Unmanned aerial vehicle road ice and snow detection, melting and evaluation method based on optimization algorithm

Abstract

The application discloses a UAV road ice and snow detection, melting and evaluation method based on an optimization algorithm, relates to the technical field of unmanned aerial vehicles, and provides a UAV equipment phase laser range finder and a camera, which are used for detecting road snow thickness and road conditions, calculating snow thickness, identifying road condition parameters through image processing, then adopting an A* algorithm for path planning, taking the starting point, the end point and the current position of the UAV group as base points, using a heuristic function and a cost function to evaluate the priority of nodes, determining the best path, and generating a three-dimensional model of road surface information, based on the three-dimensional model, evaluating the number and degree of road surface cracks and deformation, determining the urgency and method of repair, and finally, the UAV group cooperatively works to execute a flight task and spray a thawing agent, so that the safe operation of the road surface in the ice and snow environment is ensured.

Description

Technical Field

[0001] This invention relates to the field of unmanned aerial vehicle (UAV) technology, specifically to a method for detecting, melting, and evaluating road surface ice and snow using UAVs based on optimization algorithms. Background Technology

[0002] Existing snow and ice removal technologies typically have limited coverage in extreme weather conditions, requiring more time and manpower to complete the task. Snow and ice removal are usually done manually or mechanically, which is a huge workload for large roads. In severe weather conditions, personnel face safety risks such as slipping and injury when performing these tasks, posing a threat to their safety. It is also difficult for staff to cope with different weather conditions and different types of road surfaces, making road maintenance difficult and costly in varying weather conditions.

[0003] In this context, drone technology has emerged, gradually developing into more efficient, safe, and economical solutions. Drones can quickly cover large areas of roads, respond rapidly to weather changes, and improve mission efficiency. Using drones for snow removal and road surface identification can reduce the risks to personnel in adverse weather conditions and improve work safety. At the same time, drones are highly adaptable and can dynamically adjust missions according to actual conditions, adapting to different weather conditions and road types, providing greater flexibility. Although the initial investment in drone technology may be high, it can significantly reduce manpower and equipment costs, as well as mission time, saving money in the long run.

[0004] However, traditional UAV maintenance methods are mainly used for inspection tasks, such as taking photos and collecting data. These tasks are usually scattered and require different UAVs to perform, resulting in fragmented task execution, requiring more time and resources, and reducing maintenance efficiency. At the same time, effective information cannot be obtained when surveying road surface information in icy and snowy environments. Ice and snow can interfere with the performance of sensors, resulting in inaccurate information, requiring manual on-site surveys. Therefore, there is an urgent need for a UAV-based method for detecting, melting, and evaluating road surface ice and snow that can be applied to snow melting to solve these problems. Summary of the Invention

[0005] (a) Technical problems to be solved

[0006] To address the shortcomings of existing technologies, this invention provides a method for detecting, melting, and evaluating road surface ice and snow using UAVs based on optimization algorithms. This method solves the problem of fragmented task execution in existing technologies, which leads to the need for more time and resources and reduces maintenance efficiency.

[0007] (II) Technical Solution

[0008] To achieve the above objectives, the present invention provides a method for detecting, melting, and evaluating road surface ice and snow using an unmanned aerial vehicle (UAV) based on an optimization algorithm, comprising:

[0009] The process involves detecting the melting route of the spraying process and setting up a drone swarm, which includes one surveying drone and one or more structural analysis drones. One surveying drone is used for route confirmation, and the drone swarm route is planned by detecting the snow thickness and road conditions. The structural analysis drones are used to survey the road surface information and upload it to the cloud service platform for road surface structure analysis.

[0010] Ground workstations are deployed around the area to be tested to receive and process data from the cloud service platform, and output the processed route planning and road surface analysis results. The route planning is transmitted to multiple structural analysis drones via wireless network, and the analysis results are output to the staff.

[0011] Multi-machine linkage and information sharing: Multiple structural analysis drones and survey drones share route information. The survey drones upload the route information to the cloud service platform, and then the cloud service platform synchronizes the route information to each structural analysis drone.

[0012] The drone swarm detection, based on the survey results of the survey drones, is shared with the drone swarm after the ground work station analyzes the route. The drone swarm executes the flight route and sprays melting agent on the ground snow during the flight.

[0013] After the snow melts, the road surface is inspected. After the snow melts, the structural analysis drones in the drone swarm scan the road surface and transmit the acquired information to the ground work station. The ground work station analyzes the scanned information and outputs the road surface structure information.

[0014] The present invention is further configured such that: the structural analysis UAV is equipped with a dedicated mist nozzle system for spraying a melting agent to melt the ice and snow on the road;

[0015] The present invention is further configured such that the step of planning the route of the drone swarm by detecting the snow thickness and road conditions on the road surface specifically includes:

[0016] The surveying drone is equipped with a phase laser rangefinder and a camera. The phase laser rangefinder sends laser pulses, modulates the laser beam, and measures the phase delay of the modulated light traveling to and from the survey line. Based on the wavelength of the modulated light, the distance represented by the phase delay is calculated. Specifically:

[0017] Where h represents the snow thickness, Indicates the excitation wavelength. Indicates phase difference, Pi is the mathematical constant, and the round-trip time measured by a laser rangefinder.

[0018] At the same time, determine the amount of melting agent to be sprayed at each point, specifically:

[0019] ,in That is, the amount of solvent sprayed. Indicates the thickness of the snow cover. This indicates the melting agent factor, which is set according to the melting effect of the actual melting agent used.

[0020] The camera captures images of the road surface and processes them to identify road condition parameters.

[0021] Based on the path planning A* algorithm, the flight path of the UAV swarm is planned;

[0022] The present invention is further configured such that the flight path of the planned unmanned aerial vehicle (UAV) swarm is specifically as follows:

[0023] Define the starting point, ending point, and current position of the drone swarm;

[0024] Create a state space containing multiple path nodes, where nodes reflect the position and state of the drones. Assume there are currently n drones, and the position of each drone is... ;

[0025] Initialize the starting node of the A* algorithm, which is the current position of the drone;

[0026] Define the heuristic function h(n) and the cost function g(n), start the A* algorithm search process, and iteratively select the best node to expand the search tree;

[0027] In each iteration, the node with the lowest cost is selected for expansion, specifically:

[0028] The heuristic function h(n) is used to estimate the distance from the current node n to the target node;

[0029] The cost function g(n) is used to estimate the actual cost from the starting node to the current node n;

[0030] The evaluation function is f(n) = g(n) + h(n), which is used to select the node for the next expansion.

[0031] The optimal path is found by continuously evaluating the priority f(n) of the nodes;

[0032] The present invention is further configured such that the definition method of the heuristic function h(n) and the cost function g(n) is specifically as follows:

[0033] Using the Euclidean distance estimated by straight-line distance as a heuristic function, it is represented in a two-dimensional plane as the straight-line distance between two points:

[0034] ,

[0035] in This represents the distance between two points in the x-coordinate system. This represents the distance between two points in the y-coordinate system;

[0036] The cost function g(n) is defined as follows:

[0037] g(n) is used to estimate the actual cost from the starting node to the current node n, which is calculated by accumulating the costs along the path, specifically:

[0038] ,

[0039] in, This represents the actual cost from the starting node to the parent node. This represents the cost from the parent node to the current node;

[0040] The present invention is further configured such that the multiple structural analysis UAVs perform coordinated maintenance, specifically:

[0041] Each structural analysis drone takes pictures and samples of the road surface using a camera;

[0042] The sampling information is uploaded to the cloud service platform for data analysis.

[0043] All drone sampling data is integrated to generate a complete 3D model of the road surface structure;

[0044] Analysis based on 3D modeling is used to detect the current road surface maintenance risks.

[0045] The present invention is further configured such that: the route information also includes road snow thickness information, and the melting agent spraying amount of the dedicated mist nozzle system is determined based on the analysis of road snow conditions by the cloud service platform;

[0046] The present invention is further configured such that the data integration and 3D modeling steps specifically include:

[0047] Each structural analysis drone uses its onboard camera to capture and sample road surface information. Each sampled image includes parameters... ,in This represents the image pixel value, specifically the brightness of the pixel in the i-th row and j-th column. Represents the horizontal coordinates of pixels in an image. Represents the vertical coordinates of pixels in an image;

[0048] The sampling information is uploaded to the cloud service platform and stored as an image dataset D;

[0049] The uploaded image is processed, and key feature points are detected and matched using a feature matching algorithm to obtain a set of feature points. ,in This represents the horizontal coordinates of the feature point in the image. This represents the vertical coordinates of the feature point in the image. Indicates the depth of the feature point in actual three-dimensional space;

[0050] Matching-based feature points The distance between the camera and feature points is calculated using triangulation to generate a 3D point cloud. ,in This represents the horizontal coordinate of a point in three-dimensional space. Represents the vertical coordinates of a point in three-dimensional space. Represents the depth of a point in three-dimensional space;

[0051] The present invention is further configured such that the data integration and 3D modeling steps also include:

[0052] After the snow melts, a second survey of the road surface is conducted to compare the snow thickness and check the melting effect.

[0053] The melting effect was confirmed by comparing the snow thickness and performing a second melting process on any unmelted areas.

[0054] Confirmed based on snow depth and melt agent spraying amount, specifically:

[0055] Where h represents the snow thickness, Indicates the excitation wavelength. Indicates phase difference, Pi is the mathematical constant, and the round-trip time measured by a laser rangefinder.

[0056] ,in That is, the amount of solvent sprayed. Indicates the thickness of the snow cover. This indicates the melting agent factor, which is set according to the melting effect of the actual melting agent used.

[0057] Set a snow thickness threshold according to the actual scenario requirements. When the snow thickness detected in the second test is greater than the threshold, repeat the snow melting process until the thickness requirement is met.

[0058] (III) Beneficial Effects

[0059] This invention provides a method for detecting, melting, and evaluating road surface ice and snow using unmanned aerial vehicles (UAVs) based on optimization algorithms. It has the following beneficial effects:

[0060] The UAV-based method for detecting, melting, and evaluating road surface ice and snow based on optimization algorithms provided in this application is mainly used in icy and snowy environments. Through UAV technology, it achieves integrated detection and maintenance, including road snow thickness and condition detection, path planning, fog nozzle systems, data integration and 3D modeling, risk assessment, multi-UAV collaboration, and UAV swarm detection. Among these:

[0061] The drone is equipped with a phase laser rangefinder and a camera to detect the snow thickness and road conditions. The phase laser rangefinder measures the round-trip time and calculates the snow thickness, while the camera captures road images and identifies road condition parameters through image processing.

[0062] Then, the A* algorithm is used for path planning, which efficiently covers the road surface area. The starting point, ending point and current position of the drone swarm are used as the base points. Heuristic functions and cost functions are used to evaluate the priority of nodes and determine the optimal path.

[0063] The mist spraying system uses a specialized spraying system to spray a melting agent, melting ice and snow on the road and ensuring smooth traffic flow.

[0064] The data integration and 3D modeling stage involves uploading images acquired by cameras to a cloud service platform to generate a 3D model of the road surface information. Based on the 3D model, the number and extent of cracks and deformations are assessed to determine the urgency and methods of maintenance.

[0065] Finally, the drone swarm worked together to carry out flight missions and spray melting agents to ensure the safe operation of roads in icy and snowy conditions.

[0066] In summary, the UAV-based method for detecting, melting, and evaluating road surface ice and snow based on optimization algorithms provided in this application integrates detection and maintenance functions, achieving integrated UAV maintenance. This reduces operational complexity and engineering management difficulties, making the maintenance process more efficient and allowing for completion in a shorter time. Through a mist nozzle system, the UAV can spray melting agents to melt ice and snow covering roads, thereby improving road safety and facilitating image acquisition. Simultaneously, the real-time data transmission and analysis from the cloud service platform provides more timely decision support, enabling operators to better understand the road surface condition and risks. Furthermore, the UAV can adapt to various maintenance tasks, not just a single one, increasing its versatility and applicability. Finally, through 3D modeling and data integration, operators can better understand the road surface structure and condition, thus better planning and executing maintenance work.

[0067] This solves the problem of fragmented task execution in existing technologies, which leads to the need for more time and resources and reduces maintenance efficiency. Attached Figure Description

[0068] Figure 1 This is a flowchart of the UAV road surface ice and snow detection, melting and evaluation method based on optimization algorithm of the present invention.

[0069] Figure 2 The flowchart below shows the flight path determination process for the UAV road surface ice and snow detection, melting, and evaluation method based on optimization algorithms of the present invention. Detailed Implementation

[0070] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0071] Example

[0072] Please see Figures 1-2 This invention provides a method for detecting, melting, and evaluating road surface ice and snow using unmanned aerial vehicles (UAVs) based on optimization algorithms, including:

[0073] The spraying melt route detection involves setting up a drone swarm, which includes one survey drone and one or more structural analysis drones. One survey drone is used for route confirmation, and the drone swarm route is planned by detecting the snow thickness and road conditions. The structural analysis drones are used to survey the road surface information and upload it to the cloud service platform for road surface structure analysis.

[0074] The structural analysis drone is equipped with a dedicated mist nozzle system for spraying melting agents to melt ice and snow on roads;

[0075] The specific steps for planning drone swarm routes by detecting snow thickness and road conditions include:

[0076] The surveying drone is equipped with a phase laser rangefinder and a camera. The phase laser rangefinder sends laser pulses, modulates the laser beam, and measures the phase delay of the modulated light traveling to and from the survey line. Based on the wavelength of the modulated light, the distance represented by the phase delay is calculated. Specifically:

[0077] Where h represents the snow thickness, Indicates the excitation wavelength. Indicates phase difference, Pi is the mathematical constant, and the round-trip time measured by a laser rangefinder.

[0078] At the same time, determine the amount of melting agent to be sprayed at each point, specifically:

[0079] ,in That is, the amount of solvent sprayed. Indicates the thickness of the snow cover. This indicates the melting agent factor, which is set according to the melting effect of the actual melting agent used.

[0080] The camera captures images of the road surface and processes them to identify road condition parameters.

[0081] Based on the path planning A* algorithm, the flight path of the UAV swarm is planned;

[0082] The specific flight routes for the planned drone swarm are as follows:

[0083] Define the starting point, ending point, and current position of the drone swarm;

[0084] Create a state space containing multiple path nodes, where nodes reflect the position and state of the drones. Assume there are currently n drones, and the position of each drone is... ;

[0085] Initialize the starting node of the A* algorithm, which is the current position of the drone;

[0086] Define the heuristic function h(n) and the cost function g(n), start the A* algorithm search process, and iteratively select the best node to expand the search tree;

[0087] In each iteration, the node with the lowest cost is selected for expansion, specifically:

[0088] The heuristic function h(n) is used to estimate the distance from the current node n to the target node;

[0089] The cost function g(n) is used to estimate the actual cost from the starting node to the current node n;

[0090] The evaluation function is f(n) = g(n) + h(n), which is used to select the node for the next expansion.

[0091] The optimal path is found by continuously evaluating the priority f(n) of the nodes;

[0092] The specific definition methods for the heuristic function h(n) and the cost function g(n) are as follows:

[0093] Using the Euclidean distance estimated by straight-line distance as a heuristic function, it is represented in a two-dimensional plane as the straight-line distance between two points:

[0094] ,

[0095] in This represents the distance between two points in the x-coordinate system. This represents the distance between two points in the y-coordinate system;

[0096] The cost function g(n) is defined as follows:

[0097] g(n) is used to estimate the actual cost from the starting node to the current node n, which is calculated by accumulating the costs along the path, specifically:

[0098] ,

[0099] in, This represents the actual cost from the starting node to the parent node. This represents the cost from the parent node to the current node;

[0100] Ground workstations are deployed around the area to be tested to receive and process data from the cloud service platform, and output the processed route planning and road surface analysis results. The route planning is transmitted to multiple structural analysis drones via wireless network, and the analysis results are output to the staff.

[0101] Multi-machine linkage and information sharing: Multiple structural analysis drones and survey drones share route information. The survey drones upload the route information to the cloud service platform, and then the cloud service platform synchronizes the route information to each structural analysis drone.

[0102] Multiple structural analysis drones were used in coordinated maintenance, specifically:

[0103] Each structural analysis drone takes pictures and samples of the road surface using a camera;

[0104] The sampling information is uploaded to the cloud service platform for data analysis.

[0105] All drone sampling data is integrated to generate a complete 3D model of the road surface structure;

[0106] Analysis based on 3D modeling is used to detect the current road surface maintenance risks.

[0107] The route information also includes information on the thickness of snow on the road surface. Based on the analysis of the snow conditions on the road surface by the cloud service platform, the amount of melt spraying agent for the dedicated mist nozzle system is determined.

[0108] The data integration and 3D modeling steps specifically include:

[0109] Each structural analysis drone uses its onboard camera to capture and sample road surface information. Each sampled image includes parameters... ,in This represents the image pixel value, specifically the brightness of the pixel in the i-th row and j-th column. Represents the horizontal coordinates of pixels in an image. Represents the vertical coordinates of pixels in an image;

[0110] The sampling information is uploaded to the cloud service platform and stored as an image dataset D;

[0111] The uploaded image is processed, and key feature points are detected and matched using a feature matching algorithm to obtain a set of feature points. ,in This represents the horizontal coordinates of the feature point in the image. This represents the vertical coordinates of the feature point in the image. Indicates the depth of the feature point in actual three-dimensional space;

[0112] Matching-based feature points The distance between the camera and feature points is calculated using triangulation to generate a 3D point cloud. ,in This represents the horizontal coordinate of a point in three-dimensional space. Represents the vertical coordinates of a point in three-dimensional space. Represents the depth of a point in three-dimensional space;

[0113] The data integration and 3D modeling steps also include:

[0114] The data integration and 3D modeling steps also include:

[0115] After the snow melts, a second survey of the road surface is conducted to compare the snow thickness and check the melting effect.

[0116] The melting effect was confirmed by comparing the snow thickness and performing a second melting process on any unmelted areas.

[0117] Confirmed based on snow depth and melt agent spraying amount, specifically:

[0118] Where h represents the snow thickness, Indicates the excitation wavelength. Indicates phase difference, Pi is the mathematical constant, and the round-trip time measured by a laser rangefinder.

[0119] ,in That is, the amount of solvent sprayed. Indicates the thickness of the snow cover. This indicates the melting agent factor, which is set according to the melting effect of the actual melting agent used.

[0120] Set a snow thickness threshold according to the actual scenario requirements. When the snow thickness detected in the second test is greater than the threshold, repeat the snow melting process until the thickness requirement is met.

[0121] The drone swarm detection is based on the survey results of the survey drones. The ground work station analyzes the route and shares it with the drone swarm. The drone swarm executes the flight route and sprays melting agent on the ground snow during the flight.

[0122] After the snow melts, the road surface is inspected. After the snow melts, the structural analysis drones in the drone swarm scan the road surface and transmit the acquired information to the ground work station. The ground work station analyzes the scanned information and outputs the road surface structure information.

[0123] In summary, in this application:

[0124] The UAV-based method for detecting, melting, and evaluating road surface ice and snow based on optimization algorithms provided in this application is mainly used in icy and snowy environments. Through UAV technology, it achieves integrated detection and maintenance, including road snow thickness and condition detection, path planning, fog nozzle systems, data integration and 3D modeling, risk assessment, multi-UAV collaboration, and UAV swarm detection. Among these:

[0125] The drone is equipped with a phase laser rangefinder and a camera to detect the snow thickness and road conditions. The phase laser rangefinder measures the round-trip time and calculates the snow thickness, while the camera captures road images and identifies road condition parameters through image processing.

[0126] Then, the A* algorithm is used for path planning, which efficiently covers the road surface area. The starting point, ending point and current position of the drone swarm are used as the base points. Heuristic functions and cost functions are used to evaluate the priority of nodes and determine the optimal path.

[0127] The mist spraying system uses a specialized spraying system to spray a melting agent, melting ice and snow on the road and ensuring smooth traffic flow.

[0128] The data integration and 3D modeling stage involves uploading images acquired by cameras to a cloud service platform to generate a 3D model of the road surface information. Based on the 3D model, the number and extent of cracks and deformations are assessed to determine the urgency and methods of maintenance.

[0129] Finally, the drone swarm worked together to carry out flight missions and spray melting agents to ensure the safe operation of roads in icy and snowy conditions.

[0130] In summary, the UAV-based method for detecting, melting, and evaluating road surface ice and snow based on optimization algorithms provided in this application integrates detection and maintenance functions, achieving integrated UAV maintenance. This reduces operational complexity and engineering management difficulties, making the maintenance process more efficient and allowing for completion in a shorter time. Through a mist nozzle system, the UAV can spray melting agents to melt ice and snow covering roads, thereby improving road safety and facilitating image acquisition. Simultaneously, the real-time data transmission and analysis from the cloud service platform provides more timely decision support, enabling operators to better understand the road surface condition and risks. Furthermore, the UAV can adapt to various maintenance tasks, not just a single one, increasing its versatility and applicability. Finally, through 3D modeling and data integration, operators can better understand the road surface structure and condition, thus better planning and executing maintenance work.

[0131] This solves the problem of fragmented task execution in existing technologies, which leads to the need for more time and resources and reduces maintenance efficiency.

[0132] Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the implementation of the present invention. Those skilled in the art can make other variations or modifications based on the above description. It is neither necessary nor possible to exhaustively describe all embodiments here. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the claims of the present invention.

Claims

1. A method for detecting, melting and evaluating the road ice and snow based on the optimization algorithm of the unmanned aerial vehicle, characterized in that, include: The process involves detecting the melting route of the spraying process and setting up a drone swarm, which includes one surveying drone and one or more structural analysis drones. One surveying drone is used for route confirmation, and the drone swarm route is planned by detecting the snow thickness and road conditions. The structural analysis drones are used to survey the road surface information and upload it to the cloud service platform for road surface structure analysis. Ground workstations are deployed around the area to be tested to receive and process data from the cloud service platform, and output the processed route planning and road surface analysis results. The route planning is transmitted to multiple structural analysis drones via wireless network, and the analysis results are output to the staff. Multi-machine linkage and information sharing: Multiple structural analysis drones and survey drones share route information. The survey drones upload the route information to the cloud service platform, and then the cloud service platform synchronizes the route information to each structural analysis drone. The drone swarm detection, based on the survey results of the survey drones, is shared with the drone swarm after the ground work station analyzes the route. The drone swarm executes the flight route and sprays melting agent on the ground snow during the flight. After the snow melts, the road surface is inspected and a snow-melting agent is sprayed on it. Once the snow melts, a structural analysis drone from the drone swarm scans the road surface and transmits the acquired information to a ground work station. The ground work station then analyzes the scanned information and outputs the road surface structure information. The steps for planning the drone swarm route by detecting the snow thickness and road conditions specifically include: The surveying drone is equipped with a phase laser rangefinder and a camera. The phase laser rangefinder sends laser pulses, modulates the laser beam, and measures the phase delay of the modulated light traveling to and from the survey line. Based on the wavelength of the modulated light, the distance represented by the phase delay is calculated. Specifically: where h denotes the snow depth, denotes the laser wavelength, denotes the phase difference, denotes the circular constant, the round-trip time measured by the laser rangefinder; At the same time, determine the amount of melting agent to be sprayed at each point, specifically: wherein i.e. the amount of thawing agent sprayed, denotes the snow depth, denotes the thawing agent factor, which is set in accordance with the thawing effect of the actually used thawing agent; The camera captures images of the road surface and processes them to identify road condition parameters. Based on the path planning A* algorithm, the flight path of the UAV swarm is planned; The specific flight routes of the planned drone swarm are as follows: Define the starting point, ending point, and current location of the drone swarm; creating a state space comprising a plurality of path nodes, wherein the nodes reflect the position and state of the drones, given that there are n drones, each drone has a position ; Initialize the starting node of the A* algorithm, which is the current position of the drone; Define the heuristic function h(n) and the cost function g(n), start the A* algorithm search process, and iteratively select the best node to expand the search tree; In each iteration, the node with the lowest cost is selected for expansion, specifically: The heuristic function h(n) is used to estimate the distance from the current node n to the target node; The cost function g(n) is used to estimate the actual cost from the starting node to the current node n; The evaluation function is f(n) = g(n) + h(n), which is used to select the node for the next expansion. The optimal path is found by continuously evaluating the priority f(n) of nodes. 2.The UAV road ice and snow detection, melting and evaluation method based on optimization algorithm of claim 1, wherein, The structural analysis drone is equipped with a dedicated mist nozzle system for spraying melting agents to melt ice and snow on roads. 3.The UAV road ice and snow detection, melting and evaluation method based on optimization algorithm of claim 1, wherein, The specific definition methods for the heuristic function h(n) and cost function g(n) are as follows: Using the Euclidean distance estimated by straight-line distance as a heuristic function, it is represented in a two-dimensional plane as the straight-line distance between two points: ， wherein denotes the distance between two points in the x coordinate system, denotes the distance between two points in the y coordinate system; The cost function g(n) is defined as follows: g(n) is used to estimate the actual cost from the starting node to the current node n, which is calculated by accumulating the costs along the path, specifically: ， wherein, represents the actual cost from the start node to the parent node, represents the cost from the parent node to the current node.

4. The method for detecting, melting, and evaluating road surface ice and snow using an unmanned aerial vehicle (UAV) based on an optimization algorithm as described in claim 1, characterized in that, The multiple structural analysis drones are used for coordinated maintenance, specifically as follows: Each structural analysis drone takes pictures and samples of the road surface using a camera; The sampling information is uploaded to the cloud service platform for data analysis. All drone sampling data is integrated to generate a complete 3D model of the road surface structure; Analysis based on 3D modeling is used to detect road surface risks that currently require maintenance.

5. The optimization algorithm based UAV road ice and snow detection, melting and evaluation method according to claim 1, characterized in that, The route information also includes information on the thickness of snow on the road surface. Based on the analysis of the snow conditions on the road surface by the cloud service platform, the amount of melt spraying agent for the dedicated mist nozzle system is determined.

6. The method for detecting, melting, and evaluating road surface ice and snow using an unmanned aerial vehicle (UAV) based on an optimization algorithm as described in claim 1, characterized in that, The data integration and 3D modeling steps specifically include: Each structural analysis unmanned aerial vehicle shoots and samples road surface information through a camera carried by the unmanned aerial vehicle, and each sampling image parameter includes wherein represents an image pixel value, i.e., the luminance of a pixel in the i-th row and the j-th column, represents a horizontal coordinate of a pixel in the image, represents a vertical coordinate of a pixel in the image; The sampling information is uploaded to the cloud service platform and stored as an image dataset D; Image processing is performed on the uploaded image, and key feature points are detected and matched through a feature matching algorithm to obtain a feature point set wherein denotes a horizontal coordinate of the feature point in the image, denotes a vertical coordinate of the feature point in the image, denotes a depth of the feature point in the actual three-dimensional space; Matching-based feature points A three-dimensional point cloud is generated by calculating the distance between the camera and the feature points by triangulation where represents the horizontal coordinate of the point in three-dimensional space, represents the vertical coordinate of the point in three-dimensional space, represents the depth of the point in three-dimensional space.

7. The method for detecting, melting, and evaluating road surface ice and snow using an unmanned aerial vehicle (UAV) based on an optimization algorithm as described in claim 1, characterized in that, The data integration and 3D modeling steps also include: After the snow melts, a second survey of the road surface is conducted to compare the snow thickness and check the melting effect. The melting effect was confirmed by comparing the snow thickness and performing a second melting process on any unmelted areas. Confirmed based on snow depth and melt agent spraying amount, specifically: where h denotes the snow depth, denotes the laser wavelength, denotes the phase difference, denotes the circular constant, the round-trip time measured by the laser rangefinder; wherein i.e. the amount of de-icing agent sprayed, denotes the snow depth, denotes the de-icing agent factor, which is set in accordance with the melting effect of the de-icing agent actually employed; Set a snow thickness threshold according to the actual scenario requirements. When the snow thickness detected in the second test is greater than the threshold, repeat the snow melting process until the thickness requirement is met.

Images