Block chain-based shared unmanned aerial vehicle distributed traceability management system

Abstract

The invention provides a shared unmanned aerial vehicle distributed traceability management system based on a block chain. The system comprises a task data acquisition module, a cooperative trust graph construction module, a trust-task coupling enhancement module, a trust-driven dynamic consensus module and a joint traceability module. According to the method, the problems of credibility deficiency in an existing centralized system, insufficient dynamic adaptability based on a block chain system, weak privacy protection and the like are solved, and the cooperation credibility, the traceability and the privacy guarantee capability of the shared unmanned aerial vehicle in a multi-user, multi-organization and multi-task scene are improved.

Description

Technical Field

[0001] The present invention belongs to the field of blockchain, and particularly relates to a shared drone distributed traceability management system based on blockchain. Background Art

[0002] With the rapid development of the sharing economy, drones have gradually penetrated from traditional military, surveying and mapping, and logistics fields into a new business model of multi-party shared use, forming an application trend of "shared drones". In this mode, drones are dynamically rented and collaboratively used by different users, organizations or platforms in various business scenarios to serve complex and changing task requirements such as urban logistics, security patrols, and agricultural monitoring. The wide application of shared drones not only effectively improves resource utilization and reduces operating costs, but also brings unprecedented management challenges. Especially in the context of multi-party use and cross-organization cooperation, the credibility and traceability of the task execution process, data collection process, and collaboration process of shared drones have become key pain points.

[0003] Existing shared drone management systems usually adopt a centralized architecture or a semi-distributed architecture, and a single platform is used to allocate tasks, collect flight data, and manage logs for drones. However, this centralized management mode has obvious limitations. First, in a centralized system, all flight data and task logs of drones are centrally stored in the management platform, which has the risks of single-point failure, data tampering, and loss, and it is difficult to ensure data integrity and credibility during the task execution process. Second, existing systems mostly adopt simple log storage and basic encryption mechanisms, lacking the ability to dynamically model and credibly record the complex collaboration relationships between drones, and unable to truly reflect the collaboration behaviors and task dependencies of multiple drones in a shared environment. Especially in the scenario where multiple parties share drone resources and multiple organizations or users jointly schedule a group of drones to execute tasks, the collaboration relationships and trust relationships lack a transparent and verifiable mechanism, resulting in difficulties in subsequent responsibility tracing, behavior auditing, and collaboration optimization.

[0004] To overcome the deficiencies in security and credibility brought about by centralization, some research and systems have begun to introduce blockchain technology, leveraging its features such as distributed ledger, immutability, and full-process recording to ensure the traceability of the drone flight and mission execution processes. However, such blockchain-based shared drone traceability management systems still have deficiencies. Firstly, existing blockchain-based solutions usually focus on simple data uploading to the blockchain and the execution of smart contracts, lacking dynamic and intelligent description and trustworthy management of the collaborative relationships among drones. Especially in an environment where drone cluster tasks change frequently and the entry and exit of nodes are dynamically complex, it is difficult to meet the efficient and real-time traceability requirements. Secondly, most current blockchain solutions adopt fixed consensus mechanisms and fail to make intelligent and adaptive adjustments in response to changes in the number of nodes, trust relationships, and collaboration modes in the shared drone scenario, which easily leads to inefficient consensus processes or inability to maintain consistency in a dynamic environment, reducing the real-time performance and stability of the traceability system. In addition, in a shared environment, drone collaborative tasks involving multiple parties are often accompanied by risks such as data privacy and leakage of sensitive mission information. Existing systems lack a privacy-friendly trust management mechanism and it is difficult to ensure the openness, transparency, and trustworthy verification of the mission process while protecting the privacy of mission collaboration. To address the above problems, the prior art has not proposed a distributed solution for shared drones. Summary of the Invention

[0005] The objective of the present invention is to propose a blockchain-based distributed traceability management system for shared drones, which establishes a unified technical framework, solves problems such as the lack of credibility in existing centralized systems, insufficient dynamic adaptability, and weak privacy protection in blockchain-based systems, and improves the collaborative credibility, traceability ability, and privacy protection ability of shared drones in multi-user, multi-organization, and multi-task scenarios.

[0006] To achieve the above objective, the present invention provides a blockchain-based distributed traceability management system for shared drones, and the system includes the following modules:

[0007] A mission data collection module, which is used to collect drone collaboration data and generate a standardized data set; the drone collaboration data includes position coordinates, the original encoding of the current collaboration action, flight speed and battery level, timestamp, and mission number.

[0008] A collaborative trust graph construction module, which is used to map each data in the standardized data set to a node in the graph to construct a collaborative trust graph reflecting the collaborative relationships among drones, perform correlation calculation and analysis based on the collaborative trust graph to generate trust edge weights, and perform weighted fusion of the action behaviors of adjacent nodes and the trust edge weights to generate node trust embeddings, which are used to reflect the trust in the drone collaboration environment and behavior patterns.

[0009] A trust - task coupling enhancement module, which is used to dynamically enhance the node trust embedding in the trust graph by combining corresponding task features to obtain enhanced trust embedding and input it into the trust - driven dynamic consensus module;

[0010] A trust - driven dynamic consensus module, which is used to dynamically determine the consensus node set of the collaborative trust graph based on the enhanced trust embedding and assign corresponding task - aware consensus weights;

[0011] A joint traceability module, which is used to generate a multi - dimensional traceability data structure according to the consensus nodes, the types and standardized collaborative actions of the tasks executed by the corresponding nodes, the corresponding node trust embeddings, the weighted degrees of the corresponding nodes in the trust graph, the set of adjacent nodes of the corresponding nodes, and the set of edge weights between the corresponding nodes and their neighbors.

[0012] Preferably, the task data acquisition module further includes:

[0013] Mapping the original encoding and task number of the current collaborative action to the type and standardized collaborative action of the task executed by the corresponding node;

[0014] And within a preset time range, only retain the latest valid data of each drone, remove redundancy, and reduce the noise introduced by time resampling during graph modeling.

[0015] Preferably, calculating and analyzing the association degree based on the collaborative trust graph to generate trust edge weights specifically includes:

[0016] For any node pair (v i , v j ), use the association degree calculation and analysis to calculate its trust edge weight e i,j as follows:

[0017]

[0018] Among them, and are the standardized collaborative action vectors of v i and v j ; Sim(·) is the cosine similarity of the action vectors; ||pos i - pos j || is the Euclidean distance of the node positions, and Exp(-||·||) is used as the spatial trust decay factor; θ task is the task type perception factor, which is determined by the type of the task executed by the corresponding node;

[0019] According to the trust edge weight e i,j and the standardized collaborative actions of its adjacent nodes, perform weighted aggregation to generate a trust embedding reflecting the drone collaborative environment and behavior patterns.

[0020] Preferably, the trust-task coupling enhancement module specifically performs the following:

[0021] Based on the types of execution tasks in the standardized dataset and the built-in task feature table of the system, generate corresponding task feature vectors for each node, including: task risk level and task collaboration density prediction;

[0022] According to the task risk level R i and the task collaboration density prediction D i Enhance the trust embedding w i · to obtain an enhanced trust embedding Expressed as:

[0023]

[0024] Where γ1, γ2 are coupling adjustment factors, and the system dynamically configures them according to different task types;

[0025] Among them, in the high-risk and sparse collaboration task scenario, R i and 1 - D i Are both relatively large, so that Obtains a more significant enhancement, improving the system's trust response to specific task environments;

[0026] In the dense collaboration and low-risk task scenario, The enhancement effect is weaker, maintaining the original distribution rationality of the trust graph. Preferably, the conditions for the consensus node set are as follows:

[0027] Condition 1: Among them, Is the enhanced trust embedding, τ task Is the dynamic trust threshold of task perception, and the system automatically adjusts it according to task_type i Logistics task: τ task = 0.5; Security patrol task: τ task = 0.7;

[0028] Condition 2: The collaboration influence of the node is not lower than the minimum collaboration threshold κ min , κ i ≥κ min , κ i Is the weighted degree of the node in the trust graph, where κ i = ∑ j∈N(i) e i,j , j is the adjacent node, N(i) is the set of adjacent nodes, and e i,j Is the trust edge weight from node i to node j.

[0029] Preferably, perform normalized weight analysis on the enhanced trust embedding to obtain the task perception consensus weights of each consensus node.

[0030] Preferably, the task risk level includes low risk and high risk; the task collaboration density is estimated to be 0-1, where the lower the value, the sparser the collaboration of the drones under the task, and the more important the credibility of a single node.

[0031] Preferably, in the joint traceability module, a credible behavior factor θ is designed i , which is used to comprehensively characterize the collaboration credibility of v i ; the credible behavior factor θ i is expressed as:

[0032]

[0033] where, |E i | is the sum of the trust edge weights of v i ; |N(i)| is the number of adjacent nodes; η is a collaboration density sensitive factor; θ i is a composite credibility index used to reflect the trust value + the strength of the collaboration relationship.

[0034] The beneficial technical effects of the present invention are at least as follows:

[0035] The present invention proposes a distributed traceability management system for a shared drone environment, which focuses on solving the deficiencies of the prior art in aspects such as credible collaboration, dynamic traceability, and privacy protection. The present invention innovatively establishes a "dynamic trust collaboration management mechanism" for a shared drone cluster, which can real-time sense and model the collaboration relationship and trust relationship between drones, and dynamically drive the consensus process of the distributed ledger system as a core parameter. By introducing a closed-loop mechanism of "collaboration-trust-traceability" within the system, it not only ensures the collaboration transparency and credible traceability ability of drones in a multi-party shared environment, but also effectively alleviates problems such as low consensus efficiency and discontinuous traceability chains caused by dynamic environmental changes. In addition, the present invention also designs a "privacy-friendly collaboration and trust management mechanism" around the sensitive information involved in drone sharing tasks, realizing the privacy protection of task data and collaboration data, preventing the leakage of sensitive collaboration relationships, and at the same time ensuring that external users and multi-party administrators can complete complete and verifiable traceability analysis based on the system.

[0036] The present invention focuses on the core issue of "distributed traceability management in a shared drone environment", establishes a unified technical framework, solves problems such as the lack of credibility in existing centralized systems, insufficient dynamic adaptability and weak privacy protection in blockchain-based systems, and improves the collaboration credibility, traceability ability, and privacy protection ability of shared drones in multi-user, multi-organization, and multi-task scenarios. Description of the Drawings

[0037] The present invention will be further described with reference to the accompanying drawings. However, the embodiments in the drawings do not constitute any limitation to the present invention. For those of ordinary skill in the art, other drawings can also be obtained based on the following drawings without creative efforts.

[0038] Figure 1 This is a framework diagram of the blockchain-based shared drone distributed traceability management system of the present invention. Detailed implementation manners

[0039] The embodiments of the present invention will be described in detail below. The examples of the embodiments are shown in the accompanying drawings, where the same or similar reference numerals denote the same or similar elements or elements with the same or similar functions from beginning to end. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention and should not be construed as a limitation to the present invention.

[0040] As Figure 1 shown, the blockchain-based shared drone distributed traceability management system provided by the embodiment of the present invention includes:

[0041] A task data acquisition module 1, configured to collect drone collaboration data and generate a standardized data set; the drone collaboration data includes position coordinates, the original encoding of the current collaboration action, flight speed and power, timestamp, and task number.

[0042] Specifically, the following operations are performed:

[0043] In the shared drone environment, this module aims to complete the distributed acquisition and standardization of collaboration behavior data, ensuring that high-consistency and low-redundancy input data D is provided for trust graph modeling, serving the trust modeling and distributed traceability objectives of the overall patent system.

[0044] Each drone node i in the shared drone cluster, during the execution of the collaboration task, based on the local sensor and communication module, collects collaboration data in real time and generates a standardized data packet d i , which contains the following fields:

[0045] pos i = [x i , y i , z i : Position coordinates;

[0046] act i : The original encoding of the current collaboration action (such as "hand over supplies", "cooperate in cruising", etc.);

[0047] state i = [v i , b i : Flight speed and power;

[0048] t i : Timestamp;

[0049] task i : Task number.

[0050] Due to the heterogeneous characteristics of multi-tenancy, multi-task, and multi-platform in the shared drone environment, after data collection, the system starts the "task type standardization mechanism" for task i and maps task i to task_type i (such as logistics, inspection, security, etc.), and calls the corresponding "collaborative action standard dictionary " to map act i to the unified action encoding within the task

[0051] The system then applies the time window redundancy removal mechanism. Within the time range t ± Δt, only the latest valid data of each drone is retained, removing redundancy and reducing the noise introduced by time resampling during graph modeling.

[0052] The finally output standardized collaborative dataset D has the structure:

[0053]

[0054] Combined to form:

[0055] D = {d1, d2,..., d n}

[0056] This dataset D is used as the input for constructing the trust graph G(V, E) and is directly used in the subsequent "trust relationship determination and edge generation" process to complete the first step of trust graph modeling.

[0057] The collaborative trust graph construction module 2 is used to map each data in the standardized dataset to a node in the graph to construct a collaborative trust graph reflecting the collaborative relationship of drones, calculate and analyze the association degree based on the collaborative trust graph, generate the trust edge weight, and perform weighted fusion on the action behavior of adjacent nodes and the trust edge weight to generate the node trust embedding for reflecting the trust in the drone collaborative environment and behavior pattern.

[0058] Specifically, it executes as follows:

[0059] This module receives the standardized collaborative behavior dataset D = {d1, d2,..., d n} output by module 1, constructs a trust graph G(V, E) reflecting the collaborative relationship of the shared drone cluster, and generates the trust embedding {w i}, as the input for the subsequent dynamic consensus mechanism. The data D already contains the structured collaboration behavior records of each drone. This module directly maps it to the node v in the graph. i , forming a node set V = {v1, v2,..., v n}}. Subsequently, the system establishes the edge set E of the graph based on the "task - collaboration - space coupling rule".

[0060] For the characteristics of diverse, cross - organizational, and complex spatial relationships of shared drone collaboration tasks, the system proposes a "Dynamic Collaboration - Space Coupling Edge Weight Mechanism (DCSE)" for the generation of the trust edge weight e i,j . For any node pair (v i , v j ), the system calculates its trust edge weight using the correlation degree calculation and analysis as follows:

[0061]

[0062] Variable definitions:

[0063] and are the normalized collaboration action vectors of v i and v j ;

[0064] Sim(·) is the cosine similarity of the action vectors;

[0065] ||pos i - pos j || is the Euclidean distance of the node positions, and Exp(-||·||) is used as the spatial trust decay factor;

[0066] θ task is the "task type perception factor", determined by task_type, for example:

[0067] Logistics task: θ task = 0.7, with a higher weight for action similarity;

[0068] Inspection task: θ task = 0.4, with a higher weight for position adjacency.

[0069] Through this mechanism, the system ensures that the collaboration trust graph not only captures the collaboration behavior similarity but also adapts to the influence of spatial relationships according to the task characteristics, conforming to the actual scenario of shared drone task collaboration.

[0070] After completing the construction of the graph G(V, E), the system designs a "Neighbor Collaboration Aggregation Embedding Mechanism (NCAA)" based on the trust edge relationship to generate a trust embedding w for each node v i ​i , as an indication of the collaborative credibility for its subsequent participation in the consensus and traceability processes.

[0071] The trust embedding generation formula is as follows:

[0072]

[0073] Variable definition:

[0074] N(i) is the set of adjacent nodes of v i ;

[0075] e i,j is the trust edge weight;

[0076] is the standard action of the neighbor node;

[0077] σ(·) is the activation function (such as ReLU) to ensure that the trust embedding is non - negative.

[0078] This embedding mechanism innovatively performs weighted aggregation on "the action behavior of adjacent nodes" and "the strength of trust edges" to generate the trust embedding w i , which directly serves "consensus node screening" and "consensus weight assignment" in the subsequent dynamic consensus mechanism.

[0079] The trust - task coupling enhancement module 3 is used to dynamically enhance the node trust embedding in the trust graph by combining the corresponding task characteristics, obtain the enhanced trust embedding, and input it into the trust - driven dynamic consensus module. Specifically,

[0080] Execute as follows:

[0081] This module takes the trust graph G(V,E) generated by module 2 and the basic trust embedding {w i} as inputs. For the dynamic collaboration characteristics of multi - task scenarios in the shared UAV environment, it proposes a "trust - task coupling mechanism" to complete the task - aware enhancement of the trust embedding on the basis of the trust graph and generate as the input for the subsequent dynamic consensus mechanism.

[0082] The tasks of shared UAVs are diverse, including various task types such as logistics, inspection, and security. Moreover, the collaborative density and risk sensitivity of the tasks vary greatly, resulting in that a single w i is difficult to accurately reflect the true trust needs in different task scenarios. To solve this problem, this module proposes a "Task - Dynamic - Characteristic - Aware Trust Enhancement Mechanism (TDC - TEM)" to improve the sensitivity of the trust embedding to task dynamic characteristics.

[0083] First, the system is based on task_type in D i(Obtained from Module 1) and the task feature table built into the system, for each node v i Generate a task feature vector T i =[[R i ,[[D i , where:

[0084] R i is the task risk level (pre-defined by the system, e.g., 0.1 for low risk, 0.9 for high risk);

[0085] D i is the estimated task collaboration density (0 - 1, the lower the value, the sparser the collaboration of UAVs under the task, and the more important the credibility of a single node).

[0086] Subsequently, design an innovative trust - task coupling model to dynamically enhance w i :

[0087]

[0088] where γ1 and γ2 are coupling adjustment factors, and are dynamically configured by the system according to different task types. For example:

[0089] For logistics tasks, γ1 = 0.2, γ2 = 0.3; for security tasks, γ1 = 0.5, γ2 = 0.2;

[0090] R i and[[D i The introduction endows the trust embedding with the perception ability for "high - risk / low - collaboration - density" tasks.

[0091] The core innovation of this mechanism lies in task - feature sensitivity:

[0092] In high - risk and sparse - collaboration task scenarios (such as inspection tasks),[[R i and 1 - D i are both relatively large, making obtain more significant enhancement and improving the system's trust response to specific task environments;

[0093] In dense - collaboration and low - risk tasks (such as regular logistics collaboration), the enhancement effect is weaker, maintaining the original distribution rationality of the trust graph.

[0094] Finally, the system outputs the enhanced trust embedding for subsequent Module 4 as the direct input for consensus - node screening and consensus - weight allocation.

[0095] Illustrative example:

[0096] U1 and U2 are both involved in high-risk and sparse-collaboration security tasks. The original values are w1 = 0.6, R1 = 0.9, D1 = 0.3, and the security task configuration is γ1 = 0.5, γ2 = 0.2. After calculation, we get It is significantly enhanced compared to the original trust value, reflecting the dynamic adjustment effect of the task scenario.

[0097] The trust-driven dynamic consensus module 4 is used to dynamically determine the consensus node set of the collaborative trust graph based on the enhanced trust embedding and assign corresponding task-aware consensus weights.

[0098] Specifically, the following steps are executed:

[0099] This module is based on the trust graph G(V,E) generated by module 3 and the task-aware trust embedding Design a trust-driven dynamic consensus control mechanism (TDC-CM) to dynamically screen out the node set C participating in distributed consensus and assign task-aware consensus weights {ω i} as the input for the subsequent consensus process.

[0100] Shared drone systems usually have the characteristics of dynamic topology, uneven collaborative distribution, and diverse task types. The traditional static "fixed consensus nodes + fixed weights" mechanism is difficult to adapt to the real-time changing trust situation in the shared task environment. To solve this problem, this module designs a "trust-graph structure fusion mechanism" with task dynamic perception to dynamically screen consensus nodes and assign weights.

[0101] Dynamic screening of consensus nodes

[0102] The system is based on G(V,E) to dynamically determine the consensus node set C. When the node v i meets the following conditions, it joins the consensus candidate set C:

[0103] Condition 1: where τ task is the task-aware dynamic trust threshold, which is automatically adjusted by the system according to task_type i :

[0104] Logistics task: τ task = 0.5

[0105] Security patrol task: τ task = 0.7

[0106] Condition 2: The collaborative influence of the node is not lower than the minimum collaboration threshold κ min , κ i ≥ κ min , where κ i = ∑j∈N(i) e i,j 。

[0107] Consensus weight allocation mechanism

[0108] Inside C, the system designs a weight allocation mechanism of "trust-topology coupling", combines the trust distribution with the collaborative topology influence, and outputs {ω i}.

[0109]

[0110] Variable definition:

[0111] : Trust value of task perception;

[0112] κ i : Collaborative weighted degree of the node;

[0113] α∈[1,2]: Controls the sensitivity of the topology influence to the weight distribution;

[0114] ∑ i∈C ω i = 1, satisfying weight normalization.

[0115] Output content

[0116] {C}: The set of consensus candidate nodes dynamically screened;

[0117] {ω i}: Task perception consensus weights of each consensus node.

[0118] For example, in the case of a logistics task, the trust values of U1, U2, and U3 are respectively The system configures τ task = 0.5:

[0119] U1 and U2 meet the screening conditions, and U3 is excluded due to being lower;

[0120] Subsequently, the system calculates ω i according to κ i , and outputs ω U1 = 0.4, ω U2 = 0.6, which is directly passed to module 5 for the actual consensus process.

[0121] The joint traceability module 5 is used to generate a multi-dimensional traceability data structure based on the consensus nodes, the types and standardized collaborative actions of the corresponding nodes performing tasks, the trust embedding of the corresponding nodes, the weighted degrees of the corresponding nodes in the trust graph, the set of adjacent nodes of the corresponding nodes, and the set of edge weights between the corresponding nodes and their neighbors.

[0122] Specifically, it executes as follows:

[0123] In the shared drone system, based on the existing trust graph G(V, E), collaborative behavior data D, task-aware trust embedding and the dynamic consensus control results {C, ω i}, the "Trusted Collaboration-Trust-Task Joint Tracing Model (TTT-Trace)" is proposed to complete the modeling of the multi-dimensional tracing data structure R, providing basic data for subsequent on-chain consensus and tracing verification. In a multi-task and multi-tenant environment, the collaborative process of shared drones is decentralized, the trust scores are dynamic, and the behavior data and trust data are usually separated, making it difficult to trace the three-dimensional causal relationship of "task - collaboration - trust" of drone behavior.

[0124] Therefore, the system proposes to fuse three-dimensional data within each task cycle to model and generate R, ensuring that the collaborative behavior, trust status, and task background of each node are completely traceable.

[0125] Joint Tracing Structure Design

[0126] The R structure proposed by the system is as follows:

[0127]

[0128] Variable Definitions:

[0129] v i : Consensus node, from C screened by Module 4;

[0130] task_type i 、 : The type of task and standardized collaborative actions executed by node v i , from D;

[0131] : Node trust embedding, from Module 3;

[0132] κ i : The weighted degree of node in G, κ i =∑ j∈N(i) e i,j ;

[0133] N(i): The set of adjacent nodes of v i ;

[0134] E i : The set of edge weights between v i and its neighbors, E i ={e i,j |j∈N(i)}.

[0135] Special Innovation Item - Trusted Behavior Factor

[0136] To enhance the intrinsic causal interpretability of the traceability data for "collaborative behavior - trust score - task background", the system innovatively proposes the "credible behavior factor" θ i , which is used to comprehensively characterize the i collaborative credibility of v.

[0137]

[0138] Variable description:

[0139] |E i | is the sum of the trust edge weights of v; i

[0140] |N(i)| is the number of adjacent nodes;

[0141] η is the collaborative density sensitivity factor;

[0142] θ i is a composite credibility index reflecting "trust value + collaborative relationship strength";

[0143] This factor is directly used as one of the fields of R to assist subsequent on-chain verification or traceability and accountability.

[0144] Traceability link output

[0145] Finally, the system outputs the joint traceability record R, whose core function is to provide direct evidence for future anomaly detection, traceability auditing, and liability determination.

[0146] Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working processes of the above-described system, device, and unit can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0147] In several embodiments provided in the present application, it should be understood that the disclosed system, device, and method can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division, and there can be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Additionally, the couplings or direct couplings or communication connections shown or discussed with each other can be through some interfaces, and the indirect couplings or communication connections of the devices or units can be in electrical, mechanical, or other forms.

[0148] ​When the above-mentioned function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the modules of the methods described in various embodiments of this application. The foregoing storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs.

[0149] Although the embodiments of the present invention have been shown and described, those skilled in the art can understand that: various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principle and spirit of the present invention. The scope of the present invention is defined by the claims and their equivalents.

Claims

1. A blockchain-based shared drone distributed traceability management system, characterized in that, The system includes the following modules: A task data acquisition module, which is used to acquire UAV collaboration data and generate a standardized data set; the UAV collaboration data includes position coordinates, the original encoding of the current collaboration action, flight speed and battery level, time stamps, and task numbers; A collaboration trust graph construction module, which is used to map each data in the standardized data set to a node in the graph to construct a collaboration trust graph reflecting the UAV collaboration relationship, perform correlation calculation and analysis based on the collaboration trust graph, generate trust edge weights, and perform weighted fusion of the action behaviors of adjacent nodes and the trust edge weights to generate node trust embeddings, which are used to reflect the trust of the UAV collaboration environment and behavior patterns; A trust-task coupling enhancement module, which is used to dynamically enhance the node trust embeddings in the trust graph by combining the corresponding task characteristics to obtain enhanced trust embeddings and input them into a trust-driven dynamic consensus module; A trust-driven dynamic consensus module, which is used to dynamically determine the consensus node set of the collaboration trust graph based on the enhanced trust embeddings and assign corresponding task perception consensus weights; A joint traceability module, which is used to generate a multi-dimensional traceability data structure according to the consensus nodes, the types of tasks executed by the corresponding nodes and the standardized collaboration actions, the corresponding node trust embeddings, the weighted degrees of the corresponding nodes in the trust graph, the set of adjacent nodes of the corresponding nodes, and the set of edge weights between the corresponding nodes and their neighbors.

2. The blockchain-based shared drone distributed traceability management system according to claim 1, wherein, The task data acquisition module further includes: Mapping the original encoding of the current collaboration action and the task number to the types of tasks executed by the corresponding nodes and the standardized collaboration actions; And within a preset time range, only retain the latest valid data of each UAV, remove redundancy, and reduce the noise introduced by time resampling during graph modeling.

3. The blockchain-based shared drone distributed traceability management system according to claim 2, wherein, The performing the correlation calculation and analysis based on the collaboration trust graph to generate trust edge weights specifically includes: For any node pair (v i , v j ), use relevance calculation analysis to calculate its trust edge weight e i,j as follows: Among them, and are the standardized collaborative action vectors of v i and v j ; Sim(·) is the cosine similarity of the action vectors; ||pos i - pos j || is the Euclidean distance of the node positions, and Exp(-||·||) is used as the spatial trust decay factor; θ task is the task type perception factor, which is determined by the type of task executed by the corresponding node; According to the trust edge weight e i,j and the normalized collaborative actions of its adjacent nodes, weighted aggregation is performed to generate a trust embedding that reflects the UAV collaborative environment and behavior patterns.

4. The blockchain-based shared drone distributed traceability management system according to claim 2, characterized in that, The trust-task coupling enhancement module specifically executes the following: Based on the types of tasks executed in the standardized data set and the built-in task feature table of the system, generate corresponding task feature vectors for each node, including: task risk level and task collaboration density prediction; According to the task risk level R i and the estimated task collaboration density D i enhance the trust embedding w i · to obtain an enhanced trust embedding which is expressed as: Where γ1 and γ2 are coupling adjustment factors, and the system dynamically configures them according to different task types; Among them, in the high-risk and sparse-collaboration task scenarios, R i and 1-D i are both relatively large, such that is significantly enhanced, improving the system's trust response to specific task environments; In a scenario of collaborative-intensive and low-risk tasks, The enhancement effect is weaker, maintaining the rationality of the original distribution of the trust graph.

5. The blockchain-based shared drone distributed traceability management system according to claim 1, wherein, The conditions for the consensus node set are as follows: Condition 1: Among them, is the enhanced trust embedding, τ task is the task-aware dynamic trust threshold, and the system automatically adjusts according to task_type i Logistics task: τ task = 0.5; Security patrol task: τ task = 0.7; Condition 2: The collaborative influence of the node is not less than the minimum collaboration threshold κ min , κ i ≥κ min , κ i is the weighted degree of the node in the trust graph, where κ i = ∑ j∈N(i) e i,j , j is the adjacent node, N(i) is the set of adjacent nodes, and e i,j is the trust edge weight from node i to node j.

6. The blockchain-based shared drone distributed traceability management system according to claim 1, wherein Obtain the task perception consensus weights of each consensus node according to the normalized weight analysis of the enhanced trust embeddings.

7. The blockchain-based shared drone distributed traceability management system according to claim 4, wherein, The task risk level includes low risk and high risk; the task collaboration density prediction is 0-1, where a lower value represents sparser UAV collaboration under the task, and the credibility of a single node is more important.

8. The blockchain-based shared drone distributed traceability management system according to claim 1, wherein In the joint traceability module, a trusted behavior factor θ is designed i , which is used to comprehensively characterize the collaboration credibility of v i . The trusted behavior factor θ i is expressed as: Among them, |E i | is the sum of v i trust edge weights; |N(i)| is the number of adjacent nodes; η is the cooperation density sensitive factor; θ i is a composite credibility index used to reflect the trust value + the strength of the cooperation relationship.

Images