Municipal road surface intelligent inspection method and system based on unmanned aerial vehicle array

Abstract

The invention discloses a municipal road surface intelligent inspection method and system based on an unmanned aerial vehicle array, and the method comprises the steps: constructing a municipal digital twin base, fusing multi-source data, and carrying out the preliminary screening of diseases through a convolutional neural network; a quantum key and elliptic curve encryption mixed mode is adopted to perform data security enhanced transmission; obtaining the position of a relay node of the unmanned aerial vehicle based on an edge computing node, and constructing a microwave link through multi-module cooperation; measuring link quality, and dynamically planning an unmanned aerial vehicle inspection array; a three-level intelligent diagnosis system is constructed based on multi-algorithm coupling, and disease diagnosis is carried out; evaluating the inspection result, and updating the inspection scheme. According to the invention, rapid routing inspection is carried out through the unmanned aerial vehicle array, data transmission safety is guaranteed by adopting an encryption technology, transmission reliability is improved by dynamic link aggregation, a routing inspection scheme is dynamically adjusted, a routing inspection path and the unmanned aerial vehicle array are automatically optimized, a virtuous circle is formed, and intelligent development of municipal road routing inspection is promoted.

Description

Technical Field

[0001] The present invention relates to the technical field of drone intelligent inspection technology, and in particular to a municipal road intelligent inspection method and system based on a drone array. Background Art

[0002] Traditional drone inspection technology has numerous limitations. First, manual inspection, currently the primary method, is extremely inefficient, making it difficult to cover large areas quickly and virtually impossible to achieve comprehensive inspections. Manual inspections rely heavily on the inspector's experience and subjective judgment, which can easily lead to missed inspections, false detections, and other issues, delaying maintenance.

[0003] Traditional data collection relied primarily on simple measurement tools and photographic documentation. The data captured was limited and lacked in-depth information, making it difficult to fully reflect the true extent of pavement damage. Traditional technologies also lacked effective data analysis tools and algorithms, making it impossible to accurately classify damage, quantify its assessment, and predict its trends. Actions were only taken when damage became more severe. Furthermore, traditional inspection plans were typically fixed, conducted according to predetermined times and routes, and unable to dynamically adjust to real-time pavement conditions, damage levels, and environmental changes. Traditional inspection plans also lacked consideration for task priorities, making it difficult to rationally allocate inspection resources based on the severity and urgency of the damage.

[0004] Traditional technologies also present numerous problems in data transmission. The data transmission process is easily restricted by the network environment, resulting in slow transmission speeds and high data loss rates. In remote areas or those with poor network signals, data transmission stability is difficult to guarantee. The lack of effective data encryption means that initial disease screening data is at risk of being stolen or tampered with.

[0005] This solution addresses the many shortcomings of the above-mentioned traditional technologies. By building a municipal digital twin base, it collects multi-source data and integrates various information, solving the problem of insufficient data collection in traditional methods; using edge computing and encryption technology, it optimizes data preprocessing and transmission security; through dynamic planning and optimization of inspection plans, it improves the flexibility and pertinence of inspections; with the help of an AI diagnostic system, it improves the accuracy and depth of disease diagnosis, overcoming the many limitations of traditional inspections. Summary of the Invention

[0006] The present invention provides a method for intelligent municipal road inspection based on an array of drones, which includes: Integrate geographic information system map data with building information models, collect and correlate multi-source data, build a municipal digital twin foundation, perform point cloud downsampling at the edge, extract regions of interest from image data, and use lightweight convolutional neural network algorithms to perform initial disease screening. A quantum channel is established through the quantum key distribution module, quantum bits are used to generate and distribute quantum keys, and the elliptic curve encryption algorithm is called to combine the quantum key with the elliptic curve public key to generate a hybrid encryption key, thereby enhancing the security of the initial disease screening data transmission. The optimal microwave link parameters are calculated based on the mission load and network topology information, and the position of the drone relay node is calculated. Through the coordination of the flight control system, communication module and onboard sensors, the position is adjusted in combination with the path planning algorithm to build a microwave link. The encrypted data is sent via microwave. The drone relay node completes demodulation and error correction, re-encoding and modulation, and forwards it to the target node, forming a complete transmission link. Based on graph theory, a microwave link topology is constructed. Link quality is quantified using an edge weight calculation formula. A genetic algorithm and Dijkstra algorithm are integrated to combine link quality and task priority to construct a weighted graph and generate an initial inspection path set. The initial population is generated through genetic algorithm encoding. The global path is optimized through selection, crossover, and mutation iterations. The Dijkstra algorithm is used to calculate the local shortest path between nodes. The global optimization results are combined with the local optimal path to determine the globally optimal inspection route and array. Acquire array point cloud data of the target area, compare and match abnormal areas and historical features with the historical damage database, implement target detection algorithms for rapid crack detection, use point cloud neural network algorithms to assess structural safety, and combine long-short-term memory networks with time series analysis to predict damage trends; Based on the disease diagnosis, evaluate the inspection results, update the inspection plan, and repeat the first step.

[0007] The above-mentioned intelligent municipal pavement inspection method based on drone arrays integrates geographic information system map data and building information models, collects and correlates multi-source data, builds a municipal digital twin foundation, performs point cloud downsampling at the edge, extracts regions of interest from image data, and uses a lightweight convolutional neural network algorithm to perform initial disease screening, including: Build a municipal digital twin foundation, integrating geographic information system maps, building information models, and inspection data, and annotating key infrastructure information; Point cloud downsampling and image region of interest extraction are performed at the edge to retain key feature information, identify diseased areas, and generate preliminary disease screening results.

[0008] The above-mentioned method for intelligent municipal pavement inspection based on a drone array comprises: constructing a microwave link topology map based on graph theory; quantifying link quality using an edge weight calculation formula; integrating a genetic algorithm with a Dijkstra algorithm; constructing a weighted graph based on link quality and task priority to generate an initial inspection path set; generating an initial population using genetic algorithm encoding; iteratively optimizing the global path through selection, crossover, and mutation; calculating the local shortest path between nodes using the Dijkstra algorithm; and combining the global optimization result with the local optimal path to determine the globally optimal inspection route and array. The method comprises: Build microwave link topology based on graph theory, measure link quality, and adjust microwave link transmission; By integrating the genetic algorithm and the Dijkstra algorithm, the initial inspection path combination is generated by encoding and fitness evaluation is performed to generate the optimal inspection route and array.

[0009] The above-mentioned intelligent inspection method for municipal pavement based on drone arrays acquires array point cloud data of the target area, compares abnormal areas with historical characteristics with a historical damage database, implements a target detection algorithm for rapid crack detection, uses a point cloud neural network algorithm to assess structural safety, and combines long-short-term memory networks with time series analysis to predict damage trends, including: Acquire and compare array point cloud data with historical damage database, and synchronize multi-sensor data timestamps; Based on the comparison results, disease diagnosis is carried out with the help of a three-level intelligent diagnosis system.

[0010] In the above-mentioned intelligent inspection method for municipal pavement based on a drone array, a three-level intelligent diagnosis system is used to diagnose defects based on the comparison results, including: Identify crack features through target detection algorithms, calculate coordinates and boundary ranges, and screen potential damage areas; At the regional center, multi-scale feature extraction is performed using point cloud neural network algorithms, combined with finite element analysis and structural mechanics modeling to assess the overall safety status of the structure; Long-short-term memory networks combined with time series analysis are used to capture the dynamic patterns of disease development and predict future spread speed and severity.

[0011] An intelligent municipal road inspection system based on a drone array, comprising: Data acquisition module: used to integrate geographic information system map data and building information models, collect and correlate multi-source data, build the municipal digital twin foundation, complete point cloud downsampling at the edge, extract regions of interest from image data, and complete initial disease screening using a lightweight convolutional neural network algorithm; Key transmission module: used to establish a quantum channel through the quantum key distribution module, generate and distribute quantum keys using quantum bits, call the elliptic curve encryption algorithm, combine the quantum key with the elliptic curve public key to generate a hybrid encryption key, and perform security-enhanced transmission of disease initial screening data; Link construction module: This module is used to calculate the optimal microwave link parameters based on the mission load and network topology information, calculate the location of the UAV relay node, coordinate the flight control system, communication module and onboard sensors, and use the path planning algorithm to adjust the location to build the microwave link. The encrypted data is sent via microwave. The UAV relay node completes demodulation and error correction, re-encoding and modulation, and forwards it to the target node, forming a complete transmission link. Inspection planning module: This module is used to construct microwave link topology based on graph theory, quantify link quality through edge weight calculation formulas, integrate genetic algorithms with Dijkstra's algorithm, combine link quality and task priority to construct a weighted graph and generate an initial inspection path set. Genetic algorithm encoding generates an initial population, iteratively optimizes the global path through selection, crossover, and mutation, and uses Dijkstra's algorithm to calculate the local shortest path between nodes. The global optimization results are combined with the local optimal path to determine the globally optimal inspection route and array. Intelligent diagnosis module: used to obtain array point cloud data of the target area, compare and match abnormal areas and historical features with the historical disease database, implement target detection algorithms for rapid crack detection, use point cloud neural network algorithms to assess structural safety, combine long-term and short-term memory networks with time series analysis to predict disease trends, evaluate inspection results, and update inspection plans.

[0012] In the above-mentioned municipal road intelligent inspection system based on a drone array, the data acquisition module specifically includes: Twin model construction submodule: used to build the municipal digital twin base, integrate geographic information system maps, building information models and inspection data, and annotate key infrastructure information; Disease initial screening submodule: used to perform point cloud downsampling and image region of interest extraction at the edge, retain key feature information, identify diseased areas, and generate initial disease screening results.

[0013] In the above-mentioned intelligent municipal road inspection system based on a drone array, the inspection planning module specifically includes: Link quality calculation submodule: used to construct microwave link topology based on graph theory, measure link quality, and adjust microwave link transmission; Array planning submodule: used to integrate genetic algorithm and Dijkstra algorithm, encode and generate initial inspection path combination and perform fitness evaluation to generate the optimal inspection route and array.

[0014] In the above-mentioned intelligent road inspection system based on a drone array, the intelligent diagnosis module specifically includes: Array point cloud data synchronization submodule: used to obtain and compare array point cloud data with the historical disease database, and synchronize multi-sensor data timestamps; Diagnosis and prediction submodule: It is used to conduct rapid crack detection, structural safety assessment, and disease trend prediction based on the comparison results with the help of a three-level intelligent diagnosis system.

[0015] The beneficial effects achieved by the present invention are as follows: This solution uses drone arrays to achieve rapid, large-scale inspections, and combines multi-source data collection with AI diagnostic technology to significantly improve the speed and accuracy of disease detection. A hybrid mode of quantum key and elliptic curve encryption is used to enhance the security of data transmission for initial disease screening. At the same time, a dynamic link aggregation strategy is used to intelligently select transmission links to improve data transmission reliability and reduce the risk of data loss. The inspection plan is dynamically adjusted, and the robot learning algorithm and multi-objective optimization algorithm are used to determine the optimal inspection route and array. The three-level AI diagnostic system combines a variety of advanced algorithms to quickly detect diseases, assess structural safety, and predict disease trends. This solution automatically optimizes inspection routes and drone arrays based on inspection results and disease diagnosis, forming a virtuous cycle, promoting the intelligent development of municipal road inspection systems, and providing more accurate data support for urban planning, construction, and management. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments described in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0017] Figure 1 This is a flow chart of a method for intelligent municipal road inspection based on a drone array provided in Example 1 of the present application.

[0018] Figure 2 This is a schematic diagram of a municipal road intelligent inspection system based on a drone array provided in Example 2 of the present application. DETAILED DESCRIPTION

[0019] The following is a clear and complete description of the technical solutions in the embodiments of the present invention, in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0020] Example 1

[0021] like Figure 1 As shown, the first embodiment of the present application provides a municipal road intelligent inspection method based on a drone array, comprising: S110: Integrates geographic information system map data with building information models, collects and correlates multi-source data, builds a municipal digital twin foundation, performs point cloud downsampling at the edge, extracts regions of interest from image data, and uses lightweight convolutional neural network algorithms to perform initial disease screening. The multi-source inspection data includes at least drone inspection data and urban planning and deployment data.

[0022] The system builds a municipal digital twin foundation, integrates GIS maps and BIM models, and imports urban planning data. It controls drones to collect inspection data, performs preprocessing such as point cloud downsampling and image ROI extraction, and uses a lightweight CNN algorithm to perform initial disease screening. It then combines urban planning data to eliminate misjudgments, stores the results, and feeds them back to the user interface.

[0023] Building a municipal digital twin foundation, collecting multi-source data, and conducting initial disease screening includes the following sub-steps: S111: Build a municipal digital twin foundation, integrating geographic information system maps, building information models, and inspection data, and annotating key infrastructure information; High-precision Geographic Information System (GIS) map data was loaded as the spatial framework for the base. Building Information Modeling (BIM) was imported to spatially align and integrate the 3D structural information of the urban infrastructure with the GIS map, seamlessly integrating them under a unified coordinate system.

[0024] The system then imports the inspection data into the base and associates it with GIS and BIM data. Key infrastructure information, such as bridges, tunnels, and important intersections, is annotated, highlighting their locations and attributes through a visual interface.

[0025] S112: Perform point cloud downsampling and image region of interest extraction at the edge, retain key feature information, identify diseased areas, and generate preliminary disease screening results; After receiving the raw data collected by the drone at the edge, the point cloud downsampling module is started to remove redundant array point cloud data through an algorithm, reducing the data volume and retaining key feature information.

[0026] Extract regions of interest (ROI) from image data to identify and crop areas that may contain diseases, reducing the size of data for subsequent processing.

[0027] Calling the lightweight convolutional neural network (CNN) algorithm and utilizing the pre-trained model deployed on the edge, the processed data is used to identify the characteristics of defects, quickly determine whether there are defects such as cracks and potholes, and obtain preliminary screening results for the defects.

[0028] The initial disease screening results and related data are compressed and packaged, and transmitted to the central server via the network.

[0029] S120: A quantum channel is established through the quantum key distribution module. Quantum keys are generated and distributed using quantum bits. The elliptic curve encryption algorithm is called to combine the quantum key with the elliptic curve public key to generate a hybrid encryption key for enhanced security transmission of disease screening data. The system establishes a quantum channel through a quantum key distribution (QKD) module. Leveraging the unclonability and untestability of quantum bits, it generates and distributes a set of quantum keys in real time. Once QKD is complete, the system splits the quantum key into multiple sub-key fragments and stores them in a secure key pool.

[0030] The system invokes the Elliptic Curve Cryptography (ECC) module to generate a public-private key pair based on pre-negotiated elliptic curve parameters. The quantum key fragment is combined with the elliptic curve public key to generate a hybrid encryption key. During data encryption, the system first uses the hybrid encryption key to symmetrically encrypt the initial disease screening data, generating an encrypted data block.

[0031] At the same time, the system encrypts the elliptic curve private key in segments, using quantum key fragments to encrypt the private key segments. The encrypted data and private key segments are transmitted to the receiving end via a secure channel. The receiving system first decrypts the elliptic curve private key using the quantum key fragments to recover the complete private key information.

[0032] The hybrid encryption key is regenerated using the elliptic curve private key and the quantum key fragment to decrypt the encrypted data block. After verifying the data integrity and security, the data is transmitted to the link and decrypted to obtain the initial disease screening.

[0033] S130: Calculate the optimal microwave link parameters based on the mission load and network topology information, calculate the position of the UAV relay node, coordinate the flight control system, communication module and onboard sensors, and use the path planning algorithm to adjust the position and build a microwave link. Send the encrypted data via microwave. The UAV relay node completes demodulation and error correction, re-encoding and modulation, and forwards it to the target node, forming a complete transmission link. The system first calculates the optimal microwave link parameter configuration based on the current task load and network topology information through the resource management module of the edge computing node, and sends these parameter instructions to the drone relay node through control signals.

[0034] The flight control system of the drone relay node works in coordination with the communication module, uses onboard sensors to perceive the surrounding environment, and combines with the preset flight path planning algorithm to precisely adjust the position of the drone relay node and the antenna pointing angle, forming an unobstructed, high-gain microwave transmission path between the drone and the edge computing node.

[0035] At the same time, the system needs to obtain node data according to the transmission requirements of the link and environmental factors, perform normalization and other data preprocessing, and dynamically calculate the position of the drone relay node in each node. Among them, the relay node position objective function is in, represents the balance coefficient, which is the influence of the sum of squares of distances between dominant nodes; m represents the total number of existing nodes; The dynamic weight matrix element represents the inspection priority association weight between the i-th and j-th existing nodes; The asymmetric attenuation coefficient matrix element represents the anisotropic attenuation of the environment on the signal transmission from node i to j; Represents the distance between the i-th and j-th existing nodes; The impact of dominant energy consumption items; Represents the energy consumption index, coupling the UAV flight energy consumption and signal transmission power; Represents the distance from the i-th existing node to the relay node, i represents the existing node, r represents the relay node, ; represents the terrain complexity factor, which is used as a parameter to adjust the power of distance; k is the penalty coefficient, which controls the degree of delay constraint and forces relay nodes to be close to meet the delay; Represents the truncation function, when the total transmission distance Exceed When , the penalty term is activated to constrain the total delay of signal transmission and avoid communication timeout caused by relay nodes being deployed too far away; is the time delay tolerance threshold; represents the signal propagation constant.

[0036] Derivative the relay node coordinates, set the partial derivative to 0, and solve the coordinate expression. The relay node position coordinates can be obtained: At the same time, the edge computing node's communication module is activated, encoding and modulating the data to be transmitted according to the preset microwave link parameters, and sending it to the drone relay node via microwave signals. Its communication processing module demodulates and corrects the signal, then re-encodes the modulated signal based on the target node's address information and forwards the data to the target node via the microwave link.

[0037] Throughout the link's operation, the system uses a monitoring module to monitor the link status in real time, including key indicators such as signal strength and bit error rate. If an anomaly is detected, such as signal attenuation or an increase in bit error rate, the system triggers a link optimization mechanism, with the edge computing node and the drone relay node working together to adjust the link parameters.

[0038] S140: Builds a microwave link topology based on graph theory, quantifies link quality using an edge weight calculation formula, integrates a genetic algorithm with the Dijkstra algorithm, combines link quality with task priority, constructs a weighted graph, and generates an initial inspection path set. Genetic algorithm encoding generates an initial population, iteratively optimizes the global path through selection, crossover, and mutation. Dijkstra's algorithm is used to calculate the local shortest path between nodes. The global optimization results are combined with the local optimal path to determine the globally optimal inspection route and array. The system constructs a microwave link topology map in real time and measures link quality using an edge weight calculation formula. It also uses a robotic learning algorithm to analyze and predict link status, dynamically planning drone inspection plans and arrays.

[0039] Based on graph theory, a microwave link topology is constructed. Link quality is quantified using an edge weight calculation formula. A genetic algorithm and a Dijkstra algorithm are integrated to combine link quality and task priority to construct a weighted graph and generate an initial inspection path set. An initial population is generated through genetic algorithm encoding. The global path is optimized through selection, crossover, and mutation iterations. The Dijkstra algorithm is used to calculate the local shortest path between nodes. The global optimization results are combined with the local optimal path to determine the globally optimal inspection route and array. The specific steps include the following: S141: Build microwave link topology maps based on graph theory, measure link quality, and adjust microwave link transmission; Based on graph theory, a microwave link topology graph is constructed, with nodes in the link (such as edge computing nodes and drone relay nodes) as vertices and microwave connections between nodes as edges. Link quality is measured using the edge weight calculation formula, which is as follows: Where n represents the number of link quality evaluation indicators; represents the value of the i-th link quality evaluation index; Indicates the The fuzzy membership function value of each indicator; represents the Riemann curvature tensor, which reflects the curvature of the link propagation space; represents the determinant of the Riemann curvature tensor, which is used to quantify the comprehensive degree of impact of space curvature on link quality; m represents the number of bit error rate sampling points; represents the weight of the jth bit error rate sampling point; represents the bit error rate value of the jth bit error rate sampling point; The position error penalty coefficient is used to adjust the impact of the UAV relay node position error on the link quality; K represents the number of UAV relay nodes participating in signal transmission in the microwave link; represents the actual position vector of the kth UAV relay node; represents the ideal position vector of the kth UAV relay node; , where S represents the ideal position vector of the k-th UAV relay node; Represents a function that describes the distribution of signal intensity in the two-dimensional projection area.

[0040] Edges with higher weights indicate good link quality and are suitable for data transmission; edges with lower weights indicate poor link quality and require optimization or adjustment. The system dynamically adjusts the microwave link topology based on edge weights, prioritizing high-weight links for data transmission, thereby optimizing the performance and reliability of the entire microwave link.

[0041] S142: Integrate the genetic algorithm and the Dijkstra algorithm to encode and generate the initial inspection path combination and perform fitness evaluation to generate the optimal inspection route and array; The system constructs a weighted graph based on link quality indicators, using task priorities as weight adjustment coefficients to generate an initial set of inspection paths. A genetic algorithm is introduced to generate the initial population through encoding, with each chromosome representing a possible inspection path combination.

[0042] The fitness of the population is evaluated. The fitness function comprehensively considers the total path weight and task priority, prioritizing paths with high link quality and high task priority. Through selection, crossover, and mutation operations, the genetic algorithm iteratively optimizes the population, gradually approaching the global optimal solution.

[0043] At the same time, the Dijkstra algorithm is used to calculate the shortest path between each node and determine the optimal local path in the path combination optimized by the genetic algorithm.

[0044] The global optimization results obtained by the genetic algorithm are combined with the local optimal path of the Dijkstra algorithm, and the path weight is dynamically adjusted according to the task priority to finally determine the globally optimal inspection route and array.

[0045] S150: Acquire array point cloud data of the target area, compare and match abnormal areas and historical features with the historical damage database, implement target detection algorithms for rapid crack detection, use point cloud neural network algorithms to assess structural safety, and combine long-short-term memory networks with time series analysis to predict damage trends; Array point cloud data of the target area is acquired and compared with data from a historical disease database. Through a three-level intelligent diagnostic system—preliminary screening, feature analysis, and in-depth diagnosis—defects are gradually identified and classified. Preset thresholds are used to screen potential disease areas, extract features, and match them with historical data. Deep learning models are then used to accurately diagnose and output the type and severity of the disease.

[0046] Obtain array point cloud data of the target area, compare and match abnormal areas and historical features with the historical damage database, implement target detection algorithms for rapid crack detection, use point cloud neural network algorithms to assess structural safety, and combine long-short-term memory networks with time series analysis to predict damage trends. The specific steps include the following: S151: Acquire and compare array point cloud data with the historical disease database, and synchronize multi-sensor data timestamps; Collect array point cloud data from the target area, acquiring multi-sensor data in real time. Compare the array point cloud data with a historical damage database to obtain comparison results. Simultaneously, synchronize the multi-sensor data with timestamps. Accurately match abnormal areas in the array point cloud data with historical damage characteristics, and conduct a comprehensive analysis based on multi-sensor information.

[0047] S152: Based on the comparison results, disease diagnosis is performed using a three-level intelligent diagnosis system; Based on the comparison of array point cloud data with a historical defect database, defect diagnosis is performed according to a three-level intelligent diagnostic process. The object detection algorithm (YOLOv8) is used to quickly detect cracks at the edge and initially locate the defect area. A point cloud neural network algorithm (PointNet++) is used to conduct a detailed assessment of structural safety at the regional center and analyze the impact of the defect on the structure. A long short-term memory network (LSTM) is combined with time series analysis to predict the development trend of the defect.

[0048] Based on the comparison results, disease diagnosis is performed using a three-level intelligent diagnosis system, which includes the following sub-steps: S1521: Identify crack features through target detection algorithms, calculate coordinates and boundary ranges, and screen potential damage areas; At the edge, the system deploys an object detection algorithm to rapidly process collected images or array point cloud data. Crack features are identified using a pre-trained deep learning model, and convolutional neural networks are used to extract the crack's texture and shape characteristics. Once a crack is detected, its center coordinates and boundary range are precisely calculated, and its location information is annotated in the data. Real-time feedback on detection results allows for rapid screening of potential areas of damage.

[0049] S1522: In the regional center, multi-scale feature extraction is performed using a point cloud neural network algorithm, combined with finite element analysis and structural mechanics modeling to assess the overall safety status of the structure; At the regional center, the system conducts an in-depth assessment of structural safety using a point cloud neural network algorithm. It first extracts multi-scale features from the array point cloud data to identify potential defects such as tiny cracks, deformed areas, and potential stress concentration points.

[0050] Finite element analysis combined with structural mechanics principles is used to create a mechanical model of the extracted features. The stress distribution, deformation, and stability indicators of the structure under its current defect state are calculated. Based on the analysis results, a comprehensive assessment of the overall safety status of the structure is conducted to determine whether there are any safety hazards, and a detailed assessment report is generated.

[0051] S1523: Using long-short-term memory networks combined with time series analysis to capture the dynamic patterns of disease development and predict future spread and severity; Extract time series information about defect characteristics from inspection data, such as crack width, defect area, and structural deformation over time. Preprocess the time series data, including normalization and missing value filling. This processed data is then fed into an LSTM network, leveraging its powerful long-short-term memory capabilities to capture the dynamic patterns of defect development.

[0052] LSTMs learn dependencies within time series to identify key turning points and changing trends in disease development. Combined with time series analysis methods, such as the Autoregressive Moving Average (ARMA) model, they enhance the accuracy of disease trend forecasts. Ultimately, based on the results of LSTM and time series analysis, future disease trends, including their spread rate and potential severity, are predicted.

[0053] S160: Based on the disease diagnosis, the inspection results are evaluated, the inspection plan is updated, and S110 is executed repeatedly; Based on the third-level intelligent disease diagnosis results and inspection data, a comprehensive evaluation of the inspection effect is conducted; based on the evaluation results, the digital twin model of municipal facilities and the disease feature library are automatically updated, the inspection path and drone array are optimized, and S110 is continued.

[0054] Example 2

[0055] like Figure 2 As shown, the second embodiment of the present application provides a municipal road intelligent inspection system based on a drone array, including: Data Collection Module 21: This module integrates geographic information system map data with building information models, collects and correlates multi-source data, builds a municipal digital twin foundation, performs point cloud downsampling at the edge, extracts regions of interest from image data, and uses lightweight convolutional neural network algorithms to perform initial disease screening. Twin model construction submodule 211: used to build the municipal digital twin base, integrate geographic information system maps, building information models and inspection data, and annotate key infrastructure information; Disease initial screening submodule 212: used to perform point cloud downsampling and image region of interest extraction at the edge, retain key feature information, identify disease areas, and generate disease initial screening results Key transmission module 22: used to establish a quantum channel through the quantum key distribution module, use quantum bits to generate and distribute quantum keys, call the elliptic curve encryption algorithm, combine the quantum key with the elliptic curve public key, generate a hybrid encryption key, and perform security-enhanced transmission of disease initial screening data.

[0056] Link construction module 23: is used to calculate the optimal microwave link parameters based on the mission load and network topology information, calculate the position of the UAV relay node, coordinate the flight control system, communication module and airborne sensors, and adjust the position in combination with the path planning algorithm to build a microwave link. The encrypted data is sent via microwave, and the UAV relay node completes demodulation and error correction, re-encoding and modulation and forwards it to the target node to form a complete transmission link.

[0057] Inspection Planning Module 24: Used to construct a microwave link topology based on graph theory, quantify link quality using an edge weight calculation formula, integrate a genetic algorithm with the Dijkstra algorithm, combine link quality and task priority to construct a weighted graph and generate an initial inspection path set, generate an initial population using genetic algorithm encoding, iteratively optimize the global path through selection, crossover, and mutation, calculate the local shortest path between nodes using the Dijkstra algorithm, combine the global optimization results with the local optimal path, and determine the globally optimal inspection route and array; Link quality calculation submodule 241: used to construct a microwave link topology map based on graph theory, measure link quality, and adjust microwave link transmission; Array planning submodule 242: used to integrate the genetic algorithm and the Dijkstra algorithm, encode and generate the initial inspection path combination and perform fitness evaluation to generate the optimal inspection route and array.

[0058] Intelligent diagnosis module 25: used to obtain array point cloud data of the target area, compare and match abnormal areas and historical features with the historical disease database, implement target detection algorithms for rapid crack detection, use point cloud neural network algorithms to assess structural safety, combine long-term and short-term memory networks with time series analysis to predict disease trends, evaluate inspection results, update inspection plans, and execute S110 in a loop; Array point cloud data synchronization submodule 251: used to obtain and compare array point cloud data with the historical disease database, and synchronize multi-sensor data timestamps; Diagnosis prediction submodule 252: used to perform rapid crack detection, structural safety assessment, and disease trend prediction based on the comparison results with the help of a three-level intelligent diagnosis system.

[0059] Corresponding to the above embodiment, an embodiment of the present invention provides a computer storage medium, comprising: at least one memory and at least one processor; The memory is used to store one or more program instructions; The processor is used to run one or more program instructions to execute a method and system for intelligent municipal road inspection based on a drone array.

[0060] Corresponding to the above embodiments, an embodiment of the present invention provides a computer-readable storage medium, which contains one or more program instructions, and the one or more program instructions are used by a processor to execute a municipal road intelligent inspection method and system based on a drone array.

[0061] The embodiments disclosed in the present invention provide a computer-readable storage medium, in which computer program instructions are stored. When the computer program instructions are run on a computer, the computer executes the above-mentioned municipal pavement intelligent inspection method and system based on a drone array.

[0062] In the embodiments of the present invention, the processor may be an integrated circuit chip having signal processing capabilities. The processor may be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0063] The methods, steps, and logic diagrams disclosed in the embodiments of the present invention can be implemented or executed. A general-purpose processor can be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of the present invention can be directly implemented and executed by a hardware decoding processor, or by a combination of hardware and software modules within the decoding processor. The software modules can be located in a storage medium well-established in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, or registers. The processor reads the information from the storage medium and, in conjunction with its hardware, completes the steps of the aforementioned methods.

[0064] The storage medium may be a memory and may be, for example, a volatile memory or a nonvolatile memory, or may include both volatile and nonvolatile memory.

[0065] Among them, the non-volatile memory can be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory.

[0066] Volatile memory may be random access memory (RAM), which is used as an external cache memory. By way of example and not limitation, many forms of RAM are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM), and direct RAM bus random access memory (DRRAM).

[0067] The storage media described in the embodiments of the present invention are intended to include, but are not limited to, these and any other suitable types of memory.

[0068] Those skilled in the art will appreciate that in one or more of the above examples, the functions described herein can be implemented using a combination of hardware and software. When software is used, the corresponding functions can be stored in a computer-readable medium or transmitted as one or more instructions or codes on a computer-readable medium. Computer-readable media include computer storage media and communication media, wherein communication media includes any medium that facilitates the transmission of computer programs from one place to another. The storage medium can be any available medium that can be accessed by a general-purpose or special-purpose computer.

[0069] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent replacements, improvements, etc. made on the basis of the technical solution of the present invention should be included in the scope of protection of the present invention.

Claims

1. A municipal road intelligent inspection method based on drone array, characterized in that: include: Integrate geographic information system map data with building information models, collect and correlate multi-source data, build a municipal digital twin foundation, perform point cloud downsampling at the edge, extract regions of interest from image data, and use lightweight convolutional neural network algorithms to perform initial disease screening. A quantum channel is established through the quantum key distribution module, quantum bits are used to generate and distribute quantum keys, and the elliptic curve encryption algorithm is called to combine the quantum key with the elliptic curve public key to generate a hybrid encryption key, thereby enhancing the security of the initial disease screening data transmission. The optimal microwave link parameters are calculated based on the mission load and network topology information, and the position of the drone relay node is calculated. Through the coordination of the flight control system, communication module and onboard sensors, the position is adjusted in combination with the path planning algorithm to build a microwave link. The encrypted data is sent via microwave. The drone relay node completes demodulation and error correction, re-encoding and modulation, and forwards it to the target node, forming a complete transmission link. Based on graph theory, a microwave link topology is constructed. Link quality is quantified using an edge weight calculation formula. A genetic algorithm and Dijkstra algorithm are integrated to combine link quality and task priority to construct a weighted graph and generate an initial inspection path set. The initial population is generated through genetic algorithm encoding. The global path is optimized through selection, crossover, and mutation iterations. The Dijkstra algorithm is used to calculate the local shortest path between nodes. The global optimization results are combined with the local optimal path to determine the globally optimal inspection route and array. Acquire array point cloud data of the target area, compare and match abnormal areas and historical features with the historical damage database, implement target detection algorithms for rapid crack detection, use point cloud neural network algorithms to assess structural safety, and combine long-short-term memory networks with time series analysis to predict damage trends; Based on the disease diagnosis, evaluate the inspection results, update the inspection plan, and repeat the first step.

2. The method for intelligent municipal road inspection based on drone array according to claim 1 is characterized in that: Integrate geographic information system map data with building information models, collect and correlate multi-source data, build a municipal digital twin foundation, perform point cloud downsampling on the edge, extract regions of interest from image data, and use lightweight convolutional neural network algorithms to perform initial disease screening, including: Build a municipal digital twin foundation, integrating geographic information system maps, building information models, and inspection data, and annotating key infrastructure information; Point cloud downsampling and image region of interest extraction are performed at the edge to retain key feature information, identify diseased areas, and generate preliminary disease screening results.

3. The method for intelligent municipal road inspection based on drone array according to claim 1 is characterized in that: Based on graph theory, a microwave link topology is constructed. Link quality is quantified using an edge weight calculation formula. A genetic algorithm and Dijkstra algorithm are integrated to combine link quality and task priority to construct a weighted graph and generate an initial inspection path set. The initial population is generated through genetic algorithm encoding. The global path is optimized through selection, crossover, and mutation iterations. The Dijkstra algorithm is used to calculate the local shortest path between nodes. The global optimization results are combined with the local optimal path to determine the globally optimal inspection route and array. This includes: Build microwave link topology based on graph theory, measure link quality, and adjust microwave link transmission; By integrating the genetic algorithm and the Dijkstra algorithm, the initial inspection path combination is generated by encoding and fitness evaluation is performed to generate the optimal inspection route and array.

4. The method for intelligent municipal road inspection based on drone array according to claim 1 is characterized in that: Acquire array point cloud data of the target area, compare and match abnormal areas and historical features with the historical damage database, implement target detection algorithms for rapid crack detection, use point cloud neural network algorithms to assess structural safety, and combine long-short-term memory networks with time series analysis to predict damage trends, including: Acquire and compare array point cloud data with historical damage database, and synchronize multi-sensor data timestamps; Based on the comparison results, disease diagnosis is carried out with the help of a three-level intelligent diagnosis system.

5. The method for intelligent municipal road inspection based on drone array according to claim 4 is characterized in that: Based on the comparison results, disease diagnosis is performed with the help of a three-level intelligent diagnosis system, including: Identify crack features through target detection algorithms, calculate coordinates and boundary ranges, and screen potential damage areas; At the regional center, multi-scale feature extraction is performed using point cloud neural network algorithms, combined with finite element analysis and structural mechanics modeling to assess the overall safety status of the structure; Long-short-term memory networks combined with time series analysis are used to capture the dynamic patterns of disease development and predict future spread speed and severity.

6. A municipal road intelligent inspection system based on drone array, characterized by: include: Data acquisition module: used to integrate geographic information system map data and building information models, collect and correlate multi-source data, build the municipal digital twin foundation, complete point cloud downsampling at the edge, extract regions of interest from image data, and complete initial disease screening using a lightweight convolutional neural network algorithm; Key transmission module: used to establish a quantum channel through the quantum key distribution module, generate and distribute quantum keys using quantum bits, call the elliptic curve encryption algorithm, combine the quantum key with the elliptic curve public key to generate a hybrid encryption key, and perform security-enhanced transmission of disease initial screening data; Link construction module: This module is used to calculate the optimal microwave link parameters based on the mission load and network topology information, calculate the location of the UAV relay node, coordinate the flight control system, communication module and onboard sensors, and use the path planning algorithm to adjust the location to build the microwave link. The encrypted data is sent via microwave. The UAV relay node completes demodulation and error correction, re-encoding and modulation, and forwards it to the target node, forming a complete transmission link. Inspection planning module: This module is used to construct microwave link topology based on graph theory, quantify link quality through edge weight calculation formulas, integrate genetic algorithms with Dijkstra's algorithm, combine link quality and task priority to construct a weighted graph and generate an initial inspection path set. Genetic algorithm encoding generates an initial population, iteratively optimizes the global path through selection, crossover, and mutation, and uses Dijkstra's algorithm to calculate the local shortest path between nodes. The global optimization results are combined with the local optimal path to determine the globally optimal inspection route and array. Intelligent diagnosis module: used to obtain array point cloud data of the target area, compare and match abnormal areas and historical features with the historical disease database, implement target detection algorithms for rapid crack detection, use point cloud neural network algorithms to assess structural safety, combine long-term and short-term memory networks with time series analysis to predict disease trends, evaluate inspection results, and update inspection plans.

7. The municipal road intelligent inspection system based on drone array according to claim 6 is characterized in that: The data acquisition module specifically includes: Twin model construction submodule: used to build the municipal digital twin base, integrate geographic information system maps, building information models and inspection data, and annotate key infrastructure information; Disease initial screening submodule: used to perform point cloud downsampling and image region of interest extraction at the edge, retain key feature information, identify diseased areas, and generate initial disease screening results.

8. The municipal road intelligent inspection system based on drone array according to claim 6 is characterized in that: The inspection planning module specifically includes: Link quality calculation submodule: used to construct microwave link topology based on graph theory, measure link quality, and adjust microwave link transmission; Array planning submodule: used to integrate genetic algorithm and Dijkstra algorithm, encode and generate initial inspection path combination and perform fitness evaluation to generate the optimal inspection route and array.

9. The municipal road intelligent inspection system based on drone array according to claim 6 is characterized in that: The intelligent diagnosis module specifically includes: Array point cloud data synchronization submodule: used to obtain and compare array point cloud data with the historical disease database, and synchronize multi-sensor data timestamps; Diagnosis and prediction submodule: It is used to conduct rapid crack detection, structural safety assessment, and disease trend prediction based on the comparison results with the help of a three-level intelligent diagnosis system.

Images