Scheduling method and system matched with aviation luggage change information

Abstract

The invention discloses a scheduling method and system matched with air luggage change information, and relates to the technical field of air transportation intelligent scheduling, and the method comprises the steps: collecting luggage change information, passenger travel data and airport equipment state data in an air transportation scene in real time, and outputting an original data stream; an airport operation digital twinborn body is constructed in the element universe, a collaborative transfer path is imported into a digital twinborn environment for space-time preview simulation, luggage flow conflict points are detected in real time, and a physical constraint conflict area coordinate set is marked; inverse solution is carried out on conflict events and traceability strategy defect features in rehearsal simulation through an inverse reinforcement learning algorithm, a reward function space is reconstructed, and gradient compensation parameters are output; and fusing the physical constraint conflict region coordinate set and the gradient compensation parameters, generating a three-dimensional space scheduling instruction set, and executing instruction issuing. Based on the space-time rehearsal simulation method, the security boundary of the luggage transfer path is accurately identified, and the physical constraint overrun risk is effectively reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent scheduling of air transportation, and in particular to a scheduling method and system for supporting airline baggage change information. Background Art

[0002] The development of airline baggage management systems has evolved from basic automation to intelligent systems. Graph neural network technology is being applied to analyze passenger travel data and build social relationship models to support baggage dispatch decisions. Furthermore, the integration of digital twin technology into airport operations is becoming increasingly mature. By loading building information models and IoT data streams, dynamic simulation of the physical environment and conflict detection are achieved. The introduction of inverse reinforcement learning algorithms further enhances policy optimization capabilities, improving the accuracy of reward function reconstruction by inversely solving sample sets of state-action pairs.

[0003] However, existing technologies still suffer from core flaws when processing baggage change information: a lack of an efficient dynamic weighting mechanism for group baggage priority and real-time conflict simulation and optimization capabilities. This prevents accurate integration of passenger itinerary dependencies during baggage transfers and makes it difficult to accurately predict physical constraint conflict points through spatiotemporal rehearsals. This leads to increased probability of path intersection collisions, uncontrolled equipment response delays, and abnormal volume overload rates on sorting carousels, significantly increasing the risk of baggage delays and inefficient resource utilization. Summary of the Invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a scheduling method for supporting airline baggage change information to solve the problem of insufficient dynamic weighting of group baggage priority and real-time conflict simulation optimization in airline baggage change scheduling.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In the first aspect, the present invention provides a scheduling method for supporting airline baggage change information, which includes real-time collection of baggage change information, passenger itinerary data and airport equipment status data in air transportation scenarios, and output of raw data streams; inputting the raw data streams into a pre-trained graph neural network, and constructing a social relationship topology map by analyzing the co-occurring flights, seat proximity and ticket purchase time interval characteristics in passenger booking records; triggering a group baggage priority weighting mechanism based on the social relationship topology map, and generating a collaborative transfer path for passenger baggage with itinerary associations; constructing an airport operation digital twin in the metaverse, importing the collaborative transfer path into the digital twin environment for spatiotemporal rehearsal simulation, detecting baggage flow conflict points in real time and marking the coordinate set of the physical constraint conflict area; inversely solving the conflict events in the rehearsal simulation through an inverse reinforcement learning algorithm, tracing the strategy defect characteristics and reconstructing the reward function space, and outputting gradient compensation parameters; fusing the physical constraint conflict area coordinate set and the gradient compensation parameters to generate a three-dimensional space scheduling instruction set and execute the instruction issuance.

[0008] As a preferred solution of the method for dispatching supporting airline baggage change information of the present invention, the passenger itinerary data includes passenger booking records.

[0009] As a preferred solution of the method for scheduling air baggage change information according to the present invention, the specific steps of constructing a social relationship topology graph are as follows:

[0010] Parse passenger booking records in the original data stream to extract co-occurring flight numbers, adjacent seat numbers, and purchase time differences;

[0011] Applying a negative exponential decay function to the ticket purchase time difference generates a time decay coefficient, and performing an inverse Euclidean distance calculation on adjacent seat numbers generates spatial proximity;

[0012] The co-occurring flight numbers are used as the initial weights of the graph edges, and the time decay coefficient and spatial proximity are integrated to generate a dynamic social weight matrix. The passenger node features are aggregated through a graph convolutional network to output a social relationship topology graph.

[0013] As a preferred solution of the method for scheduling airline baggage change information according to the present invention, the steps of generating a coordinated transfer route for passenger baggage associated with itineraries are as follows:

[0014] When the flight status associated with any passenger node in the social relationship topology graph changes, traverse the baggage transfer status of all passengers connected to the changed node and obtain the loading status identification code and current location coordinates;

[0015] If the baggage loading status identification code is "Unloaded", a flight binding instruction is generated to forcibly synchronize the current baggage to the changed target flight;

[0016] If the baggage loading status identification code is loaded, calculate the shortest transfer time from the current location coordinates of the corresponding baggage to the target flight boarding gate, and generate a virtual synchronization time window constraint;

[0017] According to the dynamic weight values ​​of the connecting edges in the social relationship topology graph, the transfer sequence adjustment coefficients are assigned to the passenger nodes constrained by the virtual synchronization time window, and the collaborative transfer path is output.

[0018] As a preferred solution of the scheduling method for supporting airline baggage change information of the present invention, the specific steps of detecting baggage flow conflict points in real time and marking the coordinate set of the physical constraint conflict area are as follows:

[0019] Load the airport building information model and real-time device IoT data streams into the Metaverse environment to build a dynamic digital twin;

[0020] Import the collaborative transfer path into the dynamic digital twin and perform spatiotemporal simulation of baggage transfer, including sorting carousel volume overload rate, path intersection collision probability, and equipment response delay.

[0021] Activate the physics engine to simulate the effects of baggage stacking and dumping and conveyor belt vibration and deviation, and mark the coordinate sets of the physical constraint conflict areas that exceed the safety threshold.

[0022] As a preferred solution of the scheduling method for supporting air baggage change information of the present invention, the output gradient compensation parameter is specifically performed as follows:

[0023] The coordinate set of the physical constraint conflict area is combined with the digital twin environment parameters to generate a state-action pair sample set as the input data of the inverse reinforcement learning algorithm;

[0024] The probability distribution of the state-action pair sample set is calculated through the maximum entropy probability model, and the reward function corresponding to the optimal strategy is deduced;

[0025] Analyze the weight distribution defects of the reward function, identify the device response delay weights and abnormal path crossing penalty coefficients that cause conflicts, and reconstruct the reward function space structure;

[0026] Device failure simulation noise and path random offsets are injected into the state-action pair sample set to generate adversarial perturbation samples, and the policy network model is trained to output gradient compensation parameters.

[0027] As a preferred solution of the scheduling method for supporting airline baggage change information of the present invention, the steps of generating a three-dimensional spatial scheduling instruction set and executing the instruction issuance are as follows:

[0028] Calculate the start-up time offset value of the sorting equipment motor based on the gradient compensation parameter and generate the motor control instruction;

[0029] Based on the coordinate set of the physical constraint conflict area, the A* algorithm is used to re-plan the ground crew's travel path and output the coordinate sequence of the path navigation points;

[0030] The motor control instructions are aligned with the path navigation point coordinate sequence according to the timestamp to generate a structured instruction data packet. The motor control instructions are sent to the baggage sorting controller through the message queue, and the path navigation point coordinate sequence is transmitted to the ground staff handheld terminal through the wireless protocol.

[0031] In the second aspect, the present invention provides a scheduling system for supporting airline baggage change information, including a data acquisition module, a graph construction module, a path generation module, a simulation preview module, a strategy optimization module and an instruction scheduling module; the data acquisition module is used to collect baggage change information, passenger itinerary data and airport equipment status data in air transportation scenarios in real time, and output the original data stream; the graph construction module is used to input the original data stream into a pre-trained graph neural network, and construct a social relationship topology map by analyzing the co-occurring flights, seat proximity and ticket purchase time interval characteristics in passenger booking records; the path generation module is used to trigger the social relationship topology map based on the social relationship topology map. A weighted mechanism for group baggage priority is implemented to generate collaborative transfer paths for passenger baggage with itinerary associations; the simulation rehearsal module is used to construct a digital twin of airport operations in the metaverse, import the collaborative transfer paths into the digital twin environment for spatiotemporal rehearsal simulation, detect baggage flow conflict points in real time, and mark the coordinate sets of the physical constraint conflict areas; the strategy optimization module is used to inversely resolve conflict events in the rehearsal simulation through an inverse reinforcement learning algorithm, trace the characteristics of strategy defects, reconstruct the reward function space, and output gradient compensation parameters; the instruction scheduling module is used to fuse the coordinate sets of the physical constraint conflict areas with the gradient compensation parameters, generate a three-dimensional space scheduling instruction set, and execute the instructions.

[0032] In a third aspect, the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, any step of the method for scheduling matching airline baggage change information as described in the first aspect of the present invention is implemented.

[0033] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, any step of the method for scheduling matching airline baggage change information as described in the first aspect of the present invention is implemented.

[0034] The beneficial effects of the present invention are as follows: It constructs a digital twin of airport operations within a metaverse environment, uses a physics engine to accurately simulate the effects of baggage stacking and dumping and conveyor belt vibration offsets, dynamically calculates the volume overload rate of sorting carousels, the probability of path intersection collisions, and the duration of equipment response delays in real time, and generates a high-precision three-dimensional coordinate set of conflict areas. Based on a spatiotemporal rehearsal simulation method, it accurately identifies the safety boundaries of baggage transfer paths, effectively reducing the risk of physical constraint violations. By integrating real-time device IoT data streams, the spatiotemporal consistency of conflict detection is significantly improved, providing reliable data support for the generation of gradient compensation parameters, thereby enhancing the robustness and execution efficiency of scheduling instructions. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0036] Figure 1 A flowchart of the scheduling method for supporting airline baggage change information.

[0037] Figure 2 A flowchart constructed for a social relationship topology graph.

[0038] Figure 3 Flowchart of digital twin conflict detection.

[0039] Figure 4 Flowchart generated for a three-dimensional scheduling instruction. DETAILED DESCRIPTION

[0040] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0041] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0042] Secondly, the term "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in various places throughout this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive of other embodiments.

[0043] Reference Figures 1 to 4, is an embodiment of the present invention, which provides a method for scheduling airline baggage change information, including the following steps:

[0044] S1: Collect baggage change information, passenger itinerary data, and airport equipment status data in air transportation scenarios in real time, and output the original data stream.

[0045] The original data stream refers to the original data stream with timestamps and device association tags generated by performing spatiotemporal alignment and semantic annotation on baggage change information, passenger itinerary data, and airport equipment status data;

[0046] S1.1: Baggage change information includes baggage status change events (e.g., a bag changes from "unloaded" to "loaded") and the baggage's current location coordinates. During the collection process, event logs or sensor data streams are directly read to ensure real-time data.

[0047] Passenger travel data includes passenger booking records, which contain specific attributes such as flight number, seat number, and purchase time. When collecting this data, you can directly query or subscribe to update events from the booking device.

[0048] Airport equipment status data includes the equipment's current operating status (e.g., "normal" or "faulty") and physical location information (e.g., equipment coordinates). During collection, equipment status reports are directly pulled or events are pushed.

[0049] S1.2: Add a unified timestamp to each collected data point (including baggage change information, passenger itinerary data, and airport equipment status data). Timestamps should be synchronized with the system clock (e.g., Network Time Protocol (NTP)) to ensure that all data points have a consistent UTC time reference.

[0050] Based on the added timestamps, all data points are sorted in time series to ensure that events at the same moment are in the same position in the data stream;

[0051] S1.3: Associate data points with physical locations. Specifically, associate baggage change information with the device coordinates of the location where it occurred; associate passenger itinerary data with flight information; and associate flight information with boarding gate device coordinates. Airport equipment status data directly uses its own device coordinates to output a set of time-space aligned data points. Each data point contains a timestamp, location coordinates, and data content.

[0052] It should be noted that data content specifically refers to structured information fields that have been semantically annotated, including status change types in baggage change information (such as "loading status change"), flight binding relationships in passenger itinerary data (such as "flight CA1234"), and operating parameters in airport equipment status data (such as "conveyor belt speed 0.5m / s").

[0053] Semantically annotate the spatiotemporally aligned data point set and add device-associated labels.

[0054] Semantic annotation automatically generates semantic tags based on data content. Specifically, baggage change information is labeled with a "change type" and a "device association tag," passenger itinerary data is labeled with a "trip association tag" and a "device association tag," and airport equipment status data is labeled with a "device association tag." Label generation uses simple rules, such as string matching or dictionary mapping.

[0055] For example, baggage change information is marked: the "Change Type" label is "Loading Status Change", and the "Equipment Association Label" label is "Sorter 001";

[0056] Passenger itinerary data annotation: "Itinerary association tag" is labeled "Flight binding", and "Device association tag" is labeled "Gate A12";

[0057] Airport equipment status data annotation: "Equipment associated tag" is labeled "Conveyor Belt 005";

[0058] Tag generation rules: In a baggage status change event, if the status value includes "Loading", the "Change Type" tag is set to "Loading Status Change"; the device identifier is directly used as the "Device Association Tag".

[0059] S1.4: Integrate all processed data points to generate the original data stream.

[0060] The raw data stream is structured into a list of fields: timestamp, device-associated tag, data content, and semantic tag, and transmitted in real time through a data stream pipeline (such as Kafka or RabbitMQ) for use in subsequent steps.

[0061] S2: The raw data stream is input into a pre-trained graph neural network, and a social relationship topology graph is constructed by analyzing the co-occurring flights, seat proximity, and ticket purchase time interval features in passenger booking records.

[0062] S2.1: Parse the passenger booking records in the original data stream and extract the co-occurring flight numbers, adjacent seat numbers, and purchase time differences.

[0063] Specifically, from the original data stream, filter the data entries with the semantic tag "trip-related tag" and extract all passenger booking records;

[0064] Among them, each passenger booking record contains the passenger's unique identifier, flight number, seat number and ticket purchase time;

[0065] Group the passenger booking records of each flight, count the passengers who appear on the same flight, and combine them to generate the co-occurring flight numbers;

[0066] Query the flight seat physical layout database, calculate the physical adjacent relationship of passenger seat numbers on the same flight, and generate adjacent seat numbers;

[0067] For each pair of passengers on the same flight, calculate the absolute difference (in seconds) between the ticket purchase timestamps to generate the ticket purchase time difference.

[0068] S2.2: Apply a negative exponential decay function to the purchase time difference to generate a time decay coefficient, and perform an inverse Euclidean distance calculation on adjacent seat numbers to generate spatial proximity.

[0069] Specifically, the time difference for ticket purchase , apply the negative exponential decay function to calculate the time decay coefficient , the expression is:

[0070]

[0071] Where, is a predefined attenuation factor used to control the attenuation rate. is a natural constant, representing the base of the exponential function;

[0072] It should be noted that the predefined attenuation factor refers to the attenuation factor obtained by fitting the exponential attenuation curve through the historical flight passenger ticket purchase time difference data and optimizing it using the maximum likelihood estimation method. The optimal value of .

[0073] Convert adjacent seat numbers to physical coordinates (using the flight seat layout database) and calculate the Euclidean distance between the two seats , the expression is:

[0074] ;

[0075] Where, represents the horizontal coordinate of passenger A's seat, represents the horizontal coordinate of passenger B's seat, Indicates the vertical coordinate of passenger A's seat, Indicates the vertical coordinate of passenger B's seat;

[0076] Among them, passengers A and B represent two passenger nodes connected by dynamic weight edges in the social relationship topology graph. The correlation between passengers A and B is jointly defined by the co-occurrence flight, seat proximity and ticket purchase time interval features.

[0077] Applies the inverse Euclidean distance function, outputting a scalar-valued spatial proximity , the expression is:

[0078] ;

[0079] Where, Indicates a very small positive number (such as 0.01), used to prevent A division by zero error occurred.

[0080] S2.3: Use the co-occurring flight numbers as the initial weights of the graph edges, fuse the time decay coefficient and spatial proximity to generate a dynamic social weight matrix, aggregate passenger node features through a graph convolutional network, and output a social relationship topology graph.

[0081] Specifically, the co-occurring flight numbers are used as the basic weights of the graph edges to construct the initial adjacency matrix between passenger nodes. The initial adjacency matrix is ​​dynamically weighted by fusing the time decay coefficient and spatial proximity to generate a dynamic social weight matrix. ;

[0082] Count the total number of flights of each passenger in the passenger booking record as the initial node feature;

[0083] Perform graph convolution on the initial node features and output the updated node feature matrix, which is expressed as:

[0084] ;

[0085] Where, Indicates the The node feature matrix of the layer, represents a nonlinear activation function (such as ReLU), Degree matrix negative half to the power of (normalization term), Indicates the The node feature matrix of the layer, Indicates the The trainable parameter matrix of the layer;

[0086] Based on the passenger's unique identifier, the updated node feature matrix and the dynamic social weight matrix The value of is used to construct a social relationship topology structure and output the social relationship topology.

[0087] S3: Trigger the group baggage priority weighting mechanism based on the social relationship topology graph and generate a collaborative transfer path for the baggage of passengers with itinerary associations.

[0088] S3.1: When the flight dynamics associated with any passenger node in the social relationship topology graph changes, traverse all baggage transfer statuses connected to the changed node and obtain the loading status identification code and current location coordinates.

[0089] Specifically, the flight status associated with the passenger node is monitored in real time. When the flight number, boarding gate or take-off time of any passenger node changes, the corresponding passenger node is marked as a changed node.

[0090] In the social relationship topology graph, query all passenger nodes (i.e., passengers associated with itineraries) that are connected to the change node by dynamic weight edges.

[0091] Using the baggage's unique identifier (bound to the passenger's unique identifier), the loading status code ("Unloaded" or "Loaded") for each bag is retrieved from the baggage change information. The baggage's current location coordinates (such as the sorting carousel coordinates) are retrieved from the airport equipment status data using the baggage's unique identifier, and the status set of all bags associated with the change node is output.

[0092] S3.2: If the baggage loading status identification code is unloaded, a flight binding instruction is generated to forcibly synchronize the current baggage to the changed target flight.

[0093] Specifically, for the baggage status set, baggage entries with a loading status identification code of "unloaded" are filtered, and the ID of the passenger to whom the baggage belongs is extracted to generate an unloaded baggage list;

[0094] The ID of the passenger to whom the baggage belongs is the unique identification code assigned to each passenger by the airline in the passenger booking record. It is usually composed of an alphanumeric combination (such as ABC123). The unique identification code is bound to the passenger's passport information and flight itinerary data to ensure that the passenger and checked items can be accurately linked during the baggage handling process.

[0095] The unloaded baggage list includes a baggage unique identifier (identifying the baggage itself) and an associated passenger unique identifier (identifying the passenger to whom the baggage belongs);

[0096] Create structured instructions for each unloaded bag and output a flight-bound instruction set.

[0097] S3.3: If the baggage loading status identification code is loaded, calculate the shortest transfer time from the current location coordinates of the corresponding baggage to the target flight boarding gate, and generate a virtual synchronization time window constraint.

[0098] Specifically, for the baggage status set, entries with a loading status identification code of "loaded" are filtered to generate a loaded baggage list;

[0099] According to the changed target flight number, query the airport flight database to obtain the target boarding gate coordinates;

[0100] It should be noted that the airport flight database refers to a structured data set that stores flight dynamic information and airport geographic spatial data, including the relationship between flight number, planned take-off and landing time, assigned boarding gate number and the three-dimensional coordinates of the boarding gate, and is updated in real time through the airline's operating platform or the airport ground management platform.

[0101] Use the A* algorithm to calculate the shortest path distance from the current luggage location coordinates to the target boarding gate coordinates , output the shortest transit time value of each baggage , the expression is:

[0102] ;

[0103] Where, is the average speed of the baggage conveyor belt;

[0104] It should be noted that the average speed of the baggage conveyor belt is determined by measuring the actual conveyor belt operating speed at the airport (such as taking the average of multiple samples). The example value is usually 0.5 m / s (the specific value needs to be calibrated according to the airport equipment parameters).

[0105] When generating time window constraints, the flight's scheduled gate closure time is used as a benchmark, and the time difference between the shortest baggage transfer time and the gate closure time is formed to determine the latest time boundary for baggage to arrive at the boarding gate, thereby defining the effective time range for baggage transfer operations.

[0106] S3.4: According to the dynamic weight values ​​of the connecting edges in the social relationship topology graph, transfer sequence adjustment coefficients are assigned to passenger nodes constrained by the virtual synchronization time window, and the collaborative transfer path is output.

[0107] Specifically, the corresponding weight values ​​of the dynamic social weight matrix between the passenger node and the change node constrained by the time window are extracted from the social relationship topology graph. , calculate the adjustment coefficient , the expression is:

[0108] ;

[0109] Where, Represents the sum of the weight values ​​of all associated nodes;

[0110] Sort the constrained passenger nodes in descending order by the adjustment coefficient and output the priority ( The larger the value, the higher the priority);

[0111] For each bag, a collaborative transfer path is generated by combining time window constraints and priorities;

[0112] It should be noted that high-priority bags are allocated earlier conveyor slots and shorter routes, while low-priority bags are allowed to use alternative routes or delay slots.

[0113] S4: Build a digital twin of airport operations in the metaverse, import the collaborative transfer path into the digital twin environment for spatiotemporal rehearsal simulation, detect baggage flow conflict points in real time, and mark the coordinate sets of the physical constraint conflict areas.

[0114] S4.1: Load the airport building information model and real-time device IoT data streams into the metaverse environment to build a dynamic digital twin.

[0115] Specifically, the airport's three-dimensional layout data is obtained from the airport flight database. The airport's three-dimensional layout data includes building structures, equipment locations, and path networks.

[0116] Subscribe to real-time device IoT data streams using data stream pipelines;

[0117] Integrate the airport's three-dimensional layout data and real-time device IoT data streams into the metaverse environment, establish a real-time mapping relationship in an in-memory database (such as Redis), and form a digital twin to update the device status and baggage location in real time.

[0118] S4.2: Import the collaborative transfer path into the dynamic digital twin and perform spatiotemporal simulation of baggage transfer, sorting carousel volume overload rate, path intersection collision probability, and equipment response delay time.

[0119] Specifically, the co-transport pathway is loaded into the dynamic digital twin as input.

[0120] Start the spatiotemporal simulation of baggage transfer. This simulation simulates the movement of baggage along a path, sorted by a timestamp sequence. During the simulation, the baggage's movement from its current location to the destination gate is simulated, with the baggage position updated incrementally (at a rate of milliseconds to reflect real-time dynamics). Calculate the sorting carousel overload rate: Monitor the number of bags on each sorting carousel, compare it to the carousel's maximum capacity, and output the overload rate. Calculate the path intersection collision probability: Detect overlap in baggage position coordinates at the same time. Overlap indicates a collision event, and output the probability. Calculate the device response delay: Record the start time and actual completion time of the device processing the instruction and output the delay value.

[0121] Detect baggage flow conflicts. Baggage flow conflicts are locations and times in the simulation where the overload rate is too high, the collision probability is abnormal, or the delay duration exceeds the limit. Output a conflict point list. The conflict point list contains fields such as the conflict type (such as "overload"), the current baggage location coordinates (sorting carousel or path intersection), and a timestamp.

[0122] S4.3: Activate the physics engine to simulate the effects of baggage stacking and dumping and conveyor belt vibration and deviation, and mark the coordinate sets of the physical constraint conflict areas that exceed the safety threshold.

[0123] Among them, the physics engine refers to a software component that simulates physical phenomena (such as based on Newtonian mechanics), and the parameter settings include luggage weight, friction coefficient and conveyor belt vibration parameters.

[0124] Specifically, the physics engine is enabled to simulate baggage stacking and tipping, as well as conveyor belt vibration deflection. When the current coordinates of multiple bags are close to each other, the stack height and stability are calculated. If the stack height exceeds the stability angle, the physics engine predicts a tipping event. Simultaneously, in the conveyor belt vibration deflection simulation, random vibration perturbations are applied to the conveyor belt coordinates, and the deflection amplitude is calculated through integration.

[0125] Monitor whether parameters in the physics engine simulation exceed safety thresholds. These thresholds include stacking height limits and offset amplitude limits. The stacking height limit is the maximum allowable ratio of the baggage stacking height to the bottom width, and the offset amplitude limit is the maximum allowable ratio of the conveyor belt vibration displacement to the conveyor belt width. When any parameter is detected to exceed any limit, the 3D spatial coordinates and timestamp of the corresponding conflict point are recorded.

[0126] Combine the dumping event and vibration offset exceeding results, and associate them with the baggage's current location coordinates and timestamp in the conflict point list to generate a physical constraint conflict area coordinate set;

[0127] The physical constraint conflict region coordinate set is a three-dimensional coordinate set, including the conflict type (such as stacking and falling), the three-dimensional coordinates, and the associated timestamp.

[0128] S5: Use the inverse reinforcement learning algorithm to reverse the conflict events in the preview simulation, trace the characteristics of the policy defects and reconstruct the reward function space, and output the gradient compensation parameters.

[0129] S5.1: Combine the coordinate set of the physical constraint conflict area with the digital twin environment parameters to generate a state-action pair sample set as input data for the inverse reinforcement learning algorithm.

[0130] Specifically, the coordinate set of the physical constraint conflict area is aligned with the digital twin environment parameters by timestamp. After alignment, each conflict point is associated with the digital twin environment parameter value at the corresponding moment;

[0131] State: It is composed of the coordinates of the conflict point, the volume overload rate of the sorting carousel at the associated time, the path intersection collision probability, and the equipment response delay time; Action: It is defined as the instructions related to the conflict point in the collaborative transportation path (such as the path navigation point coordinate sequence or motor control instructions).

[0132] Each sample is a two-tuple (state, action), and the output is a set of state-action pair samples.

[0133] S5.2: Calculate the probability distribution of the state-action pair sample set through the maximum entropy probability model and infer the reward function corresponding to the optimal strategy.

[0134] Specifically, from the state-action pair sample set, state features (device response delay, path intersection collision probability, and sorting carousel volume overload rate) are extracted as input items of the reward function;

[0135] The state-action pair sample set is input into the maximum entropy probability model. The maximum entropy probability model is a standard inverse reinforcement learning algorithm that infers the reward function. Specifically, based on the co-occurrence frequency of states and actions in the state-action pair sample set, a probability distribution function is fitted, and the probability distribution of the state-action pair sample set is calculated. By maximizing the likelihood probability of the state-action pair sample set, the reward function form that optimizes the probability distribution is solved. The state eigenvalues ​​and weight distribution when the conflict event occurs are compared to identify defects such as too low a weight for the device response delay or insufficient penalty coefficient for path intersection collision.

[0136] The output of the reward function is a linear weighted expression, and the weight variables correspond to conflict detection indicators (such as the weight of the device response delay time and the weight of the path intersection collision probability).

[0137] S5.3: Analyze the weight distribution defects of the reward function, identify the device response delay weights and abnormal path crossing penalty coefficients that cause conflicts, and reconstruct the reward function space structure.

[0138] Specifically, the weight distribution of the reward function is analyzed to extract the weight value corresponding to the device response delay and the penalty coefficient corresponding to the path intersection collision probability;

[0139] If the device response delay weight is lower than the preset weight threshold, it is marked as a defect feature;

[0140] If the abnormal path crossing penalty coefficient deviates from the historical normal range, it is marked as a defect feature;

[0141] Increase the device response delay weight to the preset weight threshold, correct the abnormal path intersection penalty coefficient to the median of the historical normal range, keep other weights unchanged, and output the reconstructed reward function expression.

[0142] It should be noted that when determining the preset weight threshold, the P95 percentile value of the equipment response delay weight distribution in the historical operation data is used as the benchmark, and the historical normal range is determined by statistically analyzing the interquartile range (IQR) of the abnormal path cross penalty coefficient in the past three months; for example, the preset threshold range of the equipment response delay weight may be between 0.15-0.35, and the historical normal range of the abnormal path cross penalty coefficient may fluctuate within the range of 0.05-0.25.

[0143] S5.4: Inject device failure simulation noise and path random offsets into the state-action pair sample set to generate adversarial perturbation samples and train the policy network model to output gradient compensation parameters.

[0144] Specifically, we superimpose device failure simulation noise on the device response delay characteristics of the state-action pair sample set; and inject random position offsets into the path navigation point coordinate sequence of the action field to generate an extended adversarial perturbation sample set.

[0145] Use the graph convolutional network structure as the policy network model, input the state field of the adversarial perturbation sample set, and output the predicted value of the action field;

[0146] It should be noted that the training process of using a graph convolutional network structure as a policy network model is as follows: a graph data structure containing a node feature matrix and an adjacency matrix is ​​constructed, where the node feature matrix is ​​composed of the state parameters of the adversarial perturbation sample set (including dimensional data such as device response delay time and path intersection collision probability), and the adjacency matrix quantifies the correlation strength between the state parameters; the node feature matrix is ​​processed through multi-layer graph convolution operations, and each layer performs linear transformation and then propagates features with the adjacency matrix, using the ReLU activation function; finally, the processed node feature matrix is ​​mapped to the action parameter prediction value, and the mean square error is used as the loss function during training. The network parameters are optimized through backpropagation, and the Adam optimizer is used to complete the convergence of the policy network model;

[0147] Calculate the gradient of the loss function in the strategy network model with respect to the state parameters contained in the node feature matrix, normalize the gradient value into a compensation parameter vector, and output the gradient compensation parameters. The gradient compensation parameters include the device response delay compensation coefficient, the path crossing compensation coefficient, and the volume overload compensation coefficient.

[0148] S6: Fuse the coordinate set of the physical constraint conflict area with the gradient compensation parameters to generate a three-dimensional space scheduling instruction set and execute the instruction issuance.

[0149] S6.1: Calculate the start-up time offset value of the sorting equipment motor based on the gradient compensation parameter and generate a motor control instruction.

[0150] Specifically, the product of the device response delay compensation coefficient and the device response delay duration is used as the time compensation amount, and the time compensation amount is converted into a clock cycle adjustment value in combination with the pre-stored motor acceleration characteristic parameters;

[0151] Write the clock cycle adjustment value into the PLC timer register to generate the motor control instruction;

[0152] Among them, the pre-stored motor acceleration characteristic parameters refer to the inherent performance data obtained through the motor factory test, including the starting time-speed curve under rated voltage, the maximum allowable acceleration slope and the inertia compensation coefficient. They are usually stored in the device controller in the form of a lookup table for accurately calculating the clock cycle compensation value.

[0153] The motor control instruction contains the sorting device number, the adjusted start timestamp and the execution priority.

[0154] S6.2: Based on the coordinate set of the physical constraint conflict area, use the A* algorithm to replan the ground crew's travel path and output the coordinate sequence of the path navigation points.

[0155] Specifically, read the coordinate set of the physical constraint conflict area and set the path planning parameters;

[0156] Path planning parameters include: starting point: the current location coordinates of the ground crew (airport equipment status data contains physical location information); end point: the coordinates of the nearest safe point outside the conflict area (calculated by the coordinate set boundary); obstacles: all points within the coordinate set of the physical constraint conflict area;

[0157] Use the same A* algorithm as in step S3.3 to perform A* algorithm path planning and output a sequence of path navigation point coordinates.

[0158] S6.3: Align the motor control instructions with the path navigation point coordinate sequence according to the timestamp, generate a structured instruction data packet, send the motor control instructions to the baggage sorting controller through the message queue, and simultaneously transmit the path navigation point coordinate sequence to the ground staff handheld terminal through the wireless protocol.

[0159] Specifically, the motor control instructions and the path navigation point coordinate sequence are aligned by timestamp, a structured instruction data packet is generated, and the instruction is issued. The fields of the structured instruction data packet include: instruction type (motor control / path navigation), target device / personnel identification, execution time, and content data.

[0160] It should be noted that the timestamps of motor control instructions are matched to the time windows of the route navigation points; the motor control instructions are sent to the baggage sorting controller via a message queue (such as Kafka). The coordinate sequence of the route navigation points is transmitted to the ground staff handheld terminal via a wireless protocol (such as Wi-Fi 6).

[0161] This embodiment also provides a scheduling system for supporting air baggage change information, including: a data acquisition module, a graph construction module, a path generation module, a simulation preview module, a strategy optimization module and an instruction scheduling module; the data acquisition module is used to collect baggage change information, passenger itinerary data and airport equipment status data in air transportation scenarios in real time and output the original data stream; the graph construction module is used to input the original data stream into a pre-trained graph neural network and construct a social relationship topology map by analyzing the co-occurring flights, seat proximity and ticket purchase time interval characteristics in passenger booking records; the path generation module is used to trigger group communication based on the social relationship topology map A weighted baggage priority mechanism is used to generate collaborative transfer paths for passenger baggage with itinerary associations. A simulation rehearsal module is used to build a digital twin of airport operations in the metaverse, import collaborative transfer paths into the digital twin environment for spatiotemporal rehearsal simulation, detect baggage flow conflict points in real time, and mark the coordinate sets of physical constraint conflict areas. A strategy optimization module is used to inversely resolve conflict events in the rehearsal simulation through an inverse reinforcement learning algorithm, trace the characteristics of strategy defects, reconstruct the reward function space, and output gradient compensation parameters. An instruction scheduling module is used to fuse the coordinate sets of physical constraint conflict areas with gradient compensation parameters, generate a three-dimensional space scheduling instruction set, and execute and issue instructions.

[0162] This embodiment further provides a computer device suitable for use with the method for scheduling airline baggage change information, comprising: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the method for scheduling airline baggage change information as proposed in the above embodiment.

[0163] The computer device may be a terminal, comprising a processor, memory, a communication interface, a display, and an input device connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores an operating system and computer programs. The internal memory provides an environment for the operating system and computer programs stored in the non-volatile storage media. The communication interface of the computer device is used to communicate with external terminals via wired or wireless communication. Wireless communication may be achieved via Wi-Fi, a carrier network, NFC (near-field communication), or other technologies. The display of the computer device may be a liquid crystal display or an electronic ink display. The input device may be a touchscreen overlay on the display, buttons, a trackball, or a touchpad on the computer device housing, or an external keyboard, touchpad, or mouse.

[0164] This embodiment further provides a storage medium having a computer program stored thereon. When the program is executed by a processor, the program implements the method for scheduling airline baggage change information as proposed in the above embodiment. The storage medium can be implemented by any type of volatile or non-volatile storage device, or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0165] In summary, this invention constructs a digital twin of airport operations within a metaverse environment. It utilizes a physics engine to accurately simulate the effects of baggage stacking and dumping, as well as conveyor belt vibration offsets. It dynamically calculates the volume overload rate of sorting carousels, the probability of path intersection collisions, and the duration of equipment response delays in real time, generating a high-precision three-dimensional coordinate set of conflict zones. Based on a spatiotemporal rehearsal simulation method, it accurately identifies the safety boundaries of baggage transfer paths, effectively reducing the risk of physical constraint violations. By integrating real-time device IoT data streams, the spatiotemporal consistency of conflict detection is significantly improved, providing reliable data support for the generation of gradient compensation parameters, thereby enhancing the robustness and execution efficiency of scheduling instructions.

[0166] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for dispatching airline baggage change information, characterized by: include, Collect baggage change information, passenger itinerary data, and airport equipment status data in real time in air transportation scenarios, and output raw data streams; The raw data stream is fed into a pre-trained graph neural network, which constructs a social relationship topology graph by analyzing the co-occurring flights, seat proximity, and ticket purchase time interval features in passenger booking records. Triggering a group baggage priority weighting mechanism based on the social relationship topology, generating collaborative transfer routes for passenger baggage with itinerary associations; Build a digital twin of airport operations in the metaverse, import collaborative transfer routes into the digital twin environment for spatiotemporal rehearsal simulation, detect baggage flow conflict points in real time, and mark the coordinate sets of physical constraint conflict areas; Through the inverse reinforcement learning algorithm, the conflict events in the preview simulation are reversed, the strategy defect characteristics are traced, the reward function space is reconstructed, and the gradient compensation parameters are output; The coordinate set of the physical constraint conflict area and the gradient compensation parameters are integrated to generate a three-dimensional space scheduling instruction set and execute the instruction issuance.

2. The method for dispatching airline baggage change information according to claim 1, wherein: The passenger itinerary data includes passenger booking records.

3. The method for dispatching airline baggage change information according to claim 1, wherein: The specific steps of constructing the social relationship topology graph are as follows: Parse passenger booking records in the original data stream to extract co-occurring flight numbers, adjacent seat numbers, and purchase time differences; Applying a negative exponential decay function to the ticket purchase time difference generates a time decay coefficient, and performing an inverse Euclidean distance calculation on adjacent seat numbers generates spatial proximity; The co-occurring flight numbers are used as the initial weights of the graph edges, and the time decay coefficient and spatial proximity are integrated to generate a dynamic social weight matrix. The passenger node features are aggregated through a graph convolutional network to output a social relationship topology graph.

4. The method for dispatching airline baggage change information according to claim 3, wherein: The specific steps of generating a collaborative transfer route for passenger luggage with itinerary association are as follows: When the flight status associated with any passenger node in the social relationship topology graph changes, traverse the baggage transfer status of all passengers connected to the changed node and obtain the loading status identification code and current location coordinates; If the baggage loading status identification code is "Unloaded", a flight binding instruction is generated to forcibly synchronize the current baggage to the changed target flight; If the baggage loading status identification code is loaded, calculate the shortest transfer time from the current location coordinates of the corresponding baggage to the target flight boarding gate, and generate a virtual synchronization time window constraint; According to the dynamic weight values ​​of the connecting edges in the social relationship topology graph, the transfer sequence adjustment coefficients are assigned to the passenger nodes constrained by the virtual synchronization time window, and the collaborative transfer path is output.

5. The method for dispatching airline baggage change information according to claim 4, characterized in that: The specific steps of detecting baggage flow conflict points in real time and marking the coordinate set of the physical constraint conflict area are as follows: Load the airport building information model and real-time device IoT data streams into the Metaverse environment to build a dynamic digital twin; Import the collaborative transfer path into the dynamic digital twin and perform spatiotemporal simulation of baggage transfer, including sorting carousel volume overload rate, path intersection collision probability, and equipment response delay. Activate the physics engine to simulate the effects of baggage stacking and dumping and conveyor belt vibration and deviation, and mark the coordinate sets of the physical constraint conflict areas that exceed the safety threshold.

6. The method for dispatching airline baggage change information according to claim 5, characterized in that: The output gradient compensation parameters are specifically performed as follows: The coordinate set of the physical constraint conflict area is combined with the digital twin environment parameters to generate a state-action pair sample set as the input data of the inverse reinforcement learning algorithm; The probability distribution of the state-action pair sample set is calculated through the maximum entropy probability model, and the reward function corresponding to the optimal strategy is deduced; Analyze the weight distribution defects of the reward function, identify the device response delay weights and abnormal path crossing penalty coefficients that cause conflicts, and reconstruct the reward function space structure; Device failure simulation noise and path random offsets are injected into the state-action pair sample set to generate adversarial perturbation samples, and the policy network model is trained to output gradient compensation parameters.

7. The method for dispatching airline baggage change information according to claim 6, wherein: The specific steps of generating a three-dimensional space scheduling instruction set and executing the instruction issuance are as follows: Calculate the start-up time offset value of the sorting equipment motor based on the gradient compensation parameter and generate the motor control instruction; Based on the coordinate set of the physical constraint conflict area, the A* algorithm is used to re-plan the ground crew's travel path and output the coordinate sequence of the path navigation points; The motor control instructions are aligned with the path navigation point coordinate sequence according to the timestamp to generate a structured instruction data packet. The motor control instructions are sent to the baggage sorting controller through the message queue, and the path navigation point coordinate sequence is transmitted to the ground staff handheld terminal through the wireless protocol.

8. A system for dispatching airline baggage change information, based on the method for dispatching airline baggage change information according to any one of claims 1 to 7, characterized in that: Including data acquisition module, graph construction module, path generation module, simulation preview module, strategy optimization module and instruction scheduling module; The data acquisition module is used to collect baggage change information, passenger itinerary data and airport equipment status data in the air transportation scenario in real time and output the original data stream; The graph construction module is used to input the original data stream into the pre-trained graph neural network and construct a social relationship topology graph by analyzing the co-occurring flights, seat proximity and ticket purchase time interval characteristics in the passenger booking records; The path generation module is used to trigger the group baggage priority weighting mechanism based on the social relationship topology graph and generate a collaborative transfer path for the baggage of passengers with itinerary associations; The simulation rehearsal module is used to build a digital twin of airport operations in the metaverse, import the collaborative transfer path into the digital twin environment for spatiotemporal rehearsal simulation, detect baggage flow conflict points in real time, and mark the coordinate sets of the physical constraint conflict areas; The strategy optimization module is used to reversely solve the conflict events in the preview simulation through the inverse reinforcement learning algorithm, trace the characteristics of the strategy defects and reconstruct the reward function space, and output the gradient compensation parameters; The instruction scheduling module is used to fuse the physical constraint conflict area coordinate set and the gradient compensation parameters, generate a three-dimensional space scheduling instruction set and execute the instruction issuance.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method for scheduling supporting airline baggage change information according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for scheduling supporting airline baggage change information according to any one of claims 1 to 7 are implemented.

Images