Scene simulation method, simulation system, device and storage medium

Abstract

The application discloses a scene simulation method, a simulation system, a device and a storage medium. The method comprises the following steps: constructing a scene simulation model asset library in advance, screening a three-dimensional object model of a target scene object from the scene simulation model asset library to build a simulation scene, recording and caching scene evolution recording data sets of core dynamic objects and pre-broken objects in the simulation scene at the same frame rate, synchronously rendering the cached broken animation and trajectory data in a multi-channel mode, and performing real-time simulation processing and rendering on non-core dynamic objects in the simulation scene. Different scenes are constructed through the scene simulation model asset library, and the scene evolution recording data sets are directly called through each channel at the same frame rate for synchronous simulation rendering in the multi-channel simulation mode, so that the visual effect of picture synchronization is realized through the data state synchronization between channels. According to the embodiment of the application, the complex scene can be smoothly run in the multi-channel scene.

Description

Technical Field

[0001] This application belongs to the field of simulation technology, and in particular relates to a scene simulation method, simulation system, device and storage medium. Background Technology

[0002] With the continuous advancement of science and technology, simulations can be used to conduct advance rehearsals for scenarios such as natural disasters, providing important reference value for disaster prediction.

[0003] Currently, real-time simulation is typically achieved by building scenarios to obtain real-time display effects. However, for multi-channel immersive display systems, synchronization failures between multiple channels can lead to image misalignment, disrupting visual spatial continuity and significantly reducing the immersive experience. Therefore, ensuring synchronized visual effects and achieving smooth operation in complex disaster scenarios across multiple channels is a crucial problem that urgently needs to be solved. Summary of the Invention

[0004] This application provides a disaster simulation method, simulation system, device, and storage medium that can ensure synchronized visual effects and enable smooth operation of complex scenarios in multi-channel mode.

[0005] In a first aspect, embodiments of this application provide a scene simulation method, comprising: pre-constructing a scene simulation model asset library, the scene simulation model asset library including at least one 3D object model of a dynamic object and at least one 3D object model of a static object, wherein at least one dynamic object is a core dynamic object, and at least one object is a pre-fragmented object, the 3D object model of the pre-fragmented object containing a fragmented mesh cluster; selecting a 3D object model of a target scene object from the scene simulation model asset library according to the scene simulation type; constructing a simulation scene based on the 3D object model of the target scene object; performing scene evolution recording at the same frame rate on the core dynamic object and the pre-fragmented object in the simulation scene, obtaining and caching a scene evolution recording dataset, the scene evolution recording dataset including at least the fragmentation animation of the pre-fragmented object and the trajectory data of the core dynamic object; when the simulation channel mode is a multi-channel mode, synchronously rendering the cached fragmentation animation and the trajectory data through the slave nodes corresponding to each simulation channel mode, based on the synchronous rendering instructions of the master node, and performing real-time simulation processing and rendering on the non-core dynamic objects in the simulation scene.

[0006] In one possible implementation, the pre-construction of the scene simulation model asset library includes: acquiring three-dimensional object models of each dynamic object and each static object, wherein at least one object is a pre-fragmented object; setting physical simulation attributes for the three-dimensional object models; performing pre-fragmentation processing on the three-dimensional object models of the pre-fragmented objects according to a predefined fragmentation algorithm to obtain the fragmented mesh cluster; performing optimization processing on the three-dimensional object models of the static objects according to a predefined optimization algorithm to obtain optimized three-dimensional object models of the static objects; and constructing the scene simulation model asset library based on the three-dimensional object models of each dynamic object with set physical simulation attributes, the optimized three-dimensional object models of each static object, and the fragmented mesh cluster.

[0007] In one possible implementation, the step of selecting 3D object models of target scene objects from the scene simulation model asset library according to the scene simulation type includes: selecting 3D object models of target scene objects from the scene simulation model asset library according to the physical simulation attributes and pre-fragmented object attributes corresponding to the scene simulation type; the step of building a simulation scene based on the 3D object models of the target scene objects includes: configuring an initial scene according to the scene simulation type; and deploying the selected 3D object models of the target scene objects in the initial scene according to the positions where the target scene objects should appear in the scene, thereby obtaining the simulation scene.

[0008] In one possible implementation, the simulation scenario is a disaster simulation scenario; the step of recording the scene evolution at the same frame rate for the core dynamic object and the pre-fragmented object in the simulation scenario to obtain and cache the disaster evolution recording dataset includes: setting the running frame rate of the recording mode and the fragmentation disaster parameters corresponding to multiple levels of disaster intensity; for each level of disaster intensity, in the recording mode, performing simulation calculations on the disaster simulation scenario based on the fragmentation disaster parameters corresponding to that level of disaster intensity, and recording the fragmentation animation of the pre-fragmented object and the trajectory data of the core dynamic object at the same running frame rate during the simulation process to obtain and cache the scene evolution recording dataset corresponding to that level of disaster intensity.

[0009] In one possible implementation, when the simulation channel mode is a multi-channel mode, the process of synchronously rendering the cached fragmentation animation and trajectory data based on the synchronous rendering instructions of the master node through the slave nodes corresponding to each channel mode, and performing real-time simulation processing and rendering of non-core dynamic objects in the simulation scene, includes: receiving a disaster level selection instruction, determining the target level disaster intensity based on the disaster level selection instruction; when the simulation channel mode is a multi-channel mode, obtaining the trajectory data corresponding to the target level disaster intensity from the cache through the master node, and distributing the trajectory data to each of the slave nodes through the synchronous rendering instructions; obtaining the fragmentation animation corresponding to the target level disaster intensity from the cache through each of the slave nodes based on the synchronous rendering instructions distributed by the master node; synchronously rendering the fragmentation animation and trajectory data based on the same running frame rate, and performing real-time simulation processing and rendering of non-core dynamic objects in the disaster simulation scene.

[0010] In one possible implementation, when the simulation channel mode is multi-channel mode, obtaining the trajectory data corresponding to the target level disaster intensity from the cache through the master node and distributing the trajectory data to each of the slave nodes through synchronous rendering instructions includes: when the simulation channel mode is multi-channel mode, running the physical simulation engine through the master node, converting the cached trajectory data of the core dynamic object and the breaking animation of the pre-fragmented object into a binary state data sequence, packaging the binary state data sequence into a synchronous rendering instruction, and broadcasting it to each of the slave nodes through UDP.

[0011] In one possible implementation, when the simulation channel mode is a multi-channel mode, before synchronously rendering the cached fragmentation animation and trajectory data based on the synchronous rendering instructions of the master node through the slave nodes corresponding to each channel mode, and before performing real-time simulation processing and rendering of non-core dynamic objects in the disaster simulation scene, the method further includes: determining the appropriate simulation channel mode by detecting the simulation running environment, wherein the simulation channel mode includes a single-channel mode or a multi-channel mode; determining the load of a single machine when the simulation channel mode is a single-channel mode; obtaining real-time single-machine scene parameters and real-time single-machine simulation trigger instructions when the load of the single machine is not greater than a preset load threshold; performing real-time simulation of the simulation scene through the single machine according to the real-time single-machine scene parameters and the real-time single-machine simulation trigger instructions; and rendering the cached fragmentation animation and trajectory data, and performing real-time simulation processing and rendering of non-core dynamic objects in the simulation scene when the load of the single machine is greater than the load threshold.

[0012] Secondly, embodiments of this application provide a scene simulation system, comprising: an asset management module, used to pre-build a scene simulation model asset library, the scene simulation model asset library including at least one 3D object model of a dynamic object and at least one 3D object model of a static object, wherein at least one dynamic object is a core dynamic object, and at least one object is a pre-fragmented object, the 3D object model of the pre-fragmented object containing a fragmented mesh cluster; a scene construction module, used to filter out the 3D object model of a target scene object from the scene simulation model asset library according to the scene simulation type; the scene construction module is further used to, based on the 3D object model of the target scene object... The system includes a body model to build a simulation scene; a data caching and recording module for recording the scene evolution at the same frame rate for the core dynamic objects and pre-fragmented objects in the simulation scene, obtaining and caching a scene evolution recording dataset, which includes at least the breaking animation of the pre-fragmented objects and the trajectory data of the core dynamic objects; and a simulation rendering module for synchronously rendering the cached breaking animation and trajectory data based on the synchronous rendering instructions of the master node through the slave nodes corresponding to each simulation channel mode when the simulation channel mode is multi-channel mode, as well as performing real-time simulation processing and rendering of non-core dynamic objects in the simulation scene.

[0013] Thirdly, embodiments of this application provide a scene simulation device, the device comprising: a processor and a memory storing computer program instructions; the processor, when executing the computer program instructions, implements the scene simulation method as described in any one of the first aspects.

[0014] Fourthly, embodiments of this application provide a computer-readable storage medium storing computer program instructions, which, when executed by a processor, implement the scene simulation method as described in any one of the first aspects.

[0015] The disaster simulation method, simulation system, device, and storage medium of this application embodiment utilize a pre-built scene simulation model asset library. Based on the required scene simulation type, target 3D object models are selected from the disaster simulation model asset library to construct different simulation scenes. By pre-caching scene simulation data at the same frame rate, an evolution recording dataset is obtained. In multi-channel simulation mode, the master node server directly calls the scene evolution recording dataset and transmits it to the slave nodes, enabling each slave node to perform simulation rendering using the scene evolution recording dataset cached at the same frame rate. This achieves data state synchronization between channels and realizes a complex simulation scene that runs smoothly in multi-channel scenarios. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments of this application will be briefly introduced below. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a flowchart illustrating a scene simulation method provided in an embodiment of this application; Figure 2 This is a schematic diagram of an immersive multi-channel display structure provided in an embodiment of this application; Figure 3 This is a schematic diagram of a process for constructing a scene simulation model asset library provided in an embodiment of this application; Figure 4 This is a schematic diagram of an object simulation effect provided in an embodiment of this application; Figure 5 This is an example of generating a fragmented mesh cluster according to an embodiment of this application; Figure 6 This is an example of another effect diagram for generating a broken mesh cluster provided in an embodiment of this application; Figure 7 This is a schematic diagram of the structure of a scene simulation model asset library provided in an embodiment of this application; Figure 8 This is a schematic diagram of another scene simulation model asset library provided in this application embodiment; Figure 9 This is a schematic diagram of a process for building a simulation scene according to an embodiment of this application; Figure 10 This is a rendering of an earthquake simulation scenario provided in an embodiment of this application; Figure 11 This is a schematic diagram of a process for determining a scene hazard evolution recording dataset according to an embodiment of this application; Figure 12 This is a schematic diagram of the structure of fragmented cached data provided in an embodiment of this application; Figure 13 This is a schematic diagram of core object state data provided in an embodiment of this application; Figure 14 This is a schematic diagram of the structure of a cache fragmentation animation provided in an embodiment of this application; Figure 15 This is a flowchart illustrating another scene simulation method provided in an embodiment of this application; Figure 16 This is a schematic diagram illustrating a scene simulation mode selection provided in an embodiment of this application; Figure 17 This is a scene simulation effect diagram provided in an embodiment of this application; Figure 18 This is a flowchart illustrating another scene simulation method provided in an embodiment of this application; Figure 19 This is a flowchart illustrating another scene simulation method provided in an embodiment of this application; Figure 20 This is a schematic diagram of the structure of a scene simulation system provided in an embodiment of this application; Figure 21 This is a schematic diagram of another earthquake disaster simulation system provided in the embodiments of this application; Figure 22 This is a schematic diagram of the structure of a scene simulation device provided in an embodiment of this application. Detailed Implementation

[0018] The features and exemplary embodiments of various aspects of this application will be described in detail below. To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only intended to explain this application and not to limit it. For those skilled in the art, this application can be implemented without some of these specific details. The following description of the embodiments is merely to provide a better understanding of this application by illustrating examples.

[0019] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus that includes said element.

[0020] It should be noted that the acquisition, storage, use, and processing of data in this application embodiment all comply with the relevant provisions of national laws and regulations.

[0021] It should be noted that in the embodiments of this application, certain software, components, models and other existing solutions in the industry may be mentioned. These should be regarded as exemplary and are only intended to illustrate the feasibility of implementing the technical solution of this application. However, it does not mean that the applicant has used or necessarily used the solution.

[0022] Scene simulation is a technical method that uses computer technology, mathematical models, and virtual reality to simulate the development and impact of events with high precision and high fidelity. It aims to provide a scientific basis for the prediction, emergency response, and decision support of various scenario events.

[0023] To circumvent the significant technical challenges of synchronization and performance in multi-channel real-time simulation, a common approach is to use pre-made animation. This involves creating a fixed, high-precision evolution video or animation sequence for a specific scene in advance through offline rendering or pre-computation. In a multi-channel immersive simulation system, each node synchronously plays this pre-made animation file. While this method ensures visual synchronization and quality, it is extremely time-consuming and costly. Realistic animation (especially with complex physical fragmentation and particle effects) requires a large investment of art and special effects personnel, with production cycles measured in weeks or even months. Furthermore, it lacks flexibility; once the animation is complete, key parameters of the scene simulation cannot be adjusted, making the simulation a demonstration rather than a true simulation. It also suffers from poor scene scalability and a forced binding between the animation and the specific scene model. Changing the simulation scene or object almost necessitates creating new animation assets from scratch, making it impossible to reuse core logic and hindering the widespread adoption and maintenance of multi-channel simulation systems.

[0024] In traditional distributed rendering architectures, if each rendering node runs physical simulations (such as fluid dynamics, structural failure, and debris splashing) independently, differences in floating-point calculation precision, different random number seeds, or even small time differences in frame loops will lead to non-negligible deviations in the virtual object states (position, velocity, and degree of damage) between channels. For example, in earthquake disaster simulation, the front channel in a multi-channel immersive simulation system might show a collapsed building, while the side channel shows it still shaking, causing severe cognitive confusion and loss of immersion. This is also accompanied by the risk of a dramatic increase in rendering load. Because disaster effects themselves are high-load, realistic disaster simulation relies on numerous particle systems and calculations of geometric breakage and deformation. Even on a single high-performance graphics workstation, maintaining real-time disaster simulation scene rendering at over 30fps requires performance optimization. This leads to the problem of multi-channel stereoscopic rendering doubling the load. In a typical five-channel Cave Automatic Virtual Environment (CAVE) system, to achieve immersive stereoscopic simulation, each graphics rendering node (i.e., computer) needs to render both the left and right eye views simultaneously for the same channel. Each view has a resolution of 2560×1600, so the number of pixels that a single node needs to process per frame is approximately 2 × (2560 × 1600) ≈ 8.19 million pixels. A dual-point CAVE system doubles this to approximately 16.38 million pixels per node, leading to a significant increase in rendering load. High-load rendering tasks must be executed strictly synchronously across multiple nodes. The communication overhead and latency of the synchronization mechanism further consume graphics and computing resources.

[0025] Under the combined effect of the aforementioned factors, the overall effective frame rate of multi-channel immersive simulation systems often drops sharply from the level achievable by a single machine (e.g., 45-60fps) to 15fps or even lower, resulting in screen stuttering and sluggish interaction, severely undermining the smoothness and realism required for an immersive experience. To ensure multi-channel performance, a data caching mechanism is employed, sacrificing the real-time performance and flexibility of the single-machine mode. However, if real-time computing is used to pursue the best single-machine experience, it is difficult to scale to a multi-channel environment; existing technologies lack a unified data computing architecture adapted to different operating environments.

[0026] Specifically, in the scene simulation process of a multi-channel immersive simulation system, there is a problem of asynchronous images between channels. If frame data synchronization fails between multiple projection channels, it will cause misalignment of images between channels, disrupting the continuity of visual space and significantly reducing the immersive experience. At the same time, there is also a problem of asynchronous physical simulation between channels.

[0027] To address the problems of existing technologies, embodiments of this application provide a scene simulation method, simulation system, device, and storage medium. Through a pre-built scene simulation model asset library, target 3D object models are selected from the disaster simulation model asset library according to the required scene simulation type, constructing different simulation scenes. By pre-caching scene simulation data at the same frame rate, an evolution recording dataset is obtained. In multi-channel simulation mode, the master node server directly calls the scene evolution recording dataset and transmits it to the slave nodes, enabling each slave node to perform simulation rendering using the cached scene evolution recording dataset at the same frame rate. This achieves data state synchronization between channels, realizing smooth operation of complex simulation scenes in multi-channel scenarios. The following section first introduces a disaster simulation method provided by embodiments of this application.

[0028] Figure 1 This is a flowchart illustrating a scene simulation method provided in an embodiment of this application. Figure 1 The provided diagram illustrates the specific steps of the scene simulation method, including: S101~S105.

[0029] S101. Pre-build a scene simulation model asset library. The scene simulation model asset library includes a 3D object model of at least one dynamic object and a 3D object model of at least one static object. Among them, at least one dynamic object is a core dynamic object, and at least one object is a pre-fragmented object. The 3D object model of the pre-fragmented object contains a cluster of broken meshes.

[0030] This application is applied to a scene simulation system for simulating the dynamic changes of a predicted event occurring in a specified scene. The 3D object model is the morphological data of the object, including its external structure and attribute data. Dynamic objects are those whose position changes during the simulation, while pre-fragmented objects are those exhibiting destructive phenomena during the simulation. Static objects can be understood as background object models that do not change during the simulation.

[0031] Furthermore, in order to achieve simulation display of different scenarios, a simulation system that is easy to modify later is first constructed. The scene objects and corresponding object attribute data required for simulation are collected, and the constructed 3D object models and corresponding attribute data are stored in the database to build a scene simulation model asset library. Further, according to the different types of 3D object models, the images, corresponding 3D object attributes and state data of each 3D object model are stored in the scene simulation model asset library to provide simulation data for the next step of building simulation scenes.

[0032] S102. Based on the scene simulation type, select the 3D object model of the target scene object from the scene simulation model asset library.

[0033] S103. Build a simulation scene based on the 3D object model of the target scene object.

[0034] Scene simulation types include logical scenes, functional scenes, and specific scenes. Specific scene simulation types include natural disaster types and non-natural disaster types, with natural disaster types including earthquakes, floods, and fires. Target scene objects are 3D object models that need to appear in the simulation scene; for example, in earthquake simulation, it is necessary to build 3D object models including buildings, bridges, and vehicles.

[0035] Furthermore, to simulate different scenarios, target scene objects within the scenario are selected from the scenario simulation model asset library via software drag-and-drop and placed into the simulation scene. After selection, the position of each target scene object is determined according to its object attributes and state data, thus constructing the simulation scene and providing simulation data for the next step of starting the simulation operation. In the earthquake disaster simulation scenario, buildings, people, vehicles, and surrounding scenes included in the earthquake disaster scenario are selected from the scenario simulation model asset library via software drag-and-drop and placed into designated locations within the simulation scene to construct the earthquake disaster simulation scenario.

[0036] S104. Record the scene evolution at the same frame rate for the core dynamic object and the pre-fragmented object in the simulation scene, obtain and cache the scene evolution recording dataset, and the scene evolution recording dataset shall include at least the breaking animation of the pre-fragmented object and the trajectory data of the core dynamic object.

[0037] Objects in a simulation scene include core dynamic objects, non-core dynamic objects, and pre-fragmented objects. Core dynamic objects are those whose locations of primary interest in the simulation scene undergo significant movement or change, such as cars during an earthquake. Non-core dynamic objects are static objects or objects that are stationary, such as tables, chairs, or the sky background. Pre-fragmented objects refer to objects that will exhibit significant animated impact phenomena during the simulation, such as tall buildings during an earthquake.

[0038] Furthermore, after constructing the simulation scene, in order to improve the display effect of the multi-channel simulation system, the scene evolution of the core dynamic objects and pre-fragmented objects in the simulation scene is recorded in each channel at the same frame rate in advance. The simulation cached fragmentation animation and the trajectory data of the core dynamic objects are stored in a designated database to obtain the scene evolution recording dataset corresponding to the core dynamic objects and pre-fragmented objects. This provides cached data for maintaining the synchronization of multi-channel frame data when performing multi-channel scene simulation in the next step.

[0039] S105. When the simulation channel mode is multi-channel mode, the slave nodes corresponding to each simulation channel mode, based on the synchronous rendering instructions of the master node, synchronously render the cached broken animation and trajectory data, and perform real-time simulation processing and rendering of non-core dynamic objects in the simulation scene.

[0040] The simulation channel mode here can be understood as selecting one or more channels for simulation rendering based on simulation requirements. For example, Figure 2 This is a schematic diagram of an immersive multi-channel display structure provided in an embodiment of this application. According to... Figure 2 The provided diagram shows a five-channel display structure, with each channel corresponding to a display area. Data rendering is performed through the channels within each area to achieve a synchronous simulation display effect.

[0041] When the simulation channel is in multi-channel mode, to achieve synchronized display of frame data for each channel, different simulation processing is applied to the target scene objects in the multi-channel simulation scene. The core dynamic objects and pre-fragmented objects in the simulation scene are fed into a pre-stored cached scene evolution recording dataset. This cached dataset is sent from the master node server to the corresponding slave node for each channel. Each slave node renders the fragmentation animation and trajectory data at the same frame rate as the scene evolution recording dataset, achieving synchronized animation playback and multi-channel synchronized scene simulation and rendering. Simultaneously, considering the low computational requirements of non-core dynamic objects in the simulation scene, real-time simulation calculations are performed on these objects to ensure synchronized frame data display and showcase the scene simulation results. Specifically, in the earthquake disaster simulation scene, the models of objects such as tall buildings in the earthquake disaster simulation scene are fed into a pre-stored cached earthquake disaster dataset via the master node server. This dataset is then sent to the slave nodes of each channel via group nodes for synchronized animation playback at the same frame rate, achieving synchronized earthquake disaster simulation and rendering for each channel. Simultaneously considering the low computational requirements of non-core dynamic objects in earthquake disaster simulation scenarios, real-time simulation calculations are performed on non-core dynamic objects such as sky scenes, ground cracks, and street-side tables and chairs to ensure synchronized display of frame data from slave nodes in each channel, showcasing the disaster simulation results. Through a hybrid approach of pre-calculation and real-time calculation of the target disaster object model, and based on an intelligently adaptive operating strategy, simulations are performed in multi-channel scenarios to ensure smooth operation in complex multi-channel environments and strict synchronization of data states between channels.

[0042] This application provides a scene simulation method. By using a pre-built scene simulation model asset library, and selecting target 3D object models from the disaster simulation model asset library according to the scene simulation type required for simulation, different simulation scenes are constructed. By pre-caching scene simulation data at the same frame rate, an evolution recording dataset is obtained. In multi-channel simulation mode, the master node server directly calls the scene evolution recording dataset and transmits it to the slave nodes, so that each slave node can perform simulation rendering of the scene evolution recording dataset cached at the same frame rate, achieving data state synchronization between channels and realizing a complex simulation scene that runs smoothly in multi-channel scenarios.

[0043] In one possible example scenario, to further determine the construction process of the scene simulation model asset library, Figure 3 This is a schematic diagram of a process for constructing a scene simulation model asset library provided in an embodiment of this application. Figure 3 This refers to the process of building a scene simulation model asset library in step S101. Figure 3 This description is based on the previous embodiment. According to... Figure 3 The provided diagram illustrates the specific steps for constructing a scene simulation model asset library, including steps S301 to S305.

[0044] S301. Obtain the 3D object models of each dynamic object and each static object, wherein at least one object is a pre-fragmented object.

[0045] The three-dimensional object model mentioned here refers to the scene objects needed in scene simulation. These can be movable facilities or stationary objects, such as people, vehicles, building complexes, trains, rivers, forests, tables, chairs, and benches.

[0046] S302. Set physical simulation properties for the 3D object model.

[0047] Physical simulation attributes refer to the mass, shape, color, position data, material, and constraints on the changes in the shape of a 3D object model.

[0048] Furthermore, create or import 3D object models that participate in real-time physical scene simulation, and define physical simulation attributes for them as state data and object attributes of the object models, further defining the characteristics of the 3D object models and providing reference data for the next step of building a scene simulation model asset library.

[0049] In one possible example scenario, Figure 4 This is a schematic diagram illustrating an object simulation effect provided in an embodiment of this application. According to... Figure 4The provided diagrams import 3D object models for real-time physical simulation. Taking a road and street scene as an example, the models include vehicles, streetlights, trees, and street benches. The physical simulation attribute parameters are defined according to the object type, and the object models of collision objects such as bicycles are simplified to achieve preprocessing of simulation assets.

[0050] S303. Perform pre-fragmentation processing on the 3D object model of the pre-fragmented object according to the predefined fracturing algorithm to obtain a cluster of broken meshes.

[0051] S304. Optimize the 3D object model of the static object according to the predefined optimization algorithm to obtain the optimized 3D object model of the static object.

[0052] Furthermore, for earthquake disaster simulation scenarios, geometric objects such as buildings and bridges that need to demonstrate damage effects are pre-fragmented to generate fragmented mesh clusters, providing the basis for triggering fragmentation animations and achieving preprocessing of fragmented geometric objects. Static objects that do not participate in physical simulation calculations (such as distant mountains and undamaged roads) are added to enrich the scene, and multi-level detail optimization and lighting baking are performed on the static objects. After baking optimization, the number of rendering draw calls for static objects can be reduced by more than 60%. The pre-built earthquake disaster simulation model asset library can improve the efficiency of earthquake disaster scene construction, reduce real-time computing load, and achieve static scene object optimization, obtaining optimized state data for each static object, providing reference data for real-time simulation calculations of static objects in subsequent multi-channel simulations.

[0053] In one possible example scenario, Figure 5 This is an example of generating a broken mesh cluster according to an embodiment of this application. Figure 6 This is an example of another method for generating fragmented mesh clusters, provided in an embodiment of this application. According to... Figure 5 and Figure 6 The provided diagram illustrates how, in an earthquake disaster scenario, the geometry of buildings and roads is pre-fragmented to generate fragmented mesh clusters, and their respective damage parameters are set to provide the basis for triggering fragmentation animations.

[0054] S305. Construct a scene simulation model asset library based on the 3D object models with physical simulation attributes set for each dynamic object, the optimized 3D object models for each static object, and the fragmented mesh clusters.

[0055] The model assets processed in steps S301-S305, along with their respective 3D object models with physical simulation attributes, optimized 3D object models of each static object, and fragmented mesh clusters, are uniformly stored in the database for management, forming a reusable scene simulation model database, which also serves as a scene simulation model asset library.

[0056] In one possible example scenario, Figure 7This is a schematic diagram of the structure of a scene simulation model asset library provided in an embodiment of this application. Figure 8 This is a schematic diagram of the structure of another scene simulation model asset library provided in this application embodiment. According to Figure 7 and Figure 8 The provided diagram illustrates how, in an earthquake disaster simulation scenario, following the simulation asset preparation and preprocessing process, three-dimensional physical models of core dynamic objects potentially needed in the earthquake disaster simulation scenario are obtained. Figure 7 The physical simulation model asset library stores non-core dynamic objects, through Figure 8 The pre-fracture model of the core dynamic object is constructed to obtain the building fracture prefabricated model asset library, which is used to store the three-dimensional object models of the core dynamic object and the pre-fracture object.

[0057] This application embodiment determines 3D object models, performs different preprocessing on the 3D object models according to different object types, and stores the preprocessed 3D object models, corresponding object simulation attributes, optimized 3D object models of each static object, and state data of broken mesh clusters to obtain a scene simulation model asset library. By predefining physical attributes and pre-fragmentation, the realism of physical behavior and computational feasibility during simulation are ensured. By distinguishing between dynamic and static objects and performing targeted optimization, multi-channel rendering performance is improved.

[0058] Figure 9 This is a schematic diagram of a process for building a target simulation scene provided in an embodiment of this application. Figure 9 This refers to the process of building the target simulation scene in step S102. Figure 9 The description is based on the first embodiment. Figure 9 The provided diagram illustrates the specific steps for building the target simulation scene, including steps S901 to S903.

[0059] S901. Based on the physical simulation attributes and pre-fragmented object attributes corresponding to the scene simulation type, select the three-dimensional object model of the target scene object from the scene simulation model asset library.

[0060] S902. Configure the initial scene according to the scene simulation type.

[0061] S903. Based on the positions where the target scene objects should appear in the scene, deploy the 3D object models of the selected target scene objects in the initial scene to obtain the simulation scene.

[0062] The target scenarios include disaster scenarios, which can be natural or non-natural disasters. Natural disasters include earthquakes, floods, and fires.

[0063] Specifically, during disaster scenario simulation, target dynamic objects are selected and configured according to the disaster scenario type. For example, a flood scenario requires setting up water emission sources and inundation area terrain, while an earthquake scenario requires setting up affected building clusters. Specifically, core dynamic objects (e.g., vehicles) and breakable / pre-breakable objects (e.g., building clusters) are selected and placed from the scenario simulation model asset library to their corresponding locations in the scene, completing the placement of target scene objects and binding simulation objects. Optimized static objects are added from the scenario simulation model asset library to achieve the placement of static auxiliary objects while ensuring the visual richness of the simulation scene and avoiding unnecessary performance overhead. Based on the disaster simulation target objects, a simulation scene containing dynamic interactive elements and a static environment, comprising a 3D object model, is flexibly and quickly assembled from the scenario simulation model asset library.

[0064] In one possible example scenario, Figure 10 This is a rendering of an earthquake disaster simulation scenario provided in an embodiment of this application. According to... Figure 10 The provided diagrams illustrate how earthquake disaster simulation scenarios can be built based on a pre-built model database. Physical simulation models and fractured geometry are selected and combined with static auxiliary objects to enrich the simulation. Earthquake disaster simulation scenarios can be quickly constructed by dragging and dropping target scene objects from the scenario simulation model asset library. This reduces the setup time for a disaster simulation scenario (such as an earthquake in a specific neighborhood) from the traditional several weeks to 1-3 days, significantly improving the efficiency and flexibility of simulation preparation.

[0065] This application embodiment quickly selects the required 3D object models in the simulation scene from the scene simulation model asset library, and quickly builds the simulation scene by dragging and dropping them to the target location. Compared with the conventional method of building each simulation object first and then building the simulation scene, this application can greatly improve the efficiency and flexibility of simulation preparation.

[0066] Figure 11 This is a schematic diagram of a process for determining a scene evolution recording dataset provided in an embodiment of this application. Figure 11 This refers to the process of determining the scene evolution recording dataset in step S103. Figure 11 This description is based on the first embodiment. The simulation scenario is set as a disaster simulation scenario, according to... Figure 11 The provided diagram illustrates the specific steps for determining the scene evolution recording dataset, including: S1101~S1102.

[0067] S1101. Set the recording mode's running frame rate and the fracture disaster parameters corresponding to the multi-level disaster intensity.

[0068] S1102. For each level of disaster intensity, in recording mode, the disaster simulation scenario is simulated and calculated based on the fracture disaster parameters corresponding to that level of disaster intensity. During the simulation, the fracture animation of the pre-fractured object and the trajectory data of the core dynamic object are recorded at the same running frame rate to obtain and cache the scene evolution recording dataset corresponding to that level of disaster intensity.

[0069] To address the risk of asynchronous frame rates in multi-channel simulations of core dynamic and pre-fragmented objects in a multi-channel environment, a recording mode needs to be selected for these objects. On a single high-performance workstation, a cached data recording mode is enabled. In earthquake disaster simulations, for core dynamic objects, a fixed frame rate is set to ensure that the fracturing animation and physical simulation data are recorded at the same frame rate, guaranteeing consistent simulation states for each channel during data playback. For pre-fragmented objects, a disaster simulation is triggered, and the forces applied in the disaster scenario cause the fracturing and collapse of the geometry in the scene. The fracturing sequence animation is recorded, resulting in a fracturing animation. Data recording for core dynamic objects is also performed, synchronously recording the core state data (position, rotation, velocity, etc.) of each frame for dynamic simulation objects (such as the ground and vehicles), ignoring their unchanging mesh data. Lightweight processing is used for auxiliary, non-core dynamic objects; their dynamic effects are not recorded frame-by-frame. Their state data is generated synchronously during simulation using a unified parametric algorithm, and the trajectory data of the core dynamic objects is obtained and cached.

[0070] For earthquake disaster scenarios, if it is necessary to adjust the damage parameters in the earthquake disaster simulation scenario, this can be done by adjusting the earthquake disaster intensity parameters (such as magnitude from 3.0 to 8.0) and other parameters that need to be modified simultaneously, repeating the above recording process to generate cached datasets containing fragmentation animations and core object trajectories corresponding to different earthquake disaster levels. This step is a key preprocessing step that balances the requirements of smooth multi-channel operation and real-time single-machine operation. It moves the most performance-intensive geometric fragmentation calculation process and complex object motion solution process to the recording stage to generate fragmentation animation sequences and keyframe data that can be played efficiently.

[0071] In one possible example scenario, Figure 12 This is a schematic diagram of the structure of fragmented cached data provided in an embodiment of this application. Figure 13 This is a schematic diagram of core object state data provided in an embodiment of this application. Figure 14 This is a schematic diagram of the structure of a cache fragmentation animation provided in an embodiment of this application. According to... Figure 12 , Figure 13 as well as Figure 14The provided diagram illustrates a simulation on a single high-performance workstation, set to cached data recording mode, with the earthquake simulation level set to strong earthquake. Simulation calculations are performed at a fixed frame rate, simultaneously recording cached fragmentation animation data and keyframe data of core ground shaking. By adjusting the earthquake intensity parameters and repeating the cached data recording process, data is generated as shown below. Figure 14 This dataset contains cached datasets including fragmentation animations and core object trajectories, corresponding to different earthquake disaster levels. By recording only the state data of core dynamic objects and pre-fragmented objects, the amount of data recorded per frame can be reduced by 70-85% compared to recording the entire scene, which helps improve the simulation efficiency of core dynamic objects.

[0072] This application embodiment reduces the simulation calculation difficulty of core dynamic objects and pre-fragmented objects by moving the most performance-intensive geometric breaking calculation and complex object motion solution process to the recording stage. It generates breaking animation data and key frame data that can be played efficiently, provides cached data for the simulation process in multi-channel mode, reduces the amount of simulation calculation, and uses each channel to perform simulation calculation at the same running frame rate to solve the problem of asynchronous running frame rates between channels caused by multi-channel simulation.

[0073] Figure 15 This is a flowchart illustrating another scene simulation method provided in an embodiment of this application. Figure 15 This refers to the process of scene simulation method in step S105. Figure 15 The description is based on the first embodiment. Figure 15 The provided diagram illustrates the specific steps of the scene simulation method, including: S1501~S1504.

[0074] S1501. Receive disaster level selection instruction and determine the target level disaster intensity based on the disaster level selection instruction.

[0075] S1502. When the simulation channel mode is multi-channel mode, the master node obtains the trajectory data corresponding to the disaster intensity of the target level from the cache, and distributes the trajectory data to each slave node through the synchronous rendering command.

[0076] S1503. Through each slave node, based on the synchronous rendering instructions distributed by the master node, obtain the fragmentation animation corresponding to the target level disaster intensity from the cache.

[0077] S1504 synchronously renders broken animations and trajectory data based on the same running frame rate, and performs real-time simulation processing and rendering of non-core dynamic objects in disaster simulation scenarios.

[0078] Disaster level refers to the severity of a disaster when it occurs. Examples include earthquake disaster level and typhoon disaster level.

[0079] Furthermore, when the simulation channel mode is multi-channel mode, the physical simulation engine is run through the master node to convert the trajectory data of the cached core dynamic objects and the breaking animation of the pre-fragmented objects into binary state data sequences. The binary state data sequences are then packaged into synchronous rendering instructions and broadcast to each slave node via UDP.

[0080] In multi-channel simulation mode, based on the earthquake disaster level selected by the user, the corresponding fragmentation animation and core object trajectory data are loaded from the earthquake disaster scene evolution recording dataset. Simultaneously, the master node of the computer simulation system reads and distributes core object state data frame by frame, while each slave node synchronously plays locally cached fragmentation animation data at the same frame rate, achieving multi-channel synchronous simulation. Furthermore, based on the master node's simulation trigger command, consistent real-time simulation calculations are performed on other non-cached objects in the disaster simulation scene (such as water, smoke, and real-time particles), achieving physical simulation and rendering of non-core objects. Multi-channel disaster simulation is achieved through cache-driven and real-time synchronous simulation calculations.

[0081] In one possible example scenario, Figure 16 This is a schematic diagram of an earthquake disaster simulation mode selection provided in an embodiment of this application. Figure 17 This is a simulation diagram of an earthquake disaster provided in an embodiment of this application. According to... Figure 16 The provided diagram illustrates the system startup process in a multi-channel scenario. It initializes the earthquake disaster simulation scenario, automatically detects hardware configuration and network topology. If the number of channels is greater than 1, it is determined to be in multi-channel mode, and the simulation mode is set to cached data simulation. If the number of channels is equal to 1, it is determined to be in single-machine desktop mode, and the cached mode is adjusted to real-time data simulation. Based on... Figure 17 The provided diagram illustrates how, in an earthquake disaster simulation scenario, the simulation level (e.g., moderate to strong earthquake) is selected, and the simulation process begins. In a multi-channel scenario, corresponding fragmentation and ground shaking cache data are loaded. The cached data is replayed at a fixed frame rate. Non-core dynamic objects, such as ground objects affected by ground shaking, undergo real-time physical simulation and are synchronized with cluster data to obtain the earthquake simulation effect display. Figure 17In terms of deployment and debugging results of the CAVE immersive binocular stereoscopic display system, the use of cached data for simulation rendering at the same frame rate across each channel, along with real-time calculation and synchronous hybrid driving, ensures that the state of objects and broken geometry in the earthquake disaster simulation scene is synchronized across any channel. Furthermore, the frame rate has increased from the initial 0.05fps in real-time simulation to approximately 15fps, ensuring smooth operation in an immersive environment. This achieves smooth rendering at high resolution with five channels, effectively eliminating visual tearing and misalignment, guaranteeing the visual quality of the immersive experience, improving hardware resource utilization efficiency, and saving hardware costs for system construction and upgrades.

[0082] This application embodiment selects a multi-channel simulation mode. In the multi-channel mode, corresponding cached data is loaded for the core dynamic object and the pre-fragmented object and sent to each slave node. The slave node corresponding to each channel plays back the data at the same running frame rate, thereby achieving synchronous simulation effect between each channel. Real-time simulation calculation is performed for non-core dynamic objects. The adaptive data processing architecture supports intelligent switching between cache playback and real-time simulation calculation, thus ensuring smooth synchronization and consistent operation experience in the multi-channel mode.

[0083] Figure 18 This is a flowchart illustrating another scene simulation method provided in the embodiments of this application. Figure 18 The description is based on the first embodiment. Figure 18 The provided diagram illustrates the specific steps of the scene simulation method, including: S1801~S1805.

[0084] S1801. By detecting the simulation running environment, determine the appropriate simulation channel mode. The simulation channel mode includes single-channel mode or multi-channel mode.

[0085] S1802. When the simulation channel mode is single-channel mode, determine the load of a single machine.

[0086] S1803. Under the condition that the load of a single machine is not greater than the preset load threshold, obtain the real-time single-machine scene parameters and the real-time single-machine simulation trigger command.

[0087] S1804. Based on the real-time stand-alone scene parameters and the real-time stand-alone simulation trigger command, perform real-time simulation of the simulation scene using a stand-alone machine.

[0088] S1805: When the load on a single machine exceeds the load threshold, render the broken animation and trajectory data in the rendering cache, and perform real-time simulation and rendering of non-core dynamic objects in the simulation scene.

[0089] In earthquake disaster scenarios, after the earthquake disaster simulation scenario is built and the cached scenario evolution recording dataset is completed, the system simulation is started, the hardware configuration and network topology are automatically detected, and environmental awareness analysis and judgment are performed to determine whether the earthquake disaster simulation scenario is in multi-channel cluster mode or single-machine desktop mode.

[0090] In standalone mode, the load of the standalone operation mode is determined. When the load in standalone mode does not exceed a preset load threshold, the earthquake disaster simulation system merges the simulation calculations of the master and slave nodes. Based directly on the original 3D object model and physical disaster parameters, each slave node performs real-time physical simulation calculations at the same frame rate. No data caching is required; the earthquake disaster simulation can be displayed synchronously using a fully real-time simulation method. However, when the load in standalone mode exceeds the load threshold, the excessive number of objects may cause display stuttering issues if real-time simulation calculations are still performed. In this case, cached data corresponding to some core dynamic objects and pre-fragmented objects in the earthquake scene needs to be loaded, parsed, and rendered to display the earthquake disaster simulation screen.

[0091] Furthermore, the object attribute parameters corresponding to the target scene objects in the simulation scene are adjusted in real time to obtain the adjusted attribute parameters, and the target scene objects in the simulation scene are simulated and rendered based on the adjusted attribute parameters.

[0092] This application allows users to adjust object attribute parameters in scene simulation in real time, and the simulation scene responds based on the input parameters, without relying on pre-recorded fixed cache datasets. The simulation scene can be reset or modified at the end or during simulation, and the disaster level in the scene can be reselected for corresponding data loading and playback. This achieves parameterized real-time driving of disaster effects and rapid scene adaptation, enabling a single codebase for two deployment modes, reducing R&D investment and facilitating flexible deployment as needed. The system automatically selects the optimal strategy, providing full real-time interaction in single-machine mode and ensuring smooth synchronization and a consistent user experience in multi-channel mode.

[0093] In one possible example scenario, Figure 19 This is a flowchart illustrating another earthquake disaster simulation method provided in an embodiment of this application. According to... Figure 19The provided diagram illustrates the process of first preparing simulation assets. In step S191, a 3D object model is created, defining the physical simulation attributes of the earthquake scene, preprocessing the fractured geometry, and optimizing the state data of static objects to obtain an earthquake disaster simulation model asset library. Next, by defining earthquake disaster types, 3D object models of earthquake disasters are dragged and dropped directly from the earthquake disaster scene simulation model asset library. Through binding and optimization, an earthquake disaster simulation scene S291 is obtained. Before conducting the earthquake simulation, cached data for each level of disaster in the earthquake disaster simulation scene is recorded according to step S391. By selecting the cache recording mode, real-time earthquake disaster simulation calculation is initiated. The master node server sends cache instructions to the slave nodes. Each channel of the slave nodes records the fracture and collapse animations of the pre-fractured geometry at the same running frame rate to obtain fracture animation data, and records the frame state data of the core dynamic objects, generating a multi-level disaster dataset for synchronous simulation of the earthquake disaster. The final step, S491, enables dynamic switching of the earthquake disaster simulation operation mode. By detecting the operating status, mode adaptation is performed. In multi-channel cluster mode, real-time single-machine scene parameters and corresponding trigger commands are acquired, and the trigger commands are sent to slave nodes. Each slave node in each channel loads earthquake disaster cache data of a specified level, and playback and synchronization are performed according to the data recording timeline and the same running frame rate. This allows the master and slave nodes to parse and replay earthquake disaster data for core dynamic objects and pre-fragmented objects at a fixed frequency. Simultaneously, real-time simulation calculations are performed on non-core dynamic objects, displaying a smooth and synchronized multi-channel immersive earthquake disaster simulation in multi-channel mode. In single-machine mode, real-time single-machine scene parameters and trigger commands are acquired, the physical properties of the simulation model are enabled, and real-time simulation calculations are performed based on the real-time single-machine scene parameters, resulting in a highly interactive and adjustable earthquake disaster display.

[0094] This application abandons the traditional model that relies on fixed pre-made animations and constructs a technical solution that can generate scene evolution effects that conform to physical laws in real time. It supports dynamic adjustment of key parameters in the simulation scene during operation or deduction, and can flexibly and quickly load and adapt different 3D scene models, solving the shortcomings of traditional methods such as long production cycles, poor flexibility, and low scalability. Simultaneously, it overcomes the performance bottleneck of multi-channel stereo rendering by simulating at the same frame rate from nodes, achieving synchronous simulation effects between multiple channels and ensuring smooth and synchronized visuals. Addressing the enormous performance pressure brought by five-channel binocular stereo high-resolution rendering, an efficient particle and geometry optimization rendering strategy is designed to solve the problem of low frame rates caused by excessive rendering load and eliminate the resulting misalignment or tearing of screen updates between channels, ensuring a smooth and stable immersive scene simulation experience on multi-channel hardware clusters. It also achieves the effect of ensuring global consistency of physical states under multi-channel rendering. A reliable physical state synchronization mechanism is established to solve the problem of inconsistent states caused by independent calculations by each rendering node. By utilizing cached datasets, this approach ensures that all dynamic elements, such as building collapse trajectories, debris splash paths, and fluid spread ranges, are presented consistently across all channels and at all times, guaranteeing visual continuity in earthquake disaster simulations. Furthermore, this application's embodiments construct an intelligent, adaptive, unified data processing and scheduling architecture. This architecture intelligently identifies whether the current system operating environment is in single-machine or multi-channel mode and dynamically adapts to the optimal data computation and rendering strategy, achieving optimal performance and flexibility across various deployment modes.

[0095] Figure 20 This is a schematic diagram of the structure of a disaster simulation system provided in an embodiment of this application. According to... Figure 20 The provided diagram illustrates the specific structure of the disaster simulation system 1000, which includes: The asset management module S2001 is used to pre-build a scene simulation model asset library. The scene simulation model asset library includes a 3D object model of at least one dynamic object and a 3D object model of at least one static object. Among them, at least one dynamic object is a core dynamic object, and at least one object is a pre-fragmented object. The 3D object model of the pre-fragmented object contains a cluster of broken meshes.

[0096] The scene construction module S2002 is used to select the 3D object model of the target scene object from the scene simulation model asset library according to the scene simulation type.

[0097] The scene building module S2002 is also used to build simulation scenes based on the 3D object models of the target scene objects.

[0098] The data caching and recording module S2003 is used to record the scene evolution at the same frame rate for the core dynamic objects and pre-fragmented objects in the simulation scene, and to obtain and cache the scene evolution recording dataset. The scene evolution recording dataset includes at least the breaking animation of the pre-fragmented objects and the trajectory data of the core dynamic objects.

[0099] The simulation rendering module S2004 is used to perform real-time simulation processing and rendering of non-core dynamic objects in the simulation scene by using the slave nodes corresponding to each simulation channel mode, based on the synchronous rendering instructions of the master node, to synchronously render cached broken animations and trajectory data, and to perform synchronous rendering of non-core dynamic objects in the simulation scene when the simulation channel mode is multi-channel mode.

[0100] In one possible example scenario, Figure 21 This is a schematic diagram of another earthquake disaster simulation system provided in an embodiment of this application. According to... Figure 21 The provided diagram shows that the earthquake disaster simulation system includes: data source S211, data and asset layer S212, core simulation service layer S213, interaction and application layer S214, and rendering and output layer S215.

[0101] Among the data sources S211, the main components include disaster scenario evolution process data S2111, disaster-related formulas S2112, and disaster scenario data S2113.

[0102] The data and asset layer S212 serves as the system's foundational data support, centrally storing and managing all static and dynamic data assets required for earthquake simulation. These primarily include: 3D geometric models, scene data, object physical attribute parameters, and pre-calculated multi-level earthquake disaster scene evolution recording datasets (such as building fragmentation animation sequences and key object motion trajectories). This layer provides a unified and efficient data access interface for the core simulation service layer. It includes multiple types of asset libraries and databases, specifically: the physical simulation asset library S2121, the fragmentation simulation asset library S2122, the disaster simulation scene asset library S2123, and the disaster simulation database S2124.

[0103] The core simulation service layer S213 primarily carries the core simulation logic, provides tools and services, supports flexible scenario construction, and binds simulation parameters. Within S213, parameterized models of disasters such as earthquakes, fires, and floods are implemented, and their triggering conditions and processes are managed. In a single-machine high-performance environment, pre-computation tasks are executed, disaster evolution data is recorded, and structured storage is achieved in the database and asset repository. Dynamic management of cache playback and real-time computation modes is implemented. In a distributed environment, strict synchronization of the states of all simulation computing nodes is ensured.

[0104] The core simulation service layer S213 specifically includes: a rapid scenario construction and asset management module S1, a cached data recording and dataset management module S2, a core disaster simulation logic and control module S3, a physical simulation state synchronization module S4, and an adaptive rendering module for running modes S5.

[0105] The Scene Rapid Construction and Asset Management module S1 provides visualization tools that support rapid drag-and-drop construction of earthquake disaster scenes from the asset library. This primarily includes: a visual scene editor S11, which provides a graphical interface for terrain rendering, object placement, and earthquake disaster parameter binding (such as defining flood levels and ignition points); and an asset pipeline S12, which automates imported models, performing pre-fragmentation, LOD (Level of Detail) generation, collider generation, and physical attribute annotation.

[0106] The cached data recording and dataset management module S2 is primarily used in a single-machine high-performance environment. It records, encodes, stores, and manages earthquake disaster process data for efficient multi-channel playback by using the same frame rate across each channel. The cached data recording and dataset management module S2 mainly includes: a recording controller S21, which provides an earthquake graphical interface allowing users to set recording parameters (such as magnitude range, recording frame rate, etc.); a keyframe data recording module S22, used to capture and compress the state of core dynamic objects (position, rotation, physical switching state) at a fixed sampling rate during physical simulation; a fragmentation animation recording module S23, used to bake the dynamic deformation and movement process of pre-fragmented geometry under earthquake disaster action into an animation cache sequence and perform efficient compression; and a disaster simulation dataset S24, used to establish the correlation between disaster type, disaster level, keyframe data, and fragmentation animation data to generate a structured disaster dataset.

[0107] The core disaster simulation logic and control module S3 allows for the definition and management of core disaster simulation logic, parameterized models, and triggering mechanisms, controlling the time, location, and intensity of disasters. The core disaster simulation logic and control module S3 mainly includes: a disaster parameterized model library S31, used to encapsulate mathematical models of different disasters. Examples include: earthquake models with parameters such as epicenter location and magnitude; fire models with parameters such as fire source point, spread rate, wind direction, and combustible material type; and flood models with parameters such as rainfall intensity, wind force, and wind direction. It also includes disaster triggers S32, used for script-based triggering, which triggers at specific times based on the simulation script; interactive triggering, manually triggered by the user during the simulation; and conditional triggering, automatically triggered based on the simulation status (e.g., triggering dam failure when the water level exceeds the warning line). Finally, it includes disaster impact range management S32, used to define the spatial range of disaster impacts and object selection rules.

[0108] In the physical simulation state synchronization module S4, global physical states can be generated and synchronized. The physics engine runs only on the main control server, serializing state data such as the transformation matrix of dynamic objects and the animation frame index of broken objects into binary, and broadcasting the state with low latency based on the User Datagram Protocol (UDP) to achieve object state synchronization.

[0109] The adaptive rendering module S5 implements automatic detection of the runtime environment and decision-making of runtime strategies. This mainly includes: Environment Awareness S51, which collects information such as hardware configuration (number / performance of GPUs), network topology, and number of projection channels; and Strategy Decision S52, which uses built-in rules and performance models. For example, if the number of channels > 1, the runtime mode is a multi-channel cached data simulation mode. If the number of channels = 1, the runtime mode is a single-machine real-time computing simulation mode.

[0110] In the interaction and application layer S214, an intuitive and convenient human-computer interaction experience can be provided to end users, including a visual parameter configuration tool, a multi-channel immersive environment configuration, an integrated head tracking and controller interaction multi-channel stereoscopic display system, support for VR head-mounted devices, and support for users' natural interaction and navigation in the simulation environment.

[0111] In the rendering and output layer S215, rendering and output can be adapted to various display environments. In standalone mode, it directly performs high-quality real-time disaster simulation rendering. In multi-channel immersive environments (such as CAVE), a master-slave rendering node cluster works collaboratively. Each computing node renders the specified viewpoint image in parallel according to synchronous triggering instructions at the same running frame rate, and automatically completes geometric correction and synchronous output between multiple channels, realizing human-computer interaction control. It also supports rendering output for immersive display devices such as VR headsets.

[0112] The disaster simulation system provided in this embodiment can be as follows: Figure 20 The disaster simulation system 1000 shown can achieve Figures 1-19 For a detailed description of the technical effects of the disaster simulation method shown, please refer to [link / reference]. Figures 1-19 The corresponding explanation is concise and will not be elaborated upon here.

[0113] Figure 22 This is a schematic diagram of the structure of a scene simulation device provided in an embodiment of this application. The scene simulation device may include a processor 2201 and a memory 2202 storing computer program instructions.

[0114] Specifically, the processor 2201 may include a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits that can be configured to implement the embodiments of this application.

[0115] Memory 2202 may include mass storage for data or instructions. For example, and not limitingly, memory 2202 may include a hard disk drive (HDD), floppy disk drive, flash memory, optical disk, magneto-optical disk, magnetic tape, or Universal Serial Bus (USB) drive, or a combination of two or more of these. Where appropriate, memory 2202 may include removable or non-removable (or fixed) media. Where appropriate, memory 2202 may be internal or external to the integrated gateway disaster recovery device. In a particular embodiment, memory 2202 is non-volatile solid-state memory.

[0116] Memory may include read-only memory (ROM), random access memory (RAM), disk storage media devices, optical storage media devices, flash memory devices, and electrical, optical, or other physical / tangible memory storage devices. Therefore, typically, memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., memory devices) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors), it is operable to perform the operations described with reference to the method according to the first aspect of this application.

[0117] The processor 2201 reads and executes computer program instructions stored in the memory 2202 to implement any of the scene simulation methods in the above embodiments.

[0118] In one example, the scene simulation device may also include a communication interface 2203 and a bus 2210. For example, Figure 22 As shown, the processor 2201, memory 2202, and communication interface 2203 are connected through bus 2210 and complete communication with each other.

[0119] The communication interface 2203 is mainly used to realize communication between various modules, devices, units and / or equipment in the embodiments of this application.

[0120] Bus 2210 includes hardware, software, or both, that couples components of an online data traffic metering device together. For example, and not limitingly, the bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Enhanced Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), HyperTransport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an Infinite Bandwidth Interconnect, a Low Pin Count (LPC) bus, a memory bus, a Microchannel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local (VLB) bus, or other suitable buses, or combinations of two or more of these. Where appropriate, bus 2210 may include one or more buses. Although specific buses are described and illustrated in embodiments of this application, any suitable bus or interconnect is contemplated herein.

[0121] Furthermore, in conjunction with one of the scene simulation methods in the above embodiments, this application embodiment can provide a computer storage medium for implementation. The computer storage medium stores computer program instructions; when these computer program instructions are executed by a processor, they implement any of the scene simulation methods in the above embodiments.

[0122] It should be clarified that this application is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of this application is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications, and additions, or change the order of steps, after understanding the spirit of this application.

[0123] The functional blocks shown in the above-described structural diagram can be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, they can be, for example, electronic circuits, application-specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, etc. When implemented in software, the elements of this application are programs or code segments used to perform the required tasks. Programs or code segments can be stored on a machine-readable medium or transmitted over a transmission medium or communication link via data signals carried on a carrier wave. "Machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, fiber optic media, radio frequency (RF) links, etc. Code segments can be downloaded via computer networks such as the Internet, intranets, etc.

[0124] It should also be noted that the exemplary embodiments mentioned in this application describe methods or systems based on a series of steps or apparatus. However, this application is not limited to the order of the above steps; that is, the steps can be performed in the order mentioned in the embodiments, or in a different order, or several steps can be performed simultaneously.

[0125] The aspects of this application have been described above with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It should be understood that each block in the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that these instructions, executable via the processor of the computer or other programmable data processing apparatus, enable the implementation of the functions / actions specified in one or more blocks of the flowchart illustrations and / or block diagrams. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor, or a field-programmable logic circuit. It is also understood that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can also be implemented by dedicated hardware performing the specified functions or actions, or can be implemented by a combination of dedicated hardware and computer instructions.

[0126] The above description is merely a specific implementation of this application. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, modules, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here. It should be understood that the protection scope of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the protection scope of this application.

Claims

1. A scene simulation method, characterized in that, include: A scene simulation model asset library is pre-built. The scene simulation model asset library includes a 3D object model of at least one dynamic object and a 3D object model of at least one static object. The at least one dynamic object is a core dynamic object, and the at least one object is a pre-fragmented object. The 3D object model of the pre-fragmented object contains a cluster of broken meshes. Based on the scene simulation type, select the 3D object model of the target scene object from the scene simulation model asset library; A simulation scene is built based on the three-dimensional object model of the target scene object; The scene evolution recording is performed at the same frame rate on the core dynamic object and the pre-fragmented object in the simulation scene to obtain and cache the scene evolution recording dataset. The scene evolution recording dataset includes at least the breaking animation of the pre-fragmented object and the trajectory data of the core dynamic object. When the simulation channel mode is multi-channel mode, the cached broken animation and trajectory data are rendered synchronously through the slave nodes corresponding to each simulation channel mode, based on the synchronous rendering instructions of the master node, and the non-core dynamic objects in the simulation scene are simulated and rendered in real time.

2. The method according to claim 1, characterized in that, The pre-built scene simulation model asset library includes: Obtain 3D object models of each dynamic object and each static object, wherein at least one object is a pre-fragmented object; Set physical simulation properties for the three-dimensional object model; The three-dimensional object model of the pre-fragmented object is pre-fragmented according to a predefined fracturing algorithm to obtain the fracturing mesh cluster; The three-dimensional object model of the static object is optimized according to a predefined optimization algorithm to obtain the optimized three-dimensional object model of the static object. Based on the three-dimensional object models with physical simulation attributes set for each dynamic object, the optimized three-dimensional object models for each static object, and the fragmented mesh cluster, a scene simulation model asset library is constructed.

3. The method according to claim 2, characterized in that, The step of selecting 3D object models of target scene objects from the scene simulation model asset library based on the scene simulation type includes: Based on the physical simulation attributes and pre-fragmented object attributes corresponding to the scene simulation type, three-dimensional object models of the target scene objects are selected from the scene simulation model asset library. The construction of the simulation scene based on the 3D object model of the target scene object includes: Configure the initial scene according to the scene simulation type; Based on the positions where the target scene objects should appear in the scene, the 3D object models of the selected target scene objects are deployed in the initial scene to obtain the simulation scene.

4. The method according to claim 1, characterized in that, The simulation scenario is a disaster simulation scenario; The process of recording the scene evolution at the same frame rate for the core dynamic objects and pre-fragmented objects in the simulation scenario to obtain and cache the disaster evolution recording dataset includes: Set the frame rate for the recording mode and the fracture disaster parameters corresponding to the multi-level disaster intensity; For each level of disaster intensity, in the recording mode, the disaster simulation scenario is simulated and calculated based on the fragmentation disaster parameters corresponding to that level of disaster intensity. During the simulation, the fragmentation animation of the pre-fragmented object and the trajectory data of the core dynamic object are recorded at the same running frame rate to obtain and cache the scene evolution recording dataset corresponding to that level of disaster intensity.

5. The method according to claim 4, characterized in that, In the case of a multi-channel simulation channel mode, the process involves synchronously rendering the cached fragmented animation and trajectory data based on the synchronous rendering instructions of the master node, using the slave nodes corresponding to each channel mode, and performing real-time simulation processing and rendering of non-core dynamic objects in the simulation scene, including: Receive a disaster level selection instruction, and determine the target level disaster intensity based on the disaster level selection instruction; When the simulation channel mode is multi-channel mode, the master node obtains the trajectory data corresponding to the target level disaster intensity from the cache, and distributes the trajectory data to each of the slave nodes through synchronous rendering instructions; Each of the slave nodes retrieves the fragmentation animation corresponding to the target level of disaster intensity from the cache based on the synchronous rendering instructions distributed by the master node; The fragmentation animation and trajectory data are rendered synchronously based on the same running frame rate, and non-core dynamic objects in the disaster simulation scene are simulated and rendered in real time.

6. The method according to claim 5, characterized in that, In the case of a multi-channel simulation channel mode, the master node retrieves the trajectory data corresponding to the target level disaster intensity from the cache, and distributes the trajectory data to each slave node via a synchronous rendering command, including: When the simulation channel mode is multi-channel mode, the physical simulation engine is run through the master node to convert the cached trajectory data of the core dynamic object and the breaking animation of the pre-fragmented object into a binary state data sequence, and then package the binary state data sequence into a synchronous rendering instruction, which is broadcast to each of the slave nodes via UDP.

7. The method according to claim 1, characterized in that, When the simulation channel mode is a multi-channel mode, before synchronously rendering the cached fragmentation animation and trajectory data based on the synchronous rendering instructions of the master node through the slave nodes corresponding to each channel mode, and before performing real-time simulation processing and rendering of non-core dynamic objects in the disaster simulation scene, the method further includes: By detecting the simulation running environment, the appropriate simulation channel mode is determined, which includes single-channel mode or multi-channel mode; When the simulation channel mode is single-channel mode, determine the load of a single machine; If the load of a single machine is not greater than a preset load threshold, obtain the real-time single-machine scene parameters and the real-time single-machine simulation trigger command. Based on the real-time stand-alone scene parameters and the real-time stand-alone simulation trigger command, the simulation scene is simulated in real time by the stand-alone machine. When the load on the single machine exceeds the load threshold, the broken animation and trajectory data in the rendering cache are rendered, and the non-core dynamic objects in the simulation scene are simulated and rendered in real time.

8. A scene simulation system, characterized in that, include: The asset management module is used to pre-build a scene simulation model asset library. The scene simulation model asset library includes a 3D object model of at least one dynamic object and a 3D object model of at least one static object. The at least one dynamic object is a core dynamic object, and the at least one object is a pre-fragmented object. The 3D object model of the pre-fragmented object contains a cluster of broken meshes. The scene construction module is used to select the three-dimensional object model of the target scene object from the scene simulation model asset library according to the scene simulation type. The scene construction module is also used to build a simulation scene based on the three-dimensional object model of the target scene object; The data caching and recording module is used to record the scene evolution at the same frame rate for the core dynamic object and the pre-fragmented object in the simulation scene, and to obtain and cache the scene evolution recording dataset. The scene evolution recording dataset includes at least the breaking animation of the pre-fragmented object and the trajectory data of the core dynamic object. The simulation rendering module is used to synchronously render the cached fragmented animation and trajectory data, and perform real-time simulation processing and rendering of non-core dynamic objects in the simulation scene, based on the synchronous rendering instructions of the master node, through the slave nodes corresponding to each simulation channel mode, when the simulation channel mode is multi-channel mode.

9. A scene simulation device, characterized in that, The device includes: a processor and a memory storing computer program instructions; When the processor executes the computer program instructions, it implements the scene simulation method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer program instructions, which, when executed by a processor, implement the scene simulation method as described in any one of claims 1-7.

Images