A method and apparatus for generating airport pavement tasks based on dynamic association

Abstract

This invention discloses a method and apparatus for generating airport pavement tasks based on dynamic association. The method includes: providing airport pavement event information based on the airport pavement layout; determining multiple airport pavement target events based on event time matching with the target events in response to a target time; classifying the airport pavement target events according to the airport pavement layout and the event attributes of each target event, and assigning event tags; linking the airport pavement target events based on event combination strategies and event tags to generate initial airport pavement tasks; and reorganizing the initial airport pavement tasks through a real-time dynamic association strategy to generate airport pavement tasks. By analyzing and combining multiple fragmented and discrete airport pavement events in multiple ways, the final integrated airport pavement tasks are obtained, enabling the rational arrangement of tasks such as airport pavement cleaning and inspection.

Description

Technical Field

[0001] This invention belongs to the technical field of airport business management, specifically relating to a method and apparatus for generating airport pavement tasks based on dynamic association. Background Technology

[0002] Airport pavements are crucial platforms for aircraft takeoff, landing, and parking, and their service performance plays a key role in airport operational safety. Continuous, scientific, and rational pavement maintenance is of great significance for improving pavement service quality and extending its lifespan.

[0003] In the existing airport pavement task management system, multiple inspection / cleaning tasks for the same area or route are usually issued independently. When a shift passes through multiple task areas during the same trip, a separate form needs to be filled out for each task area, resulting in unnecessary duplication of work. Furthermore, multiple forms can easily lead to conflicts between the same basic information in different task records, causing difficulties for subsequent data statistics and management.

[0004] Patent application CN117745264A discloses an intelligent maintenance management method applicable to airport pavements, including acquiring intelligent pavement monitoring data, intelligent pavement inspection data, and regular pavement testing data; setting pavement structural safety risk warning indicators and standards; constructing a pavement performance evaluation system; selecting pavement performance indicators; establishing an S-curve pavement performance prediction model based on the long-term accumulation of pavement performance data; establishing a multi-objective intelligent decision-making model for pavement maintenance; formulating pavement maintenance plans based on pavement performance and operational status; and tracking the implementation effects; and establishing a two-dimensional / three-dimensional GIS map platform for pavement to achieve data visualization analysis and display.

[0005] The aforementioned technologies only plan for airport pavement maintenance and do not provide methods for handling conflicts and relationships between various maintenance tasks. How to deal with the problems of duplicate form filling, low work efficiency and inconsistent basic data caused by the discrete release of tasks, and how to achieve a reasonable arrangement of tasks such as airport pavement cleaning and inspection are the problems that need to be solved at present. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this invention provides a method and apparatus for generating airport pavement tasks based on dynamic association. The method includes: providing airport pavement event information based on the airport pavement layout; determining multiple airport pavement target events based on event time matching with target time events in response to a target time; classifying the airport pavement target events according to the airport pavement layout and their event attributes, and assigning event tags; linking the airport pavement target events using event combination strategies and event tags to generate initial airport pavement tasks; and reorganizing the initial airport pavement tasks using a real-time dynamic association strategy to generate airport pavement tasks. By analyzing and combining multiple fragmented and discrete airport pavement events, and finally integrating them into airport pavement tasks, multiple independent airport pavement events can be bundled for execution. When executing airport pavement tasks, information corresponding to multiple related airport pavement events included in the task can be uniformly filled in and reported, avoiding information duplication and achieving reasonable scheduling of airport pavement cleaning, inspection, and other tasks.

[0007] In a first aspect, the present invention provides a method for generating airport pavement tasks based on dynamic association, comprising:

[0008] Based on the airport pavement layout, provide information on airport pavement matters;

[0009] In response to the target time, multiple airport pavement target items are identified based on the time matching of items with airport pavement items;

[0010] Based on the airport pavement layout and the attributes of each target item on the airport pavement, the target items on the airport pavement are classified and labeled.

[0011] By combining the task combination strategy and task tags, the target tasks of the airport pavement are linked to give the initial tasks of the airport pavement;

[0012] By using a real-time dynamic association strategy, the initial airport pavement tasks are reorganized to generate airport pavement tasks.

[0013] Furthermore, airport pavement layout includes the arrangement information of airport pavement buildings; airport pavement event information includes event attributes and event time.

[0014] Furthermore, the item label includes attribute sub-labels and relative adjacent value sub-labels.

[0015] Furthermore, based on the airport pavement layout and the attributes of each target item on the airport pavement, the target items are categorized and labeled. This process includes the following steps:

[0016] Based on the attributes of each airport pavement target item, airport pavement target items are classified and attribute sub-labels are given;

[0017] By using the airport pavement layout, the locations of target items on the airport pavement are marked, forming location labels for each target item on the airport pavement.

[0018] Based on the location labels of each airport pavement target item, the corresponding relative proximity value sub-label is given;

[0019] The attribute sub-tags and relatively adjacent value sub-tags of target items on the pavement of each airport are integrated to form item tags.

[0020] Furthermore, based on the location labels of each airport pavement target item, corresponding relative proximity value sub-labels are given, specifically including the following steps:

[0021] Based on the location tags of target items on the pavement of each airport, a proximity relationship analysis is performed on the target items on the pavement of each airport;

[0022] Based on the first proximity threshold, the relative proximity values ​​between target items on each airport pavement are given, forming relative proximity value sub-labels.

[0023] Furthermore, based on the first proximity threshold, the relative proximity values ​​between target items on each airport pavement are given, forming relative proximity value sub-labels, specifically represented as follows:

[0024]

[0025] Among them, RPV ij Let m be the sub-label of the relative proximity value between the i-th airport pavement target item and the j-th airport pavement target item. dij PT1 is the relative proximity value between the i-th airport pavement target item and the j-th airport pavement target item, and PT1 is the first proximity threshold. The relative proximity value is the distance between the center point coordinates of the location labels in the two airport pavement target items.

[0026] Furthermore, by combining the task combination strategy and task tags, the target tasks for airport pavement are linked to provide the initial tasks for airport pavement, specifically including the following steps:

[0027] Based on the attribute sub-tags, each airport pavement target item is grouped and multiple attribute subsets are given;

[0028] In each attribute subset, an adjacency graph is constructed based on the nearest value sub-labels of each airport pavement target item, with each airport pavement target item as a node in the adjacency graph;

[0029] Set access flags for nodes in the adjacency graph and initialize the state of each node to an unvisited state;

[0030] Starting from any node, a depth-first search algorithm is used to search for all reachable nodes in the adjacency graph to form connected components and determine the initial task of the airport pavement.

[0031] Mark the origin and reachable nodes as visited;

[0032] Starting from the nodes that are not visited, repeat the above search process to form a new airport runway initial task until all nodes are included in the airport runway initial task.

[0033] Summarize all the initial tasks for airport pavement to obtain the initial tasks for airport pavement.

[0034] Furthermore, starting from any node, a depth-first search algorithm is used to search for all reachable nodes in the adjacency graph to form connected components, thus determining the initial task of the airport runway surface. This specifically includes the following steps:

[0035] Determine the second proximity threshold between each node in the adjacency graph;

[0036] Starting from any node, traverse all neighboring nodes of the starting point in the adjacency graph, where a neighboring node is a node whose relative proximity value to the starting point does not exceed the second proximity threshold.

[0037] Analyze the visit flags of all neighboring nodes, record the unvisited neighboring nodes, and form connected components;

[0038] The initial task of the airport runway surface is determined by each connected component.

[0039] Furthermore, starting from any node, a depth-first search algorithm is used to search for all reachable nodes in the adjacency graph to form connected components, specifically represented as follows:

[0040]

[0041] Among them, RNS i Let ∪ be the connected component originating from node i. v∈未访问状态的节点 This indicates that all unvisited nodes are traversed. Starting from node v based on the edge set The depth-first search result, where v is the current node, and neighbors(v) is the set of edges containing node v. Let E be the set of all neighboring nodes, and E be the original set of edges. It is the set of edges in the original edge set whose relative proximity value is less than the second proximity threshold.

[0042] Furthermore, based on the airport pavement layout, the initial airport pavement tasks are reorganized using a real-time dynamic association strategy to generate airport pavement tasks. This process includes the following steps:

[0043] Based on the attribute sub-labels of the airport pavement target items in each airport pavement initial task, determine the attribute labels of each airport pavement initial task.

[0044] The attribute labels are transformed into semantic vectors, and the semantic similarity between the initial tasks of each airport pavement is analyzed.

[0045] Iterate through all airport pavement initialization tasks that are not selected, associate airport pavement initialization tasks that meet the semantic similarity condition, and change the task status of the associated airport pavement initialization tasks to the selected state to form an airport pavement initialization task cluster.

[0046] Until all airport pavement initial tasks are associated, multiple airport pavement initial task clusters are obtained;

[0047] By integrating the processing time of airport pavement matters and the layout of airport pavement, the initial tasks of each airport pavement in each initial task cluster are sorted to form an ordered initial task cluster of airport pavement, which is then identified as airport pavement tasks.

[0048] Furthermore, the attribute labels are transformed into semantic vectors, and the semantic similarity between the initial tasks of each airport pavement is analyzed. Specifically, this includes the following steps:

[0049] The attribute tags are cleaned, segmented, and keywords are extracted. Each keyword is then mapped to form a corresponding high-dimensional semantic vector.

[0050] The high-dimensional semantic vectors of the keywords in each attribute tag are combined to form the attribute tag semantic vector;

[0051] Cosine similarity is used to calculate the similarity between the semantic vectors of each attribute label.

[0052] Furthermore, the high-dimensional semantic vectors of the keywords in each attribute tag are combined to form the attribute tag semantic vector, specifically represented as follows:

[0053]

[0054] Among them, v i Let v be the semantic vector of the attribute tag corresponding to the i-th attribute tag. ij Let be the high-dimensional semantic vector of the j-th keyword in attribute tag i, n be the number of keywords in attribute tag i, and ⊕ be the vector concatenation.

[0055] Furthermore, by integrating airport pavement processing times and airport pavement layout, the initial tasks within each airport pavement initial task cluster are sorted to form an ordered airport pavement initial task cluster. This process includes the following steps:

[0056] Analyze the total processing time and center location of the initial task area for each initial task in the airport pavement initial task cluster.

[0057] Based on the airport pavement layout, the reference orientation is determined. Combined with the center position of each initial task area, the regional distribution and azimuth of the initial tasks on each airport pavement are given, forming regional weights.

[0058] By integrating the total processing time and regional weight of the initial tasks for each airport pavement, the ranking value of the initial tasks for each airport pavement is analyzed.

[0059] Based on the sorting values ​​of the initial tasks for each airport pavement, an ordered cluster of initial tasks for the airport pavement is formed.

[0060] Furthermore, the total processing time for the initial tasks of each airport pavement is specifically expressed as follows:

[0061]

[0062] Among them, T total The total processing time for the initial airport pavement task is t. k,end Let t be the end time of the event in the k-th airport pavement target event. k,start Let K be the start time of the event in the k-th airport pavement target event, and K be the number of airport pavement target events in the initial airport pavement task.

[0063] Furthermore, based on the airport pavement layout, the reference orientation is determined. Combined with the center position of each initial mission area, the regional distribution and azimuth of the initial missions on each airport pavement are given, forming regional weights. The specific steps include the following:

[0064] Use any specific point in the airport pavement layout as a reference point to determine the reference orientation;

[0065] Using the arctangent function, the azimuth angle of the initial mission on each airport pavement is calculated based on the coordinate difference between the center position of each initial mission area and the reference point.

[0066] Based on the degree of deviation between the azimuth of the initial mission for each airport pavement and the reference direction, and combined with a preset weighting function, the azimuth is transformed into the regional weight of the initial mission for the airport pavement.

[0067] Furthermore, using the arctangent function, the azimuth angle of the initial mission on each airport pavement is calculated based on the coordinate difference between the center position of each initial mission area and the reference point. Specifically, this is expressed as follows:

[0068]

[0069] Where, θ i Let x be the azimuth angle of the initial mission on the i-th airport pavement, and arctan() be the arctangent function. i Let y be the x-coordinate of the center position of the i-th initial task region. i Let x be the ordinate of the center position of the i-th initial task region. base Let y be the x-coordinate of the reference point. base The ordinate of the reference point.

[0070] Secondly, the present invention also provides an airport pavement task generation apparatus based on dynamic association, employing an airport pavement task generation method based on dynamic association as described above, comprising:

[0071] The event generation unit is used to provide airport pavement event information based on the airport pavement layout; in response to the target time, it determines multiple airport pavement target events based on the event time matching with the airport pavement events.

[0072] The item analysis unit is used to classify airport pavement target items and provide item tags based on the item attributes of each airport pavement target item, taking into account the airport pavement layout;

[0073] The task generation unit is used to link target items on the airport pavement by combining the item combination strategy and item tags, and to give the initial tasks for the airport pavement; through the real-time dynamic association strategy, the initial tasks for the airport pavement are reorganized to generate airport pavement tasks.

[0074] The present invention provides a method and apparatus for generating airport pavement tasks based on dynamic association, which has at least the following beneficial effects:

[0075] (1) By obtaining the airport pavement layout, the airport pavement item information is obtained and multiple airport pavement target items matching the target time are identified. The items are classified and labeled according to their item attributes. By combining the item combination strategy and the real-time dynamic association strategy, the airport pavement target items are linked and reorganized to generate airport pavement tasks. This enables multiple fragmented and discrete airport pavement items to be analyzed and combined multiple times to finally integrate them into airport pavement tasks. This allows multiple independent airport pavement items to be bundled and executed. When executing airport pavement tasks, the information corresponding to multiple related airport pavement items included in the airport pavement tasks can be uniformly filled in and reported to avoid information duplication and realize the reasonable arrangement of airport pavement cleaning, inspection and other tasks.

[0076] (2) Through the analysis and judgment of attribute sub-labels and relative neighbor values, the target items of the airport pavement were initially divided, and multiple initial tasks of the airport pavement were obtained. At the same time, in order to avoid many problems caused by the discrete release of tasks, the initial tasks of the airport pavement were further reorganized and / or associated to minimize the number of tasks released, improve the efficiency of task release, and reduce the probability of repeated task release. Attached Figure Description

[0077] Figure 1 A flowchart illustrating an airport pavement task generation method based on dynamic association, provided in an embodiment of the present invention;

[0078] Figure 2 A flowchart for providing item labels is provided as an embodiment of the present invention;

[0079] Figure 3 A flowchart of the initial airport pavement task is provided for embodiments of the present invention;

[0080] Figure 4 A schematic diagram of an adjacency graph provided in an embodiment of the present invention;

[0081] Figure 5 A flowchart for generating airport pavement tasks provided in an embodiment of the present invention;

[0082] Figure 6 A flowchart for determining the initial task cluster of an airport pavement according to an embodiment of the present invention;

[0083] Figure 7 A flowchart for determining region weights provided in an embodiment of the present invention;

[0084] Figure 8 This is a structural block diagram of an airport pavement task generation device based on dynamic association, provided in an embodiment of the present invention.

[0085] Among them, 201 is the item generation unit; 202 is the item analysis unit; and 203 is the task generation unit. Detailed Implementation

[0086] To better understand the above technical solutions, a detailed description of the solutions will be provided below in conjunction with the accompanying drawings and specific embodiments. Obviously, the described embodiments are merely some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0087] The terminology used in the embodiments of this invention is for the purpose of describing particular embodiments only and is not intended to limit the invention. The singular forms “a,” “the,” and “the” as used in the embodiments of this invention and the appended claims are also intended to include the plural forms, and “multiple” generally includes at least two unless the context clearly indicates otherwise.

[0088] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that an article or device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such an article or device. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the article or device that includes said element.

[0089] like Figure 1 As shown in the figure, this embodiment of the invention provides a method for generating airport pavement tasks based on dynamic association, and the specific steps are as follows:

[0090] S101: Based on the airport pavement layout, provide information on airport pavement matters.

[0091] Specifically, airport pavement layout includes the arrangement information of airport pavement buildings, and airport pavement event information includes event attributes and event time.

[0092] Understandably, based on the layout of various pavement facilities within an airport (i.e., airport pavement layout), comprehensive inspection or cleaning plans covering future days, weeks, months, and even years can be formulated. Each area's inspection or cleaning plan for each time period can be considered a single airport pavement event record. Each record includes at least the event time and event attributes. The event time indicates the time period during which the corresponding inspection or cleaning plan will occur, and the event attributes include the type of plan, which can include tasks such as inspection, cleaning, pavement repair, runway adhesive removal, and pavement maintenance. When formulating plans, users can flexibly specify the corresponding execution area, execution time, and execution cycle for each plan, and assign responsible execution teams to ensure that all pavement areas within the airport are included in the plan.

[0093] S102: In response to the target time, identify multiple airport pavement target events based on the event time matching with airport pavement events.

[0094] Understandably, the target time is the time when airport pavement tasks need to be issued. It can be the current time or any time when task issuance is required, specified by the user according to the actual situation. Based on the foregoing description, each airport pavement item includes planned items for different areas of the airport pavement at different times. Therefore, at the same time point corresponding to the target time, there are multiple airport pavement items. By comparing and matching the target time with the item times in each airport pavement item, multiple airport pavement target items corresponding to the target time are identified from all airport pavement items.

[0095] S103: Based on the airport pavement layout and the attributes of each airport pavement target item, classify the airport pavement target items and assign item labels.

[0096] The item label includes attribute sub-labels and relative adjacent value sub-labels.

[0097] Furthermore, based on the airport pavement layout and the attributes of each target item on the airport pavement, the target items are categorized and tagged. Figure 2 Specifically, it includes the following steps:

[0098] Based on the attributes of each airport pavement target item, airport pavement target items are classified and attribute sub-labels are given;

[0099] By using the airport pavement layout, the locations of target items on the airport pavement are marked, forming location labels for each target item on the airport pavement.

[0100] Based on the location labels of each airport pavement target item, the corresponding relative proximity value sub-label is given;

[0101] The attribute sub-tags and relatively adjacent value sub-tags of target items on the pavement of each airport are integrated to form item tags.

[0102] In one specific implementation, the attributes of airport pavement target items are first classified and determined. Airport pavement target items with the same attribute are grouped together, and the same attribute sub-label is set for the corresponding class. In a specific example, if two airport pavement target items both have the attribute "sweeping," then these two airport pavement target items, along with other airport pavement target items with the attribute "sweeping," are grouped together, and an attribute sub-label is configured for this class. For example, the attribute sub-label corresponding to "sweeping" is "sweep01." It is understood that different values ​​of the attribute correspond to different attribute sub-labels, and the actual values ​​of the attribute sub-labels can be selected according to actual needs and are not limited thereto.

[0103] Then, based on the airport pavement layout, the corresponding areas for the target items on the airport pavement are given and the locations are marked according to the corresponding areas. Location labels are set for each target item on the airport pavement. In this example, the location labels include the location coordinates of each point on the boundary of the corresponding area and the coordinates of the center point.

[0104] After obtaining the location labels of each airport pavement target item, the distance between the corresponding areas of each airport pavement target item is analyzed based on the location labels, and the relative proximity value sub-label of each pair of airport pavement target items is given.

[0105] Furthermore, based on the location labels of each airport pavement target item, corresponding relative proximity value sub-labels are given, specifically including the following steps:

[0106] Based on the location tags of target items on the pavement of each airport, a proximity relationship analysis is performed on the target items on the pavement of each airport;

[0107] Based on the first proximity threshold, the relative proximity values ​​between target items on each airport pavement are given, forming relative proximity value sub-labels.

[0108] Specifically, based on the first proximity threshold, the relative proximity values ​​between target items on each airport pavement are given, forming relative proximity value sub-labels, specifically represented as follows:

[0109]

[0110] Among them, RPV ij Let m be the sub-label of the relative proximity value between the i-th airport pavement target item and the j-th airport pavement target item. dij PT1 is the relative proximity value between the i-th airport pavement target item and the j-th airport pavement target item, and PT1 is the first proximity threshold. The relative proximity value is the distance between the center point coordinates of the location labels in the two airport pavement target items.

[0111] In one specific implementation, after obtaining the location labels of each airport pavement target item, the distance between corresponding center point coordinates is calculated based on the center point coordinates in the location labels to obtain the relative proximity value between each airport pavement target item. The relative proximity value is then judged in conjunction with a first proximity threshold. When the relative proximity value is less than the first proximity threshold, it indicates that the corresponding areas of the corresponding airport pavement target items are relatively close, and the relative proximity value is used as the corresponding relative proximity value sub-label. Conversely, if the relative proximity value is greater than or equal to the first proximity threshold, it indicates that the corresponding areas of the corresponding airport pavement target items are relatively far apart, and the relative proximity value sub-label is assigned a value of zero, meaning that airport pavement target items with a relative proximity value sub-label of zero will not be grouped together for execution. Finally, the attribute sub-labels and relative proximity value sub-labels of each airport pavement target item are combined to obtain the item label for each airport pavement target item.

[0112] In other implementations, assigning a value of zero to a relative neighbor sub-label can be considered as the absence of a relative neighbor sub-label.

[0113] S104: Combining the item combination strategy and item tags, link the target items of the airport pavement and give the initial tasks of the airport pavement.

[0114] Furthermore, by combining the task combination strategy and task tags, the target tasks for airport pavement are linked to provide the initial tasks for airport pavement, as referenced. Figure 3 Specifically, it includes the following steps:

[0115] Based on the attribute sub-tags, each airport pavement target item is grouped and multiple attribute subsets are given;

[0116] In each attribute subset, an adjacency graph is constructed based on the nearest value sub-labels of each airport pavement target item, with each airport pavement target item as a node in the adjacency graph;

[0117] Set access flags for nodes in the adjacency graph and initialize the state of each node to an unvisited state;

[0118] Starting from any node, a depth-first search algorithm is used to search for all reachable nodes in the adjacency graph to form connected components and determine the initial task of the airport pavement.

[0119] Mark the origin and reachable nodes as visited;

[0120] Starting from the nodes that are not visited, repeat the above search process to form a new airport runway initial task until all nodes are included in the airport runway initial task.

[0121] Summarize all the initial tasks for airport pavement to obtain the initial tasks for airport pavement.

[0122] In one specific implementation, each airport pavement target item is grouped according to its attribute sub-labels. Airport pavement target items with the same attribute sub-label values ​​are grouped together, and each group is an attribute subset, resulting in multiple attribute subsets. An adjacency graph is constructed for each airport pavement target item within each attribute subset. Starting from a given airport pavement target item, connections are established with other airport pavement target items based on the values ​​of their nearest neighbor sub-labels. The nodes in the adjacency graph represent the airport pavement target items, and the distance between nodes is the value of their nearest neighbor sub-labels. It is understood that when the nearest neighbor sub-label value is zero, no connection is established between airport pavement target items. After the adjacency graph is constructed, an access flag is set for each node in the adjacency graph. The access flag indicates the access status of each airport pavement target item, and the state of each node is initialized to an unaccessed state, indicating that none of the airport pavement target items have been accessed. An example diagram of the adjacency graph is shown below. Figure 4 As shown, it can be understood that nodes 1, 2, 3, 4, 5, and 7 belong to the same attribute subset, and nodes 6 and n belong to the same attribute subset. The attribute sub-labels of each node in each attribute subset are the same.

[0123] After initializing the adjacency graph, starting from any node, a depth-first search algorithm is used to search for all reachable nodes in the adjacency graph, forming connected components. These connected components then determine the initial airport pavement tasks. Each connected component includes multiple airport pavement target items that are connected in the adjacency graph. Each airport pavement target item contained in a connected component is identified as an initial airport pavement task. Simultaneously, the visit flags of the starting point and reachable nodes in the connected components are set to "visited," indicating that the corresponding nodes have already been assigned tasks, preventing duplicate assignments of the corresponding airport pavement target items. The search process is repeated, starting from an unvisited node, to form new initial airport pavement tasks. This process continues until all nodes are included in the initial airport pavement tasks. Finally, all initial airport pavement tasks are summarized to obtain the final initial airport pavement task.

[0124] Specifically, starting from any node, a depth-first search algorithm is used to search for all reachable nodes in the adjacency graph to form connected components, thus determining the initial task of the airport pavement. This includes the following steps:

[0125] Determine the second proximity threshold between each node in the adjacency graph;

[0126] Starting from any node, traverse all neighboring nodes of the starting point in the adjacency graph, where a neighboring node is a node whose relative proximity value to the starting point does not exceed the second proximity threshold.

[0127] Analyze the visit flags of all neighboring nodes, record the unvisited neighboring nodes, and form connected components;

[0128] The initial task of the airport runway surface is determined by each connected component.

[0129] Understandably, the nodes in the adjacency graph represent various airport pavement target items. Since the location of the corresponding area for each airport pavement target item within the airport differs, different second proximity thresholds can be set based on the actual situation, or the same second proximity threshold can be set; there is no limitation on this. After determining the second proximity threshold, starting from any unvisited node in the adjacency graph, based on the connection between the starting point and other nodes, the adjacency graph is traversed. It is determined whether the relative proximity value between each node and the starting point reaches the corresponding second proximity threshold. If it does, it means that the current node is too far from the starting point and cannot form a connected component corresponding to the current starting point. If it does not, it means that the current node is too close to the starting point and can be added to the connected component corresponding to the current starting point. This process continues until the adjacency graph has been traversed and the connected component corresponding to the current starting point is obtained. The above process is repeated, and during the node traversal, only unvisited nodes are checked for relative proximity values ​​until all nodes in the adjacency graph are updated to the visited state, resulting in multiple connected components. The nodes in each connected component form an initial task for an airport pavement sub-task.

[0130] Furthermore, starting from any node, a depth-first search algorithm is used to search for all reachable nodes in the adjacency graph to form connected components, specifically represented as follows:

[0131]

[0132] Among them, RNS i Let ∪ be the connected component originating from node i. v∈未访问状态的节点 This indicates that all unvisited nodes are traversed. Starting from node v based on the edge set The depth-first search result, where v is the current node, and neighbors(v) is the set of edges containing node v. Let E be the set of all neighboring nodes, and E be the original set of edges. It is the set of edges in the original edge set whose relative proximity value is less than the second proximity threshold.

[0133] S105: By using a real-time dynamic association strategy, the initial airport pavement task is reorganized to generate an airport pavement task.

[0134] Furthermore, based on the airport pavement layout, and through a real-time dynamic association strategy, the initial airport pavement tasks are reorganized to generate airport pavement tasks, referring to... Figure 5 Specifically, it includes the following steps:

[0135] Based on the attribute sub-labels of the airport pavement target items in each airport pavement initial task, determine the attribute labels of each airport pavement initial task.

[0136] The attribute labels are transformed into semantic vectors, and the semantic similarity between the initial tasks of each airport pavement is analyzed.

[0137] Iterate through all airport pavement initialization tasks that are not selected, associate airport pavement initialization tasks that meet the semantic similarity condition, and change the task status of the associated airport pavement initialization tasks to the selected state to form an airport pavement initialization task cluster.

[0138] Until all airport pavement initial tasks are associated, multiple airport pavement initial task clusters are obtained;

[0139] By integrating the processing time of airport pavement matters and the layout of airport pavement, the initial tasks of each airport pavement in each initial task cluster are sorted to form an ordered initial task cluster of airport pavement, which is then identified as airport pavement tasks.

[0140] Understandably, through the analysis of attribute sub-tags and relative proximity values, the target items for airport pavement were initially divided, resulting in multiple initial airport pavement tasks. To avoid numerous problems caused by the discrete deployment of tasks, the initial airport pavement tasks were further reorganized and / or associated to minimize the number of task deployments, improve the efficiency of task deployment, and reduce the probability of duplicate task deployments.

[0141] Among them, the airport pavement event processing time is the time spent from the start to the end of each airport pavement event. It can be understood as a higher-level concept of the total processing time of the initial tasks of each airport pavement.

[0142] In one specific implementation, the attribute sub-tags of airport pavement target items in each airport pavement initial task are concatenated and fused to obtain attribute tags for each airport pavement initial task. Then, semantic vector transformation is performed on the attribute tags of each airport pavement initial task, and the semantic similarity between the airport pavement initial tasks is analyzed by calculating the similarity of the attribute tags. When the semantic similarity between two airport pavement initial tasks reaches a preset semantic similarity threshold, the two airport pavement initial tasks are associated, and the task status of the airport pavement initial tasks is set to the selected state. Based on any airport pavement initial task or an associated airport pavement initial task, other airport pavement initial tasks with an unselected task status are traversed. Airport pavement initial tasks that meet the semantic similarity condition are associated, and the task status of the associated airport pavement initial tasks is changed to the selected state. Multiple airport pavement initial tasks with semantic similarity reaching the preset semantic similarity threshold are combined to form an airport pavement initial task cluster. This process continues until all airport pavement initial tasks are associated, resulting in multiple airport pavement initial task clusters.

[0143] After obtaining multiple airport pavement initial task clusters, each airport pavement initial task in each airport pavement initial task cluster is sorted in ascending order according to the time of the task. If the time of the task is the same, it is sorted in ascending order according to the relative proximity value. Finally, all airport pavement initial tasks are sorted to form an ordered airport pavement initial task cluster, thus obtaining the airport pavement tasks.

[0144] Furthermore, the attribute labels are transformed into semantic vectors, and the semantic similarity between the initial tasks of each airport pavement is analyzed. Specifically, this includes the following steps:

[0145] The attribute tags are cleaned, segmented, and keywords are extracted. Each keyword is then mapped to form a corresponding high-dimensional semantic vector.

[0146] The high-dimensional semantic vectors of each keyword in each attribute tag are combined to form the attribute tag semantic vector;

[0147] Cosine similarity is used to calculate the similarity between the semantic vectors of each attribute label.

[0148] Furthermore, the high-dimensional semantic vectors of the keywords in each attribute tag are combined to form the attribute tag semantic vector, specifically represented as follows:

[0149]

[0150] Among them, v i Let v be the semantic vector of the attribute tag corresponding to the i-th attribute tag. ijLet be the high-dimensional semantic vector of the j-th keyword in attribute tag i, n be the number of keywords in attribute tag i, and ⊕ be the vector concatenation.

[0151] In one specific implementation, each attribute label is first cleaned to remove meaningless characters, punctuation marks, and stop words. Then, word segmentation is performed, and keywords are extracted from the segmentation results. Common stop words (such as "and") are typically removed, and the text is divided into words or phrases. In a specific example, attribute label 1 is "airport pavement cleaning and inspection," and attribute label 2 is "airport pavement maintenance and repair." After cleaning and segmentation, the resulting attributes are: attribute label 1 ["airport", "pavement", "cleaning", "and", "inspection"], and attribute label 2 ["airport", "pavement", "maintenance", "and", "repair"]. Keywords are extracted from the segmentation results to obtain the keywords for attribute label 1 ["airport", "pavement", "cleaning", "inspection"], and the keywords for attribute label 2 ["airport", "pavement", "maintenance", "repair"]. Then, the keyword set is mapped to a high-dimensional semantic vector. A pre-trained word embedding model (such as Word2Vec, GloVe, or BERT) can be used to map each keyword to a high-dimensional semantic vector. For example, using a word embedding model, the vectors for each keyword are obtained as follows: airport [0.1, 0.2, 0.3], pavement [0.4, 0.5, 0.6], cleaning [0.7, 0.8, 0.9], inspection [0.1, 0.2, 0.3], maintenance [0.4, 0.5, 0.6], repair [0.7, 0.8, 0.9]. Then, the high-dimensional semantic vectors of the keywords for each attribute tag are combined to form the attribute tag semantic vector. Finally, cosine similarity is used to calculate the similarity between the semantic vectors of each attribute tag.

[0152] Furthermore, by integrating airport pavement processing times and pavement layout, the initial tasks within each airport pavement initial task cluster are sorted to form an ordered airport pavement initial task cluster, referring to... Figure 6 Specifically, it includes the following steps:

[0153] Analyze the total processing time and center location of the initial task area for each initial task in the airport pavement initial task cluster.

[0154] Based on the airport pavement layout, the reference orientation is determined. Combined with the center position of each initial task area, the regional distribution and azimuth of the initial tasks on each airport pavement are given, forming regional weights.

[0155] By integrating the total processing time and regional weight of the initial tasks for each airport pavement, the ranking value of the initial tasks for each airport pavement is analyzed.

[0156] Based on the sorting values ​​of the initial tasks for each airport pavement, an ordered cluster of initial tasks for the airport pavement is formed.

[0157] Furthermore, the total processing time for the initial tasks of each airport pavement is specifically expressed as follows:

[0158]

[0159] Among them, T total The total processing time for the initial airport pavement task is t. k,end Let t be the end time of the event in the k-th airport pavement target event. k,start Let K be the start time of the event in the k-th airport pavement target event, and K be the number of airport pavement target events in the initial airport pavement task.

[0160] The initial task area refers to the sum of the areas corresponding to each airport pavement target item in the initial task of each airport pavement. The center position of the initial task area refers to the center coordinates of the initial task area. In a specific example, the average of the maximum and minimum x-coordinates in the initial task area can be selected as the x-coordinate of the center coordinate, and the average of the maximum and minimum y-coordinates in the initial task area can be selected as the y-coordinate of the center coordinate. In other embodiments, other methods can be used to determine the center position of the initial task area, and there is no limitation on this.

[0161] In one specific implementation, after obtaining the total processing time and the center position of the initial task area for each airport pavement, a reference azimuth is determined based on the airport pavement layout. Combined with the center positions of each initial task area, the regional distribution and azimuth of the initial tasks for each airport pavement are given, forming regional weights. The reference azimuth is set according to the actual airport pavement layout and is not limited thereto. In a specific example, the reference azimuth is the origin of the two-dimensional coordinate system where the center position of the initial task area is located. The azimuth of the initial tasks on the airport pavement can be obtained using the arctangent function based on the reference azimuth and the center position of the initial task area. Based on the correspondence between the azimuth and the regional weights, the regional weights of the initial tasks on each airport pavement are given. Finally, based on the total processing time and regional weights of each initial task on the airport pavement, the ranking value of each initial task is analyzed and ranked accordingly, forming an ordered cluster of initial tasks on the airport pavement. In a specific example, the initial tasks are sorted in descending order based on their regional weights; if the regional weights are the same, they are sorted in ascending order based on their total processing time, thus completing the ranking of the initial tasks on the airport pavement. In another specific example, the total processing time and regional weights are weighted and summed according to a preset sorting coefficient to obtain the sorting value of each airport pavement initial task. The tasks are then sorted according to the sorting value to obtain the corresponding airport pavement initial task cluster.

[0162] Furthermore, based on the airport pavement layout, a reference orientation is determined. Combined with the center location of each initial mission area, the regional distribution and azimuth of the initial missions on each airport pavement are given, forming regional weights, which are then used as a reference. Figure 7 Specifically, it includes the following steps:

[0163] Use any specific point in the airport pavement layout as a reference point to determine the reference orientation;

[0164] Using the arctangent function, the azimuth angle of the initial mission on each airport pavement is calculated based on the coordinate difference between the center position of each initial mission area and the reference point.

[0165] Based on the degree of deviation between the azimuth of the initial mission for each airport pavement and the reference direction, and combined with a preset weighting function, the azimuth is transformed into the regional weight of the initial mission for the airport pavement.

[0166] Furthermore, using the arctangent function, the azimuth angle of the initial mission on each airport pavement is calculated based on the coordinate difference between the center position of each initial mission area and the reference point. Specifically, this is expressed as follows:

[0167]

[0168] Where, θ i Let x be the azimuth angle of the initial mission on the i-th airport pavement, and arctan() be the arctangent function. i Let y be the x-coordinate of the center position of the i-th initial task region. i Let x be the ordinate of the center position of the i-th initial task region. base Let y be the x-coordinate of the reference point. base The ordinate of the reference point.

[0169] Regional weights, specifically, are expressed as follows:

[0170]

[0171] Among them, W i Let θ be the region weight for the initial task of the i-th airport pavement, a and m be the transformation coefficients, and θ be the region weight. i Let |·| be the azimuth angle of the initial task for the i-th airport pavement, where |·| is the absolute value.

[0172] To address the problems of repetitive form filling, low operational efficiency, and inconsistent basic data caused by the discrete nature of task deployment, this invention provides a method and apparatus for generating airport pavement tasks based on dynamic association. The method includes: providing airport pavement item information based on the airport pavement layout; determining multiple airport pavement target items based on time matching with the target time, in response to a target time; classifying the target items according to their attributes and the airport pavement layout, and assigning item tags; linking the target items using item combination strategies and item tags to generate initial airport pavement tasks; and reorganizing the initial tasks using a real-time dynamic association strategy to generate airport pavement tasks. By analyzing and combining multiple fragmented and discrete airport pavement items, and finally integrating them into airport pavement tasks, multiple independent airport pavement items can be bundled for execution. When executing an airport pavement task, information on multiple related airport pavement items included in the task can be uniformly filled in and reported, avoiding information duplication and enabling the rational arrangement of tasks such as airport pavement cleaning and inspection.

[0173] Reference Figure 8 This invention provides an airport pavement task generation device based on dynamic association, comprising:

[0174] The event generation unit 201 is used to provide airport pavement event information based on the airport pavement layout; in response to the target time, it determines multiple airport pavement target events based on the event time matching with the airport pavement events.

[0175] The item analysis unit 202 is used to classify the target items of the airport pavement based on the airport pavement layout and the item attributes of each target item of the airport pavement, and give item tags.

[0176] The task generation unit 203 is used to link the target items of the airport pavement by combining the item combination strategy and item tags, and give the initial tasks of the airport pavement; through the real-time dynamic association strategy, the initial tasks of the airport pavement are reorganized to generate airport pavement tasks.

[0177] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working process of the described module can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0178] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention. Clearly, those skilled in the art can make various alterations and modifications to the invention without departing from its spirit and scope. Thus, if these modifications and variations of the invention fall within the scope of the claims and their equivalents, the invention is also intended to include these modifications and variations.

Claims

1. A method for generating airport pavement tasks based on dynamic association, characterized in that, Specifically, the steps include the following: Based on the airport pavement layout, provide information on airport pavement matters; In response to the target time, multiple airport pavement target items are identified based on the time matching of items with airport pavement items; Based on the airport pavement layout and the attributes of each airport pavement target item, the airport pavement target items are classified and given item labels. The item labels include attribute sub-labels and relative adjacent value sub-labels. Based on attribute sub-labels, each airport pavement target item is grouped, resulting in multiple attribute subsets. Within each attribute subset, an adjacency graph is constructed based on the nearest value sub-labels of each airport pavement target item, with each airport pavement target item as a node in the adjacency graph. Access markers are set for the nodes in the adjacency graph, and the state of each node is initialized to unvisited. Starting from any node, a depth-first search algorithm is used to search for all reachable nodes in the adjacency graph, forming connected components and determining the initial airport pavement task. The access markers of the starting point and reachable nodes are set to visited. The above search process is repeated, starting again from unvisited nodes, to form new initial airport pavement tasks until all nodes are included in the initial airport pavement task. All initial airport pavement tasks are summarized to obtain the initial airport pavement task. Based on the attribute sub-labels of airport pavement target items in each airport pavement initial task, the attribute labels of each airport pavement initial task are determined; each attribute label is transformed into a semantic vector, and the semantic similarity between each airport pavement initial task is analyzed; all airport pavement initial tasks in the unselected state are traversed, and airport pavement initial tasks that meet the semantic similarity condition are associated, and the task status of the associated airport pavement initial tasks is changed to the selected state, forming an airport pavement initial task cluster; until all airport pavement initial tasks are associated, multiple airport pavement initial task clusters are obtained. By integrating airport pavement processing time and airport pavement layout, the initial tasks of each airport pavement initial task cluster are sorted to form an ordered airport pavement initial task cluster. Specifically, this includes: analyzing the total processing time and the center position of the initial task area for each initial task in the airport pavement initial task cluster; determining the reference orientation based on the airport pavement layout, and combining the center position of each initial task area to give the regional distribution and azimuth of each initial task, forming regional weights; integrating the total processing time and regional weights of each initial task to analyze the sorting value of each initial task; and forming an ordered airport pavement initial task cluster based on the sorting value of each initial task, which is then identified as an airport pavement task. The airport pavement processing time is the time spent from start to finish for each item in the airport pavement event.

2. The airport pavement task generation method based on dynamic association as described in claim 1, characterized in that, Based on the airport pavement layout and the attribute of each target item on the airport pavement, the target items on the airport pavement are classified and labeled. The specific steps include the following: Based on the attributes of each airport pavement target item, airport pavement target items are classified and attribute sub-labels are given; By using the airport pavement layout, the locations of target items on the airport pavement are marked, forming location labels for each target item on the airport pavement. Based on the location labels of each airport pavement target item, the corresponding relative proximity value sub-label is given; The attribute sub-tags and relatively adjacent value sub-tags of target items on the pavement of each airport are integrated to form item tags.

3. The airport pavement task generation method based on dynamic association as described in claim 2, characterized in that, Based on the location labels of each airport pavement target item, the corresponding relative proximity value sub-label is given, which includes the following steps: Based on the location tags of target items on the pavement of each airport, a proximity relationship analysis is performed on the target items on the pavement of each airport; Based on the first proximity threshold, the relative proximity values ​​between target items on each airport pavement are given, forming relative proximity value sub-labels.

4. The airport pavement task generation method based on dynamic association as described in claim 1, characterized in that, Starting from any node, a depth-first search algorithm is used to search for all reachable nodes in the adjacency graph to form connected components, thus determining the initial task of the airport runway surface. This includes the following steps: Determine the second proximity threshold between each node in the adjacency graph; Starting from any node, traverse all neighboring nodes of the starting point in the adjacency graph, where a neighboring node is a node whose relative proximity value to the starting point does not exceed the second proximity threshold. Analyze the visit flags of all neighboring nodes, record the unvisited neighboring nodes, and form connected components; The initial task of the airport runway surface is determined by each connected component.

5. The airport pavement task generation method based on dynamic association as described in claim 1, characterized in that, The various attribute labels are transformed into semantic vectors, and the semantic similarity between the initial tasks of each airport pavement is analyzed. The specific steps include the following: The attribute tags are cleaned, segmented, and keywords are extracted. Each keyword is then mapped to form a corresponding high-dimensional semantic vector. The high-dimensional semantic vectors of the keywords in each attribute tag are combined to form the attribute tag semantic vector; Cosine similarity is used to calculate the similarity between the semantic vectors of each attribute label.

6. The airport pavement task generation method based on dynamic association as described in claim 1, characterized in that, Based on the airport pavement layout, the baseline orientation is determined. Combined with the center position of each initial mission area, the regional distribution and azimuth of the initial missions on each airport pavement are given, forming regional weights. The specific steps include the following: Use any specific point in the airport pavement layout as a reference point to determine the reference orientation; Using the arctangent function, the azimuth angle of the initial mission on each airport pavement is calculated based on the coordinate difference between the center position of each initial mission area and the reference point. Based on the degree of deviation between the azimuth of the initial mission for each airport pavement and the reference direction, and combined with a preset weighting function, the azimuth is transformed into the regional weight of the initial mission for the airport pavement.

7. An airport pavement task generation device based on dynamic association, characterized in that, The airport pavement task generation method based on dynamic association as described in any one of claims 1-6 specifically includes: The event generation unit is used to provide airport pavement event information based on the airport pavement layout; in response to the target time, it determines multiple airport pavement target events based on the event time matching with the airport pavement events. The event analysis unit is used to classify airport pavement target events based on the airport pavement layout and the event attributes of each airport pavement target event, and to give event labels. The event labels include attribute sub-labels and relative proximity value sub-labels. The task generation unit is used to group each airport pavement target item according to attribute sub-labels, providing multiple attribute subsets; within each attribute subset, an adjacency graph is constructed based on the nearest value sub-labels of each airport pavement target item, with each airport pavement target item as a node in the adjacency graph; access flags are set for the nodes in the adjacency graph, and the state of each node is initialized to unvisited; starting from any node, a depth-first search algorithm is used to search for all reachable nodes in the adjacency graph to form connected components, determining the airport pavement sub-initial task; the access flags of the starting point and reachable nodes are set to visited state; the above search process is repeated again, starting from an unvisited node, to form a new airport pavement sub-initial task, until all nodes are included in the airport pavement initial task; all airport pavement initial tasks are summarized to obtain the airport pavement initial task. Based on the attribute sub-labels of airport pavement target items in each airport pavement initial task, the attribute labels of each airport pavement initial task are determined; each attribute label is transformed into a semantic vector, and the semantic similarity between each airport pavement initial task is analyzed; all airport pavement initial tasks in the unselected state are traversed, and airport pavement initial tasks that meet the semantic similarity condition are associated, and the task status of the associated airport pavement initial tasks is changed to the selected state, forming an airport pavement initial task cluster; until all airport pavement initial tasks are associated, multiple airport pavement initial task clusters are obtained. By integrating airport pavement processing time and airport pavement layout, the initial tasks of each airport pavement initial task cluster are sorted to form an ordered airport pavement initial task cluster. Specifically, this includes: analyzing the total processing time and the center position of the initial task area for each initial task in the airport pavement initial task cluster; determining the reference orientation based on the airport pavement layout, and combining the center position of each initial task area to give the regional distribution and azimuth of each initial task, forming regional weights; integrating the total processing time and regional weights of each initial task to analyze the sorting value of each initial task; and forming an ordered airport pavement initial task cluster based on the sorting value of each initial task, which is then identified as an airport pavement task. The airport pavement processing time is the time spent from start to finish for each item in the airport pavement event.

Images