Mobile digital patrol path optimization method, system and device

Abstract

The present application is applicable to the field of path optimization technology, and in particular to a method, system and device for optimizing a mobile digital patrol path, the method comprising: obtaining construction area information, human-machine density information and weather information; wherein the construction area information comprises the geographic information of each construction area, the progress information of each construction area and the risk information of each construction area, and the human-machine density information is used to characterize the distribution of construction personnel and construction equipment in each construction area; based on the construction area information, the human-machine density information and the weather information, determining the risk value of each construction area; generating the optimal patrol path for mobile digital patrol according to the risk value of each construction area and the geographic information of each construction area. This method improves the efficiency of mobile digital patrol and the dynamic adaptability of mobile digital patrol.

Description

Technical Field

[0001] The present application belongs to the field of path optimization technology, and in particular, to a method, system and device for optimizing a mobile digital patrol path. Background Art

[0002] Mobile digital inspection combines mobile devices (such as smartphones, tablets, portable terminals, etc.), digital platforms and intelligent technologies to achieve inspection and monitoring anytime and anywhere. Mobile digital inspection breaks the limitations of traditional inspections and makes inspection work more flexible, efficient and convenient through mobile devices and intelligent technologies.

[0003] In existing technologies, the complexity and dynamic changes of construction sites may exceed the preset capabilities of traditional path optimization methods. For example, in the reconstruction and expansion project of the Yongtaiwen Expressway, the construction site includes multiple complex areas such as bridges, tunnels, roadbeds, and demolition of old bridges. The geographical conditions and construction conditions vary greatly. Traditional path optimization methods may not be able to respond to changes in construction status in a timely manner, resulting in high management costs and low inspection efficiency.

[0004] To sum up, in the process of optimizing the mobile digital inspection path, there is a problem of low inspection efficiency due to the inability to dynamically adjust the inspection path in a timely manner due to changes in the construction site. Summary of the invention

[0005] The embodiments of the present application provide a method, system and device for optimizing a mobile digital patrol path, which can solve the problem of low patrol efficiency caused by the inability to dynamically adjust the patrol path in a timely manner due to changes in the construction site during the process of optimizing the mobile digital patrol path in the related technology.

[0006] In a first aspect, an embodiment of the present application provides a method for optimizing a mobile digital patrol path, including:

[0007] Acquire construction area information, human-machine density information and weather information; wherein the construction area information includes geographic information of each construction area, progress information of each construction area and risk information of each construction area, and the human-machine density information is used to characterize the distribution of construction personnel and construction equipment in each construction area;

[0008] Determine a risk value for each construction area based on the construction area information, the human-machine density information and the weather information;

[0009] Based on the risk value of each construction area and the geographic information of each construction area, the optimal patrol route for mobile digital patrol is generated.

[0010] The above technical solutions in the embodiments of the present application have at least the following technical effects:

[0011] The mobile digital patrol path optimization method provided in this application first obtains construction area information, human-machine density information and weather information, which is conducive to understanding the geographical information of each construction area, the progress information of each construction area, the risk information of each construction area, and the distribution of construction personnel and construction equipment in each construction area. Then, based on the construction area information, human-machine density information and weather information, the risk value of each construction area is determined. Finally, according to the risk value of each construction area and the geographical information of each construction area, the optimal patrol path for mobile digital patrol is generated. This method reduces unnecessary detours and repeated patrols by planning the optimal patrol path, improves patrol efficiency, enables patrol personnel to complete tasks more efficiently, and reduces time and resource costs. This method can adjust the patrol path in real time based on the dynamic environment, and improve the dynamic adaptability of mobile digital patrol. This method combines multi-source information, and through risk calculation and optimal path generation, it helps construction managers obtain data support in the decision-making process, makes the patrol process more intelligent and scientific, and improves management efficiency.

[0012] In a second aspect, an embodiment of the present application provides a mobile digital intelligence patrol path optimization system, including:

[0013] An acquisition unit is used to acquire construction area information, human-machine density information and weather information; wherein the construction area information includes geographical information of each construction area, progress information of each construction area and risk information of each construction area, and the human-machine density information is used to characterize the distribution of construction personnel and construction equipment in each construction area;

[0014] a risk value determination unit, configured to determine a risk value of each construction area based on the construction area information, the human-machine density information and the weather information;

[0015] The path generation unit is used to generate an optimal patrol path according to the risk value of each construction area and the geographical information of each construction area.

[0016] In the third aspect, an embodiment of the present application provides a mobile digital patrol path optimization device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method described in any one of the embodiments of the first aspect when executing the computer program.

[0017] It can be understood that the beneficial effects of the second to third aspects mentioned above can be found in the relevant description of the first aspect mentioned above, and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0019] Figure 1 It is a flowchart of a mobile digital patrol path optimization method provided by an embodiment of the present application;

[0020] Figure 2 It is a schematic diagram of the implementation process of step S200 in the mobile digital intelligence patrol path optimization method provided in an embodiment of the present application;

[0021] Figure 3 It is a schematic diagram of the implementation process of step S220 in the mobile digital intelligence patrol path optimization method provided in one embodiment of the present application;

[0022] Figure 4 2 is a schematic diagram of the implementation flow of step S220 in the mobile digital intelligence patrol path optimization method provided in another embodiment of the present application;

[0023] Figure 5 It is a schematic diagram of the implementation process of step S300 in the mobile digital intelligence patrol path optimization method provided in an embodiment of the present application;

[0024] Figure 6 It is a structural diagram of a mobile digital intelligence patrol path optimization system provided in an embodiment of the present application;

[0025] Figure 7 It is a structural schematic diagram of the mobile digital patrol path optimization device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0026] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.

[0027] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, wholes, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or combinations thereof.

[0028] It should also be understood that the term “and / or” used in the specification and appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0029] As used in the specification and appended claims of this application, the term "if" can be interpreted as "when" or "uponce" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" can be interpreted as meaning "uponce it is determined" or "in response to determining" or "uponce [described condition or event] is detected" or "in response to detecting [described condition or event]", depending on the context.

[0030] In addition, in the description of the present application specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0031] References to "one embodiment" or "some embodiments" etc. described in the specification of this application mean that one or more embodiments of the present application include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0032] In related technologies, the complexity and dynamic changes of construction sites may exceed the preset capabilities of traditional path optimization methods. For example, in the reconstruction and expansion project of the Yongtaiwen Expressway, the construction site includes multiple complex areas such as bridges, tunnels, roadbeds, and demolition of old bridges. The geographical conditions and construction conditions vary greatly. Traditional path optimization methods may not be able to respond to changes in construction status in a timely manner, resulting in high management costs and low inspection efficiency.

[0033] To solve the above problems, the embodiments of the present application provide a method, system and device for optimizing the mobile digital patrol path. In this method, by first obtaining the construction area information, human-machine density information and weather information, it is helpful to know the geographical information of each construction area, the progress information of each construction area and the risk information of each construction area, as well as the distribution of construction personnel and construction equipment in each construction area. Then, based on the construction area information, human-machine density information and weather information, the risk value of each construction area is determined. Finally, according to the risk value of each construction area and the geographical information of each construction area, the optimal patrol path for mobile digital patrol is generated. This method reduces unnecessary detours and repeated patrols by planning the optimal patrol path, improves patrol efficiency, enables patrol personnel to complete tasks more efficiently, and reduces time and resource costs. This method can adjust the patrol path in real time based on the dynamic environment, and improve the dynamic adaptability of mobile digital patrol. This method combines multi-source information, and through risk calculation and optimal path generation, it helps construction managers obtain data support in the decision-making process, makes the patrol process more intelligent and scientific, and improves management efficiency.

[0034] The mobile digital and intelligent patrol path optimization method provided in the embodiment of the present application can be applied to a mobile digital and intelligent patrol path optimization device. At this time, the mobile digital and intelligent patrol path optimization device is the executor of the mobile digital and intelligent patrol path optimization method provided in the embodiment of the present application. The embodiment of the present application does not impose any restrictions on the specific type of the mobile digital and intelligent patrol path optimization device.

[0035] For example, the mobile digital patrol path optimization device can be a mobile phone, a tablet computer, a wearable device, a vehicle-mounted device, a laptop computer, a netbook, a personal digital assistant (PDA), a desktop computer, a smart large screen, a smart TV, a handheld device with wireless communication function, an Internet of Vehicles terminal, a computer, a laptop computer, a handheld computing device, a satellite wireless device, a user premise equipment (customer premise equipment, CPE) and / or other equipment for communicating on a wireless system and a next-generation communication system, for example, a mobile terminal in a 5G network or a mobile terminal in a future evolved public land mobile network (Public Land Mobile Network, PLMN), etc.

[0036] In order to better understand the mobile digital and intelligent patrol path optimization method provided in the embodiment of the present application, the specific implementation process of the mobile digital and intelligent patrol path optimization method provided in the embodiment of the present application is exemplarily introduced below.

[0037] Figure 1A schematic flow chart of a mobile digital intelligence patrol path optimization method provided in an embodiment of the present application is shown, and the mobile digital intelligence patrol path optimization method includes:

[0038] S100, obtaining construction area information, human-machine density information and weather information. The construction area information includes geographical information of each construction area, progress information of each construction area and risk information of each construction area. The human-machine density information is used to characterize the distribution of construction personnel and construction equipment in each construction area.

[0039] It is understood that the geographic information of each construction area may include the precise geographic location of each construction area, which may be in the form of GPS coordinates (such as latitude and longitude) or polygonal coordinates of the area boundary. The progress information of each construction area may include the completion progress percentage of each construction area, the current task stage (such as foundation, structure, decoration, etc.). The risk information of each construction area may include the current risk level of each construction area (such as low, medium, high) and possible risk factors (such as bridge aerial work, equipment hoisting, concrete pouring, etc.).

[0040] Human-machine density information may include: personnel density, the number and distribution of personnel in each construction area, calculated by the number of people per square meter to identify areas where personnel are concentrated; equipment density, calculated by the number of equipment per square meter, especially large equipment (such as cranes, mixers) and the activity radius of large equipment.

[0041] Weather information may include weather conditions, such as sunny, cloudy, rainy, snowy, etc.; temperature, high or low temperatures may affect construction materials and worker safety; humidity, which affects the drying and hardening of materials such as concrete; and wind speed, strong winds may bring risks to aerial operations and equipment lifting.

[0042] For example, the geographic information of each construction area can be collected through GPS or GIS (Geographic Information System) to draw a regional distribution map. GIS data can also provide auxiliary information such as the size, shape and surrounding environment of the construction area.

[0043] The construction progress of each construction area (progress information of each construction area) can be obtained through the construction management system. Real-time monitoring of the construction progress can help understand the construction status and progress changes.

[0044] The risk information of each construction area can be automatically calculated through the risk assessment system, and the risk level can be calculated based on factors such as geographical location, project type, and construction tasks.

[0045] The real-time location of personnel and equipment in the construction area can be obtained through technical means such as GPS positioning systems and RFID tags, and then the human-machine density (human-machine density information) can be calculated. The distribution of personnel and equipment in the construction area can be monitored by combining video surveillance systems, infrared sensors, Bluetooth beacons and other technologies to achieve real-time density statistics.

[0046] You can use third-party weather APIs (such as China Weather API or OpenWeather API) to obtain real-time weather conditions (weather information) in the construction area, including temperature, humidity, wind speed, precipitation probability and other meteorological information.

[0047] Through this step, the construction area information, human-machine density information and weather information of each construction area can be obtained, providing basic data for formulating efficient patrol routes.

[0048] S200, determining the risk value of each construction area based on the construction area information, human-machine density information and weather information.

[0049] Exemplarily, a multi-factor risk scoring model can be established, and each risk factor (various information) can be scored through the risk scoring model, that is, the information obtained through step S100 is quantified into a score value. For example, the scoring range is 1-10, and the human-machine density information, weather information, and the geographic information, progress information, and risk information of a certain construction area are input into the risk scoring model. After the risk scoring model scores each information, the score corresponding to the human-machine density information is output as 8 points, the score corresponding to the weather information is 5 points, the score corresponding to the geographic information of the construction area is 4 points, the score corresponding to the construction progress is 7 points, and the score corresponding to the risk information is 6 points.

[0050] The weight of each risk factor (various information) in the risk value calculation can be determined through expert scoring or historical data analysis. According to the weights and scores of each risk factor, the risk value of each construction area can be calculated using a weighted formula. For example, the weight value of each construction area's geographic information is 0.15, the weight value of each construction area's progress information is 0.2, the weight value of each construction area's risk information is 0.1, the weight value of human-machine density information is 0.3, and the weight value of weather information is 0.25. The score of human-machine density information is 8 points, the score of weather information is 5 points, the score of geographic information of a construction area is 4 points, the score of construction progress is 7 points, and the score of risk information is 6 points. Then the risk value of the construction area is (0.15×4)+(0.2×7)+(0.1×6)+(0.3×8)+(0.25×5)=0.61.

[0051] Through this step, the risk value of each construction area can be dynamically adjusted and calculated so that managers can promptly identify potential risks during the construction process and carry out targeted management and prevention.

[0052] In one possible implementation, see Figure 2 , S200, based on the construction area information, human-machine density information and weather information, determine the risk value of each construction area, including:

[0053] S210, constructing a construction area relationship diagram according to the construction area information and the human-machine density information.

[0054] Exemplarily, each construction area can be regarded as a node, and the characteristics of each node can include the location coordinates of each construction area (such as longitude and latitude), the progress information of each construction area, and the risk information of each construction area. Based on the geographic information of each construction area, the actual distance between each construction area can be calculated to determine the adjacent construction areas. The Euclidean distance or Manhattan distance can be used for calculation. Construction areas with closer distances can be regarded as having a direct relationship. The nodes corresponding to each construction area are connected to the adjacent nodes through edges to construct a construction area relationship graph. Each edge of the graph represents the relationship between the construction areas. A weight can be set for each edge. The weight of the edge represents the strength of the relationship between the nodes. The weight can be set according to the physical distance between the nodes. The closer the distance, the stronger the relationship and the smaller the weight. The weight can be set according to the similarity of human-machine density. The similar human-machine density of adjacent construction areas means similar resource pressure, which may cause potential risk transmission.

[0055] Optionally, S210, constructing a construction area relationship diagram according to the construction area information and the human-machine density information, including:

[0056] S211, integrating the progress information of each construction area, the risk information of each construction area, and the human-machine density information to obtain a second risk feature of each construction area.

[0057] For example, the progress information, risk information and human-machine density information of each construction area can be standardized, and then integrated into a vector by weighted averaging or using a machine learning model as the second risk feature of each construction area.

[0058] Progress information can be expressed in terms of completion (such as 0% to 100%) or remaining time. Assuming that the progress information is expressed in terms of completion percentage, the completion percentage can be directly divided by 100 to map the progress information to between 0 and 1: ,in, Represents standardized progress information, Represents the percentage of completion. Assuming that the progress information is expressed in remaining time (such as days), the minimum and maximum remaining time in all construction areas can be calculated, using Min-Max normalization: ,in, Indicates the remaining time. Indicates the minimum remaining time in all construction areas. Indicates the maximum remaining time among all construction areas.

[0059] If the risk information is a discrete level (for example, 1 represents low risk, 2 represents medium risk, and 3 represents high risk), it can be directly mapped to a value between 0 and 1 (such as low = 0.33, medium = 0.67, high = 1). If the risk information is a specific risk score, Min-Max normalization can be used to map the score to between 0 and 1: ,in, Represents standardized risk information, represents the risk score, Represents the lowest risk score among all construction areas, Indicates the highest risk score among all construction areas.

[0060] The human-machine density information can be a specific numerical value. Min-Max normalization can be used to convert the human-machine density into a range between 0 and 1: ,in, Indicates the standardized human-machine density information, Indicates the human-machine density, Indicates the lowest human-machine density in all construction areas, Indicates the highest human and machine density among all construction areas.

[0061] After standardizing the progress information, risk information, and human-machine density information of each construction area, the weight values ​​of the progress information, risk information, and human-machine density information can be set manually to characterize their relative importance to the overall risk. The standardized information can be weighted and integrated according to the weights to obtain the second risk characteristics of each construction area. Calculation formula: ,in, represents the second risk characteristic, Indicates the weight corresponding to the progress information, represents the weight corresponding to the risk information, Indicates the weight corresponding to the human-machine density information.

[0062] S212: Determine each construction area as a node of the construction area relationship graph and determine the second risk feature of each construction area as a first node feature of the node corresponding to each construction area.

[0063] Exemplarily, each construction area can be used as a node in the construction area relationship graph, and the second risk characteristic of each construction area can be used as the first node characteristic of the node corresponding to each construction area. Each node can include the basic geographical location and the second risk characteristic of the corresponding construction area.

[0064] S213: Determine the edge of the construction area relationship graph according to the geographic information of each construction area in the construction area information, wherein the edge of the construction area relationship graph is used to represent the geographic distance between each construction area and the relative position relationship of each construction area.

[0065] For example, the neighbor relationship can be calculated based on the geographic information of each construction area (such as latitude and longitude coordinates or plane coordinates). If the geographic information is longitude and latitude coordinates, the Haversine formula can be used to calculate the spherical distance between two construction areas (applicable to distance calculation on the surface of the earth); if the geographic information is plane coordinates, the Euclidean distance calculation can be used directly. The geographic distances between all construction areas are calculated according to the selected calculation method to generate a distance matrix, in which each element in the distance matrix represents the distance between the corresponding areas.

[0066] In order to simplify the graph structure and avoid a fully connected graph, a distance threshold can be set to establish edges only between areas whose distance is less than the distance threshold. The distance threshold can be set according to the scale of the construction site or actual needs. For each node corresponding to the construction area, nodes whose geographical distance is less than the distance threshold are selected as adjacent nodes, and adjacent nodes can be connected through edges. The geographical distance can be directly used as the weight of the edge. The smaller the distance, the greater the weight, so that spatially close areas are given priority in subsequent calculations.

[0067] S214, constructing a construction area relationship graph based on the nodes of the construction area relationship graph, the first node features of the nodes corresponding to each construction area, and the edges of the construction area relationship graph.

[0068] For example, all nodes (construction areas) and edges (spatial relationships) can be integrated together to form a relationship graph ,in, is a set of nodes, representing each construction area; It is an edge set, representing the geographical connection relationship between nodes, and the edge weight reflects the geographical distance.

[0069] The construction area relationship map constructed through the above steps can effectively characterize the spatial relationship between different construction areas and provide a complete risk characteristic description for each construction area, making the overall risk assessment more comprehensive.

[0070] S220, performing spatial analysis on the construction area relationship map to obtain a first risk feature of each construction area.

[0071] Exemplarily, a graph attention network (GAT) or a graph convolutional network (GCN) in a graph neural network (GNN) can be used to perform spatial analysis on the construction area relationship graph.

[0072] The graph attention network can learn the attention weight of each node based on the features of the node and its neighboring nodes, and is suitable for analyzing graph structure data with uneven connection relationships. The node features in the construction area relationship graph can be input into the graph attention network, which uses the attention mechanism to update the node features through the weighted combination of neighbor node features. In the construction area relationship graph, the regional features of high risk or high human-machine density will have a greater impact on the adjacent nodes. The graph attention network aggregates the features of the neighboring nodes according to the attention weight of each node, thereby updating the features of each node. Each node integrates the features from the adjacent nodes to form the updated first risk feature of each node.

[0073] Graph convolutional networks are suitable for capturing local smooth information in graph structures. They can be used for spatial analysis of construction area relationship graphs and generate the first risk feature by aggregating neighbor information. The node features in the construction area relationship graph are input into the graph convolutional network, which aggregates the features of each node with the features of adjacent nodes through one or more layers of convolution operations. In this process, the edge weight is used as a weighting factor for the features of adjacent nodes, so that the degree of influence between different nodes is different. After several convolution operations, the feature vector of each node will fuse the risk features from the neighboring nodes, and the risk features will be comprehensively updated to obtain the first risk feature of each node.

[0074] Optionally, see Figure 3 , S220, spatially analyze the construction area relationship map to obtain the first risk characteristics of each construction area, including:

[0075] S221, using the graph attention network to perform feature linear transformation on the first node feature of each node in the construction area relationship graph to obtain the second node feature of each node.

[0076] For example, the first node feature of each node in the construction area relationship graph can be input into the graph attention network. In the graph attention network, a learnable linear mapping matrix can be set for the first node feature of each node. According to the linear mapping matrix corresponding to each node, the first node feature of each node is linearly transformed. The calculation formula of the linear transformation is: ,in, Indicates nodes, Indicates The first node feature of the node, Indicates The second node feature of the node, Represents the linear mapping matrix. An activation function (such as ReLU) can be added after the linear transformation to introduce nonlinearity, thereby increasing the expressive power of the graph attention network, that is, After linear transformation and activation function, the second node feature of each node is obtained. The second node feature of each node is used as the updated feature of each node, which contains the information of the node itself, and after linear transformation, it has a new feature representation, which is suitable for subsequent attention calculation and information aggregation.

[0077] S222, extracting the second node features of the neighboring nodes of each node according to the construction area relationship graph and the second node features of each node.

[0078] For example, according to the edge structure of the construction area relationship graph, all neighbor nodes of each node can be extracted to form a collection of neighbor nodes of each node, and the second risk features corresponding to all neighbor nodes of each node can be formed into a set or matrix for subsequent attention mechanism calculation. Neighbor node collection , extract each neighbor node The second risk characteristic , and combined into a collection .

[0079] S223, according to the second node feature of each node and the second node feature of the neighbor node of each node, the attention coefficient of the neighbor node of each node is calculated using the attention mechanism of the graph attention network.

[0080] For example, the second node feature of each node and the second node feature of each node's neighbor node can be concatenated, the product of the concatenated feature and a learnable weight vector is calculated, and the LeakyReLU activation function is applied to nonlinearize the product to obtain the attention coefficient of each node's neighbor node. The second eigenvector of and the node Neighbor nodes The second risk characteristic After splicing, the vector after splicing is , Represents vector concatenation. Calculate the concatenated vector With a learnable weight vector The product of , and apply the LeakyReLU activation function to nonlinearize the product and obtain the node Neighbor nodes The attention coefficient is ,in Represents the attention coefficient of neighbor nodes.

[0081] S224, normalize the attention coefficients of the neighboring nodes of each node to obtain the weight values ​​of the neighboring nodes of each node.

[0082] For example, Softmax normalization can be used to normalize the attention coefficients of all neighboring nodes of each node to obtain the normalized attention coefficients of all neighboring nodes of each node, that is, the weights of all neighboring nodes of each node, and the sum of the weights of all neighboring nodes of each node is 1. That is, for node Each neighbor node of The attention coefficient Perform Softmax normalization to get the node Each neighbor node of The weight of ,in Represents weight.

[0083] S225 , performing weighted summation on the second node features of the neighboring nodes of each node according to the weight values ​​of the neighboring nodes of each node, to obtain the third node feature of each node.

[0084] For example, the node Each neighbor node of The second risk characteristic and nodes Each neighbor node of Weight Perform weighted summation to get the node The third node feature , weighted sum calculation formula: Through this step, the third node feature of each node can be calculated.

[0085] S226: Determine the third node feature of each node as the first risk feature of the construction area corresponding to each node.

[0086] For example, the node The third node feature As a node The first risk characteristic of the corresponding construction area. Through this step, the first risk characteristic of the construction area corresponding to each node can be determined. The first risk characteristic combines the characteristics of the construction area itself and the influence of its surrounding neighboring areas.

[0087] Through graph attention network processing, the first risk feature of each node can effectively integrate the risk information of itself and the surrounding construction areas, making the risk feature more comprehensive and providing an important basis for subsequent safety management and path optimization.

[0088] Optionally, see Figure 4 , S220, spatially analyze the construction area relationship map to obtain the first risk characteristics of each construction area, including:

[0089] S2201: construct an adjacency matrix of the construction area relationship graph according to the construction area relationship graph, wherein the adjacency matrix of the construction area relationship graph is used to represent the connection relationship between each construction area.

[0090] For example, it can be assumed that there are N nodes in the construction area relationship graph, that is, N construction areas. Then the adjacency matrix It can be an N×N matrix, which is used to represent the connection relationship between each node. and nodes If there is an edge between ; If the node and nodes If there is no edge between The diagonal elements of the adjacency matrix can be Set to 1 to introduce self-connections in the graph convolution operation so that each node also includes its own information when aggregating neighbor information.

[0091] For example, there are 4 construction areas (nodes), and the connection relationship between the construction areas is node 1 is connected to node 2 and node 3, node 2 is connected to node 1 and node 3, node 3 is connected to node 1, node 2 and node 4, and node 4 is connected to node 3. Then the adjacency matrix .

[0092] S2202: Perform a convolution operation on the first node feature of each node in the construction area relationship graph according to the adjacency matrix of the construction area relationship graph to obtain a fourth node feature of each node.

[0093] For example, the adjacency matrix Normalization is performed to balance the feature aggregation between nodes and avoid the imbalance caused by the difference in node degrees. Calculate the degree matrix ,in Representation Node The degree of the node The number of adjacent nodes (including self-connections). According to the degree matrix Adjacency Matrix Normalized, we get .

[0094] The first node feature of each node in the construction area relationship graph can be constructed into a first node feature matrix , each row of the matrix represents a node The first node feature .

[0095] You can initialize a learnable weight matrix , the weight matrix is the parameter of graph convolution, which is used to linearly transform node features, weight matrix It will be updated during the graph convolutional network training process. According to the first node feature matrix And the normalized adjacency matrix , perform graph convolution operation on node features and calculate the fourth node feature matrix , the calculation formula is ,in, It is a non-linear activation function (such as ReLU), which is used to introduce non-linearity to make the model more expressive; Indicates that the features of neighboring nodes are aggregated to the central node. The fourth node feature matrix obtained by convolution is Each row in represents a node The fourth node feature .

[0096] S2203: Determine the fourth node feature of each node as the first risk feature of the construction area corresponding to each node.

[0097] For example, the node The fourth node feature As a node The first risk characteristic of the corresponding construction area.

[0098] Through the above steps, the node features can be aggregated using the adjacency matrix and graph convolution operations. The fourth node feature of each node not only contains its own risk information, but also integrates the impact of adjacent areas. The final first risk feature fully reflects the overall risk characteristics of the construction area, providing strong support for subsequent analysis and decision-making.

[0099] S230: Determine a risk value of each construction area according to the weather information and the first risk characteristic of each construction area.

[0100] Exemplarily, weather information and the first risk feature are converted into a risk value for each construction area, and a regression model or a neural network can be used to establish a relationship to map different types of inputs into numerical outputs of risk values.

[0101] Weather information can be converted into numerical features, and the numerical features can be converted into relatively consistent scales through standardization or normalization. For example, the numerical features corresponding to weather information can be [temperature, humidity, wind speed, rainfall]. The numerical features corresponding to weather information and the first risk feature vector can be combined into a complete feature vector. For example, if weather information contains m features, and the first risk feature of a construction area is a vector of length n, then the feature of the construction area is a vector of length m+n.

[0102] Different models can be selected to predict the risk value of the construction area based on the input feature vector. The models may include: linear regression, which is suitable for simple scenarios, such as the linear relationship between the impact of weather and the first risk feature on the risk value; multi-layer perceptron (MLP), which is a simple neural network structure that can handle nonlinear relationships and can use several hidden layers to map input features to risk values; XGBoost or random forest regression, which is suitable for more complex data distributions and nonlinear feature relationships and has strong explanatory power.

[0103] Collect sample data, where each piece of data contains a weather information sample, a first risk feature sample, and the corresponding actual risk value (label). Divide the sample data set into a training set and a validation set, and train the selected model through supervised learning so that the model can learn the relationship between weather information and the first risk feature vector and the risk value. After the model training is completed, the new weather information and the first risk feature of each construction area are input into the model respectively to obtain the risk value prediction of each construction area.

[0104] Through the above steps, the risk value of each construction area can be adjusted in real time with the dynamic changes of weather and the first risk characteristics of each construction area, thereby providing accurate and reliable risk assessment information to help construction managers take appropriate measures.

[0105] Optionally, S230, determining a risk value of each construction area according to the weather information and the first risk characteristic of each construction area, includes:

[0106] S231, extracting features from weather information to obtain weather features.

[0107] Exemplarily, a simple neural network can be used to extract weather features. Weather information (such as temperature, humidity, wind speed, precipitation, etc.) is input into the network. The network can use one or more hidden layers to perform linear transformation and nonlinear activation (such as ReLU) on the weather information to capture the nonlinear relationship between weather features. The output of the hidden layer can be used as the extracted weather feature vector, which can be a fixed-length vector, such as a length of k.

[0108] Principal component analysis (PCA) can be used to extract weather features, standardize each data in the weather information so that the mean of each data is 0 and the variance is 1, which is conducive to PCA processing data of different scales, calculate the covariance matrix of the standardized data, perform eigenvalue decomposition on the covariance matrix, find the main components (i.e. the direction with the largest eigenvalue), select the first k principal components (which can be the first few components that can explain most of the variance), and project the weather information onto these principal components to obtain the weather features after dimensionality reduction. PCA compresses the weather information from the original high-dimensional space to a low-dimensional feature space, retains the most important variation direction in the data, and generates a weather feature vector with a length of k.

[0109] S232, performing a feature splicing operation on the weather feature and the first risk feature of each construction area to obtain a third risk feature of each construction area.

[0110] For example, the weather feature vector and the first risk feature vector of the construction area can be concatenated to form a new feature vector, namely, the third risk feature. , the weather characteristic vector is , then the third risk characteristic of the construction area is ,in, Represents vector concatenation.

[0111] S233, inputting the third risk feature of each construction area into a risk assessment model so that the risk assessment model calculates the risk value of each construction area. The risk assessment model is a deep learning model.

[0112] For example, a suitable deep learning model can be selected according to the data features. The deep learning model may include: multi-layer perceptron (MLP), which is suitable for processing fixed-length input feature vectors; convolutional neural network (CNN), which can capture local correlations between features; recurrent neural network (RNN, LSTM, GRU), which is suitable for processing time series data. The selected model is trained to obtain a risk assessment model, and the third risk feature of each construction area is input into the trained risk assessment model. The model can output a scalar value corresponding to each construction area as the risk value of each construction area under the current conditions.

[0113] Training process of risk assessment model: Historical data of the construction area can be collected, including the third risk characteristics of the construction area and the corresponding actual risk values. Standardize or normalize the feature data to make the scales of different features consistent. Min-Max standardization or Z-score standardization can be used. Divide the historical data into training set, validation set and test set, which can be divided into 70%, 20% and 10% ratios to facilitate the generalization ability of the model.

[0114] Input the data into the selected model and calculate the predicted value through forward propagation. The loss value is calculated through the loss function (such as mean squared error (MSE) and mean absolute error (MAE), and the model parameters are updated through back propagation to minimize the loss function. The training process can be carried out for multiple rounds, and each round will traverse the entire training set, gradually reducing the loss value and improving the accuracy of the model. After each round of training, the model is evaluated on the validation set to track the model performance in order to monitor the progress of training and prevent overfitting. If the loss is observed on the validation set to no longer decrease (that is, the validation error begins to increase), the early stopping mechanism can be enabled to stop training in advance to avoid overfitting. After the training is completed, the model can be finally evaluated using the test set to verify the generalization ability of the model and calculate the error indicators (such as MSE, MAE, R2 score) on the test set to determine the performance of the model on unseen data. Based on the test results, the hyperparameters of the model (such as learning rate, number of layers, number of neurons, etc.) are further tuned and retrained to optimize the model performance.

[0115] Through the above steps, the risk value of each construction area can be obtained, providing data support for construction management, inspection scheduling and risk warning.

[0116] In a possible implementation, S200 determines the risk value of each construction area based on the construction area information, the human-machine density information, and the weather information, including:

[0117] S201, obtaining a risk value mapping table, wherein the risk value mapping table includes a risk value mapping table corresponding to the progress information of the construction area, a risk value mapping table corresponding to the risk information of the construction area, and a risk value mapping table corresponding to the weather information.

[0118] Exemplarily, a risk value mapping table corresponding to different information may be constructed artificially, and the risk value mapping table corresponding to different information may be input and stored in a local database, and may be directly called from the local database when needed.

[0119] The risk value mapping table corresponding to the progress information can map different progress stages of the construction area to corresponding risk values.

[0120] As shown in Table 1:

[0121]

[0122] Table 1

[0123] The risk value mapping table corresponding to the risk information can map known risk sources (such as high-altitude operations on bridges, equipment lifting, etc.) into specific risk values ​​based on the inherent risk information of each construction area.

[0124] As shown in Table 2:

[0125]

[0126] Table 2

[0127] The risk value mapping table corresponding to weather information can set corresponding risk values ​​according to different weather conditions (such as sunny days, rainy days, strong winds, etc.).

[0128] As shown in Table 3:

[0129]

[0130] Table 3

[0131] S202, based on the risk value mapping table, digitize the progress information of each construction area, the risk information of each construction area, and the weather information to obtain the progress risk value of each construction area, the basic risk value of each construction area, and the weather risk value.

[0132] For example, the risk value mapping table corresponding to the progress information can be searched according to the progress information to obtain the progress risk value. For example, the progress of a construction area is 70%, and the corresponding progress risk value in the search table is 3.

[0133] The risk value mapping table corresponding to the risk information can be used to convert the risk information into a numerical basic risk value. For example, if the risk source of a construction area is high-altitude bridge operation, the corresponding basic risk value in the lookup table is 5.

[0134] The weather status is converted into a numerical weather risk value through the risk value mapping corresponding to the weather information. For example, if the current weather is sunny, the corresponding weather risk value in the lookup table is 1.

[0135] S203, digitizing the man-machine density information according to the number of construction personnel and the number of construction equipment in each construction area in the man-machine density information to obtain a man-machine density risk value for each construction area.

[0136] For example, weights may be assigned to the number of construction workers and the number of construction equipment, and the human-machine density risk value of the construction area may be calculated based on the number of construction workers and the number of construction equipment in the construction area and the weights corresponding to the number of construction workers and the weights corresponding to the number of construction equipment. The formula for calculating the human-machine density risk value is: ,in, Indicates the risk value of human-machine density, Indicates the number of construction workers, represents the weight corresponding to the number of construction workers, Indicates the number of construction equipment, Indicates the weight corresponding to the number of construction equipment.

[0137] S204, obtaining progress weight, basic risk weight, weather weight and human-machine density weight. The progress weight is used to characterize the weight corresponding to the progress information of the construction area, the basic risk weight is used to characterize the weight corresponding to the risk information of the construction area, the weather weight is used to characterize the weight corresponding to the weather information, and the human-machine density weight is used to characterize the weight corresponding to the human-machine density information.

[0138] For example, the progress weight, basic risk weight, weather weight and human-machine density weight can be set manually according to the actual situation. For example, the progress weight =0.2, basic risk weight =0.3, weather weight =0.2, human-machine density weight =0.3.

[0139] S205, according to the progress weight, basic risk weight, weather weight and human-machine density weight, the progress risk value of each construction area, the basic risk value of each construction area, the weather risk value and the human-machine density risk value of each construction area are weighted and summed to obtain the risk value of each construction area.

[0140] For example, the risk values ​​of the construction area and the weights corresponding to the risk values ​​are weighted and summed, and the risk value calculation formula is: ,in, represents the risk value, represents the progress risk value, represents the basic risk value, represents the weather risk value, Indicates the risk value of human-machine density.

[0141] Through the above steps, the risk value of each construction area can comprehensively consider factors such as progress, basic risk, weather, and human-machine density, providing a scientific risk assessment basis for the generation of subsequent inspection routes.

[0142] S300, generating the optimal patrol route for mobile digital patrol according to the risk value of each construction area and the geographic information of each construction area.

[0143] For example, the geographic information of each construction area (such as GPS coordinates, area boundaries) and the corresponding risk value can be integrated into a map or grid. Inspection strategies can be set, such as giving priority to high-risk areas, or inspecting level by level according to risk classification.

[0144] You can choose a suitable path planning algorithm, which may include: heuristic search algorithm (such as A* algorithm or Dijkstra algorithm). If the geographical distribution of the construction area is complex and the distance factor needs to be considered, you can use A* algorithm or Dijkstra algorithm to generate the shortest path, using distance and risk value as weights. The higher the weight, the higher the priority of the path; ant colony algorithm or genetic algorithm. For complex path optimization problems with multiple nodes, you can consider using ant colony algorithm or genetic algorithm. Such algorithms can find the overall optimal path through iterative optimization; Traveling Salesman Problem (TSP) algorithm, which is suitable for covering all construction areas and returning to the starting point, and can find the minimum cost path covering each construction area.

[0145] Based on the map containing the geographic information of each construction area and the corresponding risk value and the patrol strategy, the selected algorithm is used to calculate the optimal patrol path so that the patrol path can give priority to high-risk areas and minimize the driving distance or time. The optimal patrol path is visualized on the map to generate a patrol navigation route. The visualization tool can help patrol personnel intuitively see the patrol sequence and specific path. If there is an automatic patrol device (such as a drone or robot), the path can be directly input into the device so that the device can perform the patrol task according to the planned path.

[0146] Through this step, the patrol efficiency and safety can be effectively improved, and real-time path adjustment can be achieved in a dynamic environment, thereby improving the response and adaptability of the mobile digital patrol system.

[0147] In one possible implementation, see Figure 5 , S300, generates the optimal patrol path for mobile digital patrol according to the risk value of each construction area and the geographical information of each construction area, including:

[0148] S310, determining a path cost function according to the risk value of each construction area and the geographical information of each construction area, wherein the path cost function is used to calculate the path cost between each construction area.

[0149] It can be understood that the path cost function is used to measure the cost of traveling from one construction area to another construction area.

[0150] Exemplarily, the variables of the path cost function may include distance cost and risk cost. The distance cost may be the physical distance between the construction areas, and the distance between each construction area is calculated based on the coordinate data (such as longitude and latitude) in the geographic information. The risk cost may be the risk value of the construction area introduced into the cost function, and the high-risk area has a higher cost in the path.

[0151] Distance cost can be calculated using Euclidean distance, which is applicable to smaller construction areas; it can also be calculated using the Haversine formula, which is applicable to geographic coordinates (latitude and longitude), and can more accurately reflect geographic distances in larger areas.

[0152] The risk cost reflects the accumulation of risks from one construction area to another, and the average risk value from one construction area to another can be calculated. The formula for calculating the risk cost is: ,in, Indicates area To area The risk cost, and Indicates area and Region risk value.

[0153] The path cost function can be ,in, Indicates area To area Path cost, represents the weight of the distance cost, Indicates area To area The distance cost, Represents the weight of risk cost.

[0154] S320, based on the path cost function, calculating the path cost between each construction area to obtain a cost matrix.

[0155] Exemplarily, according to the geographic coordinates and distance cost calculation formula in the geographic information of each construction area, the distance between a construction area and other construction areas, that is, the distance cost, is calculated; according to the risk value of each construction area and the risk cost calculation formula, the risk cost of the construction area to other construction areas is calculated; the calculated distance cost and risk cost are substituted into the path cost function to calculate the path cost from the construction area to other construction areas. Repeating these steps can obtain the path cost between each construction area. A symmetric matrix is ​​formed with the number of each construction area as the rows and columns of the cost matrix, and the calculated path cost is filled into the matrix to obtain the cost matrix. The path cost of each construction area to itself can be set to 0.

[0156] Assuming there are n construction areas, the cost matrix is an n×n matrix, the cost matrix Each element in can be For each pair of construction areas , calculate the construction area To the construction area Distance cost and risk costs , and the calculated distance cost and risk costs Substitute into the path cost function , the construction area can be calculated To the construction area The path cost ,Right now , and each pair of construction areas The path cost Fill in the cost matrix , for the construction area The path cost to itself, fill 0 into the cost matrix ,Right now .

[0157] S330, according to the cost matrix, use the path optimization algorithm to calculate the optimal path of the mobile digital intelligent patrol and obtain the optimal patrol path.

[0158] For example, assuming that the goal is to start from a starting point, pass through all construction areas and return to the starting point, the Traveling Salesman Problem (TSP) algorithm can be used. The cost matrix is ​​input into the TSP algorithm, and a starting construction area (for example, area 1) can be selected. According to the cost matrix, the TSP algorithm is used to calculate the total cost of the path starting from the starting point, traversing all construction areas, and finally returning to the starting point. During the calculation process, the TSP algorithm can record the patrol path with the minimum total path cost and output the optimal path. The optimal path can include the patrol path sequence and the total path cost.

[0159] Through the above steps, the mobile digital patrol system can efficiently complete patrol tasks in high-risk areas and improve the efficiency and safety of construction management.

[0160] Optionally, S330, calculating the optimal path of the mobile digital intelligent patrol using a path optimization algorithm according to the cost matrix to obtain the optimal patrol path includes:

[0161] S331, based on a genetic algorithm, a collection of patrol paths is generated according to the risk value of each construction area and the geographic information of each construction area.

[0162] For example, each construction area can be regarded as a gene, which can be represented by a number, such as [1, 2, 3, ..., N], indicating the order of the construction areas. A random generation or heuristic method (such as the nearest neighbor algorithm) can be used to randomly disrupt the order of the construction areas to generate multiple patrol paths to form an initial path collection (patrol path collection).

[0163] S332, calculating the total cost of each patrol path in the patrol path collection according to the cost matrix.

[0164] For example, each pair of adjacent areas in a patrol path in the patrol path collection can be traversed, and for each pair of adjacent areas, the corresponding value in the cost matrix is ​​searched, and the corresponding value is accumulated to the total cost of the path. If the path is closed (i.e., returns to the starting area), the cost between the starting area and the end area can be added. After the traversal is completed, the total cost of the patrol path is recorded. Repeating these steps can obtain the total cost of each patrol path in the patrol path collection.

[0165] Suppose there is a path [area 1, area 3, area 2], and the cost matrix is , the cost of finding the distance from region 1 to region 3 in the cost matrix is , the cost of finding the distance from region 3 to region 2 in the cost matrix is , the cost of finding the distance from region 2 to region 1 in the cost matrix is , the total cost of this patrol path is .

[0166] S333: Calculate the fitness of each patrol path according to the total cost of each patrol path in the patrol path collection.

[0167] Exemplarily, fitness is an indicator for evaluating the quality of each patrol path, and can be expressed as the inverse of the total cost, that is, the smaller the total cost, the higher the fitness, indicating that the path is better.

[0168] S334, according to the fitness of each patrol path, a genetic algorithm is used to perform selection operation, crossover operation and mutation operation on the patrol path collection to generate an optimal patrol path.

[0169] It can be understood that the selection operation aims to select a path with higher fitness from the patrol path collection to generate the next generation path collection. The selection method may include roulette selection, which calculates the sum of the fitness of all paths, assigns a fitness weight to each path, generates a random number, and selects a path according to the weight; tournament selection, which randomly selects several paths from the path collection, compares the fitness of these paths, and selects the one with the highest fitness.

[0170] The crossover operation is used to generate a new path by combining the features of two parent paths. The crossover methods can include Order Crossover (OX), which randomly selects two intersection points, defines the crossover interval of the parent path, copies the crossover interval of the first parent path to the child path, and fills in non-repeated elements from the second parent path to maintain the order; Partially Mapped Crossover (PMX), which randomly selects two intersection points, copies the part, and fills in the unselected elements in the second parent path to prevent elements from being repeated and maintain path features.

[0171] Mutation operations are used to introduce new features and increase the diversity of the path. Mutation methods can include exchange mutation, randomly selecting two construction areas in the path and swapping their positions; reversal mutation, randomly selecting a section of the path and reversing the elements in this section of the path; and insertion mutation, randomly selecting a construction area in the path and inserting the area into a random position in the path.

[0172] Exemplarily, appropriate selection methods, crossover methods, and mutation methods may be selected. Based on the selection method, the patrol path collection is selected to generate a parent path; the parent path is crossovered to generate a child path; the child path is mutated to increase diversity; the newly generated child path collection replaces the old path collection to prepare for the next generation. The selection operation, crossover operation, and mutation operation are repeated until the termination condition is met (such as reaching the set number of generations or the fitness is no longer significantly improved). In the last generation, the fitness of all patrol paths is evaluated, and the patrol path with the highest fitness is selected as the optimal patrol path.

[0173] The optimal patrol path can be visualized on the construction area map system to facilitate actual patrol operations. The effectiveness of the optimal path can be evaluated based on the feedback from the actual patrol, and the parameters of the genetic algorithm (such as selection pressure, mutation rate, etc.) can be adjusted based on the feedback to optimize the next round of path generation process.

[0174] Through the above steps, the genetic algorithm can be used to effectively optimize the patrol path, which is beneficial to the patrol efficiency and safety between construction areas.

[0175] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0176] Corresponding to the mobile digital and intelligent patrol path optimization method described in the above embodiments, the embodiments of the present application also provide a mobile digital and intelligent patrol path optimization system, and each unit of the system can implement each step of the mobile digital and intelligent patrol path optimization method. Figure 6A structural block diagram of a mobile digital patrol path optimization system provided in an embodiment of the present application is shown. For ease of explanation, only the parts related to the embodiment of the present application are shown.

[0177] Reference Figure 6 , the system comprises:

[0178] The acquisition unit is used to acquire construction area information, human-machine density information and weather information. The construction area information includes the geographical information of each construction area, the progress information of each construction area and the risk information of each construction area. The human-machine density information is used to characterize the distribution of construction personnel and construction equipment in each construction area.

[0179] The risk value determination unit is used to determine the risk value of each construction area based on the construction area information, human-machine density information and weather information.

[0180] The path generation unit is used to generate an optimal patrol path according to the risk value of each construction area and the geographical information of each construction area.

[0181] It should be noted that the information interaction, execution process and other contents between the above-mentioned units are based on the same concept as the method embodiment of the present application. Their specific functions and technical effects can be found in the method embodiment part and will not be repeated here.

[0182] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated into a processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned device can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0183] The present application also provides a mobile digital intelligence patrol path optimization device, Figure 7 This is a schematic diagram of the structure of a mobile digital intelligence patrol path optimization device provided in one embodiment of the present application. Figure 7 As shown, the mobile digital intelligence patrol path optimization device 7 of this embodiment includes: at least one processor 60 ( Figure 7 Only one is shown), at least one memory 61 ( Figure 7Only one is shown in the figure) and a computer program 62 stored in the at least one memory 61 and executable on the at least one processor 60. When the processor 60 executes the computer program 62, the mobile digital intelligent patrol path optimization device 6 implements the steps in any of the above-mentioned mobile digital intelligent patrol path optimization device embodiments, or the mobile digital intelligent patrol path optimization device 6 implements the functions of each unit in the above-mentioned device embodiments.

[0184] Exemplarily, the computer program 62 may be divided into one or more modules / units, which are stored in the memory 61 and executed by the processor 60 to complete the present application. The one or more modules / units may be a series of computer program instruction segments capable of completing specific functions, which are used to describe the execution process of the computer program 62 in the mobile digital intelligent patrol path optimization device 6.

[0185] The mobile digital intelligence patrol path optimization device 6 can be a computing device such as a desktop computer, a notebook, a PDA, and a cloud server. The mobile digital intelligence patrol path optimization device can include, but is not limited to, a processor 60 and a memory 61. Those skilled in the art can understand that Figure 7 It is only an example of the mobile digital intelligent patrol path optimization device 6 and does not constitute a limitation of the mobile digital intelligent patrol path optimization device 6. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, it may also include input and output devices, network access devices, buses, etc.

[0186] The processor 60 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0187] In some embodiments, the memory 61 may be an internal storage unit of the mobile digital and intelligent patrol path optimization device 6, such as a hard disk or memory of the mobile digital and intelligent patrol path optimization device 6. In other embodiments, the memory 61 may also be an external storage device of the mobile digital and intelligent patrol path optimization device 6, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the mobile digital and intelligent patrol path optimization device 6. Furthermore, the memory 61 may also include both the internal storage unit and the external storage device of the mobile digital and intelligent patrol path optimization device 6. The memory 61 is used to store operating systems, applications, boot loaders (BootLoader), data and other programs, such as the program code of the computer program. The memory 61 can also be used to temporarily store data that has been output or is to be output.

[0188] An embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in any of the above method embodiments are implemented.

[0189] An embodiment of the present application provides a computer program product. When the computer program product runs on a mobile digital and intelligent patrol path optimization device, the mobile digital and intelligent patrol path optimization device implements the steps in any of the above-mentioned method embodiments.

[0190] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present application implements all or part of the processes in the above-mentioned embodiment method, which can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may at least include: any entity or device that can carry the computer program code to the mobile digital intelligent patrol path optimization device, a recording medium, a computer memory, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), an electric carrier signal, a telecommunication signal, and a software distribution medium. For example, a USB flash drive, a mobile hard disk, a magnetic disk or an optical disk. In some jurisdictions, according to legislation and patent practice, computer-readable media cannot be electric carrier signals and telecommunication signals.

[0191] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0192] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0193] In the embodiments provided in the present application, it should be understood that the disclosed mobile digital intelligent patrol path optimization system, device and method can be implemented in other ways. For example, the mobile digital intelligent patrol path optimization system and device embodiments described above are merely schematic. For example, the division of the modules or units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0194] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0195] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A mobile digital intelligence patrol path optimization method, characterized in that: include: Acquire construction area information, human-machine density information and weather information; wherein the construction area information includes geographic information of each construction area, progress information of each construction area and risk information of each construction area, the human-machine density information is used to characterize the distribution of construction personnel and construction equipment in each construction area, and the progress information of each construction area, the risk information of each construction area, the human-machine density information and the weather information are all real-time data; Determine a risk value for each construction area based on the construction area information, the human-machine density information and the weather information; Generate the optimal patrol route for mobile digital patrol based on the risk value and geographic information of each construction area; Wherein, determining the risk value of each construction area based on the construction area information, the human-machine density information and the weather information includes: Constructing a construction area relationship diagram according to the construction area information and the human-machine density information; Performing spatial analysis on the construction area relationship map to obtain a first risk feature of each construction area; Determining a risk value for each construction area based on the weather information and the first risk characteristic of each construction area; The step of constructing a construction area relationship diagram according to the construction area information and the human-machine density information includes: Integrate the progress information of each construction area, the risk information of each construction area and the human-machine density information to obtain a second risk feature of each construction area; Determine each construction area as a node of the construction area relationship graph and determine the second risk feature of each construction area as a first node feature of the node corresponding to each construction area; Determine the edge of the construction area relationship graph according to the geographic information of each construction area in the construction area information; wherein the edge of the construction area relationship graph is used to represent the geographic distance between each construction area and the relative position relationship of each construction area; The construction area relationship graph is constructed based on the nodes of the construction area relationship graph, the first node features of the nodes corresponding to each construction area, and the edges of the construction area relationship graph.

2. The mobile digital intelligence patrol path optimization method according to claim 1, characterized in that: The spatial analysis of the construction area relationship map is performed to obtain the first risk feature of each construction area, including: Using a graph attention network to perform a feature linear transformation on a first node feature of each node in the construction area relationship graph to obtain a second node feature of each node; Extracting the second node features of neighboring nodes of each node according to the construction area relationship graph and the second node features of each node; According to the second node feature of each node and the second node feature of the neighbor node of each node, the attention coefficient of the neighbor node of each node is calculated using the attention mechanism of the graph attention network; Normalize the attention coefficients of each node's neighbor nodes to obtain the weight values ​​of each node's neighbor nodes; According to the weight values ​​of the neighboring nodes of each node, the second node features of the neighboring nodes of each node are weighted summed to obtain the third node feature of each node; The third node feature of each node is determined as the first risk feature of the construction area corresponding to each node.

3. The mobile digital intelligence patrol path optimization method according to claim 1, characterized in that: The spatial analysis of the construction area relationship map is performed to obtain the first risk feature of each construction area, including: According to the construction area relationship graph, construct an adjacency matrix of the construction area relationship graph; wherein the adjacency matrix of the construction area relationship graph is used to characterize the connection relationship between each construction area; According to the adjacency matrix of the construction area relationship graph, a convolution operation is performed on the first node feature of each node in the construction area relationship graph to obtain a fourth node feature of each node; The fourth node feature of each node is determined as the first risk feature of the construction area corresponding to each node.

4. The mobile digital intelligence patrol path optimization method according to claim 1, characterized in that: Determining the risk value of each construction area according to the weather information and the first risk characteristic of each construction area includes: Extracting features from the weather information to obtain weather features; Performing a feature splicing operation on the weather feature and the first risk feature of each construction area to obtain a third risk feature of each construction area; The third risk feature of each construction area is input into the risk assessment model so that the risk assessment model calculates the risk value of each construction area; wherein the risk assessment model is a deep learning model.

5. The mobile digital intelligence patrol path optimization method according to claim 1, characterized in that: The method of generating an optimal patrol path for mobile digital patrol according to the risk value of each construction area and the geographic information of each construction area includes: Determine a path cost function according to the risk value of each construction area and the geographical information of each construction area; wherein the path cost function is used to calculate the path cost between each construction area; Based on the path cost function, the path cost between each construction area is calculated to obtain a cost matrix; According to the cost matrix, a path optimization algorithm is used to calculate the optimal path for mobile digital intelligent patrol to obtain the optimal patrol path.

6. The mobile digital intelligence patrol path optimization method according to claim 5, characterized in that: The step of calculating the optimal path of the mobile digital intelligent patrol using a path optimization algorithm according to the cost matrix to obtain the optimal patrol path includes: Based on the genetic algorithm, a collection of patrol routes is generated according to the risk value of each construction area and the geographical information of each construction area; Calculating the total cost of each patrol path in the patrol path collection according to the cost matrix; Calculating the fitness of each patrol path according to the total cost of each patrol path in the patrol path collection; According to the fitness of each patrol path, the genetic algorithm is used to perform selection operation, crossover operation and mutation operation on the patrol path collection to generate the optimal patrol path.

7. A mobile digital intelligence patrol path optimization system, characterized in that: Used to implement the mobile digital intelligence patrol path optimization method according to any one of claims 1 to 6, the mobile digital intelligence patrol path optimization system comprises: An acquisition unit is used to acquire construction area information, human-machine density information and weather information; wherein the construction area information includes geographical information of each construction area, progress information of each construction area and risk information of each construction area, and the human-machine density information is used to characterize the distribution of construction personnel and construction equipment in each construction area; a risk value determination unit, configured to determine a risk value of each construction area based on the construction area information, the human-machine density information and the weather information; The path generation unit is used to generate an optimal patrol path according to the risk value of each construction area and the geographical information of each construction area.

8. A mobile digital intelligent patrol path optimization device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 6 is implemented.

Images