PKS-based multi-domain environment construction method, device and storage medium

Abstract

The present invention provides a PKS-based multi-domain environment construction method, device, and storage medium, belonging to the field of environmental simulation and construction technology. The method comprises: preprocessing multi-domain environmental information to generate complete environmental description data; mapping and analyzing the complete environmental description data with the environmental partitioning strategy; and sending the data to the corresponding PKS node of the environmental construction system. Each PKS node is then controlled to sequentially call a construction sub-process to perform multi-threaded parallel construction based on the sent content to obtain the corresponding sub-construction environment components; integrating and splicing all sub-construction environment components, parsing the update parameters of the operation instructions, and interactively rendering the updated parameters with the integrated and spliced ​​components to obtain and output the final construction results. This method enables the rapid construction of multi-dimensional, multi-temporal, multi-scale, and highly realistic environments, improving the user experience.

Description

Technical Field

[0001] The present invention relates to the field of environmental simulation and construction technology, and in particular to a PKS-based multi-domain environment construction method, device, and storage medium. Background Art

[0002] As application scenarios across various industries continue to expand, the requirements for environment construction are becoming increasingly stringent. Traditional methods for constructing environments often have numerous limitations. For example, it's difficult to effectively integrate diverse battlefield information from different data sources, in different formats, and representing different physical meanings. This results in fragmented environmental information and an inability to fully reflect complex real-world scenarios. Furthermore, computing performance bottlenecks restrict the scale and complexity of environment construction. High-fidelity environment construction involves processing massive amounts of data, such as high-precision rendering of detailed terrain and real-time calculation of complex physical phenomena (such as explosion effects and airflow simulations). Existing centralized computing architectures suffer from slow rendering speeds when dealing with large-scale scenarios, failing to meet the real-time demands of environmental display and interaction. This results in a poor user experience and limited application scenarios.

[0003] Therefore, the present invention proposes a PKS-based multi-domain environment construction method, device and storage medium. Summary of the Invention

[0004] The present invention provides a PKS-based multi-domain environment construction method, device, and storage medium to address the deficiencies in information fusion and computing performance of environment construction in the prior art, achieve rapid construction of multi-dimensional, multi-temporal, multi-scale, and highly realistic environments, and improve user experience.

[0005] The present invention provides a PKS-based multi-domain environment construction method, comprising:

[0006] Step 1: Start the environment and build the system based on the PKS computing platform, load the pre-configured system parameters, and initialize the PKS cluster resources.

[0007] Step 2: Collect multi-domain environment information of the desired construction scene, and pre-process the multi-domain environment information to generate complete environment description data, wherein the multi-domain environment information is sub-environment information of different types of environmental elements involved in the desired construction scene;

[0008] Step 3: Map and analyze the complete environment description data with the environment partitioning strategy, and send it to the corresponding PKS node of the environment construction system. Then control each PKS node to call the construction sub-process in sequence according to the sent content to perform multi-threaded parallel construction to obtain the corresponding sub-construction environment components;

[0009] Step 4: Integrate and splice all sub-build environment components. At the same time, monitor the operation instructions input by the user and parse the update parameters of the operation instructions to interactively render the update parameters with the integrated and spliced ​​components to obtain and output the final build result.

[0010] Preferably, step 1 comprises:

[0011] Store pre-configured system parameters in configuration files;

[0012] Use PKSCLI to build a computing platform startup environment;

[0013] Load the system parameters in the configuration file in the startup environment;

[0014] Use the PKS API to initialize PKS cluster resources.

[0015] Preferably, the multi-domain environment information is pre-processed to generate complete environment description data, including:

[0016] The required construction scene is roughly divided into several coarse scene construction items, and each coarse scene construction item is deeply mined to obtain a refined construction set;

[0017] Matching the environmental element types involved in the multi-domain environmental information with each refined construction set respectively to obtain a matching vector for each environmental element type, and constructing an initial self-map of the corresponding environmental element type;

[0018] Construct a global map of all environmental element types based on the mutual causal relationships between them;

[0019] Based on the global graph, the interaction set of each environmental element type is locked, and the corresponding initial self-graph is adjusted according to the interaction set to obtain the current self-graph, wherein the nodes in the current self-graph represent the refined environmental entities, and the edges represent the association relationship between the corresponding environmental element type and the refined environmental entity;

[0020] Determine the type weight of each environmental element type based on all current self-maps, and generate sub-environment description data by combining the sub-environment information under the corresponding environmental element type of the multi-domain environmental information;

[0021] The complete environment description data is constructed based on all sub-environment description data.

[0022] Preferably, a global map of all environmental factor types is constructed based on the mutual causal relationships between the environmental factor types, including:

[0023] Obtain several simulation instances of all environmental element types involved in the required construction scenario from the simulation database, and run each simulation instance separately to obtain the corresponding operation association matrix ,in, represents the running correlation matrix under the k-th simulation instance; represents the interaction intensity between the first environmental element type obtained based on the k-th simulation instance and each remaining environmental element type; represents the interaction intensity between the N1th environmental element type obtained based on the kth simulation instance and each remaining environmental element type;

[0024] The mutual causal relationship between the corresponding environmental element type and each remaining environmental element type is determined based on all running correlation matrices.

[0025] Preferably, determining the mutual causal relationship between the corresponding environmental factor type and each remaining environmental factor type includes:

[0026] Determine the associated broad value for each environmental element type;

[0027]

[0028] in, Indicates the associated broad value of the i-th environmental element type; Indicates the number of simulation instances; Indicates the row vector of the simulation matrix for the i-th environmental element type under the k-th simulation instance the number of Indicates the total number of environmental element types present; represents the interaction intensity between the i-th environmental element type and the j-th environmental element type in the simulation matrix under the k-th simulation instance; Represents all the row vectors corresponding to the i-th environmental element type in the simulation matrix under the k-th simulation instance The maximum value in ;

[0029] According to the association broadness value and a first number of positive interaction types, a second number of negative interaction types and a third number of no interaction types of the corresponding environmental element type and each remaining environmental element type in all operation association matrices;

[0030] Selecting a maximum number from the first number, the second number, and the third number under the same environmental factor type; if the maximum number is the first number and / or the third number, determining that the corresponding environmental factor type is a positive interaction type;

[0031] If the maximum number is the second number and / or the third number, determining that the corresponding environmental element type is a negative interaction type;

[0032] If the maximum number is the second number and / or the first number, determining that the corresponding environmental element type is a negative interaction type;

[0033] If the maximum number is the third number, then if the associated broadness value is greater than the preset broadness threshold, then the type corresponding to the maximum value is selected from the second number and the third number as the final interaction type, wherein if the second number is consistent with the third number, then the corresponding environmental element type is determined to be a positive-negative interaction type;

[0034] If the associated broadness value is not greater than the preset broadness threshold, then the corresponding environmental element type is determined to be an irrelevant interaction type.

[0035] Preferably, after determining the interaction type corresponding to the environmental element type, the method further includes:

[0036] When the corresponding environmental element type is an irrelevant interaction type, it is determined that the mutual causal relationship between the corresponding environmental element type and each remaining environmental element type does not exist;

[0037] Extract the sum of the interaction intensities between the same environmental factor type and the corresponding remaining environmental factor types from all running correlation matrices;

[0038] If the interaction strength sum is consistent with the interaction type of the corresponding environmental element type, then the corresponding interaction strength sum is used as the value of the mutual causal relationship between the corresponding environmental element type and the corresponding remaining environmental element types;

[0039] If the interaction strength and the interaction type of the corresponding environmental element type are inconsistent, then the corresponding interaction strength and the interaction type are attenuated, and the attenuated value is used as the value of the mutual causal relationship between the corresponding environmental element type and the corresponding remaining environmental element types.

[0040] Preferably, performing attenuation processing on the corresponding interaction strength and the interaction type includes:

[0041]

[0042]

[0043] in, Indicates the value of attenuation processing; represents a symbolic function; represents the interaction strength before the corresponding attenuation processing; It represents the sum of the interaction intensities greater than 0 involved in all operation matrices of the corresponding environmental element type; 2 represents the sum of the interaction intensities less than 0 involved in all operation matrices for the corresponding environmental element type.

[0044] The present invention provides a multi-domain environment construction device based on PKS, comprising:

[0045] The system construction module is used to start the environment construction system based on the PKS computing platform, load pre-configured system parameters, and initialize the PKS cluster resources;

[0046] A preprocessing module is used to collect multi-domain environmental information of the required construction scene, and preprocess the multi-domain environmental information to generate complete environmental description data;

[0047] A component building module is used to map and analyze the complete environment description data with the environment partitioning strategy, and send it to the corresponding PKS node of the environment construction system, and control each PKS node to call the construction sub-process in sequence according to the sent content to perform multi-threaded parallel construction to obtain the corresponding sub-construction environment components;

[0048] The integration and splicing module is used to integrate and splice all sub-construction environment components. At the same time, it monitors the operation instructions input by the user and parses the update parameters of the operation instructions to interactively render the update parameters with the integrated and spliced ​​components to obtain and output the final construction results.

[0049] The present invention provides a storage medium on which a computer program is stored. When the computer program is executed by a processor, it implements any one of the methods for constructing a multi-domain environment based on PKS.

[0050] Compared with the prior art, the present invention has the following advantages:

[0051] Solve the deficiencies of existing technologies in environment construction in terms of information fusion and computing performance, realize the rapid construction of multi-dimensional, multi-temporal, multi-scale, and highly realistic environments, and improve user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the technical solutions in the present invention or the prior art, a brief introduction is given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0053] Figure 1 This is a flow chart of a method for constructing a multi-domain environment based on PKS in an embodiment of the present invention;

[0054] Figure 2 This is a structural diagram of a PKS-based multi-domain environment construction device in an embodiment of the present invention. DETAILED DESCRIPTION

[0055] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the embodiments described are only some of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0056] The present invention provides a method for constructing a multi-domain environment based on PKS. Figure 1 As shown, including:

[0057] Step 1: Start the environment and build the system based on the PKS computing platform, load the pre-configured system parameters, and initialize the PKS cluster resources.

[0058] Step 2: Collect multi-domain environment information of the desired construction scene, and pre-process the multi-domain environment information to generate complete environment description data, wherein the multi-domain environment information is sub-environment information of different types of environmental elements involved in the desired construction scene;

[0059] Step 3: Map and analyze the complete environment description data with the environment partitioning strategy, and send it to the corresponding PKS node of the environment construction system. Then control each PKS node to call the construction sub-process in sequence according to the sent content to perform multi-threaded parallel construction to obtain the corresponding sub-construction environment components;

[0060] Step 4: Integrate and splice all sub-build environment components. At the same time, monitor the operation instructions input by the user and parse the update parameters of the operation instructions to interactively render the update parameters with the integrated and spliced ​​components to obtain and output the final build result.

[0061] In this embodiment, an environment construction system is started on a computing platform based on PKS (Pivotal Container Service, a container orchestration and management platform) (taking a cloud computing platform as an example, the system is responsible for preparing the basic operating system, network configuration, etc. of the server in preparation for subsequent application deployment), and pre-configured system parameters are loaded, including but not limited to the multi-domain type of the target environment (such as a military battlefield, an urban construction scene), the desired spatiotemporal scale range, the required information source access address, etc. At the same time, the PKS cluster resources are initialized and allocated, and various key modules are started, such as the multi-information fusion engine, the distributed environment construction algorithm module, the real-time rendering engine, etc., to put them in standby state.

[0062] In this embodiment, according to initialization settings, the system collects environmental information from a variety of data sources. For example, by establishing data connections with military intelligence databases, geographic information system servers, and meteorological monitoring networks, it obtains the latest battlefield intelligence, high-definition satellite imagery, and real-time weather data in real time. This data continuously flows into the system at varying rates and in various formats, and is uniformly processed by a multi-information fusion engine.

[0063] In this embodiment, after multi-domain information is fused and processed, complete environmental description data is generated, and the distributed environment construction process is then initiated. Based on the partitioning strategy, the large-scale environment is split into numerous sub-environment modules and distributed to each computing node in the PKS cluster. Within each node, a hierarchical construction process is followed, sequentially invoking the basic terrain construction tool, special effects generation module, entity loading module, and other modules. Each module operates in parallel and multi-threaded to rapidly build the sub-module environment. During this process, nodes continuously communicate with each other to synchronize boundary data. Once all sub-modules are constructed, the entire model is spliced ​​and integrated to form a preliminary complete environmental model. It should be noted that multi-domain environmental information is a collection of information covering different types of environmental elements, i.e., multi-domain information after fusion. For example, when constructing an urban ecological simulation scenario, multi-domain environmental information includes information such as temperature, humidity, and wind speed in the meteorological domain; topography and land use type information in the geographic domain; and plant and animal species and distribution information in the biological domain. For example, when constructing a virtual park environment, the complete environmental description data includes the location and shape of buildings within the park, road distribution, the types and coverage of green vegetation, as well as real-time information such as temperature, humidity, and light intensity.

[0064] An environment partitioning strategy is a method for dividing the overall environment according to certain rules. For example, when building a large game scene, the environment partitioning strategy can be divided by geographical region, such as urban area, forest area, desert area, etc., and each area is assigned to a different PKS node for construction.

[0065] A PKS node is a single server or computing unit in a PKS cluster. Each server is a PKS node, each of which is responsible for part of the computing and storage tasks.

[0066] A construction sub-process is a series of operations that complete the construction of a specific sub-environment. For example, when building a 3D virtual scene, the construction sub-process may include a model loading sub-process, a material rendering sub-process, a lighting setup sub-process, etc. Each sub-process is responsible for completing a specific part of the scene construction.

[0067] Multi-threaded parallel construction can improve construction efficiency. For example, when building a complex industrial plant model, one thread is responsible for building the main structure of the plant, and another thread is responsible for adding internal equipment models. Multiple threads work simultaneously, greatly shortening the overall construction time.

[0068] In this embodiment, integration and splicing is to combine the various sub-construction environment components into a complete environment model. For example, when constructing a virtual environment of a large shopping mall, the previously constructed shopping mall building main components, internal store components, personnel model components, etc. are integrated and spliced ​​according to the design plan to form a complete shopping mall virtual scene.

[0069] For example, when a user is building a large shopping mall environment, he or she enters the command "change the wall color to light blue", where "light blue" is the updated parameter, and the effect rendering is subsequently achieved through color rendering and component placement during the rendering process.

[0070] In this example, the constructed environment model is input into a real-time rendering engine, which uses the PKS GPU cluster for high-speed rendering and outputs a visual image to a display terminal. Simultaneously, an interactive response system monitors user input in real time. Once an input command is captured, it immediately triggers the corresponding computing task, updates the environment state, and the rendering engine rapidly redraws the image, enabling real-time interaction between the user and the virtual environment. Users can freely explore and simulate operations within the environment, and the system continuously optimizes the environment's presentation based on user feedback until the task is completed.

[0071] The beneficial effects of the above technical solution are: by starting the environment construction system to process the information of the required construction scene, and performing set mapping analysis to run the process to obtain components, the reliability of the construction of each component is ensured, and subsequently through integration, splicing and updating rendering, not only fast construction is achieved, but also the realism of the construction environment is guaranteed, thereby improving the user experience.

[0072] The present invention provides a method for constructing a multi-domain environment based on PKS, wherein step 1 comprises:

[0073] Store pre-configured system parameters in configuration files;

[0074] Use PKS CLI to build a computing platform startup environment;

[0075] Load the system parameters in the configuration file in the startup environment;

[0076] Use the PKS API to initialize PKS cluster resources.

[0077] In this embodiment, the code for building the platform startup environment is as follows:

[0078] # Log in to the PKS API

[0079] pks login -a $pks_api_endpoint -u $pks_username -p $pks_password -k

[0080] # Create a PKS cluster

[0081] pks create-cluster $cluster_name -p small -n3

[0082] The beneficial effect of the above technical solution is that it facilitates the reasonable construction of the subsequent environment by configuring files, system parameters, loading parameters, and initializing the system.

[0083] The present invention provides a PKS-based multi-domain environment construction method, which pre-processes the multi-domain environment information to generate complete environment description data, including:

[0084] The required construction scene is roughly divided into several coarse scene construction items, and each coarse scene construction item is deeply mined to obtain a refined construction set;

[0085] Matching the environmental element types involved in the multi-domain environmental information with each refined construction set respectively to obtain a matching vector for each environmental element type, and constructing an initial self-map of the corresponding environmental element type;

[0086] Construct a global map of all environmental element types based on the mutual causal relationships between them;

[0087] Based on the global graph, the interaction set of each environmental element type is locked, and the corresponding initial self-graph is adjusted according to the interaction set to obtain the current self-graph, wherein the nodes in the current self-graph represent the refined environmental entities, and the edges represent the association relationship between the corresponding environmental element type and the refined environmental entity;

[0088] Determine the type weight of each environmental element type based on all current self-maps, and generate sub-environment description data by combining the sub-environment information under the corresponding environmental element type of the multi-domain environmental information;

[0089] The complete environment description data is constructed based on all sub-environment description data.

[0090] In this embodiment, the relationships between environmental factors (such as the mutual influence of temperature and humidity, the connection between pollution sources and polluted areas), the causal relationship between environmental events (such as heavy rain leading to floods), etc. are constructed by using nodes to represent environmental entities (such as a pollutant, a geographical area, or a meteorological phenomenon), and edges to represent the relationships between entities, thus constructing a knowledge graph.

[0091] In this embodiment, the interaction intensity ranges from (-1, 1), the value of M1 is greater than 10, and the number of environmental element types is greater than 5, that is, at least 5 dimensions are used to construct the environment model related to the scene.

[0092] In this embodiment, the required construction scene refers to a complete and specific scene to be constructed. For example, if a simulation scene of a smart city is to be constructed, the smart city scene is the required construction scene.

[0093] The coarse division is to preliminarily determine the required items based on the scenario to be constructed, for example, including coarse scenario construction items such as transportation, energy, and public services. Specifically, deep mining of the coarse scenario construction item of transportation may obtain detailed content such as bus system, subway system, road planning, etc. At this time, the obtained detailed construction set may include bus route planning, bus stop distribution, subway line direction, distribution of different levels of roads, etc.

[0094] In this embodiment, the multi-domain environmental information is more specifically environmental information in different fields and dimensions. For smart city scenarios, the multi-domain environmental information may include information in different fields such as geographic information, meteorological information, and population information.

[0095] In this embodiment, the environmental factor types are different types of factors in multi-domain environmental information, such as topography and landforms in geographic information, temperature and humidity in meteorological information, and population density and age distribution in population information.

[0096] The match vector is a vector obtained through the matching process that represents the degree of match between an environmental feature type and each refinement element in the refinement set. Assuming that a given environmental feature type has multiple matching items in the refinement set, each matching item can be represented by a numerical value. The vector composed of these numerical values ​​is the match vector, which is directly calculated based on the similarity function and ranges from 0 to 1.

[0097] In this embodiment, the initial self-graph is a graph constructed based on environmental factor types. The nodes and edges in the graph reflect the initial association between the environmental factor types and the related refined content. For example, for the environmental factor type of temperature, the nodes in the initial self-graph may include refined content related to temperature (such as air conditioning usage in energy consumption), and the edges represent the association between temperature and these refined content.

[0098] In this embodiment, the mutual causal relationship is positive, negative, irrelevant, etc., which is mainly used to determine the direction between different types of environmental elements, and then the global map is the connection between elements.

[0099] In this embodiment, an interaction set is a collection of other environmental factor types that have an interaction relationship with a particular environmental factor type, as determined based on the global graph. For example, for the environmental factor type of temperature, its interaction set might include humidity, light intensity, etc., because these factors interact with each other and are directly derived from the global graph.

[0100] In this embodiment, the current self-map is obtained by adjusting the edge length based on the initial self-map, because the edge length represents the correlation situation. For example, considering the impact of humidity on temperature, in the initial self-map of temperature, the original correlation strength is 0.85, and the adjusted correlation strength is 0.82.

[0101] In this embodiment, the type weight=the sum of the corresponding lengths of all association relationships involved in the current self-map under the corresponding environmental element type / the sum of the corresponding lengths of all association relationships under all environmental element types.

[0102] In this embodiment, the sub-environment information is a part of the multi-domain environment information.

[0103] In this embodiment, the sub-environment description data is used to describe the sub-environment conditions under a certain environmental factor type. For example, for the temperature environmental factor type, the sub-environment description data may include the weighted average temperature of different areas in different time periods, temperature change trends, etc.

[0104] In this embodiment, the complete environmental description data is the comprehensive data obtained by integrating the sub-environmental description data of all environmental element types, which can fully describe the constructed scene environment. For smart city scenarios, the complete environmental description data will include detailed descriptions of all environmental element types such as geography, weather, population, and energy, providing a comprehensive environmental basis for scene construction.

[0105] The beneficial effects of the above technical solution are: by roughly dividing the scenes and deeply mining the items, the comprehensiveness of the details can be guaranteed, and then by performing matching analysis and mutual causal relationship analysis, the self-map can be adjusted to ensure the comprehensiveness of the complete environment description data, provide a complete foundation for environment construction, and thus improve user experience.

[0106] The present invention provides a PKS-based multi-domain environment construction method, which constructs a global map of all environmental element types based on the mutual causal relationship between environmental element types, including:

[0107] Obtain several simulation instances of all environmental element types involved in the required construction scenario from the simulation database, and run each simulation instance separately to obtain the corresponding operation association matrix ,in, represents the running correlation matrix under the k-th simulation instance; represents the interaction intensity between the first environmental element type obtained based on the k-th simulation instance and each remaining environmental element type; represents the interaction intensity between the N1th environmental element type obtained based on the kth simulation instance and each remaining environmental element type;

[0108] The mutual causal relationship between the corresponding environmental element type and each remaining environmental element type is determined based on all running correlation matrices.

[0109] In this embodiment, the simulation database contains different combinations of environmental element types and corresponding simulation instances, which can be directly retrieved.

[0110] Suppose we build a smart home scenario, the types of environmental factors involved include temperature, humidity, light intensity and electrical equipment status.

[0111] Example 1: Simulates an indoor environment with a temperature of 25°C, a humidity of 60%, a light intensity of 500 lux, the air conditioner on, and the lights off.

[0112] Example 2: Simulates the indoor environment with a temperature of 28°C, a humidity of 70%, a light intensity of 800 lux, the air conditioner off, and the lights on.

[0113] Example 3: Simulates the indoor environment with a temperature of 22°C, a humidity of 50%, a light intensity of 300 lux, and air conditioning and lights on.

[0114] Run these three instances in the simulation environment respectively, and record the changes of each environmental factor in different instances.

[0115] The operation correlation matrix can be directly obtained from the operation results. For example, for example 1:

[0116]

[0117] The mutual causal relationship can be obtained by further analyzing the operation correlation matrix.

[0118] The beneficial effect of the above technical solution is that by obtaining instances related to the elements and simulating them in sequence, the interactive relationship between the elements can be obtained, providing a basis for determining the mutual causal relationship.

[0119] The present invention provides a multi-domain environment construction method based on PKS, which determines the mutual causal relationship between the corresponding environmental element type and each remaining environmental element type, including:

[0120] Determine the associated broad value for each environmental element type;

[0121]

[0122] in, Indicates the associated broad value of the i-th environmental element type; Indicates the number of simulation instances; Indicates the row vector of the simulation matrix for the i-th environmental element type under the k-th simulation instance the number of Indicates the total number of environmental element types present; represents the interaction intensity between the i-th environmental element type and the j-th environmental element type in the simulation matrix under the k-th simulation instance; Represents all the row vectors corresponding to the i-th environmental element type in the simulation matrix under the k-th simulation instance The maximum value in ;

[0123] According to the association broadness value and a first number of positive interaction types, a second number of negative interaction types and a third number of no interaction types of the corresponding environmental element type and each remaining environmental element type in all operation association matrices;

[0124] Selecting a maximum number from the first number, the second number, and the third number under the same environmental factor type; if the maximum number is the first number and / or the third number, determining that the corresponding environmental factor type is a positive interaction type;

[0125] If the maximum number is the second number and / or the third number, determining that the corresponding environmental element type is a negative interaction type;

[0126] If the maximum number is the second number and / or the first number, determining that the corresponding environmental element type is a negative interaction type;

[0127] If the maximum number is the third number, then if the associated broadness value is greater than the preset broadness threshold, then the type corresponding to the maximum value is selected from the second number and the third number as the final interaction type, wherein if the second number is consistent with the third number, then the corresponding environmental element type is determined to be a positive-negative interaction type;

[0128] If the associated broadness value is not greater than the preset broadness threshold, then the corresponding environmental element type is determined to be an irrelevant interaction type.

[0129] In this embodiment, the preset broad threshold is 0.3.

[0130] In this embodiment, the positive interaction type refers to an interaction intensity of a positive number, the negative interaction type refers to an interaction intensity of a negative number, and the non-interaction type refers to an interaction intensity of 0.

[0131] In this embodiment, for example, the interaction between type 1 and type 2 is a positive interaction type of 0. At this time, type 1 points to type 2. If it is a negative interaction type, then type 2 points to type 1. If it is a non-interaction type, then there is no pointing between type 1 and type 2. If it is a positive or negative interaction type, then type 1 points to type 2 and type 2 points to type 1.

[0132] The beneficial effect of the above technical solution is: by calculating the associated broad value of each environmental element type and combining it with the size judgment of the quantity under different interaction types, the final interaction type of the environmental element type can be effectively determined, providing a basis for constructing the map.

[0133] The present invention provides a PKS-based multi-domain environment construction method, which, after determining the interaction type corresponding to the environmental element type, further includes:

[0134] When the corresponding environmental element type is an irrelevant interaction type, it is determined that the mutual causal relationship between the corresponding environmental element type and each remaining environmental element type does not exist;

[0135] Extract the sum of the interaction intensities between the same environmental factor type and the corresponding remaining environmental factor types from all running correlation matrices;

[0136] If the interaction strength sum is consistent with the interaction type of the corresponding environmental element type, then the corresponding interaction strength sum is used as the value of the mutual causal relationship between the corresponding environmental element type and the corresponding remaining environmental element types;

[0137] If the interaction strength and the interaction type of the corresponding environmental element type are inconsistent, then the corresponding interaction strength and the interaction type are attenuated, and the attenuated value is used as the value of the mutual causal relationship between the corresponding environmental element type and the corresponding remaining environmental element types.

[0138] In this embodiment, the attenuation processing is performed on the corresponding interaction strength and the interaction type, including:

[0139]

[0140]

[0141] in, Indicates the value of attenuation processing; represents a symbolic function; represents the interaction strength before the corresponding attenuation processing; It represents the sum of the interaction intensities greater than 0 involved in all operation matrices of the corresponding environmental element type; 2 represents the sum of the interaction intensities less than 0 involved in all operation matrices for the corresponding environmental element type.

[0142] In this embodiment, if the sum is a value greater than 0 and the interaction type is a positive interaction type, then it is considered consistent. If the sum is a value less than 0 and the interaction type is a negative interaction type, then it is considered inconsistent.

[0143] The beneficial effect of the above technical solution is: by determining the interaction strength and the consistent relationship between the value and the type, the reasonable processing of the value is achieved, thereby ensuring the reliability of the graph constructed by the mutual causal relationship.

[0144] The present invention provides a multi-domain environment construction device based on PKS, such as Figure 2 As shown, including:

[0145] The system construction module is used to start the environment construction system based on the PKS computing platform, load pre-configured system parameters, and initialize the PKS cluster resources;

[0146] A preprocessing module is used to collect multi-domain environmental information of the required construction scene, and preprocess the multi-domain environmental information to generate complete environmental description data;

[0147] A component building module is used to map and analyze the complete environment description data with the environment partitioning strategy, and send it to the corresponding PKS node of the environment construction system, and control each PKS node to call the construction sub-process in sequence according to the sent content to perform multi-threaded parallel construction to obtain the corresponding sub-construction environment components;

[0148] The integration and splicing module is used to integrate and splice all sub-construction environment components. At the same time, it monitors the operation instructions input by the user and parses the update parameters of the operation instructions to interactively render the update parameters with the integrated and spliced ​​components to obtain and output the final construction results.

[0149] The beneficial effects of the above technical solution are: by starting the environment construction system to process the information of the required construction scene, and performing set mapping analysis to run the process to obtain components, the reliability of the construction of each component is ensured, and subsequently through integration, splicing and updating rendering, not only fast construction is achieved, but also the realism of the construction environment is guaranteed, thereby improving the user experience.

[0150] The present invention provides a storage medium on which a computer program is stored. When the computer program is executed by a processor, it implements any one of the methods for constructing a multi-domain environment based on PKS.

[0151] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A method for constructing a multi-domain environment based on PKS, characterized in that: include: Step 1: Start the environment and build the system based on the PKS computing platform, load the pre-configured system parameters, and initialize the PKS cluster resources. Step 2: Collect multi-domain environment information of the desired construction scene, and pre-process the multi-domain environment information to generate complete environment description data, wherein the multi-domain environment information is sub-environment information of different types of environmental elements involved in the desired construction scene; Step 3: Map and analyze the complete environment description data with the environment partitioning strategy, and send it to the corresponding PKS node of the environment construction system. Then control each PKS node to call the construction sub-process in sequence according to the sent content to perform multi-threaded parallel construction to obtain the corresponding sub-construction environment components; Step 4: Integrate and splice all sub-build environment components. At the same time, monitor the operation instructions input by the user and parse the update parameters of the operation instructions to interactively render the updated parameters with the integrated and spliced ​​components to obtain and output the final build result; The multi-domain environment information is pre-processed to generate complete environment description data, including: The required construction scene is roughly divided into several coarse scene construction items, and each coarse scene construction item is deeply mined to obtain a refined construction set; Matching the environmental element types involved in the multi-domain environmental information with each refined construction set respectively to obtain a matching vector for each environmental element type, and constructing an initial self-map of the corresponding environmental element type; Construct a global map of all environmental element types based on the mutual causal relationships between them; Based on the global graph, the interaction set of each environmental element type is locked, and the corresponding initial self-graph is adjusted according to the interaction set to obtain the current self-graph, wherein the nodes in the current self-graph represent the refined environmental entities, and the edges represent the association relationship between the corresponding environmental element type and the refined environmental entity; Determine the type weight of each environmental element type based on all current self-maps, and generate sub-environment description data by combining the sub-environment information under the corresponding environmental element type of the multi-domain environmental information; The complete environment description data is constructed based on all sub-environment description data.

2. The method for constructing a multi-domain environment based on PKS according to claim 1, characterized in that: Step 1 includes: Store pre-configured system parameters in configuration files; Use PKSCLI to build a computing platform startup environment; Load the system parameters in the configuration file in the startup environment; Use the PKS API to initialize PKS cluster resources.

3. The method for constructing a multi-domain environment based on PKS according to claim 1, characterized in that: Based on the mutual causal relationships between environmental factor types, a global map of all environmental factor types is constructed, including: Obtain several simulation instances of all types of environmental elements involved in the required construction scenario from the simulation database, and run each simulation instance separately to obtain the corresponding operation association matrix ,in, represents the running correlation matrix under the k-th simulation instance; represents the interaction intensity between the first environmental element type obtained based on the k-th simulation instance and each remaining environmental element type; represents the interaction intensity between the N1th environmental element type obtained based on the kth simulation instance and each remaining environmental element type; The mutual causal relationship between the corresponding environmental element type and each remaining environmental element type is determined based on all running correlation matrices.

4. The method for constructing a multi-domain environment based on PKS according to claim 3, characterized in that: Determine the mutual causal relationship between the corresponding environmental factor type and each remaining environmental factor type, including: Determine the associated broad value for each environmental element type; in, Indicates the associated broad value of the i-th environmental element type; Indicates the number of simulation instances; Indicates the row vector of the simulation matrix for the i-th environmental element type under the k-th simulation instance the number of Indicates the total number of environmental element types present; represents the interaction intensity between the i-th environmental element type and the j-th environmental element type in the simulation matrix under the k-th simulation instance; Represents all the row vectors corresponding to the i-th environmental element type in the simulation matrix under the k-th simulation instance The maximum value in ; According to the association broadness value and a first number of positive interaction types, a second number of negative interaction types and a third number of no interaction types of the corresponding environmental element type and each remaining environmental element type in all operation association matrices; Selecting a maximum number from the first number, the second number, and the third number under the same environmental factor type; if the maximum number is the first number and the third number, determining that the corresponding environmental factor type is a positive interaction type; If the maximum number is the second number and the third number, it is determined that the corresponding environmental element type is a negative interaction type; If the maximum number is the second number and the first number, determining that the corresponding environmental element type is a negative interaction type; If the maximum number is the third number, then if the associated broadness value is greater than the preset broadness threshold, then the type corresponding to the maximum value is selected from the second number and the third number as the final interaction type, wherein if the second number is consistent with the third number, then the corresponding environmental element type is determined to be a positive-negative interaction type; If the associated broadness value is not greater than the preset broadness threshold, then the corresponding environmental element type is determined to be an irrelevant interaction type.

5. The method for constructing a multi-domain environment based on PKS according to claim 4, characterized in that: After determining the interaction type corresponding to the environmental element type, it also includes: When the corresponding environmental element type is an irrelevant interaction type, it is determined that the mutual causal relationship between the corresponding environmental element type and each remaining environmental element type does not exist; Extract the sum of the interaction intensities between the same environmental factor type and the corresponding remaining environmental factor types from all running correlation matrices; If the interaction strength sum is consistent with the interaction type of the corresponding environmental element type, then the corresponding interaction strength sum is used as the value of the mutual causal relationship between the corresponding environmental element type and the corresponding remaining environmental element types; If the interaction strength and the interaction type of the corresponding environmental element type are inconsistent, then the corresponding interaction strength and the interaction type are attenuated, and the attenuated value is used as the value of the mutual causal relationship between the corresponding environmental element type and the corresponding remaining environmental element types.

6. The method for constructing a multi-domain environment based on PKS according to claim 5, characterized in that: Attenuation processing is performed on the corresponding interaction strength and interaction type, including: in, Indicates the value of attenuation processing; represents a symbolic function; represents the interaction strength before the corresponding attenuation processing; It represents the sum of the interaction intensities greater than 0 involved in all operation matrices of the corresponding environmental element type; 2 represents the sum of the interaction intensities less than 0 involved in all operation matrices for the corresponding environmental element type.

7. A multi-domain environment construction device based on PKS, characterized in that: include: The system construction module is used to start the environment construction system based on the PKS computing platform, load pre-configured system parameters, and initialize the PKS cluster resources; A preprocessing module is used to collect multi-domain environmental information of the required construction scene, and preprocess the multi-domain environmental information to generate complete environmental description data; A component building module is used to map and analyze the complete environment description data with the environment partitioning strategy, and send it to the corresponding PKS node of the environment construction system, and control each PKS node to call the construction sub-process in sequence according to the sent content to perform multi-threaded parallel construction to obtain the corresponding sub-construction environment components; The integration and splicing module is used to integrate and splice all sub-build environment components. At the same time, it monitors the operation instructions input by the user and parses the update parameters of the operation instructions to interactively render the updated parameters with the integrated and spliced ​​components to obtain and output the final build result; The multi-domain environment information is pre-processed to generate complete environment description data, including: The required construction scene is roughly divided into several coarse scene construction items, and each coarse scene construction item is deeply mined to obtain a refined construction set; Matching the environmental element types involved in the multi-domain environmental information with each refined construction set respectively to obtain a matching vector for each environmental element type, and constructing an initial self-map of the corresponding environmental element type; Construct a global map of all environmental element types based on the mutual causal relationships between them; Based on the global graph, the interaction set of each environmental element type is locked, and the corresponding initial self-graph is adjusted according to the interaction set to obtain the current self-graph, wherein the nodes in the current self-graph represent the refined environmental entities, and the edges represent the association relationship between the corresponding environmental element type and the refined environmental entity; Determine the type weight of each environmental element type based on all current self-maps, and generate sub-environment description data by combining the sub-environment information under the corresponding environmental element type of the multi-domain environmental information; The complete environment description data is constructed based on all sub-environment description data.

8. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the PKS-based multi-domain environment construction method according to any one of claims 1 to 6 is implemented.

Images