Simulation data generation method and device, electronic equipment and storage medium

By training the converter model to generate prediction models, and automatically generate simulation data, it solves the high threshold, low efficiency and low precision problems of simulation data generation in the existing technology, realizes the intelligence and flexibility of simulation data, and expands the simulation technology Application scope.

CN118965719APending Publication Date: 2024-11-15SHENZHEN TIANHAI CHENGUANG TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410984122.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-22
Publication Date
2024-11-15

AI Technical Summary

Technical Problem

The existing simulation data generation technology has high threshold, low efficiency, limited accuracy and lack of flexibility, which limits the popularization and application of simulation technology.

Method used

By obtaining the training set related to the target simulation, the decoder of the training converter model obtains the prediction model, obtains the input text for word segmentation and word vector conversion, and uses the prediction model to generate the target simulation data to realize the automation and intelligence of the simulation data.

Benefits of technology

It lowers the threshold for the use of simulation systems, improves the efficiency and accuracy of simulation data generation, realizes automatic generation from input text to target simulation data, shortens simulation preparation time, and improves user experience.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118965719A_ABST
    Figure CN118965719A_ABST
Patent Text Reader

Abstract

The embodiment of the invention discloses a simulation data generation method and device, electronic equipment and a storage medium, and relates to the technical field of artificial intelligence and emulation.The method comprises the steps that a training set related to target simulation is obtained, a prediction model is obtained for a decoder of a converter model according to the training set, an input text is obtained, and the input text is input into the converter model; and performing word segmentation and word vector conversion on the input text to obtain input data, and generating simulation data corresponding to target simulation according to the input data through the prediction model. According to the method, the simulation parameter generation prediction model based on the large model is introduced, automation and intelligence of simulation data generation are achieved, the use threshold of a simulation system is lowered, the simulation data generation efficiency is improved, and the problems that in the prior art, the threshold is high, efficiency is low, precision is limited, and flexibility is lacked are solved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence and simulation technology, and in particular to a simulation data generation method, device, electronic equipment and storage medium. Background Art

[0002] Currently, the generation of simulation data mainly relies on users to manually configure the parameters of each link by being familiar with the simulation software. This process usually requires users to have deep professional knowledge and be able to deeply understand the operating mechanism of the simulation system and the specific impact of each parameter on the simulation results. After determining the approximate parameter range, users also need to gradually adjust the parameters through continuous debugging and trial and error to approach the expected simulation effect.

[0003] As can be seen from the above, the existing technology requires extremely high professional skills from users. Users must have an in-depth understanding of the current simulation system and rich operating experience. This not only increases the learning cost, but also limits the popularization and application scope of simulation technology. Moreover, since the adjustment of simulation parameters depends entirely on human settings and there are often complex interactions between parameters, it is difficult to quickly find the optimal parameter combination. Users need to spend a lot of time on repeated debugging and trial and error, which not only prolongs the simulation data generation cycle, but also reduces work efficiency.

[0004] In addition, although users can approach the expected simulation effect through continuous debugging, it is difficult to ensure the precision and accuracy of the simulation results due to the interference of human factors and the limitations of parameter adjustment. Existing technologies can usually only perform parameter configuration and debugging for specific simulation systems, lacking flexibility and scalability.

[0005] In summary, existing technical solutions have many shortcomings in the process of generating simulation data, including high threshold, low efficiency, limited accuracy and lack of flexibility. These problems limit the application and development of simulation technology in a wider range of fields. Therefore, it is necessary to explore new technical solutions to overcome these shortcomings and improve the efficiency and accuracy of simulation data generation.

[0006] Therefore, there is a need for a simulation data generation method that can lower the usage threshold, improve simulation accuracy, and increase generation efficiency and flexibility. Summary of the invention

[0007] Each embodiment of the present invention provides a simulation data generation method to solve the problems of high threshold, low efficiency, limited accuracy and lack of flexibility in the prior art. The technical solution is as follows:

[0008] According to one aspect of the present invention, a simulation data generation method comprises: obtaining a training set related to a target simulation, and obtaining a prediction model for a decoder of a converter model according to the training set; obtaining input text, and performing word segmentation and word vector conversion on the input text to obtain input data; the input text comprises simulation requirements and prompts; and generating simulation data corresponding to the target simulation according to the input data through the prediction model; the simulation data comprises scenes, targets, environments and camera positions.

[0009] In one of the embodiments, obtaining a prediction model for the decoder of the converter model according to the training set is achieved by the following steps: performing text preprocessing and word vector generation on the data in the training set to obtain a corresponding word vector set; and training the decoder of the converter model with the word vector set to obtain a prediction model.

[0010] In one of the embodiments, text preprocessing and word vector generation are performed on the data in the training set to obtain the corresponding word vector set by the following steps: dividing the long documents in the training set into paragraphs and sentences; performing word segmentation on each of the sentences, converting the continuous text in the paragraph into a sequence of words and phrases to obtain a word segmentation result; using the word2vec algorithm to convert the word segmentation result into a representation in a high-dimensional vector space to obtain a corresponding word vector set.

[0011] In one of the embodiments, the prediction model is obtained by training the decoder of the converter model with the word vector set through the following steps: the word vector set and the corresponding simulation parameter labels are combined and input into the decoder of the converter model, and the BCE l oss function is used as the loss function to perform multiple rounds of training to obtain the prediction model, and the parameters of the prediction model are optimized by the back propagation algorithm.

[0012] In one of the embodiments, obtaining a prediction model for a decoder of a converter model based on the training set further comprises the following steps: adding a start tag and an end tag to the decoder of the converter model and the input and output of the prediction model.

[0013] In one of the embodiments, the method further includes the following steps: running a simulation environment of the target simulation according to the simulation data to obtain a simulation result, and optimizing the prediction model according to the simulation result.

[0014] In one of the embodiments, the method further includes the following steps: adding an enhanced model MOD corresponding to a theme and a scene to the simulation environment according to the simulation data, and setting parameters of the simulation environment according to the simulation data.

[0015] According to one aspect of the present invention, a simulation data generating device comprises: a model training module, used to obtain a training set related to a target simulation, and obtain a prediction model for a decoder of a converter model according to the training set; an input acquisition module, used to obtain input text, and obtain input data by performing word segmentation and word vector conversion on the input text; and a simulation data generating module, used to generate simulation data corresponding to the target simulation according to the input data through the prediction model.

[0016] According to one aspect of the present invention, an electronic device includes at least one processor and at least one memory, wherein the memory stores computer-readable instructions; the computer-readable instructions are executed by one or more of the processors, so that the electronic device implements the simulation data generation method as described above.

[0017] According to one aspect of the present invention, a storage medium stores computer-readable instructions thereon, wherein the computer-readable instructions are executed by one or more processors to implement the simulation data generating method as described above.

[0018] The beneficial effects brought by the technical solution provided by the present invention are:

[0019] In the above technical scheme, the present invention first obtains a training set related to the target simulation, obtains a prediction model for the decoder of the converter model according to the training set, obtains an input text, performs word segmentation and word vector conversion on the input text to obtain input data, generates simulation data corresponding to the target simulation according to the input data through the prediction model, and introduces a simulation parameter generation prediction model based on a large model to achieve automation and intelligence of simulation data generation, lower the use threshold of the simulation system, improve the efficiency of simulation data generation, and realize automatic generation of simulation data corresponding to the target simulation from the input text, so that the user does not need to directly operate the complex simulation system, and can generate simulation parameters through natural language, which greatly shortens the simulation preparation time. The natural language interaction method enables the system to understand user needs more quickly and accurately, improve user experience, and effectively solve the problems of high threshold, low efficiency, limited accuracy and lack of flexibility in the prior art. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in describing the embodiments of the present invention. Obviously, the drawings described below are only some embodiments of the present invention, and those skilled in the art can obtain other drawings based on these drawings without creative work.

[0021] Figure 1 is a flow chart of a method for generating simulation data according to an exemplary embodiment;

[0022] Figure 2 yes Figure 1 Flow chart of obtaining a word vector set in the corresponding embodiment;

[0023] Figure 3 It is a flowchart of a simulation data generation method in an application scenario;

[0024] Figure 4 yes Figure 3 Schematic diagram of the process of generating simulation data in the corresponding application scenario;

[0025] Figure 5 is a block diagram of a simulation data generating device according to an exemplary embodiment;

[0026] Figure 6 is a hardware structure diagram of an electronic device according to an exemplary embodiment;

[0027] Figure 7 It is a block diagram of an electronic device according to an exemplary embodiment. DETAILED DESCRIPTION

[0028] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be interpreted as limiting the present invention.

[0029] It will be understood by those skilled in the art that, unless expressly stated, the singular forms "one", "said", and "the" used herein may also include plural forms. It should be further understood that the term "comprising" used in the specification of the present disclosure refers to the presence of the features, integers, steps, operations, elements, and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof. It should be understood that when we refer to an element as being "connected" or "coupled" to another element, it may be directly connected or coupled to the other element, or there may be intermediate elements. In addition, the "connection" or "coupling" used herein may include wireless connection or wireless coupling. The term "and / or" used herein includes all or any unit and all combinations of one or more associated listed items.

[0030] Prompts: Refers to a carefully designed text input that is used to stimulate or guide a pre-trained model (such as the GPT series model) to generate a specific type of output. By changing the prompt, the model can be guided to complete different tasks, such as text generation, question answering, translation, etc. This technology is called "prompt engineering" or "prompt tuning", which helps researchers and developers leverage the capabilities of pre-trained models in a more flexible and controllable way.

[0031] Simulation system: a tool that simulates the real-world system process through a computer program. The simulation system is a technology that models the object system and uses the model as an experimental device to analyze existing or planned systems. It is an important tool for engineers, managers and decision makers to test, evaluate and visualize the proposed solutions for operations, processes, or dynamic systems.

[0032] The present invention provides a simulation data generation method, which realizes the automation and intelligence of simulation data generation by introducing a simulation parameter generation prediction model based on a large model, reduces the use threshold of the simulation system, improves the efficiency of simulation data generation, and solves the problems of high threshold, low efficiency, limited accuracy and lack of flexibility in the prior art. The simulation data generation method is suitable for a simulation data generation device, which can be an electronic device. The simulation data generation method in the embodiment of the present invention can be applied to a variety of scenarios, such as parameter setting of a simulation system.

[0033] See also Figure 1 , an embodiment of the present invention provides a simulation data generation method, which is applicable to electronic equipment.

[0034] In the following method embodiments, for ease of description, the execution subject of each step of the method is taken as an electronic device as an example for illustration, but this does not constitute a specific limitation.

[0035] like Figure 1 As shown, the method may include the following steps:

[0036] Step 110, obtaining a training set related to the target simulation, and obtaining a prediction model for the decoder of the converter model according to the training set.

[0037] In one possible implementation, text preprocessing and word vector generation are performed on the data in the training set to obtain a corresponding word vector set, and the decoder of the converter model is trained with the word vector set to obtain a prediction model.

[0038] In one possible implementation, such as Figure 2As shown, performing text preprocessing and word vector generation on the data in the training set to obtain the corresponding word vector set may include the following steps:

[0039] Step 210, segment the long documents in the training set into paragraphs and sentences.

[0040] Step 230, segment each sentence, convert the continuous text in the paragraph into a sequence of words and phrases to obtain the segmentation result.

[0041] Step 250: Use the word2vec algorithm to convert the word segmentation results into representations in a high-dimensional vector space to obtain a corresponding word vector set.

[0042] In one possible implementation, the word vector set and the corresponding simulation parameter label are combined and input into the decoder of the converter model, and the BCE Loss function is used as the loss function to perform multiple rounds of training to obtain the prediction model, and the parameters of the prediction model are optimized through the back propagation algorithm.

[0043] In one possible implementation, a start tag and an end tag are added to the input and output of the decoder and prediction model of the converter model.

[0044] Step 130, obtain input text, perform word segmentation and word vector conversion on the input text to obtain input data.

[0045] Step 150, generating simulation data corresponding to the target simulation according to the input data through the prediction model.

[0046] In a possible implementation, a simulation environment for target simulation is run according to simulation data to obtain simulation results, and the prediction model is optimized according to the simulation results.

[0047] In a possible implementation, an enhanced model MOD corresponding to a theme and a scene is added to the simulation environment according to the simulation data, and parameters of the simulation environment are set according to the simulation data.

[0048] Through the above process, the present invention first obtains a training set related to the target simulation, obtains a prediction model for the decoder of the converter model according to the training set, obtains an input text, performs word segmentation and word vector conversion on the input text to obtain input data, generates simulation data corresponding to the target simulation according to the input data through the prediction model, and introduces a simulation parameter generation prediction model based on a large model to achieve automation and intelligence of simulation data generation, lower the use threshold of the simulation system, improve the efficiency of simulation data generation, and realize automatic generation from input text to simulation data corresponding to the target simulation, so that users do not need to directly operate a complex simulation system, and can generate simulation parameters through natural language, which greatly shortens the simulation preparation time. The natural language interaction method enables the system to understand user needs more quickly and accurately, improve user experience, and thus effectively solve the problems of high threshold, low efficiency, limited accuracy and lack of flexibility in the prior art.

[0049] In one application scenario, Figure 3 A flow chart showing the simulation data generation method.

[0050] like Figure 3 As shown, a flowchart of simulation data generation in the application scenario of simulation game development is presented, which mainly focuses on scenario construction, goal setting, simulation parameter adjustment, environment configuration and final user feedback.

[0051] First, the user describes the requirements in natural language, including the name and number of scenes, the name and number of targets, the name and status of the environment, and the spatial three-dimensional point of the camera position. These natural language requirements are converted into prompts and input into the trained prediction model ParamgenMode l to obtain simulation data. Simulation parameters are obtained based on these simulation data and input into the simulation platform to guide the construction of the simulation environment. After the simulation environment is built and run, new simulation data is obtained. Based on this simulation data, feedback on the need for two-way lanes and some different types of vehicles is obtained. These feedback contents are combined with the user's need to generate a highway scene from a bird's-eye view and input into the prediction model ParamgenMode l again to obtain simulation data. Each time simulation data is obtained through the prediction model, a suitable Mod is selected from the MOd library according to the specific content of the simulation data and added to the simulation environment. Through multiple simulation data acquisition and Mod addition, a real simulation scene that highly meets user requirements is finally obtained.

[0052] Specifically, simulation parameters also include various factors that may affect the gaming experience, such as weather and lighting. These factors are crucial to creating a realistic gaming environment. The camera position is a point in three-dimensional space, so in game development, the camera position and angle need to be precisely controlled to provide players with the best gaming perspective and experience.

[0053] In addition, users provide feedback on the collected samples, and the model will combine the new natural language for word segmentation and vectorization processing and then adjust the generated parameters to achieve the goal of optimizing the generated data.

[0054] like Figure 4 As shown in the figure, the training process of the prediction model ParamgenMode l includes a fine-tuning training stage and a prediction stage. In the fine-tuning training stage, first collect a large amount of industry-related document information, segment the text and divide it into sentences, perform word segmentation on each sentence, and then use word2vec to obtain word vectors. Load the decoder module of the open source transformer, use the word vectors as model input, fine-tune the training of ParamgenMode l, use 200 training epochs, use the predicted classification loss - BCE l oss, and add start and end to indicate the start and end of the prediction.

[0055] Furthermore, in the prediction stage: input the prompt text and use word2vec to get the vector after word segmentation + start input ParamgenMode l until the prediction ends, and the prediction result will output the camera height parameter.

[0056] Through the above process, an embodiment of the present invention provides a simulation data generation method. By introducing a simulation parameter generation prediction model based on a large model, the automation and intelligence of simulation data generation is realized, the use threshold of the simulation system is lowered, the efficiency of simulation data generation is improved, and the automatic generation of simulation data corresponding to the target simulation from the input text is realized, so that the user does not need to directly operate the complex simulation system, and the simulation parameters can be generated through natural language, which greatly shortens the simulation preparation time. The natural language interaction mode enables the system to understand user needs more quickly and accurately, improve user experience, and thus effectively solve the problems of high threshold, low efficiency, limited accuracy and lack of flexibility in the prior art.

[0057] The following is an embodiment of the device of the present invention, which can be used to execute the simulation data generation method involved in the present invention. For details not disclosed in the embodiment of the device of the present invention, please refer to the method embodiment of the simulation data generation method involved in the present invention.

[0058] See also Figure 5 , an embodiment of the present invention provides a simulation data generating device 800.

[0059] The device 800 includes but is not limited to: a model training module 810 , an input acquisition module 830 and a simulation data generation module 850 .

[0060] The model training module 810 is used to obtain a training set related to the target simulation, and obtain a prediction model based on the decoder of the converter model according to the training set.

[0061] The input acquisition module 830 is used to acquire input text, perform word segmentation and word vector conversion on the input text to obtain input data.

[0062] The simulation data generation module 850 is used to generate simulation data corresponding to the target simulation according to the input data through the prediction model.

[0063] It should be noted that, when the simulation data provided in the above embodiment is generated, only the division of the above functional modules is used as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the simulation data generating device will be divided into different functional modules to complete all or part of the functions described above.

[0064] In addition, the simulation data generation device and the simulation data generation method provided in the above embodiments belong to the same concept, wherein the specific manner in which each module performs operations has been described in detail in the method embodiments and will not be repeated here.

[0065] Figure 6 The present invention schematically illustrates the structure of an electronic device according to an exemplary embodiment.

[0066] It should be noted that the electronic device is only an example adapted to the present invention and cannot be considered to provide any limitation on the scope of use of the present invention. The electronic device cannot be interpreted as requiring dependence on or having Figure 6 One or more components of exemplary electronic device 2000 are shown.

[0067] The hardware structure of the electronic device 2000 may vary greatly due to differences in configuration or performance, such as Figure 6 As shown, the electronic device 2000 includes: a power supply 210, an interface 230, at least one memory 250, and at least one central processing unit (CPU) 270.

[0068] Specifically, the power supply 210 is used to provide operating voltage for each hardware device on the electronic device 2000 .

[0069] The interface 230 includes at least one wired or wireless network interface 231 for interacting with external devices. Of course, in other examples adapted by the present invention, the interface 230 may further include at least one serial-to-parallel conversion interface 233, at least one input-output interface 235, and at least one USB interface 237, etc. Figure 6As shown, this is not intended to be a specific limitation.

[0070] The memory 250 is a carrier for storing resources, which may be a read-only memory, a random access memory, a disk or an optical disk, etc. The resources stored thereon include an operating system 251, an application 253 and data 255, etc. The storage method may be temporary storage or permanent storage.

[0071] Among them, the operating system 251 is used to manage and control various hardware devices and application programs 253 on the electronic device 2000 to enable the central processing unit 270 to calculate and process the massive data 255 in the memory 250. It can be Windows ServerTM, Mac OS XTM, UnixTM, LinuxTM, FreeBSDTM, etc.

[0072] The application program 253 is a computer-readable instruction that performs at least one specific task based on the operating system 251, and may include at least one module ( Figure 6 (not shown), each module may respectively include computer-readable instructions for the electronic device 2000. For example, the simulation data generating device may be regarded as an application 253 deployed on the electronic device 2000.

[0073] The data 255 may be signal information, etc., stored in the memory 250 .

[0074] The central processor 270 may include one or more processors and is configured to communicate with the memory 250 through at least one communication bus to read the computer-readable instructions stored in the memory 250, thereby realizing the operation and processing of the mass data 255 in the memory 250. For example, the simulation data generation method is completed by the central processor 270 reading a series of computer-readable instructions stored in the memory 250.

[0075] In addition, the present invention can also be implemented by hardware circuits or hardware circuits combined with software. Therefore, the implementation of the present invention is not limited to any specific hardware circuits, software, or a combination of the two.

[0076] See also Figure 7 In an embodiment of the present invention, an electronic device 4000 is provided. The electronic device 4000 may include: a desktop computer, a laptop computer, a server, etc. with sensor recognition capability.

[0077] exist Figure 7 In the embodiment, the electronic device 4000 includes at least one processor 4001 and at least one memory 4003.

[0078] The data interaction between the processor 4001 and the memory 4003 can be realized through at least one communication bus 4002. The communication bus 4002 may include a path for transmitting data between the processor 4001 and the memory 4003. The communication bus 4002 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industrial Standard Architecture) bus. The communication bus 4002 can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 7 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0079] Optionally, the electronic device 4000 may further include a transceiver 4004, which may be used for data interaction between the electronic device and other electronic devices, such as data transmission and / or data reception, etc. It should be noted that in actual applications, the transceiver 4004 is not limited to one, and the structure of the electronic device 4000 does not constitute a limitation on the embodiments of the present invention.

[0080] Processor 4001 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array) or other programmable logic devices, transistor logic devices, hardware components or any combination thereof. It may implement or execute various exemplary logic blocks, modules and circuits described in conjunction with the disclosure of the present invention. Processor 4001 may also be a combination that implements a computing function, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0081] The memory 4003 can be a ROM (Read On Memory) or other types of static storage devices that can store static information and instructions, a RAM (Random Access Memory) or other types of dynamic storage devices that can store information and instructions, or an EEPROM (Electric Erasable Programmable Read On Memory), a CD-ROM (Compact Disk Read On Memory) or other optical disk storage, optical disk storage (including compressed optical disk, laser disk, optical disk, digital versatile disk, Blu-ray disk, etc.), magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store desired program instructions or codes in the form of instructions or data structures and can be accessed by the electronic device 4000, but is not limited to this.

[0082] Computer-readable instructions are stored in the memory 4003 , and the processor 4001 can read the computer-readable instructions stored in the memory 4003 through the communication bus 4002 .

[0083] The computer-readable instructions are executed by one or more processors 4001 to implement the simulation data generating method in the above-mentioned embodiments.

[0084] In addition, an embodiment of the present invention provides a storage medium on which computer-readable instructions are stored. The computer-readable instructions are executed by one or more processors to implement the simulation data generating method as described above.

[0085] A computer program product is provided in an embodiment of the present invention. The computer program product includes computer-readable instructions. The computer-readable instructions are stored in a storage medium. One or more processors of an electronic device read the computer-readable instructions from the storage medium, load and execute the computer-readable instructions, so that the electronic device implements the simulation data generation method as described above.

[0086] Compared with the related art, the present invention has the following beneficial effects:

[0087] 1. The present invention first obtains a training set related to the target simulation, obtains a prediction model for the decoder of the converter model according to the training set, obtains an input text, performs word segmentation and word vector conversion on the input text to obtain input data, generates simulation data corresponding to the target simulation according to the input data through the prediction model, and introduces a simulation parameter generation prediction model based on a large model to achieve automation and intelligence of simulation data generation, lower the use threshold of the simulation system, improve the efficiency of simulation data generation, and achieve automatic generation from input text to simulation data corresponding to the target simulation, so that users do not need to directly operate a complex simulation system, and can generate simulation parameters through natural language, which greatly shortens the simulation preparation time. The natural language interaction method enables the system to understand user needs more quickly and accurately, improve user experience, and thus effectively solve the problems of high threshold, low efficiency, limited accuracy and lack of flexibility in the prior art.

[0088] 2. The present invention can lower the threshold for use: by introducing intelligent generation models, users do not need to have an in-depth understanding of the complex parameter configuration and debugging process of the simulation system. Users only need to input simulation requirements in natural language, and the system can automatically output simulation parameters that meet the requirements. This greatly lowers the threshold for use of simulation technology, making it easy for non-professionals or beginners to get started, thereby expanding the application scope of simulation technology.

[0089] 3. The present invention can improve generation efficiency: the intelligent generation model can quickly respond to user input and automatically find the optimal or near-optimal combination of simulation parameters based on a large amount of training data and optimization algorithms. This avoids the tedious process of repeated debugging and trial and error in traditional methods and significantly improves the generation efficiency of simulation data. This advantage is particularly evident in urgent or high-demand projects.

[0090] 4. The present invention can improve simulation accuracy: The intelligent generation model can capture the complex relationships between parameters through advanced technologies such as deep learning, and generate more accurate and reliable simulation parameters based on these relationships. Compared with manually set parameters, these automatically generated parameters are more likely to reflect the actual situation, thereby improving the precision and accuracy of the simulation results.

[0091] 5. The present invention can enhance flexibility: This solution is not only applicable to specific simulation systems, but can also adapt to new simulation scenarios and requirements by retraining or adjusting the model. With the continuous development of simulation technology and the emergence of new application scenarios, users can achieve support for new systems and parameter generation through simple model updates or configuration adjustments. This flexibility makes this solution have a wider application prospect.

[0092] 6. The present invention can optimize user experience: Through natural language input and visual display of simulation results, the present solution provides users with a more intuitive and convenient operation experience. Users do not need to face complex interfaces and parameter settings, but only need to focus on their own simulation needs and results. This user-centric design concept helps to improve user satisfaction and loyalty.

[0093] It should be understood that, although the steps in the flowchart of the accompanying drawings are displayed in sequence as indicated by the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least a part of the steps in the flowchart of the accompanying drawings may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be executed in turn or alternately with other steps or at least a part of the sub-steps or stages of other steps.

[0094] The above descriptions are only some embodiments of the present invention. It should be pointed out that, for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A method for generating simulation data, characterized in that: The method comprises: Obtaining a training set related to the target simulation, and obtaining a prediction model for a decoder of the converter model according to the training set; Obtain input text, perform word segmentation and word vector conversion on the input text to obtain input data; the input text includes simulation requirements and prompts; The prediction model generates simulation data corresponding to the target simulation according to the input data; the simulation data includes a scene, a target, an environment and a camera position.

2. The simulation data generation method according to claim 1, characterized in that: The step of obtaining a prediction model for a decoder of a converter model according to the training set comprises: Performing text preprocessing and word vector generation on the data in the training set to obtain a corresponding word vector set; The decoder of the converter model is trained by the word vector set to obtain a prediction model.

3. The simulation data generation method according to claim 2, characterized in that: The text preprocessing and word vector generation of the data in the training set to obtain a corresponding word vector set includes: Segmenting the long documents in the training set into paragraphs and sentences; Segmenting each of the sentences, converting the continuous text in the paragraph into a sequence of words and phrases to obtain a segmentation result; The word2vec algorithm is used to convert the word segmentation results into representations in a high-dimensional vector space to obtain the corresponding word vector set.

4. The simulation data generation method according to claim 2, characterized in that: The method of obtaining a prediction model by training a decoder of a converter model through the word vector set includes: The word vector set and the corresponding simulation parameter label are combined and input into the decoder of the converter model, and the BCEloss function is used as the loss function for multiple rounds of training to obtain the prediction model, and the parameters of the prediction model are optimized by the back propagation algorithm.

5. The simulation data generation method according to claim 1, characterized in that: The decoder of the converter model is used to obtain a prediction model according to the training set, and further comprises: Add start tags and end tags to the input and output of the decoder of the converter model and the prediction model.

6. The simulation data generation method according to claim 1, characterized in that: The method further comprises: A simulation environment for the target simulation is run according to the simulation data to obtain a simulation result, and the prediction model is optimized according to the simulation result.

7. The simulation data generation method according to claim 6, characterized in that: The method further comprises: According to the simulation data, an enhanced model MOD corresponding to a theme and a scene is added to the simulation environment, and parameters of the simulation environment are set according to the simulation data.

8. A simulation data generating device, characterized in that: The device comprises: A model training module, used to obtain a training set related to the target simulation, and obtain a prediction model for the decoder of the converter model according to the training set; An input acquisition module is used to acquire input text, perform word segmentation and word vector conversion on the input text to obtain input data; A simulation data generation module is used to generate simulation data corresponding to the target simulation according to the input data through the prediction model.

9. An electronic device, characterized in that: include: at least one processor and at least one memory, wherein: The memory has computer-readable instructions stored thereon; The computer-readable instructions are executed by one or more of the processors so that the electronic device implements the simulation data generating method according to any one of claims 1 to 7.

10. A storage medium having computer-readable instructions stored thereon, characterized in that: The computer-readable instructions are executed by one or more processors to implement the simulation data generating method according to any one of claims 1 to 7.