribution of the defect can be calculated.

[0073] In addition, the analysis module (220) can evaluate the progression of a defect by comparing it with the quantitative analysis history stored for the same structure in the past. For example, if there is a history of damage with a length of 8.2 cm, an average width of 0.3 cm, and a depth of 1.5 mm for a crack in the bridge pier foundation photographed 3 months ago, damage with a length of 10.1 cm, an average width of 0.45 cm, and a depth of 2.1 mm can be detected at the same location in the currently photographed image. The analysis module (220) calculates the increments of these figures to quantify the speed and trend of the defect's progression. For example, it can generate indicators such as a 23% increase in length, a 50% increase in width, and a 40% increase in depth, and from this, the rate of damage progression can be organized into grades such as 'severe', 'warning', and 'stable', visualized, or transmitted to the control module (230).

[0074] In addition, in evaluating the progression of defects, the analysis module (220) can consider not only a single defect but also the spatial distribution and temporal change of multiple defects within the same structure. For example, the entire surface of the structure can be divided into a grid of 0.5m × 0.5m, and within each grid, crack density, average defect size, and recent occurrence frequency can be calculated to generate a space-based deterioration map. Through this, the deterioration state of the entire structure can be intuitively understood, and it can be estimated whether damage is progressing intensively in a specific area.

[0075] These evaluation results can be directly utilized for various decision-making processes, such as adjusting inspection cycles, setting maintenance priorities, and determining whether to respond urgently. If necessary, the analysis module (220) can automatically generate a re-inspection command for the corresponding part to the sub-control tower (300) or multiple moving bodies (100) when the damage progression rate exceeds a threshold.

[0076] For example, if cracks in the same area are detected five or more times in one year at a specific location on the lower part of the bridge deck and the damage index exceeds 0.75, the analysis module (220) can designate the area as an 'immediate inspection required' zone and assign it as a priority mission to multiple mobile bodies (100) through the control module (230).

[0077] Furthermore, the analysis module (220) according to one embodiment may further perform a quantitative evaluation by calculating values ​​such as the depth, width, and area of ​​cracks and deformation defects within the structure using deep learning-based object recognition.

[0078] For example, the analysis module (220) can automatically detect defects such as cracks, corrosion, deformation, and peeling in specific areas of a structure within an image by applying a deep learning-based object recognition algorithm, and can extract the location and size of each defect and represent them numerically.

[0079] More specifically, damaged areas of structures can be automatically identified by utilizing the latest object detection and segmentation models such as CNN (Convolutional Neural Network), YOLO (You Only Look Once), and Mask R-CNN. For example, areas such as walls, beams, and columns within an image can be classified according to predefined classes, pixel clusters where defects occur within each area can be extracted, and based on this, the location coordinates, area, length, width, and depth of the defects can be quantified.

[0080] In addition, to improve the accuracy of the analysis, the analysis module (220) can estimate the depth of the defect by merging distance information (depth map) obtained through LiDAR or a stereo camera in addition to 2D image information. This allows for the evaluation of not only simple surface cracks but also high-severity defects that have penetrated into the structure. For example, a vertical crack in a wall is detected in the image as 35px wide and 120px high, and based on the distance information of that area, it can be quantified as a crack with an actual length of 12.3cm, a maximum width of 1.2cm, and an estimated depth of 5mm at a distance of 1.8m.

[0081] In addition, the analysis module (220) can classify types of defects (e.g., surface microcracks, exposed rebar, corrosion penetration, etc.) by ensembling multiple deep learning models or utilizing multiple classification models, and can quantify the risk level according to each type. This quantitative evaluation can be used as an important criterion when generating control commands in the control module later, and can be effectively applied to setting inspection priorities for multiple moving bodies or automatic mission distribution.

[0082] Here, the aforementioned deep learning-based object recognition can automatically detect defects such as cracks, corrosion, deformation, and detachment in the corresponding image data, and can classify the risk level by type of defect and include it in the analysis results of the analysis module (220).

[0083] For example, the analysis module (220) can perform a deep learning-based evaluation process that automatically classifies various types of defects according to the use and environment of each structure and quantifies and grades the risk level for each type. For example, since the risk evaluation criteria differ for each structure, such as bridge decks, railway piers, and dam embankments, the analysis module (220) can utilize a pre-trained risk classification model to classify defects such as cracks, corrosion, rebar exposure, deformation, delamination, and displacement in detail.

[0084] Specifically, the analysis module (220) can calculate a risk score for an object (defect) detected in an image using a plurality of defect recognition models and display it in grades such as 'normal', 'minor', 'caution', and 'severe'. For example, if corrosion is detected, it can be graded as 'minor' if the area of ​​the corrosion site is less than 10㎠ and there are no signs of deformation or moisture in the surroundings, and conversely, if the same corrosion pattern exists in the structural connection and the reflectivity reduction or rust pattern is widely spread, it can be evaluated as 'severe'.

[0085] Such classification can be achieved by considering not only the size or numerical characteristics of the defect, but also the structural importance of the defect's location and surrounding environmental information. For example, even if a crack in a tunnel ceiling is less than 1 mm wide, it may be classified as having a risk level of 'caution' or higher if it is located at a high-load point or in an area where leakage is possible, whereas a crack of the same width may be evaluated as 'normal' or 'minor' if it is on a floor section or an auxiliary structure.

[0086] Additionally, the analysis module (220) according to one embodiment may express the risk level for each defect type as a numerical value between 0.0 and 1.0 based on deep learning-based classification results, and if it exceeds a certain threshold (e.g., risk level ≥ 0.75), it may automatically generate a notification signal or transmit a re-inspection request command to a sub-control tower (300) or multiple mobile bodies (100). For example, if a delamination defect classified as having a risk level of 0.83 occurs in a concrete column, it may order immediate re-inspection and additional detailed analysis of the relevant area, and the control module (230) may reset the mission priority based on this.

[0087] Furthermore, the analysis module (220) can visualize graded risk levels to provide the damage distribution of the entire structure in the form of a “risk map,” and can intuitively monitor the condition of large-scale infrastructure such as bridges, tunnels, and reservoirs through color-based visualization (e.g., green=normal, yellow=minor, orange=caution, red=severe). This function can be particularly useful in real-time inspection dashboards or remote monitoring systems.

[0088] For example, the analysis module (220) can generate risk statistics by integrating image data received from multiple moving bodies into structural units and subdividing defects within each structure. The figures, such as “A Bridge: 2 serious cases, 5 caution cases, 7 minor cases” and “B Tunnel: 0 serious cases, 1 caution case, 3 minor cases,” can serve as key grounds for control judgment and determining emergency response priorities.

[0089] Additionally, the analysis module (220) may track the trend of risk change of a specific structure or the same defect by linking the accumulated risk analysis results with time series data. For example, if the risk of a rebar exposure defect gradually increases from “March 2024: 0.45 → June 2024: 0.52 → September 2024: 0.61”, it is determined that the defect is progressing and the priority can be raised from ‘regular inspection’ to ‘priority inspection’.

[0090] In this way, the analysis module (220) can go beyond simply detecting and quantifying defects and automatically classify the risk level of each type of defect through deep learning-based object recognition, thereby providing key information for the judgment logic of the control module and the automatic assignment of missions for multiple moving bodies.

[0091] In one embodiment, the control module (230) can generate control commands according to the four stages of disaster safety (prevention, preparation, response, and recovery) based on the quantitative evaluation results performed by the analysis module (220).

[0092] For example, the control module (230) can make a judgment corresponding to one or more of the four disaster safety stages (prevention, preparedness, response, recovery) based on the type of defect and severity level of the structure calculated by the analysis module (220), and generate a control command corresponding to each judgment stage.

[0093] For example, if the analysis results show that a minor crack has newly occurred on the concrete surface of a specific structure and the width and depth of the crack are lower than a reference threshold, the control module (230) may determine that the defect corresponds to a "prevention stage." In this case, the control module (230) may generate a control command to additionally assign a precision shooting mission for the location to some of the multiple mobile bodies (100), or to automatically set a periodic monitoring schedule for the same location.

[0094] In addition, if the area of ​​the crack exceeds a certain level and the defect at the same location is clearly progressing compared to the past inspection history, the control module (230) may determine this as a "preparation stage" and automatically transmit a warning signal for the structure to a central control server or relevant agency, while also generating a command to relocate multiple movable bodies in the surrounding area to perform additional inspections on adjacent structures.

[0095] Meanwhile, if the crack depth or corrosion has expanded to the point where it threatens the safety of the structure, or if multiple types of defects are identified simultaneously in the same structure, the control module (230) may classify the situation as a "response phase." In this case, the control module (230) may control the immediate deployment of multiple mobile units to the site to perform real-time video streaming, multi-angle inspection, or search for personnel within the site, along with an emergency situation notification. At this time, if necessary, a command to delegate control of the area to the sub-control tower (300) may also be included.

[0096] Additionally, when the disaster situation has ended to a certain level and the location and characteristics of the remaining defects have been sufficiently identified, the control module (230) can automatically generate a control command to switch the area to a "recovery phase," create a list of structures requiring detailed inspection or additional monitoring, reduce the activity range of the mobile body, or reassign the mission to a new target area.

[0097] In this way, the control module (230) according to one embodiment classifies the situation based on the analyzed quantitative evaluation results and can transmit various control commands, such as mission resetting, warning notifications, transfer of control, concentrated deployment of mobiles, enhanced surveillance, or reduction of range, to multiple mobiles individually or in groups for each disaster safety stage. In addition, in complex sites where multiple situations overlap, a multi-layered control strategy may be established by simultaneously considering multiple disaster safety stages.

[0098] Accordingly, a control module (230) according to one embodiment can transmit control commands related to four stages of disaster safety (prevention, preparation, response, and recovery) generated based on quantitative evaluation results to a plurality of mobile bodies (100) through a communication module (210).

[0099] For example, in the case of a minor defect determined to be a “prevention stage,” the control module (230) may transmit a regular inspection route and shooting command to an easily accessible mobile body (100) among a plurality of mobile bodies (100) to continuously perform monitoring tasks for the location. At this time, the mobile body (100) takes detailed photos of micro-cracks or early corrosion areas using a high-resolution camera or an infrared thermal imaging camera, and the captured data is transmitted back to the main control tower (200) to be used for future analysis.

[0100] For example, if moderate damage or changes corresponding to the “preparation phase” are detected, the control module (230) may assign an emergency inspection mission to the structure and adjacent areas. To this end, an optimal combination of mobile units is selected by considering the remaining battery level, communication status, and movable range of the mobile units, and a control command is generated to distribute detailed missions to each mobile unit. For example, some mobile units may be commanded to focus on inspecting major damaged areas of the structure, while other mobile units may be commanded to monitor the surrounding environment or detect obstacles on the movement path.

[0101] For example, in the event of severe damage or a disaster judged to be a “response phase,” the control module (230) may command multiple mobile units to be rapidly assembled at the site and simultaneously broadcast real-time video of the site. In this process, control authority is transferred to a sub-control tower (300) to support independent and rapid response at the site, and the roles of the mobile units can be dynamically reassigned to mission execution, disaster response, rescue of personnel, fire suppression, etc. Additionally, commands to automatically send emergency alerts or request additional personnel support may be included in response to the emergency situation.

[0102] Finally, in the “recovery phase,” the control module (230) can generate control commands to lower the mission priority and reduce the inspection frequency for structures that have already been addressed, while releasing or switching resource-deficient mobiles to a standby state. At the same time, it is also possible to monitor the progress of the recovery operation and, if necessary, send redeployment commands to support the recovery.

[0103] In this way, the control module (230) according to one embodiment performs specific task assignment and role adjustment for each of the four disaster safety levels based on the quantitative evaluation results, thereby effectively managing multiple mobile bodies and enabling appropriate response to on-site conditions.

[0104] FIG. 4 is a diagram further illustrating a plurality of mobile body configurations for performing a collaboration-based inspection algorithm according to an embodiment of the present invention.

[0105] Referring to FIG. 4, a plurality of mobile bodies (100) according to one embodiment may further include a zone identification module (140), a zone division module (150), and a path reconstruction module (160) implemented with a collaboration-based inspection algorithm.

[0106] In one embodiment, the zone identification module (140) can perform a collaborative inspection algorithm that exchanges image data and location information acquired through the sensor module (110) and GPS among a plurality of moving bodies (100), analyzes the overlapping area of ​​the field of view between moving bodies based on the exchanged image data and location information, and identifies an area having an overlap density below a certain standard as a zone where inspection has been omitted.

[0107] For example, the area identification module (140) can exchange image data and location data acquired through the camera and GPS module mounted on each of the multiple moving bodies (100), and align this data spatiotemporally to analyze the overlapping field of view between the moving bodies. For example, if three drones A, B, and C fly around the same structure and each collect image data, the field of view for each drone at a specific time of shooting can be calculated by comparing metadata such as the time of shooting, GPS coordinates, altitude, and camera direction of each image.

[0108] Subsequently, the area identification module (140) according to one embodiment models the area where the field of view of each drone overlaps in a vector form and calculates the number of overlapping fields of view (coverage redundancy level) within a specific area. For example, if the same area is photographed by drones A and B but not by drone C, the overlap density of the area is calculated as 2, and if it is determined to be less than a standard value (e.g., 3) set by the system, it may be identified as an 'area where inspection omission is a concern'.

[0109] This overlap density can be corrected by taking into account not only simple field of view overlap but also the quality of the captured video (e.g., resolution, illumination, shaking, etc.). For example, even if two fields of view overlap, if one video is severely shaking, a lower weight is applied to that field of view, and the actual overlap density can be adjusted to, for instance, 1.3.

[0110] In addition, the zone identification module (140) calculates blind spots on the surface of the structure based on 3D terrain information, and can identify not only areas without overlapping views but also blind spots caused by the shape of the structure as areas where inspection is missed. In this way, the zone identification module (140) can perform complex analysis based on spatial analysis beyond simple comparison of views to precisely identify areas with a high probability of being missed in inspection.

[0111] In one embodiment, the zone division module (150) can perform a collaborative inspection algorithm that automatically divides the zone where an identified inspection omission occurred, identified by the zone identification module (140), among a plurality of mobile bodies to determine the inspection range of each mobile body.

[0112] For example, the zone division module (150) can automatically divide the zone identified as missing inspection by the zone identification module (140) among a plurality of mobile bodies (100) and allocate an optimal inspection range by considering the current location, flight path, remaining battery level, communication status, etc. of each mobile body. For example, if the identified missing inspection zone consists of an area of ​​300 m², and mobile body A is located within 20 m of the inspection site and has a remaining battery level of 80%, while mobile body B is located within 100 m of the inspection site and has a remaining battery level of 40%, the zone division module (150) can allocate more than 70% of the area to A and allocate the remaining 30% to B.

[0113] More specifically, the zone division module (150) subdivides the zone into grid units or polygon units and assigns areas considering the maneuverability or turning radius of each moving body. For example, for an underwater drone with a large turning radius, a wide and simple structured area may be assigned rather than a narrow and complex area, while conversely, a complex frame section or an obstacle proximity section may be assigned to a small drone capable of precise maneuvering.

[0114] In addition, the zone division module (150) according to one embodiment may also take into account differences in sensor performance of the mobile bodies. For example, inspection utilizing a wide aerial field of view is suitable for a mobile body equipped with a high-resolution camera, and a mobile body equipped with an infrared sensor that performs well in low-light environments may be preferentially placed in areas with weak lighting or night inspection areas. By integrating and analyzing these factors, the zone division module (150) can reallocate areas that are missed during inspection to each mobile body in the most efficient manner, thereby contributing to reducing inspection time and preventing duplicate inspections.

[0115] Furthermore, based on real-time feedback regarding the area assigned to each mobile unit (e.g., inspection progress, presence of unexpected obstacles, etc.), the area division module (150) can dynamically readjust the inspection range. For example, if a specific mobile unit is unable to perform an inspection due to a communication failure or mechanical defect, the area is reassigned to another adjacent mobile unit to minimize the missed section. This redistribution maximizes the inspection efficiency of the entire system by forming a real-time intelligent collaboration system.

[0116] In one embodiment, the path reconstruction module (160) can reconstruct the inspection path of each mobile body to maximize inspection efficiency by considering operational status information such as the remaining battery level, available distance, and communication status of each mobile body for the area where inspection omissions occurred, which was divided by the area division module (150).

[0117] For example, the route reconstruction module (160) can reconstruct the inspection route by collecting operational status information collected in real time from each of the multiple mobile bodies (100), calculating the inspection range that each mobile body can perform, and then prioritizing the area closest to the area where the inspection was missed and which each mobile body can safely reach and return to.

[0118] More specifically, the path reconstruction module (160) first determines the remaining battery capacity of each vehicle in percentage units and calculates the maximum available distance for each vehicle in real time (e.g., distance travelable based on remaining battery capacity). In addition, by comprehensively considering the distance from the current location to each zone where inspection has been missed, the expected flight or travel time within the zone, and regional communication stability, it can determine in advance whether inspection is possible.

[0119] For example, if the battery level of a specific drone-type vehicle is 30% and inspection is possible up to a distance of about 1.2 km based on the average consumption rate, the path reconstruction module (160) can first select one of the inspection omission zones located within a radius of 1.2 km from the vehicle, and if there are multiple zones satisfying the same conditions, the priority can be adjusted based on communication strength, obstacle density, or history of recent access attempts to assign an optimal zone.

[0120] In addition, the path reconstruction module (160) according to one embodiment can automatically optimize the path for each moving body by applying a minimum path search algorithm (e.g., Dijkstra algorithm, A* algorithm) to each divided inspection omission area to prevent duplicate inspections among multiple moving bodies and to minimize the total inspection time. At this time, constraints such as obstacles or no-fly zones can also be reflected to construct a practically safe and efficient path.

[0121] For example, if zones A, B, and C where inspection is omitted are adjacent, and mobile 1 has a battery of 60% and good communication, and mobile 2 has a battery of 35% and partially overlaps with a section with poor communication, the path reconstruction module (160) may assign a path to mobile 1 to continuously inspect A and B, and restrict the path to mobile 2 to inspect only the sub-area of ​​zone C where communication is good.

[0122] Additionally, if there is a possibility that the battery level of the vehicle may drop below a threshold during inspection, the path reconstruction module (160) can prevent deviation or data loss during the mission by terminating the inspection of the vehicle early based on a preset warning threshold and prioritizing the resetting of the path to return to the nearest landing point or charging point.

[0123] FIG. 5 is a diagram illustrating an additional configuration of a main control tower for automatically assigning or switching control authority with a sub-control tower according to an embodiment of the present invention.

[0124] Referring to FIG. 5, a main control tower (200) according to one embodiment may further include a control authority distribution module (240).

[0125] In one embodiment, the control authority distribution module (240) can automatically assign or switch control authority for each of the plurality of mobile bodies (100) according to the communication status with at least one sub-control tower (300), location precision, and control priority.

[0126] For example, the control authority distribution module (240) can dynamically manage control authority for multiple mobile bodies (100) by comprehensively analyzing status information received from each of the main control tower (200) and at least one sub-control tower (300). Specifically, the control authority distribution module (240) can determine the optimal control entity in real time by considering the wireless communication quality (RSSI, packet loss rate, etc.) between each control tower and multiple mobile bodies, GPS-based positional precision (Horizontal Dilution of Precision, HDOP), the relative distance between the mobile body and the control tower, and the control priority index assigned to each control tower.

[0127] For example, in a situation where a main control tower (200) is controlling multiple drone-type mobiles (100) near a specific disaster site, if a sub-control tower (300) capable of entering the area is additionally deployed, the control authority distribution module (240) can automatically transfer control authority for said mobiles from the main control tower (200) to the sub-control tower (300) if it is determined that the communication quality between the sub-control tower (300) and the mobiles is excellent and the positional accuracy is also improved.

[0128] Additionally, a control authority distribution module (240) according to one embodiment may be configured to assign control authority to the control tower with the higher priority when multiple sub-control towers can simultaneously communicate with a single mobile body, according to a predefined control priority assigned to each sub-control tower (300). For example, if there are sub-control towers A and B located in the same area, and A has a higher priority and a lower communication delay, control authority for the mobile body is preferentially assigned to sub-control tower A.

[0129] In addition, the control authority distribution module (240) can enable immediate response by forcibly switching the control authority to the most appropriate alternative control tower nearby when it is determined that the response of the currently controlling control tower is delayed or unstable, in cases where emergency measures are required due to changes in the state of the mobile body, such as a decrease in battery level or communication instability. This process can be automatically performed based on predefined threshold conditions (e.g., packet loss rate of 30% or more, response time exceeding 500ms, etc.).

[0130] Additionally, the control authority distribution module (240) can perform a smooth handover of authority so that the nearest control tower becomes the controlling entity when a change in location within the mission section of the mobile body occurs above a certain level, and in this case, a temporary co-control mode may be utilized to prevent duplicate transmission or conflict of control commands.

[0131] In this way, the control authority distribution module (240) according to one embodiment performs the function of efficiently distributing or concentrating control over a plurality of mobile bodies in various situations, thereby improving the reliability and response speed of the entire system and strengthening real-time situational response capabilities at disaster sites.

[0132] FIG. 6 is a diagram illustrating an additional configuration of a main control tower for generating a re-shooting command by comparing the measured values ​​of the camera's shooting angle, illumination level, and focal length at the time of shooting with reference values, respectively, according to an embodiment of the present invention, and dynamically reassigning the role of each moving body according to priority.

[0133] Referring to FIG. 6, the analysis module (220) of the main control tower (200) according to one embodiment may include a re-shooting command generation module (221), and the control module (230) of the main control tower (200) may include a moving object reassignment module (231).

[0134] In one embodiment, the reshoot command generation module (221) compares the measured values ​​of the camera's shooting angle, illumination level, and focal length at the time of shooting an image using a camera (not shown) with reference values, and if any of them do not meet the reference values, it can automatically generate a reshoot command for the corresponding image among the captured image data.

[0135] For example, the reshoot command generation module (221) can automatically analyze the quality of each frame based on sensor-based measurements collected at the time of shooting along with image data received from a plurality of moving bodies (100). The main quality judgment factors used at this time are the shooting angle, illumination level, and focal length, each of which is evaluated by comparison with a preset reference value.

[0136] First, the standard for the shooting angle may be set to a range of ±10 degrees relative to the front of the subject. The reshoot command generation module (221) determines the camera's orientation through gyroscope sensor or IMU sensor data at the time of video shooting and calculates the relative angle with respect to the subject's center axis. For example, in a video for inspecting cracks in a specific transmission tower, if the camera is tilted more than 25 degrees relative to the horizontal, distortion may occur in determining the depth or direction of the cracks, so the frame is determined to have exceeded the standard and becomes subject to a reshoot command.

[0137] In the case of illuminance levels, the average luminance value or histogram distribution of images collected from the camera sensor is analyzed to determine whether a preset minimum illuminance threshold (e.g., 120 Lux) is met.

[0138] For example, if the average illuminance value of a video captured in an environment with insufficient natural light, such as at sunset or inside a tunnel, is below a standard, the outline of the subject or the defective area is not clearly identified, so the video is considered to be of insufficient quality and becomes subject to re-capture. At this time, the re-capture command generation module (221) can generate a re-capture instruction packet including the coordinates, time, and moving body ID of the video frame and transmit it to the moving body or an adjacent moving body.

[0139] In addition, the focal length can be determined based on the subject's boundary sharpness or the lens position value at the point of autofocus (AF) operation.

[0140] For example, if the autofocus does not function properly in a video captured while a moving body is performing vibration or a sharp turn, and the edges of the subject are blurry, and the result of the resolution analysis within the video does not meet the standard resolution (e.g., edge clarity of 80% or more at 1080p), it is determined that reshooting is necessary. In this case, the reshooting command generation module (221) may include a command to approach the same location again or to reshoot under stable shooting conditions.

[0141] Additionally, the reshoot command generation module (221) can set a reshoot flag for the corresponding video frame if it fails to meet any of the three criteria above, and determine the reshoot priority based on this.

[0142] For example, in the case where a specific image satisfies the shooting angle and illumination standards but is only out of focus, or in the case where both the shooting angle and illumination are poor, the latter frame can be designed to be processed first by setting the re-shooting priority to 1.

[0143] In addition, if a video that fails to meet the standard is detected consecutively among multiple frames, the reshoot command generation module (221) can automatically determine to expand the reshoot area to a certain range rather than a single frame, and generate a reshoot command for the entire section, and can even include a pre-roll or post-roll shooting section to prevent video omission.

[0144] In this way, the reshoot command generation module (221) can comprehensively determine multiple quality standards and automatically generate a reshoot instruction based on whether the standards are not met, thereby maintaining consistent image collection quality of the moving object and improving the reliability of the final data.

[0145] Accordingly, the re-shooting command generation module (221) according to one embodiment may transmit re-shooting commands generated in various forms as described above to a plurality of corresponding mobile bodies (100) via wireless communication.

[0146] In one embodiment, the mobile reassignment module (231) can automatically calculate the priority of the mission to be assigned to each mobile based on the result data of the quantitative evaluation performed in the analysis module (220) of FIG. 3, and can dynamically reassign the role of each mobile according to the automatically calculated priority.

[0147] For example, the mobile reassignment module (231) can calculate the priority for mission redistribution by integrating multiple indicators such as inspection accuracy, response speed, battery level, communication stability, and mission performance rate based on the quantitative evaluation score for each mobile generated by the analysis module (220).

[0148] More specifically, for example, the vehicle reassignment module (231) can calculate a comprehensive score by applying indicator values ​​calculated as a result of quantitative evaluation for each vehicle (100) to a pre-set weight-based evaluation formula, and determine whether to assign a priority mission or a secondary mission to the vehicle based on this score.

[0149] For example, among the evaluation indicators, weights are assigned to inspection accuracy (30%), communication stability (25%), remaining battery (20%), mission completion rate (15%), response delay time (10%), etc., and based on this, mobile units with a final score of 85 points or higher are classified as priority mission targets, while those with a score of less than 60 points may be switched to standby or support roles.

[0150] Additionally, for example, the mobile object reassignment module (231) may prioritize assigning a reshooting mission to a mobile object that has the highest quantitative evaluation score among multiple mobile objects located nearby in order to resolve image quality issues in the area when a specific reshooting command occurs. For example, if the image illumination value of a mobile object that previously performed the shooting falls below a standard value, the mission may be reassigned to a mobile object that has high communication stability and sufficient battery capacity among other mobile objects located within 100m of the vicinity.

[0151] In addition, for example, the mobile object reassignment module (231) can be automatically configured to be excluded from the same type of inspection mission if a specific mobile object has a high error rate recorded during consecutive mission execution. For example, if a specific mobile object has a history of re-shooting due to focal length errors or shooting angle errors in 2 out of the last 3 inspection missions, the mobile object may be excluded from future precision shooting missions and instead redistributed to wide area search missions or data relay missions.

[0152] Furthermore, for example, the vehicle reassignment module (231) may also consider the topographical conditions where each vehicle is located and the availability of a communication network. For example, in mountainous terrain, a specific vehicle equipped with a high-power communication module is advantageous for real-time video transmission, so that vehicle is prioritized for high-resolution real-time monitoring missions, and low-spec vehicles may be reassigned to roles such as search assistance or rear standby.

[0153] Additionally, for example, in a disaster response scenario where priority search for human rescue is required, the mobile reassignment module (231) can prioritize assigning search missions to mobiles that can quickly access areas with a high probability of human survival by considering the thermal detection sensor sensitivity, search speed, and nearby location of each mobile.

[0154] In this way, the mobile reassignment module (231) according to one embodiment can maximize the utilization efficiency of resources within the entire system by dynamically making a decision based on quantitative evaluation data, and can adjust each mobile to perform an optimal role according to its performance characteristics. This automatic reassignment function can reduce the mission failure rate and serve as a key function for rapid disaster response and recovery support.

[0155] As described above, the functional operations of each configuration described according to various embodiments can be implemented in the form of program instructions and recorded on a computer-readable recording medium and / or memory, etc.

[0156] The aforementioned computer-readable recording medium may include program instructions, data files, data structures, etc., either individually or in combination. The program instructions recorded on the computer-readable recording medium may be those specifically designed and configured for the present invention, or they may be those known and available to those skilled in the art of computer software.

[0157] Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes; optical recording media such as CD-ROMs and DVDs; magneto-optical media such as floptical disks; and hardware devices specifically configured to store and execute program instructions such as ROM, RAM, and flash memory.

[0158] Examples of program instructions include machine code, such as that generated by a compiler, as well as high-level language code that can be executed by a computer using an interpreter, etc. The hardware device may be configured to operate as one or more software modules to perform processing according to the present invention, and vice versa.

[0159] Although the present invention has been described above with reference to preferred embodiments, those skilled in the art will understand that various modifications and changes can be made to the invention without departing from the spirit and scope of the invention as described in the following claims. Explanation of the symbols

[0160] 100: Multiple mobile bodies 110: Sensor module 120: Drive Module 200: Main Control Tower 210: Communication Module 220: Analysis Module 230: Control Module 240: Control Authority Distribution Module 300: Sub Control Tower

Claims

Claim 1 A smart control tower system for operating mobiles capable of wireless control in aerial, land, and underwater environments, comprising: a plurality of mobiles including a sensor module for collecting image data, location information, and environmental data, and a drive module for autonomous movement by receiving control commands from a main control tower or a sub-control tower below; a main control tower that generates the control commands for controlling the plurality of mobiles and performs disaster response judgments by analyzing the collected image data, location information, and environmental data; and a sub-control tower that receives control over the plurality of mobiles from the main control tower via a wireless network to respond to disaster sites that are difficult for the main control tower to access, and controls the plurality of mobiles independently or in cooperation based on the collected image data, location information, and environmental data. Claim 2 A smart control tower system according to claim 1, wherein each of the main and sub control towers comprises: an analysis module that performs a quantitative evaluation by quantifying the surface of a structure through a Pix-based quantification algorithm based on image data and environmental data, and calculating numerical values ​​such as the depth, width, and area of ​​cracks and deformation defects within the structure using deep learning-based object recognition; and a control module that generates a control command according to the four stages of disaster safety (prevention, preparation, response, and recovery) based on the results of the quantitative evaluation performed and transmits it to the plurality of mobile bodies. Claim 3 A smart control tower system according to claim 1, wherein the plurality of mobile bodies exchange image data and location information with each other, analyze the field of view overlap area between the mobile bodies based on the exchanged image data and location information to identify an area having an overlap density below a certain standard as an area where inspection omissions have occurred, and perform a collaborative inspection algorithm that automatically divides the identified area where inspection omissions have occurred among the plurality of mobile bodies to determine the inspection range of each mobile body. Claim 4 A smart control tower system according to claim 1, wherein the main control tower includes a control authority distribution module that automatically assigns or switches control authority to each of the plurality of mobile bodies according to the communication status with the sub-control tower, positional accuracy, and control priority. Claim 5 A smart control tower system according to paragraph 2, wherein the control module automatically calculates the priority of the mission to be assigned to each mobile body based on the result data of the quantitative evaluation performed, and dynamically reassigns the role of each mobile body according to the calculated priority.

Images