Airport luggage tracking system and method based on block chain

Abstract

The invention relates to the technical field of airport baggage management, in particular to an airport baggage tracking system and method based on a block chain, and the method comprises the steps: baggage information collection and identification: collecting basic baggage information through an intelligent tag, encrypting the basic baggage information to generate a unique identifier, and recording the unique identifier in the block chain; luggage state updating: collecting state change information by using the Internet of Things equipment, and writing the state change information into the block chain after verification of the smart contract; anomaly detection and prediction: simulating and analyzing a luggage state by means of a generative adversarial network, and triggering an alarm and updating the block chain if an anomaly is found; path optimization: adopting a reinforcement learning algorithm to optimize a transportation path according to a real-time state and historical data and recording the transportation path; and performing anomaly analysis and response, analyzing path anomaly based on the graph convolutional network, and processing and recording according to a strategy. The data security and transparency are guaranteed by using the block chain technology, the luggage management efficiency is improved by combining various advanced algorithms, the abnormal risk is reduced, the service experience is optimized, and the method is suitable for the field of airport luggage management.

Description

Technical Field

[0001] The present invention relates to the technical field of airport baggage management, and in particular to a blockchain-based airport baggage tracking system and method. Background Art

[0002] Traditional airport baggage tracking methods have exposed many problems in response to growing business needs and are in urgent need of improvement.

[0003] Traditional baggage tracking methods mainly rely on manual operations and decentralized information systems, and information acquisition is obviously untimely and inaccurate. It is difficult for passengers to grasp the dynamics of their luggage in real time, and they often need to spend a lot of time and energy to inquire about the location of their luggage at the airport. Moreover, due to the lack of effective integration and real-time sharing of data in various links, errors or delays are prone to occur during the information transmission process, resulting in the inability of airport staff to make accurate decisions in a timely manner, further exacerbating the uncertainty of luggage transportation.

[0004] The data management model of the existing baggage management system also has many drawbacks. The centralized database architecture exposes data security to potential risks. At the same time, this architecture limits the efficient sharing and interaction of data. The data flow between different departments or systems is not smooth, making it difficult to achieve collaborative operations, affecting the overall efficiency of baggage handling.

[0005] Traditional systems have limited capabilities in anomaly detection and processing. They are mostly based on simple rules and manual experience, making it difficult to accurately predict and provide early warnings for complex and ever-changing baggage transportation conditions. For potential anomalies, such as the risk of baggage congestion caused by unreasonable allocation of transportation resources, the system is unable to detect and take effective measures in a timely manner, and often responds passively after the problem occurs, resulting in delayed processing time and increasing the possibility of baggage delay or loss.

[0006] Therefore, in response to the above problems, a blockchain-based airport baggage tracking system and method are proposed. Summary of the invention

[0007] The purpose of the present invention is to provide a blockchain-based airport baggage tracking system and method to solve the problems raised in the above background technology.

[0008] To achieve the above object, the present invention provides the following technical solutions:

[0009] A blockchain-based airport baggage tracking method includes the following steps:

[0010] Step S1, baggage information collection and identification: when the baggage enters the airport baggage handling process, the basic information of the baggage is collected, identification information uniquely identifying the baggage is generated, and the basic information and identification information are associated and recorded in the blockchain network;

[0011] Step S2, luggage status update: obtaining the luggage status change information in real time during the luggage circulation process in the airport, and updating the luggage status change information to the corresponding luggage record in the blockchain;

[0012] Step S3, anomaly detection and prediction: Generative adversarial networks are used to simulate and analyze the luggage status to determine whether there is a potential anomaly in the luggage. Once an anomaly is determined, a corresponding alarm is triggered and the status record of the luggage in the blockchain is updated;

[0013] Step S4, route optimization: The luggage transportation route is optimized and adjusted according to the real-time status and historical data of the luggage through a reinforcement learning algorithm, and the optimized route information is recorded in the corresponding position of the blockchain;

[0014] Step S5, abnormal analysis and response: Analyze the baggage transportation path data, detect abnormal conditions that occur during transportation, take countermeasures according to preset strategies, and record the abnormal conditions and countermeasures taken in the blockchain;

[0015] Step S6, personalized service optimization: Based on the characteristics of the airport and passengers, an adaptive meta-learning algorithm is used to achieve personalized adaptation of the baggage tracking service, and the relevant personalized configuration information is stored in the corresponding baggage record in the blockchain.

[0016] As a preferred solution, baggage information collection and identification includes:

[0017] Step S1-1, information collection: When the luggage is checked in, the basic information of the luggage is collected with the help of the smart tag installed on the luggage. The basic information includes the physical attribute information of the luggage, itinerary related information and passenger related information. The physical attribute information of the luggage is weight. , size information length ,Width ,high , itinerary information including departure place ,destination , passenger-related information includes passenger identification code , integrating the above information to form a vector ;

[0018] Step S1-2, encryption processing: using symmetric encryption algorithm Pair Vector Encryption is performed, and the encryption operation is expressed as ,in, The encrypted information is obtained by distributing and storing encryption keys through a secure key management mechanism ;

[0019] Step S1-3, logo generation: Encrypted information Upload to the blockchain network and use the hash function , SHA-256 hash function pair Perform calculations to generate a unique identifier for the luggage in the blockchain network , following the formula , using this The luggage is uniquely identified in the blockchain and is associated with all its status information and operation records.

[0020] As a preferred solution, the operation of updating the baggage status includes:

[0021] Step S2-1, status information collection: When the luggage is in the process of security inspection, boarding, transfer, and arrival at the airport, the IoT devices distributed at key locations are used to collect the luggage status change information in real time. The status change information includes location coordinate information, status category, and location information. And the corresponding timestamp , the position coordinate information is expressed in Cartesian coordinate system as ,in, Indicates the collection time. The values ​​of include normal, delayed, and suspected loss, denoted as , combining the above information into a state update vector ;

[0022] Step S2-2: Writing based on smart contracts: Write Write to the blockchain, the data verification rules defined in the smart contract Covers the verification logic of data format, source legitimacy, integrity, and timing rationality. satisfy When the preset write logic , involving block generation rules and chain structure addition rules, can successfully Write to the blockchain to ensure the real-time and unchangeable nature of the data. This process is described as follows: , then execute Operation will Write to the blockchain.

[0023] As a preferred solution, the operations of anomaly detection and prediction include:

[0024] Step S3-1, Generative Adversarial Network (GAN) construction and training: Use the Generative Adversarial Network (GAN) to simulate and predict the luggage status. The GAN consists of a generator and the discriminator Composition, Generator By preset dimensions And obey uniform distribution , and is the random noise vector of the distribution interval boundary value , , as input, the operation is optimized and adjusted by its training effect, and the simulated baggage status data is output , and the output dimension is consistent with the actual collected state update vector Dimensionality consistent; Discriminator It is a neural network structure used to distinguish the real and generated , its loss function The definition is as follows: ;

[0025] in, Express expectations, is the distribution of real baggage status data, is a random noise vector The distribution of and parameters to minimize ,make The generated simulated data is close to the real data;

[0026] Step S3-2, abnormality determination and record update: During actual operation, compare the currently collected and , choose Euclidean distance as the similarity measurement index ,for and , is the vector dimension, and the Euclidean distance calculation formula is , set the similarity threshold ,when When the baggage is judged to have potential abnormality, the system automatically triggers an abnormality alarm and updates the rules. Based on the blockchain data update interface and format requirements, update the abnormal related information to the corresponding baggage record on the blockchain and execute Operation, exception-related information includes exception type and discovery time.

[0027] As a preferred solution, the path optimization operation includes:

[0028] Step S4-1, reinforcement learning RL model setting: Use reinforcement learning RL algorithm to realize intelligent optimization of baggage transportation path and define state space , whose elements are composed of the current location of the luggage, the transportation link it is in, and the status of surrounding transportation resources. A state in Expressed as ,in, Indicates location information. Indicates the transport link. Represents the state of surrounding transportation resources and action space , which includes the different transportation path adjustment actions that the baggage can choose, and the transfer to different conveyor belts is recorded as Waiting for the next batch of transfer is recorded as ,Right now , and the reward function , the reward function takes into account the transportation time , luggage congestion Possibility of delay , the specific form is ,in, are the corresponding weight coefficients and are all greater than 0. The weight coefficients are dynamically adjusted according to the actual operation conditions of the airport in different periods and areas;

[0029] Step S4-2: Path optimization based on Q-learning: The core of the Q-learning algorithm Value Function Indicates in status Take action The expected long-term return, The value update formula is: ,in, is the learning rate, ranging from between, is the discount factor, and its value range is between, To take action Then transfer to the next state, Indicates in status Next, for all possible actions The corresponding The value takes the maximum value and is updated through continuous iteration , in each state Next select The biggest move To optimize the luggage transportation path and follow the data recording rules of the blockchain Update the optimized path information to the corresponding baggage record in the blockchain, that is, execute operate.

[0030] As a preferred solution, the abnormal analysis and response operations include:

[0031] Step S5-1, Graph Convolutional Network (GCN) construction and operation: Build a graph convolutional network (GCN) to analyze the abnormalities of the baggage transportation path and build a graph for the baggage transportation path in the airport. ,in, Represents a node set, which covers the boarding gates and baggage sorting areas in the airport. The number of nodes is , Represents the edge set, corresponding to the connection channels between each location point, and the number of edges is recorded as , each edge Association attribute vector ,in, Indicates the channel capacity, Indicates the current flow rate. Represents the average transportation time, and the inter-layer propagation formula of the graph convolutional network is as follows: ,in, , is the adjacency matrix, is the identity matrix, for The degree matrix of For the Layer feature matrix, For the The layer weight matrix, is the activation function, and the feature representation of each node is obtained through multi-layer graph convolution operation to analyze the changes of various parameters on the path;

[0032] Step S5-2, abnormal determination and response: set the normal threshold range corresponding to each attribute, such as the normal flow threshold corresponding to the channel capacity The reasonable range of average transportation time is , when the analysis finds that the edge Properties If the corresponding normal threshold value is exceeded, it is determined that an abnormal situation exists and the preset emergency response process is followed. Automatically take corresponding measures and follow the abnormal recording rules of the blockchain Update the abnormal situation and the measures taken to the corresponding baggage record on the blockchain, and execute operate.

[0033] As a preferred solution, the operations of personalized service optimization include:

[0034] Step S6-1, feature quantification and meta-learning model: Using the adaptive meta-learning algorithm, the personalized adaptation of the baggage tracking service is realized according to the environmental characteristics of different airports and the personalized needs of passengers. The environmental characteristics include the size of the airport , Flight traffic distribution , layout structure , passengers’ personalized needs include the expected frequency of checking baggage status , the types of key information to focus on , quantify the airport environment characteristics into vectors , the personalized needs of passengers are quantified as vectors , the model-independent meta-learning MAML algorithm is used for meta-training and parameter update, and its meta-parameter update formula is: ,in, is the meta-learning rate, For the The test set loss function of the airport data set is used to obtain the adapted meta parameters by performing meta-training on multiple airport data sets. ;

[0035] Step S6-2, personalized service adaptation and record update: In response to new airport scenarios or changes in passenger demand, the sample data of the airport is used based on meta parameters Fine-tune and record rules according to the personalized configuration of blockchain Update the configuration information related to the personalized service to the corresponding baggage record in the blockchain, that is, execute operate.

[0036] An airport baggage tracking system based on blockchain, the airport baggage tracking system implements an airport baggage tracking method based on blockchain.

[0037] It can be seen from the technical solution provided by the present invention that the airport baggage tracking system and method based on blockchain provided by the present invention has the following beneficial effects:

[0038] 1. Improved data security and credibility:

[0039] Data encryption and non-tamperability: During the baggage information collection and identification stage, a symmetric encryption algorithm is used to encrypt the basic information of the baggage to ensure the confidentiality of the data during transmission and storage. A unique identifier is then generated through a hash function, and all information is associated and recorded in the blockchain network. The distributed ledger technology of the blockchain ensures that once the data is recorded, it is difficult to tamper with, providing a highly reliable data foundation for baggage tracking. Whether it is passengers, airport staff or other relevant parties, they can trust the authenticity and integrity of the information such as baggage status and transportation route recorded in the system, effectively avoiding the problem of chaotic baggage management.

[0040] Enhanced data transparency and traceability: The open and transparent nature of blockchain makes every status update and operation record of the luggage in the entire transportation process clear and traceable; anyone with access to the blockchain can trace the entire process of luggage from entering the airport to final delivery, including location changes and status change time at each link; this high degree of transparency not only helps optimize the internal management process of the airport, but also can quickly and accurately locate the responsible link when problems arise, improving the clarity of responsibility and management efficiency in the baggage handling process;

[0041] 2. Accurate anomaly detection and efficient processing:

[0042] Predict potential anomalies in advance: Generative adversarial networks (GANs) are used to simulate baggage status. By training the generator and discriminator, the generator can output simulated data with a distribution similar to that of real baggage status data. In actual operation, the real-time collected baggage status information is compared with the simulated data, and the Euclidean distance is used as the similarity measurement indicator. When the difference between the two is less than the set threshold, potential abnormalities of baggage can be discovered in time, such as the baggage transportation trajectory deviating from expectations, the status change not conforming to the normal mode, etc., so as to provide early warning, buy precious time for the airport to take corresponding measures, and greatly reduce the risk of baggage loss or delay;

[0043] Optimize the abnormal response mechanism: Once an abnormality is detected, the system automatically triggers an alarm and updates the abnormality-related information (such as abnormality type, discovery time, etc.) to the blockchain according to the preset rules; at the same time, the abnormality analysis model built based on the graph convolutional network (GCN) can deeply analyze the attribute changes of each node and edge on the baggage transportation path, and accurately locate the specific location and cause of the abnormality (such as channel congestion, equipment failure, etc.); according to the preset emergency response process, the airport can quickly take targeted measures, such as adjusting the baggage transportation route, deploying human and material resources, etc., which improves the efficiency and accuracy of abnormal handling, reduces the impact of abnormal situations on baggage transportation, and ensures the smooth progress of passengers' journeys;

[0044] 3. Intelligent path optimization and improved operational efficiency:

[0045] Real-time dynamic path adjustment: Reinforcement learning (RL) algorithms play a key role in optimizing baggage transportation paths. By defining a state space that includes elements such as the current location of the baggage, transportation links, and the status of surrounding resources, as well as an action space that covers different path adjustment actions, and combining a reward function that comprehensively considers transportation time, baggage congestion, and the possibility of delay, the system can select the optimal transportation path adjustment action at each decision point based on the real-time status of the baggage and historical data. For example, during the baggage transportation process, when a conveyor belt is found to be congested, the system can make a timely decision to transfer the baggage to other idle or more efficient conveyor belts, or wait for the next batch of transfers, so as to avoid long-term detention of baggage, reduce the overall transportation time, and improve the timeliness and reliability of baggage transportation.

[0046] Improve resource utilization and overall operational efficiency: The route optimization process not only focuses on the transportation efficiency of individual luggage, but also considers the reasonable allocation and utilization of the airport's overall transportation resources. By continuously optimizing the luggage transportation route, it can effectively balance the load of each transportation route, avoid excessive tension or idle waste of local resources, and improve the overall operational efficiency of the airport's baggage handling system. At the same time, the optimized route information is recorded on the blockchain, providing data reference for subsequent decision-making in similar situations, which helps the airport further optimize its operating strategy, reduce operating costs, and improve its ability to cope with large-scale luggage transportation during busy periods.

[0047] 4. Enhanced personalized service experience:

[0048] Meeting diverse passenger needs: The adaptive meta-learning algorithm achieves personalized adaptation of baggage tracking services based on the environmental characteristics of different airports (such as airport size, flight flow distribution, layout structure, etc.) and the personalized needs of passengers (such as the expected baggage status query frequency, the type of key information they pay attention to, etc.); passengers can customize baggage tracking services according to their own needs. For example, frequent business travelers may be more concerned about whether their luggage can arrive at the destination on time so that they can arrange subsequent trips reasonably. They can set a higher query frequency to obtain real-time baggage status; while leisure travelers may be more concerned about the safety of their luggage during transportation and hope to focus on whether there are any abnormalities with their luggage. By quantifying these demand characteristics and conducting meta-training, the system can provide different passengers with a baggage tracking service experience that meets their personalized expectations, greatly improving passenger satisfaction with airport services;

[0049] Flexible adaptation to changes in different scenarios: Whether it is the operational differences between different airports, or the changes in passenger flow and baggage handling needs faced by the same airport in different periods and seasons, the system of the present invention can adapt quickly; by conducting meta-training on multiple airport data sets to obtain adaptation meta-parameters, the system can use the real-time sample data of the airport to make fine-tuning and quickly adjust service strategies and configurations for new airport scenarios or changes in passenger demand; for example, during the peak tourist season, the airport passenger flow increases significantly, and the system can dynamically adjust the resource allocation and push frequency of the baggage tracking service according to the real-time situation, ensuring that stable and personalized services can still be provided to passengers under high load conditions, thereby enhancing the flexibility and adaptability of airport services and helping to improve the competitiveness of airports in different operating scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 This is a flowchart of a blockchain-based airport baggage tracking method of the present invention. DETAILED DESCRIPTION

[0051] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0052] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0053] like Figure 1 As shown, an embodiment of the present invention provides a blockchain-based airport baggage tracking method, comprising the following steps:

[0054] Step S1, baggage information collection and identification: when the baggage enters the airport baggage handling process, the basic information of the baggage is collected, identification information uniquely identifying the baggage is generated, and the basic information and identification information are associated and recorded in the blockchain network;

[0055] Step S1-1, information collection: When the luggage is checked in, the basic information of the luggage is collected with the help of the smart tag installed on the luggage. The basic information includes the physical attribute information of the luggage, itinerary related information and passenger related information. The physical attribute information of the luggage is weight. , size information length ,Width ,high , itinerary information including departure place ,destination , passenger-related information includes passenger identification code , integrating the above information to form a vector ;

[0056] Step S1-2, encryption processing: using symmetric encryption algorithm Pair Vector Encryption is performed, and the encryption operation is expressed as ,in, The encrypted information is obtained by distributing and storing encryption keys through a secure key management mechanism ;

[0057] Step S1-3, logo generation: Encrypted information Upload to the blockchain network and use the hash function , SHA-256 hash function pair Perform calculations to generate a unique identifier for the luggage in the blockchain network , following the formula , using this Uniquely refers to the baggage in the blockchain and associates all its status information and operation records;

[0058] Step S2, luggage status update: obtaining the luggage status change information in real time during the luggage circulation process in the airport, and updating the luggage status change information to the corresponding luggage record in the blockchain;

[0059] Step S2-1, status information collection: When the luggage is in the process of security inspection, boarding, transfer, and arrival at the airport, the IoT devices distributed at key locations are used to collect the luggage status change information in real time. The status change information includes location coordinate information, status category, and location information. And the corresponding timestamp , the position coordinate information is expressed in Cartesian coordinate system as ,in, Indicates the collection time. The values ​​of include normal, delayed, and suspected loss, denoted as , combining the above information into a state update vector ;

[0060] Step S2-2: Writing based on smart contracts: Write Write to the blockchain, the data verification rules defined in the smart contract Covers the verification logic of data format, source legitimacy, integrity, and timing rationality. satisfy When the preset write logic , involving block generation rules and chain structure addition rules, can successfully Write to the blockchain to ensure the real-time and unchangeable nature of the data. This process is described as follows: , then execute Operation will Write to blockchain;

[0061] Step S3, anomaly detection and prediction: Generative adversarial networks are used to simulate and analyze the luggage status to determine whether there is a potential anomaly in the luggage. Once an anomaly is determined, a corresponding alarm is triggered and the status record of the luggage in the blockchain is updated;

[0062] Step S3-1, Generative Adversarial Network (GAN) construction and training: Use the Generative Adversarial Network (GAN) to simulate and predict the luggage status. The GAN consists of a generator and the discriminator Composition, Generator By preset dimensions And obey uniform distribution , and is the random noise vector of the distribution interval boundary value , , as input, the operation is optimized and adjusted by its training effect, and the simulated baggage status data is output , and the output dimension is consistent with the actual collected state update vector Dimensionality consistent; Discriminator It is a neural network structure used to distinguish the real and generated , its loss function The definition is as follows: ;

[0063] in, Express expectations, is the distribution of real baggage status data, is a random noise vector The distribution of and parameters to minimize ,make The generated simulated data is close to the real data;

[0064] Step S3-2, abnormality determination and record update: During actual operation, compare the currently collected and , choose Euclidean distance as the similarity measurement index ,for and , is the vector dimension, and the Euclidean distance calculation formula is , set the similarity threshold ,when When the baggage is judged to have potential abnormality, the system automatically triggers an abnormality alarm and updates the rules. Based on the blockchain data update interface and format requirements, update the abnormal related information to the corresponding baggage record on the blockchain and execute Operation, exception-related information includes exception type and discovery time;

[0065] Step S4, route optimization: The luggage transportation route is optimized and adjusted according to the real-time status and historical data of the luggage through a reinforcement learning algorithm, and the optimized route information is recorded in the corresponding position of the blockchain;

[0066] Step S4-1, reinforcement learning RL model setting: Use reinforcement learning RL algorithm to realize intelligent optimization of baggage transportation path and define state space , whose elements are composed of the current location of the luggage, the transportation link it is in, and the status of surrounding transportation resources. A state in Expressed as ,in, Indicates location information. Indicates the transport link. Represents the state of surrounding transportation resources and action space , which includes the different transportation path adjustment actions that the baggage can choose, and the transfer to different conveyor belts is recorded as Waiting for the next batch of transfer is recorded as ,Right now , and the reward function , the reward function takes into account the transportation time , luggage congestion Possibility of delay , the specific form is ,in, are the corresponding weight coefficients and are all greater than 0. The weight coefficients are dynamically adjusted according to the actual operation conditions of the airport in different periods and areas;

[0067] Step S4-2: Path optimization based on Q-learning: The core of the Q-learning algorithm Value Function Indicates in status Take action The expected long-term return, The value update formula is: ,in, is the learning rate, ranging from between, is the discount factor, and its value range is between, To take action Then transfer to the next state, Indicates in status Next, for all possible actions The corresponding The value takes the maximum value and is updated through continuous iteration , in each state Next select The biggest move To optimize the luggage transportation path and follow the data recording rules of the blockchain Update the optimized path information to the corresponding baggage record in the blockchain, that is, execute operate;

[0068] Step S5, abnormal analysis and response: Analyze the baggage transportation path data, detect abnormal conditions that occur during transportation, take countermeasures according to preset strategies, and record the abnormal conditions and countermeasures taken in the blockchain;

[0069] Step S5-1, Graph Convolutional Network (GCN) construction and operation: Build a graph convolutional network (GCN) to analyze the abnormalities of the baggage transportation path and build a graph for the baggage transportation path in the airport. ,in, Represents a node set, which covers the boarding gates and baggage sorting areas in the airport. The number of nodes is , Represents the edge set, corresponding to the connection channels between each location point, and the number of edges is recorded as , each edge Association attribute vector ,in, Indicates the channel capacity, Indicates the current flow rate. Represents the average transportation time, and the inter-layer propagation formula of the graph convolutional network is as follows: ,in, , is the adjacency matrix, is the identity matrix, for The degree matrix of For the Layer feature matrix, For the The layer weight matrix, is the activation function, and the feature representation of each node is obtained through multi-layer graph convolution operation to analyze the changes of various parameters on the path;

[0070] Step S5-2, abnormal determination and response: set the normal threshold range corresponding to each attribute, such as the normal flow threshold corresponding to the channel capacity The reasonable range of average transportation time is , when the analysis finds that the edge Properties If the corresponding normal threshold value is exceeded, it is determined that an abnormal situation exists and the preset emergency response process is followed. Automatically take corresponding measures and follow the abnormal recording rules of the blockchain Update the abnormal situation and the measures taken to the corresponding baggage record on the blockchain, and execute operate;

[0071] Step S6, personalized service optimization: according to the characteristics of the airport and passengers, the adaptive meta-learning algorithm is used to realize the personalized adaptation of the baggage tracking service, and the relevant personalized configuration information is stored in the corresponding baggage record in the blockchain;

[0072] Step S6-1, feature quantification and meta-learning model: Using the adaptive meta-learning algorithm, the personalized adaptation of the baggage tracking service is realized according to the environmental characteristics of different airports and the personalized needs of passengers. The environmental characteristics include the size of the airport , Flight traffic distribution , layout structure , passengers’ personalized needs include the expected frequency of checking baggage status , the types of key information to focus on , quantify the airport environment characteristics into vectors , the personalized needs of passengers are quantified as vectors , the model-independent meta-learning MAML algorithm is used for meta-training and parameter update, and its meta-parameter update formula is: ,in, is the meta-learning rate, For the The test set loss function of the airport data set is used to obtain the adapted meta parameters by performing meta-training on multiple airport data sets. ;

[0073] Step S6-2, personalized service adaptation and record update: In response to new airport scenarios or changes in passenger demand, the sample data of the airport is used based on meta parameters Fine-tune and record rules according to the personalized configuration of blockchain Update the configuration information related to the personalized service to the corresponding baggage record in the blockchain, that is, execute operate.

[0074] An airport baggage tracking system based on blockchain, the airport baggage tracking system implements an airport baggage tracking method based on blockchain, the airport baggage tracking system includes:

[0075] Luggage collection module:

[0076] Smart tag: installed on luggage to collect basic information such as weight Dimensions (Length Width high Departure ,destination and Traveler Identification Number wait;

[0077] Information collection unit: When checking in luggage, the smart tag is triggered to collect information and integrate the information into a vector ;

[0078] Data encryption and identification module:

[0079] Encryption unit: using symmetric encryption algorithm , using a secure key management mechanism to distribute and store encryption keys , for vector Encrypt and get the encrypted information ;

[0080] Identification generation unit: using hash function (such as SHA-256 hash function) Perform calculations to generate a unique identifier for the luggage in the blockchain network , the formula is ; This identifier is used to uniquely identify the luggage in the blockchain and associate all its status information and operation records;

[0081] Baggage status update module:

[0082] IoT devices: distributed in key locations such as security check, boarding, transfer, and arrival in the airport, collecting baggage status change information in real time, including location coordinates Status Category (values ​​include normal, delayed, suspected lost) and corresponding timestamp , combined into a state update vector ;

[0083] Smart contracts: define data validation rules ,covers the verification logic of data format, source legitimacy, integrity, and temporal rationality; satisfy When the preset write logic (Involving block generation rules and chain structure addition rules), Write into the blockchain to ensure the real-time and tamper-proof nature of the data;

[0084] Anomaly detection and prediction module:

[0085] Generative Adversarial Networks (GANs):

[0086] Generator By preset dimensions And obey uniform distribution A random noise vector The input is the calculation optimized and adjusted after training effect, and the output is the simulated baggage status data , whose output dimension is consistent with the actual collected state update vector Dimensionality is consistent;

[0087] Discriminator Neural network structure for distinguishing real and generated , its loss function Defined as , during the training phase, by alternating optimization and The parameters make The generated simulated data is close to the real data;

[0088] Abnormal judgment and recording unit: During actual operation, compare the currently collected and , using Euclidean distance As a similarity measure ( , n is the vector dimension), set the similarity threshold ,when When the baggage is judged to have potential abnormality, the system automatically triggers an abnormality alarm and updates the rules. (Based on the blockchain data update interface and format requirements), update the abnormality-related information (abnormality type, discovery time) to the corresponding baggage record on the blockchain;

[0089] Path optimization module:

[0090] Reinforcement Learning (RL) Model:

[0091] State Space : Element by baggage current position , Transportation link , Status of surrounding transportation resources constitute, a state Expressed as ;

[0092] Action Space :Contains the different transportation path adjustment actions that baggage can choose, such as transferring to different conveyor belts Waiting for the next batch of transfer ,Right now ;

[0093] Reward Function :Comprehensive consideration of transportation time , luggage congestion Possibility of delay , the specific form is ,in, are the corresponding weight coefficients and are all greater than . The weight coefficients are dynamically adjusted according to the actual operation conditions of the airport in different periods and areas;

[0094] Q-learning algorithm unit: core Value Function Indicates in status Take action The expected long-term return, The value update formula is , is the learning rate, ranging from between; is the discount factor, and its value range is between; To take action The next state to transition to, Indicates in status Next, for all possible actions The corresponding The value takes the maximum value; through continuous iteration update , in each state Next select The biggest move To optimize the luggage transportation path and follow the data recording rules of the blockchain Update the optimized path information to the corresponding baggage record on the blockchain;

[0095] Abnormal analysis and response module:

[0096] Graph Convolutional Network (GCN):

[0097] Graph construction unit: Build a graph for the baggage transportation path within the airport ,in, Represents a node set, covering the boarding gates, baggage sorting areas, etc. in the airport. The number of nodes is recorded as ; Represents the edge set, corresponding to the connection channels between each location point, and the number of edges is recorded as ; Each edge Association attribute vector ,in, Indicates the channel capacity, Indicates the current flow represents the average transportation time; the inter-layer propagation formula of the graph convolutional network is ,in, , is the adjacency matrix, is the identity matrix, for The degree matrix of For the Layer feature matrix, For the The layer weight matrix, is the activation function, and the feature representation of each node is obtained through multi-layer graph convolution operations to analyze the changes of various parameters on the path;

[0098] Abnormal judgment and response unit: set the normal threshold range corresponding to each attribute, such as the normal flow threshold corresponding to the channel capacity The reasonable range of average transportation time is , when the analysis finds that the edge Properties If the corresponding normal threshold value is exceeded, it is determined that an abnormal situation exists and the preset emergency response process is followed. Automatically take corresponding measures and follow the abnormal recording rules of the blockchain Update the abnormal situation and the measures taken to the corresponding baggage record on the blockchain, and execute operate;

[0099] Personalized service optimization module:

[0100] Feature quantization and meta-learning model unit:

[0101] Feature quantification subunit: using the adaptive meta-learning algorithm, based on the size of the airport , flight traffic distribution , layout structure Environmental characteristics and the frequency of baggage status inquiries expected by passengers , the types of key information to focus on The airport environment characteristics are quantified into vectors based on personalized needs , the personalized needs of passengers are quantified as vector ;

[0102] Meta-learning subunit: The model-independent meta-learning MAML algorithm is used for meta-training and parameter update. The meta-parameter update formula is: (in, is the meta-learning rate, For the By conducting meta-training on multiple airport datasets, we can obtain the adapted meta-parameters. ;

[0103] Personalized service adaptation and recording unit: In response to new airport scenarios or changes in passenger demand, the sample data of the airport is used to adapt and record services based on meta parameters. Fine-tune and record rules according to the personalized configuration of blockchain Update the configuration information related to the personalized service to the corresponding baggage record in the blockchain.

[0104] The present invention will be described in detail below with reference to specific embodiments:

[0105] Example scenario setting:

[0106] Assume that in a busy international airport, there are a large number of flights taking off and landing every day, a large passenger flow, and a heavy baggage handling task; the airport hopes to improve the efficiency of baggage management, reduce baggage loss and delays, and improve the quality of passenger service by introducing the baggage tracking system of the present invention;

[0107] Baggage information collection and identification:

[0108] Passenger Xiao Li arrives at the airport to check in. He places his luggage on the conveyor belt at the check-in counter. The smart tag installed on the luggage automatically collects the weight of the luggage. kg, size information length cm, width cm, height Centimeters, and itinerary information Departure place ,destination Shanghai, and obtain the passenger-related information Xiao Li’s identity code ; This information is integrated into a vector ,Beijing,Shanghai,123456789];

[0109] The system uses a symmetric encryption algorithm , obtain encryption keys through a secure key management mechanism , for vector Encrypt and get the encrypted information ;

[0110] Encrypted Uploaded to the blockchain network, using the SHA-256 hash function Calculate it to generate a unique identifier for the luggage in the blockchain network ; Afterwards, this The luggage of Xiao Li will be uniquely identified in the blockchain and associated with all its subsequent status information and operation records;

[0111] Baggage status update:

[0112] Xiao Li's luggage enters the security check stage, and the IoT devices distributed in the security check area collect the luggage status change information in real time; at this time, the luggage location coordinate information is (Assume that the origin of the Cartesian coordinate system is located at a fixed point in the airport terminal), state category Normal, timestamp , combined into a state update vector ,normal, ;

[0113] The smart contract mechanism of the blockchain network receives After that, according to the defined data validation rules Verify it, including checking whether the data format is correct, whether the source is legal (i.e. whether it comes from an IoT device authorized by the airport), whether the information is complete, and whether the time sequence of the timestamp is reasonable; assuming satisfy , then according to the preset write logic ,Will Successfully written into the blockchain; this process ensures that the luggage status information can be recorded on the blockchain in real time, accurately and in an unalterable manner;

[0114] As the luggage goes through boarding, transfer and other stages, the IoT device continuously collects status information and updates it to the blockchain in the same process; for example, during the transfer process, the location coordinates of the luggage become , the status category remains , the timestamp is , the corresponding state update vector Written into the blockchain;

[0115] Anomaly Detection and Prediction:

[0116] During the luggage transportation process, the system uses the pre-trained generative adversarial network (GAN) to detect and predict anomalies; the generator in GAN By preset dimensions And obey uniform distribution A random noise vector As input, after training and adjustment, simulated baggage status data is output , whose output dimension is consistent with the actual collected state update vector Same; Discriminator It is used to distinguish the real and generated ; During the training phase, a large amount of real luggage status data is used to continuously optimize and Parameters such that The generated simulated data is as close to the real data as possible, thus improving the accuracy of anomaly detection;

[0117] When the baggage arrives at a key node, the system compares the current collected status update vector and ; Assuming that the currently collected ,normal, , calculate its Euclidean distance , set the similarity threshold ,if , it is determined that the luggage has potential abnormality;

[0118] Assuming that the calculation result meets the abnormal judgment conditions, the system automatically triggers an abnormal alarm and notifies airport staff to pay attention; at the same time, according to the update rules , and send the abnormal related information (such as abnormal type, possible delay, discovery time ) is updated to the corresponding luggage record on the blockchain;

[0119] Path Optimization:

[0120] The airport uses reinforcement learning (RL) algorithm to achieve intelligent optimization of baggage transportation paths; define the state space , where the baggage's current location (Corresponding to the above location information in the Transportation link , surrounding transportation resource status , then the current state ; Action Space ;

[0121] Reward Function Considering the transportation time , luggage congestion Possibility of delay , assuming the current weight coefficient ; For example, if you select action , estimated shipping time Minutes (estimated based on historical data and real-time monitoring), baggage congestion (indicates congestion), possibility of delay , then reward ;

[0122] The learning algorithm is based on Value Function To select the optimal action; initially, The value is randomly initialized and then iteratively updated during the baggage transportation process; for example, the current , , according to the formula (Assuming the learning rate , discount factor ), Indicates in status Next, for all possible actions The corresponding The value is taken as the maximum value, and the calculation is performed after selecting different actions. Value update; after calculation and comparison, select The action with the largest value is used to optimize the luggage transportation path; assuming that the action is selected this time , the luggage is transferred to another conveyor belt and recorded according to the data rules of the blockchain Update the optimized path information to the corresponding baggage record in the blockchain;

[0123] Abnormal analysis and response:

[0124] Construct a graph convolutional network (GCN) to analyze abnormalities in baggage transportation routes; construct a graph for baggage transportation routes within the airport , where the node set Including boarding gate and luggage separation area etc., number of nodes ; Edge set Indicates the connection channels between various location points, such as Indicates the boarding gate and The number of channels and edges between ; Each edge Association attribute vector ,like express The channel capacity is 100 pieces hour, the current flow is 80 items hours, with an average transport time of 15 minutes;

[0125] Set the normal flow threshold corresponding to the channel capacity , the reasonable range of average transportation time is minutes; when the luggage is in transit, the system uses GCN to analyze the attribute changes of each edge on the path; assuming that the monitoring Current traffic Suddenly increased to 110 (exceeding the normal threshold ) or average shipping time Changed to 25 minutes (beyond the reasonable range ), it is determined that there is an abnormal situation;

[0126] Once an abnormality is determined, the preset emergency response process , airport staff will receive notifications and take corresponding measures, such as adjusting baggage transportation routes, increasing manpower to handle congestion, etc. At the same time, according to the abnormal recording rules of the blockchain The abnormal situation (such as channel congestion) and measures taken (such as adjusting luggage to alternate channels ) is updated to the corresponding baggage record on the blockchain;

[0127] Personalized service optimization:

[0128] The adaptive meta-learning algorithm is used to realize the personalized adaptation of the baggage tracking service; for this airport, its environmental characteristics are quantified as vector , personalized demand of passenger Xiao Li Convert to vector ;

[0129] The model-independent meta-learning MAML algorithm is used for meta-training and parameter updating. By collecting historical data sets from multiple airports, meta-training is carried out on these data sets to obtain the adapted meta-parameters. ; For example, the trained It may include parameters such as the frequency of baggage status push notifications and the display priority of key information according to different passenger needs;

[0130] When Xiao Li is waiting at the airport, the system Provide him with personalized services; for example, according to his expected high-frequency query needs, the latest location and status information of the luggage is automatically pushed to his mobile app every 15 minutes; if Xiao Li's needs change, such as he suddenly pays more attention to the transportation route planning of the luggage, the system uses the real-time sample data of the airport based on the meta parameters Fine-tune and adjust the service configuration to meet his new needs, and follow the personalized configuration record rules of the blockchain Update the relevant configuration information to the corresponding baggage record in the blockchain.

[0131] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A blockchain-based airport baggage tracking method, characterized by: The following steps are involved: Step S1, baggage information collection and identification: when the baggage enters the airport baggage handling process, the basic information of the baggage is collected, identification information uniquely identifying the baggage is generated, and the basic information and identification information are associated and recorded in the blockchain network; Step S2, luggage status update: obtaining the luggage status change information in real time during the luggage circulation process in the airport, and updating the luggage status change information to the corresponding luggage record in the blockchain; Step S3, anomaly detection and prediction: Generative adversarial networks are used to simulate and analyze the luggage status to determine whether there is a potential anomaly in the luggage. Once an anomaly is determined, a corresponding alarm is triggered and the status record of the luggage in the blockchain is updated; Step S4, route optimization: The luggage transportation route is optimized and adjusted according to the real-time status and historical data of the luggage through a reinforcement learning algorithm, and the optimized route information is recorded in the corresponding position of the blockchain; Step S5, abnormal analysis and response: Analyze the baggage transportation path data, detect abnormal conditions that occur during transportation, take countermeasures according to preset strategies, and record the abnormal conditions and countermeasures taken in the blockchain; Step S6, personalized service optimization: Based on the characteristics of the airport and passengers, an adaptive meta-learning algorithm is used to achieve personalized adaptation of the baggage tracking service, and the relevant personalized configuration information is stored in the corresponding baggage record in the blockchain.

2. The blockchain-based airport baggage tracking method according to claim 1, characterized in that: The baggage information collection and identification includes: Step S1-1, information collection: When the luggage is checked in, the basic information of the luggage is collected with the help of the smart tag installed on the luggage. The basic information includes the physical attribute information of the luggage, itinerary related information and passenger related information. The physical attribute information of the luggage is weight. , size information length ,Width ,high , itinerary information including departure place ,destination , passenger-related information includes passenger identification code , integrating the above information to form a vector ; Step S1-2, encryption processing: using symmetric encryption algorithm Pair Vector Encryption is performed, and the encryption operation is expressed as ,in, The encrypted information is obtained by distributing and storing encryption keys through a secure key management mechanism ; Step S1-3, logo generation: Encrypted information Upload to the blockchain network and use the hash function , SHA-256 hash function pair Perform calculations to generate a unique identifier for the luggage in the blockchain network , following the formula , using this The luggage is uniquely identified in the blockchain and is associated with all its status information and operation records.

3. The blockchain-based airport baggage tracking method according to claim 1, characterized in that: The operation of updating the baggage status includes: Step S2-1, status information collection: When the luggage is in the process of security inspection, boarding, transfer, and arrival at the airport, the status change information of the luggage is collected in real time using IoT devices distributed at key locations. The status change information includes location coordinate information, status category, and And the corresponding timestamp , the position coordinate information is expressed in Cartesian coordinate system as ,in, Indicates the collection time. The values ​​of include normal, delayed, and suspected loss, denoted as , combining the above information into a state update vector ; Step S2-2: Writing based on smart contracts: Write Write to the blockchain, the data verification rules defined in the smart contract Covers the verification logic of data format, source legitimacy, integrity, and timing rationality. satisfy When the preset write logic , involving block generation rules and chain structure addition rules, can successfully Write to the blockchain to ensure the real-time and unchangeable nature of the data. This process is described as follows: , then execute Operation will Write to the blockchain.

4. The blockchain-based airport baggage tracking method according to claim 1, characterized in that: The anomaly detection and prediction operations include: Step S3-1, Generative Adversarial Network (GAN) construction and training: Use the Generative Adversarial Network (GAN) to simulate and predict the luggage status. The GAN consists of a generator and the discriminator Composition, Generator By preset dimensions And obey uniform distribution , and is the random noise vector of the distribution interval boundary value , , as input, the operation is optimized and adjusted by its training effect, and the simulated baggage status data is output , and the output dimension is consistent with the actual collected state update vector Dimensionality consistent; Discriminator It is a neural network structure used to distinguish the real and generated , its loss function The definition is as follows: ； in, Express expectations, is the distribution of real baggage status data, is a random noise vector The distribution of and parameters to minimize ,make The generated simulated data is close to the real data; Step S3-2, abnormality determination and record update: During actual operation, compare the currently collected and , choose Euclidean distance as the similarity measurement index ,for and , is the vector dimension, and the Euclidean distance calculation formula is , set the similarity threshold ,when When the baggage is judged to have potential abnormality, the system automatically triggers an abnormality alarm and updates the rules. Based on the blockchain data update interface and format requirements, update the abnormal related information to the corresponding baggage record on the blockchain and execute Operation, exception-related information includes exception type and discovery time.

5. The blockchain-based airport baggage tracking method according to claim 1, characterized in that: The path optimization operation includes: Step S4-1, reinforcement learning RL model setting: Use reinforcement learning RL algorithm to realize intelligent optimization of baggage transportation path and define state space , whose elements are composed of the current location of the luggage, the transportation link it is in, and the status of surrounding transportation resources. A state in Expressed as ,in, Indicates location information. Indicates the transport link. Represents the state of surrounding transportation resources and action space , which includes the different transportation path adjustment actions that the baggage can choose, and the transfer to different conveyor belts is recorded as Waiting for the next batch of transfer is recorded as ,Right now , and the reward function , the reward function takes into account the transportation time , luggage congestion Possibility of delay , the specific form is ,in, are the corresponding weight coefficients and are all greater than 0. The weight coefficients are dynamically adjusted according to the actual operation conditions of the airport in different periods and areas; Step S4-2: Path optimization based on Q-learning: The core of the Q-learning algorithm Value Function Indicates in status Take action The expected long-term return, The value update formula is: ,in, is the learning rate, ranging from between, is the discount factor, and its value range is between, To take action Then transfer to the next state, Indicates in status Next, for all possible actions The corresponding The value takes the maximum value and is updated through continuous iteration , in each state Next select The biggest move To optimize the luggage transportation path and follow the data recording rules of the blockchain Update the optimized path information to the corresponding baggage record in the blockchain, that is, execute operate.

6. The blockchain-based airport baggage tracking method according to claim 1, characterized in that: The abnormal analysis and response operations include: Step S5-1, Graph Convolutional Network (GCN) construction and operation: Build a graph convolutional network (GCN) to analyze the abnormalities of the baggage transportation path and build a graph for the baggage transportation path in the airport. ,in, Represents a node set, which covers the boarding gates and baggage sorting areas in the airport. The number of nodes is , Represents the edge set, corresponding to the connection channels between each location point, and the number of edges is recorded as , each edge Association attribute vector ,in, Indicates the channel capacity, Indicates the current flow rate. Represents the average transportation time, and the inter-layer propagation formula of the graph convolutional network is as follows: ,in, , is the adjacency matrix, is the identity matrix, for The degree matrix of For the Layer feature matrix, For the The layer weight matrix, is the activation function, and the feature representation of each node is obtained through multi-layer graph convolution operation to analyze the changes of various parameters on the path; Step S5-2, abnormal determination and response: set the normal threshold range corresponding to each attribute, such as the normal flow threshold corresponding to the channel capacity The reasonable range of average transportation time is , when the analysis finds that the edge Properties If the corresponding normal threshold value is exceeded, it is determined that an abnormal situation exists and the preset emergency response process is followed. Automatically take corresponding measures and follow the abnormal recording rules of the blockchain Update the abnormal situation and the measures taken to the corresponding baggage record on the blockchain, and execute operate.

7. The blockchain-based airport baggage tracking method according to claim 1, characterized in that: The personalized service optimization operation includes: Step S6-1, feature quantification and meta-learning model: Using the adaptive meta-learning algorithm, the personalized adaptation of the baggage tracking service is realized according to the environmental characteristics of different airports and the personalized needs of passengers. The environmental characteristics include the size of the airport , Flight traffic distribution , layout structure , passengers’ personalized needs include the expected frequency of checking baggage status , the types of key information to focus on , quantify the airport environment characteristics into vectors , the personalized needs of passengers are quantified as vectors , the model-independent meta-learning MAML algorithm is used for meta-training and parameter update, and its meta-parameter update formula is: ,in, is the meta-learning rate, For the The test set loss function of the airport data set is used to obtain the adapted meta parameters by performing meta-training on multiple airport data sets. ; Step S6-2, personalized service adaptation and record update: In response to new airport scenarios or changes in passenger demand, the sample data of the airport is used based on meta parameters Fine-tune and record rules according to the personalized configuration of blockchain Update the configuration information related to the personalized service to the corresponding baggage record in the blockchain, that is, execute operate.

8. A blockchain-based airport baggage tracking system, characterized by: The airport baggage tracking system implements the blockchain-based airport baggage tracking method described in any one of claims 1-7.

Images