A high-precision real-time transmission method for release decision information based on digital twinning

Abstract

The application discloses a kind of based on digital twinning's high-precision real-time transmission method of release decision information, first based on static environment generates three-dimensional environment pedestal, to test machine entity is parameterized modeling, semantic segmentation and LOD light weight reconstruction generates multi-detail level digital mockup, common constitute multilevel high fidelity release scene geometry twin body;Through uniform abstract data interface access multi-source release data, utilize semantic anchoring to realize the accurate addressing of business instruction in three-dimensional space, and construct multidimensional release readiness real-time evaluation model, output comprehensive readiness state value as core driving variable;Adopt programmable rendering pipeline technology, based on state-driven color intensity and transparency calculation generates dynamic visual primitive;Through cloud edge collaborative architecture based on state increment, to realize cross-terminal adaptive differential transmission and multi-party collaborative decision closed loop with light weight instruction frame.The application can significantly improve the real-time, intuitiveness and cross-terminal synchronization accuracy of release decision information.

Description

Technical Field

[0001] This invention belongs to the interdisciplinary field of aviation flight test support and industrial digital twin technology, specifically involving a high-precision real-time transmission method for flight release decision information based on digital twin. Background Technology

[0002] In the aviation industry, test flight missions for new aircraft models are characterized by high risk, high technical difficulty, and high coordination. The "release decision" before a test flight is a crucial step in ensuring flight safety, involving a comprehensive assessment of multi-dimensional information such as the technical status of the test aircraft (e.g., completion of maintenance work, fault retention status), takeoff line support resources (e.g., power supply vehicles, towing vehicles, fire-fighting facilities), and the hangar environment. Currently, existing release decision support and information presentation methods have the following limitations: ① Traditional methods rely on two-dimensional charts or paper work cards to display information, making it difficult for commanders to intuitively grasp the test aircraft's attitude, the distribution of support vehicles, and potential spatial conflicts. The information presentation is abstract, resulting in low cognitive efficiency. ② Although existing maintenance and scheduling systems have achieved digitization, task progress data (e.g., work card status) is not accurately linked to the aircraft's three-dimensional model, leading to a disconnect between "data" and "physical reality," making it difficult to quickly locate specific physical areas where tasks have not been completed. Therefore, there is an urgent need for a system that integrates digital twin technology to achieve high-precision three-dimensional presentation and real-time transmission of key release information to solve the above problems. Summary of the Invention

[0003] The purpose of this invention is to provide a high-precision real-time transmission method for flight decision information based on digital twins, so as to solve the problems existing in the prior art.

[0004] To achieve the above objectives, the present invention employs the following technical solution: A method for high-precision real-time transmission of launch decision information based on digital twins, comprising: Step 1: For the static environmental facilities and test aircraft entities in the physical launch scenario, point cloud data is collected, parametric modeling is performed, and virtual and real space consistency alignment is achieved to obtain a multi-level high-fidelity geometric twin of the launch scenario; Step 2: Based on the multi-level high-fidelity launch scene geometric twin and multi-source heterogeneous launch data, unified access, semantic binding, readiness assessment and quantization dimensionality upgrade are performed to realize the precise addressing capability of business instructions in three-dimensional space and obtain the core state variables used to drive the rendering engine. Step 3: Using core state variables and multi-level high-fidelity flying scene geometric twins, perform dynamic visual mapping and color intensity calculation to obtain the rendered 3D visual primitives. Step 4: Using the rendered 3D visual primitives and combined with a multi-detail digital prototype, perform state difference extraction, instruction frame broadcasting and distribution, local terminal redrawing, and collaborative decision-making closed loop to achieve high-precision real-time synchronization of flight decision information and collaborative interaction across terminals.

[0005] Furthermore, the construction process of the multi-level high-fidelity flight scene geometric twin includes: High-density point cloud data of static environmental facilities is acquired, outliers are removed by statistical filtering algorithm, and the iterative nearest point algorithm is applied to complete the fine registration of multi-site point cloud data. The registered point cloud topology is transformed into a triangular mesh model by Poisson reconstruction algorithm, and texture mapping is performed by combining high-definition panoramic images to generate a three-dimensional environmental base with realistic material texture. The original CAD design data of the test machine entity is imported into the conversion engine to generate a basic digital prototype. Semantic segmentation is performed on it to obtain a set of interactive parts with independent logical identifiers. Local coordinate systems and parent-child hierarchical kinematic chains are established for each part to obtain a kinematic digital prototype. A LOD adaptive algorithm based on quadratic error metric is used to perform lightweight reconstruction of the kinematic digital prototype to generate a multi-detail level digital prototype. By setting up ground control points, a spatial mapping model is established between the physical space positioning reference coordinate system and the digital space fuselage coordinate system. The transformation matrix is ​​solved to drive the digital prototype to accurately coincide with the center of the physical parking position. The three-dimensional environment base and the aligned multi-detail digital prototype together constitute a multi-level high-fidelity geometric twin of the launch scene.

[0006] Furthermore, for the key interactive components of the testing machine, their integrated readiness state values ​​are dynamically calculated using time slices as the period. The calculation formula is: ; in, Indicates association with key interactive components The Electronic work card task, The total number of tasks. For task state functions, These are the weighting coefficients. To ensure resources and key interactive components The real-time Euclidean distance, The preset safety threshold distance, These are normalized evaluation values ​​for environmental parameters. These are weighting coefficients. This is the sensitivity coefficient.

[0007] Furthermore, for each element on the surface of the multi-level detailed digital prototype... Read its texture base color And retrieve the current overall readiness status value of the component in real time. As a driving parameter; based on the fragment Spatial position calculation of rendering color intensity vector And transparency channel Defined as: ; in, For color mixing operators, The degree of difference between the current state and the release standard. This is a heatmap mapping function based on a Gaussian distribution. For visual enhancement accommodation coefficient, This is the maximum transparency value. This is the minimum transparency value. The preset ready state threshold is used; the rendering of three-dimensional visual primitives is performed based on the rendering color intensity vector and transparency channel.

[0008] Furthermore, the process of achieving high-precision real-time synchronization of launch decision information and collaborative interaction across terminals includes: Establish a global state matrix on the server side. As the sole truth center for all data, it monitors the overall readiness status value in real time. When a numerical change is detected, the attribute increment is extracted. And generate lightweight state instruction frames. ,in A unique identifier for each interactive component. For timestamps, For collaborative operation verification codes, the WebSocket long-connection protocol is used to broadcast and distribute command frames to all subscribed terminals. The terminals then parse the command frames. Locate the target component and generate incremental data Injecting a local GPU rendering pipeline drives material changes, UI text refreshes, and floating dashboard data updates for local multi-level detail digital mockups, achieving a high degree of synchronization of decision-making information across different terminal perspectives.

[0009] Furthermore, achieving precise addressing capability for business instructions in three-dimensional space includes: By establishing a standardized API gateway through a unified abstract data interface in the data access layer, the system can acquire the task status stream of the electronic work card system, the resource coordinate stream of the UWB / GPS positioning system, and the time-series data stream of the environmental sensors in real time. The system can also pre-build a "business object-geometric primitive" mapping knowledge base using an ontology-based semantic anchoring mechanism, and use a hash indexing algorithm to rigidly bind the functional location tags in the electronic work card to the grid node IDs of the multi-level detail digital prototype, so as to achieve precise addressing of business instructions in three-dimensional space.

[0010] Furthermore, obtain the comprehensive readiness status value of key interactive components. Programmable rendering pipeline technology is used to dynamically inject business logic during the fragment shading stage, establishing a mapping relationship between state values ​​and 3D geometric primitives; for each fragment on the surface of the multi-level detail digital prototype... Read its corresponding texture base color It also retrieves the current overall readiness status value of the component in real time. As driving parameters.

[0011] Furthermore, a cross-terminal synchronization mechanism enables commanders, maintenance personnel, and on-site support personnel to obtain consistent 3D visual primitives on their respective terminals; when the commander triggers the "agree to launch" command, the system uses a collaborative operation verification code. The decision-making instructions are atomically issued to all terminals, driving the test aircraft status lights and runway light strips in the entire digital twin scenario to synchronously turn green; on-site maintenance personnel execute physical launch operations based on the decision results received by handheld terminals, and synchronize the operation feedback back to the server in real time, completing a collaborative interactive closed loop from data transmission to auxiliary decision-making to action execution; simultaneously, the system background automatically generates the global state matrix at the moment of this decision. Snapshots are stored in a time-series database to enable historical tracing and archiving of the entire launch decision-making process.

[0012] A terminal device includes a processor, a memory, and a computer program stored in the memory; when the processor executes the computer program, it implements the method for high-precision real-time transmission of flight decision information based on digital twins.

[0013] A computer-readable storage medium storing a computer program; when the computer program is executed by a processor, it implements the method for high-precision real-time transmission of flight decision information based on digital twins.

[0014] Compared with the prior art, the present invention has the following technical features: Compared with existing technologies, this invention breaks through the spatial cognitive limitations of traditional flight launch decisions, which rely on two-dimensional charts and discrete information presentation. By constructing a high-fidelity digital twin scenario and a precise semantic anchoring mechanism for multi-source heterogeneous data, it achieves rigid binding of the takeoff line mission state flow, the spatiotemporal trajectory flow of support resources, and environmental parameters with the three-dimensional geometric model and its logical components, effectively solving the problem of the separation between physical entities and digital information. Simultaneously, based on a cloud-edge collaborative adaptive differential transmission architecture and a multi-dimensional flight readiness quantification calculation model, this invention compresses cross-terminal collaborative response latency to the millisecond level and transforms passive data monitoring into algorithm-based proactive risk warning. Thus, while ensuring a high degree of consistency between the command center and mobile terminal perspectives, it significantly improves the intuitiveness of flight launch decisions, end-to-end collaborative efficiency, and inherent safety in high-intensity flight test missions. Attached Figure Description

[0015] Figure 1 This is a flowchart illustrating the alignment between the LOD adaptive algorithm and the geometric consistency of virtual and real spaces. Detailed Implementation

[0016] In the highly dynamic and complex scenarios of flight test mission release decisions, there is often a challenge of spatiotemporal discreteness of multi-source heterogeneous data (such as the mission status of aircraft maintenance electronic work cards, the spatial distribution and movement trajectory of support resources, hangar environmental parameters, etc.). The traditional interaction mode that relies on two-dimensional charts or voice reports results in a serious semantic disconnect between the "data flow" and the "physical entities" on site. This makes it difficult for decision-makers to intuitively perceive spatial conflicts and the status of key nodes when facing massive concurrent tasks. Furthermore, the lag in on-site information feedback and the inconsistency of multi-terminal data are coupled together, resulting in cognitive blind spots and response delays in flight release decisions, which cannot meet the stringent requirements of precise coordination for high-intensity flight test missions.

[0017] This invention provides a high-precision real-time transmission method for flight decision information based on digital twins. The method constructs a bidirectional mapping architecture of "physical-digital" coexistence. It establishes a digital skeleton of physical entities through high-fidelity geometric modeling, and uses multi-source heterogeneous data spatiotemporal anchoring technology to inject discrete aircraft maintenance card status, support resource location, and environmental parameters into the digital skeleton to form a digital twin with real-time perception capabilities. Then, it adopts a semantically driven dynamic rendering engine to transform abstract decision data into intuitive visual primitives in three-dimensional space, and achieves high-precision real-time synchronization across terminals through an adaptive differential transmission mechanism.

[0018] This invention is mainly applied to scenarios such as takeoff line preparation, maintenance command, and multi-party collaborative decision-making at aircraft flight test sites. It is particularly suitable for occasions that require spatial correlation mapping, real-time dynamic rendering, and high-precision synchronous display across terminals of multi-source heterogeneous flight test mission data, support resource status, and flight decision results with the 3D model of the flight hangar and test aircraft. The technical solution of this invention is as follows: Step 1: For the static environmental facilities and test aircraft entities in the physical launch scenario, point cloud data is collected, parametric modeling is performed, and virtual-real space consistency alignment is achieved to obtain a multi-level high-fidelity geometric twin of the launch scenario.

[0019] Step 1.1: Collect and register point cloud data for static environmental facilities.

[0020] Static environmental facilities such as hangars, parking spaces, and maintenance platforms are placed, and point cloud data is collected, filtered, registered, reconstructed, and textured to obtain a three-dimensional environmental base with realistic material texture.

[0021] (1) For static environmental facilities such as aircraft hangars, parking positions and maintenance platforms, a stand-up 3D laser scanner is used to collect all-round spatial data and obtain high-density point cloud data.

[0022] (2) Statistical filtering algorithm is used to remove outliers in point cloud data, and iterative nearest point algorithm (ICP) is applied to complete fine registration of multi-site point cloud data, so as to obtain a global point cloud model with spatial relative positioning error controlled within millimeters.

[0023] (3) The registered point cloud topology is transformed into a triangular mesh model using the Poisson Reconstruction algorithm to obtain a high-precision three-dimensional geometric framework.

[0024] (4) Combine the high-definition panoramic images collected on site to perform texture mapping on the triangular mesh model to generate a three-dimensional environmental base with realistic material texture.

[0025] Step 1.2: Perform parametric digital prototype modeling for the physical test machine.

[0026] This step involves parametric modeling, semantic segmentation, motion chain establishment, LOD lightweight reconstruction, and spatial positioning of the physical prototype to obtain a digital prototype with multiple levels of detail.

[0027] (1) First, import the original CAD design data of the test machine entity into the conversion engine to generate a basic digital prototype containing geometric features and logical attributes.

[0028] (2) The basic digital prototype is semantically segmented to meet the interactive requirements of the launch mission. It is decomposed into a set of interactive components with independent logical identifiers, covering key interactive components such as landing gear, radome, and movable control surfaces, to obtain a semantically segmented digital prototype.

[0029] (3) Establish a local coordinate system for each component, including the rotation axis and the degree of freedom of motion, and establish a parent-child hierarchical kinematic chain to support real-time simulation of physical actions and obtain a kinematic digital prototype.

[0030] (4) A lightweight reconstruction of the kinematic digital prototype is performed using an LOD adaptive algorithm based on a quadratic error metric. The specific implementation of the algorithm is as follows: (4-1) For the kinematic digital prototype, the quadratic error metric matrix of the surface vertices is calculated. Extract geometric feature information; (4-2) Constructing the vertex shrinkage cost function The specific formula is as follows: ; in, This represents the coordinate vector of a vertex on the model surface. This represents the quadratic error metric matrix corresponding to that vertex. Represents the vertex coordinate vector The transpose of .

[0031] (4-3) Using the vertex shrinkage cost function described above, calculate the geometric deformation error generated when each edge of the kinematic digital prototype shrinks; by setting a shrinkage cost threshold, for the function value... Non-critical areas below a preset threshold are subjected to mesh extraction and vertex merging, thereby reducing the number of facets in the kinematic digital prototype.

[0032] (4-4) For the simplified kinematic digital prototype, the LOD adaptive processing module inside the system is based on the vertex shrinkage cost function. Based on the set thresholds, generate digital prototypes with multiple levels of detail, including fine level (LOD0), standard level (LOD1), and contour level (LOD2).

[0033] (4-5) Real-time acquisition of viewpoint distance, and dynamic invocation of the corresponding level of multi-detail digital prototype into the rendering engine using the LOD switcher, so as to ensure that the real-time rendering frame rate of more than 60fps can be maintained under different viewing scales.

[0034] Step 1.3: Perform geometric consistency alignment between virtual and real spaces.

[0035] This sub-step performs spatial coordinate mapping and transformation matrix solving on the 3D environment base generated in step 1.1 and the multi-level detail digital prototype generated in step 1.2, to obtain a multi-level high-fidelity geometric twin of the flight scene composed of the 3D environment base and the aligned multi-level detail digital prototype.

[0036] (1) Based on the environmental base and digital prototype established in steps 1.1 and 1.2, ground control points (GCPs) are set up on the floor of the hangar where the physical parking position is located to establish a positioning reference coordinate system in physical space. .

[0037] (2) Determine the fuselage coordinate system of the multi-detail digital prototype in digital space. And define the midpoint of the digital space. Midpoint of physical space Spatial mapping model between: ; in, for The homogeneous transformation matrix is ​​specifically represented as: ; In the above formula, for The rotation matrix represents the attitude deviation between the physical entities of the multi-level detail digital prototype and the test machine entity; for The translation vector represents the spatial displacement between the digital origin and the center of the physical parking position; for The zero vector transpose.

[0038] (3) Obtain the coordinates of the ground control points in physical space using measuring equipment. And extract the corresponding feature anchor point coordinates from the multi-level detail digital prototype. The optimal transformation matrix is ​​solved using the least squares method. .

[0039] (4) Using the transformation matrix obtained by solving The origin of the fuselage coordinate system of the multi-level detail digital prototype is precisely aligned with the center of the physical parking position to achieve alignment of the multi-level detail digital prototype; then the three-dimensional environment base and the aligned multi-level detail digital prototype together constitute a multi-level high-fidelity launch scene geometric twin.

[0040] In this step, this solution designs a multi-level modeling strategy and a LOD adaptive mechanism based on a quadratic error metric. The aim is to establish a digital base for the launch scene with physical simulation capabilities and realistic material texture through millimeter-level point cloud reconstruction and parametric prototype disassembly. At the same time, by using LOD technology, the number of facets is significantly reduced while retaining key geometric features such as bomb mounting points and air inlets. This resolves the contradiction between high-precision models and real-time rendering performance on the terminal, ensuring that complex digital twin scenes can maintain a high frame rate of over 60fps on heterogeneous devices, providing a stable geometric carrier for subsequent readiness data mapping.

[0041] Step 2: Based on the multi-level high-fidelity launch scene geometric twin and multi-source heterogeneous launch data (including electronic work card task status, support resource location and environmental parameters), unified access, semantic binding, readiness assessment and quantization are performed to realize the precise addressing capability of business instructions in three-dimensional space and obtain the core state variables used to drive the rendering engine.

[0042] Step 2.1: Access and binding of multi-source data based on a unified interface and semantic anchoring.

[0043] This sub-step implements unified interface access and semantic anchoring binding for the task status flow of the electronic work card system, the resource coordinate flow of the UWB / GPS positioning system, and the time-series data flow of the environmental sensors, thereby enabling precise addressing of business instructions (specific operation or inspection instructions issued by the electronic work card system related to the launch mission) in three-dimensional space.

[0044] (1) First, a standardized API gateway is established through the data access layer using the Unified Abstract Data Interface (UDAI).

[0045] (2) Use the API gateway to obtain the task status stream of the electronic work card system, the resource coordinate stream of the UWB / GPS positioning system, and the time-series data stream of the environmental sensor in real time.

[0046] (3) To address the issue of inconsistency between business data and 3D model structure, an ontology-based semantic anchoring mechanism is used to pre-construct a knowledge base for mapping "business objects-geometric primitives".

[0047] (4) Using the hash index algorithm, the functional location tags in the electronic work card are rigidly bound one-to-one with the grid node IDs of the multi-detail level digital prototype, so as to realize the precise addressing of business instructions in three-dimensional space.

[0048] Step 2.2: Construct a multi-dimensional real-time evaluation model for flight readiness.

[0049] Based on the binding results of step 2.1, this sub-step calculates the comprehensive readiness status value for the key interactive components of the test aircraft using time slices as the period, and obtains the real-time evaluation results of multi-dimensional flight readiness.

[0050] (1) Targeting the key interactive components of the testing machine A multi-dimensional real-time calculation model for flight readiness is deployed; this model dynamically calculates the comprehensive readiness state value using time slices as the period. .

[0051] (2) The comprehensive ready state value The calculation formula is as follows: ; in: Indicates association with component The Electronic work card task; Indicates association with component The total number of electronic work card tasks; This is a task status function. It takes the value 1 when the task is completed, 0.5 when the task is in progress, and 0 when the task has not started. The weight coefficients for electronic work card tasks are as follows: tasks on the critical path have higher weights. To ensure resources and components The real-time Euclidean distance; The distance is the preset safety threshold. Normalized evaluation values ​​for environmental parameters (whether temperature and humidity are suitable); , , These are the weighting coefficients for the work card task item, resource distance item, and environmental parameter item, respectively. This is the sensitivity coefficient; It is a natural constant.

[0052] Step 2.3: Quantize and upgrade the state information and output it.

[0053] The overall readiness value calculated by this step-by-step needle. Quantization and dimensionality enhancement are performed to obtain the core state variables used to drive the subsequent rendering engine.

[0054] (1) Obtain the key interactive components associated with it The basic score of the task dimension for all electronic work card tasks, i.e. .

[0055] (2) Monitor the location of security resources in real time, and when resources intrude into the security threshold... At that time, the Sigmoid function in the formula of step 2.2 is used. Quantify the risks of physical interference.

[0056] (3) The calculated integrated readiness state value As the core state variable driving the subsequent rendering engine, it is output to realize the dimensionality upgrade processing from discrete data to fused information.

[0057] In this step, a data binding mechanism based on semantic anchoring and a multi-dimensional quantitative evaluation model were designed. The aim is to eliminate access barriers to multi-source heterogeneous data through the UDAI interface and to solve the semantic gap between "business tags" and "physical components" using hash indexes. Simultaneously, by introducing mathematical models that include mission progress, space safety, and environmental indicators, the originally abstract and fragmented flight test support information is transformed into quantitative values ​​with spatial awareness capabilities. This provides a precise driving force for subsequent dynamic enhancement rendering, and can proactively identify physical interference risks, providing decision-makers with deeper support.

[0058] Step 3: Using core state variables and multi-level high-fidelity geometric twins of the flight scene, perform dynamic visual mapping and color intensity calculation to obtain rendered 3D visual primitives (including color enhancement and transparency changes).

[0059] Step 3.1: Design a dynamic visual mapping mechanism based on a rule engine.

[0060] This sub-step focuses on key interactive components. Overall readiness value The rule engine performs reasoning and fragment shader injection to obtain the mapping relationship between state values ​​and three-dimensional geometric primitives.

[0061] (1) First, obtain the key interactive components. Overall readiness value .

[0062] (2) Programmable rendering pipeline technology is used to dynamically inject business logic during the fragment shading stage and establish a mapping relationship between state values ​​and three-dimensional geometric primitives.

[0063] (3) For each element on the surface of the multi-detail digital prototype Read its corresponding texture base color It also retrieves the current overall readiness status value of the component in real time. As driving parameters.

[0064] Step 3.2: Perform state-driven color intensity and transparency calculations.

[0065] This step is for the fragment. Texture base color and components Overall readiness value The color intensity and transparency are calculated to obtain the rendered color intensity vector. and transparency channel Based on this, three-dimensional visual primitives are rendered.

[0066] Based on multi-level detail digital prototype surface tessellation Spatial location, calculate the rendering color intensity vector : ; And transparency channel Defined as: ; in: The original texture colors for multi-level detail digital mockups; Represents the color mixing operator; The degree of difference between the current state and the release standard. ; This is a Gaussian distribution-based heatmap mapping function used to visually highlight the core locations of unfinished areas; , The adjustment factor is used to enhance visual perception. This is the maximum transparency value (completely opaque). This is the minimum transparency value (completely transparent). This is the preset ready state threshold.

[0067] This algorithm ensures that when a component's task is not completed or there is a risk of resource conflict (i.e., ... When the color intensity vector is lower, the calculated rendering color intensity vector is... This shifts the fragment color towards red and increases its saturation, while also adjusting the opacity channel. Take the maximum value (Completely opaque), the system automatically issues a warning using a high-saturation, high-opacity red heatmap; as the task progresses... Increased Decrease Gradually restored to the original texture base color, and according to By linearly increasing the transparency, the color gradually turns into a semi-transparent green, providing intuitive visual feedback and thus completing the rendering of the three-dimensional visual primitives.

[0068] In this step, a dynamic visual enhancement mechanism based on a rule engine and a nonlinear color mapping model were designed. The aim is to transform "electronic work card progress" and "resource location" data into visual signals (such as heatmaps and flashing warnings) that conform to human spatial cognition by injecting business state variables into the rendering pipeline in real time. Through this dimensionality reduction presentation, decision-makers can instantly locate the physical areas of incomplete tasks and potential spatial conflicts in a 3D scene without consulting complex 2D charts, significantly shortening the cognitive path for decision-making and improving response speed in highly dynamic scenarios.

[0069] Step 4: Using the rendered 3D visual primitives and combined with a multi-detail digital prototype, perform state difference extraction, instruction frame broadcasting and distribution, local terminal redrawing, and collaborative decision-making closed loop to achieve high-precision real-time synchronization of flight decision information and collaborative interaction across terminals.

[0070] Step 4.1: Extraction and high-precision transmission of state differences based on cloud-edge collaboration.

[0071] This sub-step targets the global state matrix on the server side. and the overall readiness status value output in step 2.3 The numerical changes are processed to extract attribute increments and generate instruction frames, which are then broadcast via WebSocket to obtain lightweight state instruction frames. Real-time transmission.

[0072] (1) Establish a global state matrix on the server side ,in To store the state information of all digital twin entities in the entire venue (including the comprehensive readiness state value of each key interactive component). A multidimensional matrix (including parameters such as location and material properties) serves as the sole truth center for all data.

[0073] (2) Real-time monitoring of the overall readiness status value The numerical change; when a numerical jump is detected, its attribute increment is extracted. And generate lightweight state instruction frames. ;

[0074] in: A unique identifier for each interactive component; For the attribute increment of the ready state, , Indicates the time increment; For high-precision timestamps used for multi-end clock alignment; It is a collaborative operation check code used to verify the integrity and legitimacy of the decision-making instructions.

[0075] (4) Use the WebSocket long connection protocol to transmit status command frames It is distributed to all subscribed terminals in the form of broadcast, and achieves high-precision real-time transmission with extremely low bandwidth consumption by transmitting only "incremental" rather than "full model".

[0076] Step 4.2, Terminal adaptive parsing and local lightweight redraw.

[0077] This sub-step is designed for terminal devices with different performance levels and the instruction frames generated in step 4.1. The system performs computational power negotiation, LOD model preloading, instruction frame parsing, and local rendering pipeline injection to obtain highly synchronized launch decision information from the perspectives of each terminal.

[0078] (1) For terminal devices with different performance (such as command screens, mobile terminals, etc.), computing power negotiation is automatically performed during the access phase, and digital prototypes of the corresponding level of detail are preloaded according to the negotiation results (such as loading fine-level LOD0 on command screens and standard-level LOD1 or outline-level LOD2 on mobile terminals).

[0079] (2) Terminal acquires status command frame Then, firstly through timestamps Verify the timeliness of the data, then parse it. Locate the target component and increment its attributes. Inject into the local GPU rendering pipeline.

[0080] (3) The terminal drives the material changes of the local multi-detail digital prototype in real time based on the incremental data (calculated according to the color and transparency in step 3.2), refreshes the UI text and updates the floating dashboard data, so as to achieve a high degree of synchronization of decision information from the perspective of each terminal without relying on real-time rendering and streaming in the cloud.

[0081] Step 4.3: Multi-party collaborative decision-making interaction and instruction closed loop.

[0082] This sub-step addresses the collaborative decision-making needs of commanders, maintenance personnel, and on-site support personnel, as well as the visual information synchronized in step 4.2. It performs cross-terminal visual early warning synchronization, atomic issuance of decision commands, physical operation feedback transmission, and status snapshot archiving, resulting in a collaborative interactive closed loop from data transmission to auxiliary decision-making and then to action execution.

[0083] (1) First, through the cross-terminal synchronization mechanism, the commander, maintenance personnel and on-site support personnel can obtain consistent three-dimensional visual primitives on their respective terminals.

[0084] (2) When the commander triggers the "agree to launch" command based on the synchronized high-precision scenario, the system verifies the code. The decision-making instructions are atomized and issued to all terminals, driving the status lights of the test machine and the light strips of the runway in the entire digital twin scenario to turn green synchronously.

[0085] (3) On-site maintenance personnel perform physical launch operations based on the decision results received by the handheld terminal and synchronize the operation feedback back to the server in real time, completing the collaborative interactive closed loop from "data transmission" to "assistive decision-making" to "action execution".

[0086] (4) Finally, the system background automatically generates the global state matrix at the moment of this decision. Snapshots are stored in a time-series database to enable historical tracing and archiving of the entire launch decision-making process.

[0087] In this step, our solution designs a cloud-edge collaborative architecture based on state increments and a cross-terminal adaptive synchronization mechanism. The aim is to transmit KB-level instruction frames. Replacing GB-level model data solves the high latency problem in complex digital twin scenarios under wireless network environments, ensuring the "real-time" nature of transmission. Simultaneously, by establishing a "server truth center - multi-terminal local redrawing" logic, decision-makers in different spaces using different devices can conduct efficient analysis based on the same digital model at the same time and in the same dimension. This design not only achieves "high-precision transmission" of information but also realizes a collaborative interactive closed loop between the physical site and the digital space through instruction synchronization and state backtracking, thus fully supporting the high-precision real-time transmission method described in this invention.

[0088] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A method for high-precision real-time transmission of flight decision information based on digital twins, characterized in that, include: Step 1: For the static environmental facilities and test aircraft entities in the physical launch scenario, point cloud data is collected, parametric modeling is performed, and virtual and real space consistency alignment is achieved to obtain a multi-level high-fidelity geometric twin of the launch scenario; Step 2: Based on the multi-level high-fidelity launch scene geometric twin and multi-source heterogeneous launch data, unified access, semantic binding, readiness assessment and quantization dimensionality upgrade are performed to realize the precise addressing capability of business instructions in three-dimensional space and obtain the core state variables used to drive the rendering engine. Step 3: Using core state variables and multi-level high-fidelity flying scene geometric twins, perform dynamic visual mapping and color intensity calculation to obtain the rendered 3D visual primitives. Step 4: Using the rendered 3D visual primitives and combined with a multi-detail digital prototype, perform state difference extraction, instruction frame broadcasting and distribution, local terminal redrawing, and collaborative decision-making closed loop to achieve high-precision real-time synchronization of flight decision information and collaborative interaction across terminals.

2. The method for high-precision real-time transmission of flight decision information based on digital twins according to claim 1, characterized in that, The construction process of a multi-level high-fidelity flight scene geometric twin includes: High-density point cloud data of static environmental facilities is acquired, outliers are removed by statistical filtering algorithm, and the iterative nearest point algorithm is applied to complete the fine registration of multi-site point cloud data. The registered point cloud topology is transformed into a triangular mesh model by Poisson reconstruction algorithm, and texture mapping is performed by combining high-definition panoramic images to generate a three-dimensional environmental base with realistic material texture. The original CAD design data of the test machine entity is imported into the conversion engine to generate a basic digital prototype. Semantic segmentation is performed on it to obtain a set of interactive parts with independent logical identifiers. Local coordinate systems and parent-child hierarchical kinematic chains are established for each part to obtain a kinematic digital prototype. A LOD adaptive algorithm based on quadratic error metric is used to perform lightweight reconstruction of the kinematic digital prototype to generate a multi-detail level digital prototype. By setting up ground control points, a spatial mapping model is established between the physical space positioning reference coordinate system and the digital space fuselage coordinate system. The transformation matrix is ​​solved to drive the digital prototype to accurately coincide with the center of the physical parking position. The three-dimensional environment base and the aligned multi-detail digital prototype together constitute a multi-level high-fidelity geometric twin of the launch scene.

3. The method for high-precision real-time transmission of flight decision information based on digital twins according to claim 1, characterized in that, For the key interactive components of the testing machine, their overall readiness status value is dynamically calculated using time slices as the period. The calculation formula is: ； in, Indicates association with key interactive components The Electronic work card task, The total number of tasks. For task state functions, These are the weighting coefficients. To ensure resources and key interactive components The real-time Euclidean distance, The preset safety threshold distance, These are normalized evaluation values ​​for environmental parameters. These are weighting coefficients. This is the sensitivity coefficient.

4. The method for high-precision real-time transmission of flight decision information based on digital twins according to claim 1, characterized in that, For each element on the surface of a multi-level digital prototype Read its texture base color And retrieve the current overall readiness status value of the component in real time. As a driving parameter; based on the fragment Spatial position calculation of rendering color intensity vector And transparency channel Defined as: ； in, For color mixing operators, The degree of difference between the current state and the release standard. This is a heatmap mapping function based on a Gaussian distribution. For visual enhancement accommodation coefficient, This is the maximum transparency value. This is the minimum transparency value. The preset ready state threshold is used; the rendering of three-dimensional visual primitives is performed based on the rendering color intensity vector and transparency channel.

5. The method for high-precision real-time transmission of flight decision information based on digital twins according to claim 1, characterized in that, The process of achieving high-precision real-time synchronization of launch decision information and collaborative interaction across terminals includes: Establish a global state matrix on the server side. As the sole truth center for all data, it monitors the overall readiness status value in real time. When a numerical change is detected, the attribute increment is extracted. And generate lightweight state instruction frames. ,in A unique identifier for each interactive component. For timestamps, For collaborative operation verification codes, the WebSocket long-connection protocol is used to broadcast and distribute command frames to all subscribed terminals. The terminals then parse the command frames. Locate the target component and generate incremental data Injecting a local GPU rendering pipeline drives material changes, UI text refreshes, and floating dashboard data updates for local multi-level detail digital mockups, achieving a high degree of synchronization of decision-making information across different terminal perspectives.

6. The method for high-precision real-time transmission of flight decision information based on digital twins according to claim 1, characterized in that, To achieve precise addressing of business instructions in three-dimensional space, including: By establishing a standardized API gateway through a unified abstract data interface in the data access layer, the task status stream of the electronic work card system, the resource coordinate stream of the UWB / GPS positioning system, and the time-series data stream of the environmental sensors are obtained in real time. A "business object-geometric primitive" mapping knowledge base is pre-constructed using an ontology-based semantic anchoring mechanism, and a hash index algorithm is used to rigidly bind the functional location tags in the electronic work card to the grid node IDs of the multi-level detail digital prototype, so as to achieve precise addressing of business instructions in three-dimensional space.

7. The method for high-precision real-time transmission of flight decision information based on digital twin as described in claim 1, characterized in that, Obtain the overall readiness status value of key interactive components. The programmable rendering pipeline technology is used to dynamically inject business logic during the fragment shading stage to establish a mapping relationship between state values ​​and three-dimensional geometric primitives. For each element on the surface of a multi-level digital prototype Read its corresponding texture base color It also retrieves the current overall readiness status value of the component in real time. As driving parameters.

8. The method for high-precision real-time transmission of flight decision information based on digital twins according to claim 1, characterized in that, A cross-terminal synchronization mechanism enables commanders, maintenance personnel, and on-site support personnel to obtain consistent 3D visual primitives on their respective terminals; when the commander triggers the "authorize launch" command, the system uses a collaborative operation verification code. The decision-making instructions are atomized and issued to all terminals, driving the status lights of the test machine and the light strips of the runway in the entire digital twin scenario to turn green synchronously; On-site maintenance personnel execute physical launch operations based on the decision results received via handheld terminals, and synchronize the operation feedback back to the server in real time, completing a collaborative interactive closed loop from data transmission to decision support and action execution; simultaneously, the system backend automatically generates the global state matrix at the moment of this decision. Snapshots are stored in a time-series database to enable historical tracing and archiving of the entire launch decision-making process.

9. A terminal device, comprising a processor, a memory, and a computer program stored in the memory; characterized in that, When the processor executes the computer program, it implements the method for high-precision real-time transmission of flight decision information based on digital twin as described in any one of claims 1-8.

10. A computer-readable storage medium storing a computer program; characterized in that, When the computer program is executed by the processor, it implements the method for high-precision real-time transmission of flight decision information based on digital twin as described in any one of claims 1-8.

Images