A digitized yard test verification system

Abstract

This invention provides a digital vehicle testing and verification system, comprising a test task injection module, a multi-physics synchronous acquisition module, a physical information fusion modeling module, an uncertainty quantification assessment module, and a verification result feedback module, sequentially connected to form a closed-loop feedback loop. Based on a digital prototype of a general-purpose ground mobility platform, this invention extracts boundary conditions and generates test parameters, synchronously acquires measured data from multiple physics fields, and constructs a behavioral proxy model using a physical information neural network combined with first-principles control equations, eliminating the need to rely on historical operational data. The Monte Carlo Dropout method is used to calculate the 95% confidence interval width as a reliability index, automatically triggering parameter adjustment and closed-loop verification. This invention achieves high-reliability performance verification of a newly developed general-purpose ground mobility platform and the first set of general-purpose ground mobility platforms, transforming verification conclusions from fuzzy judgments into quantitative assessments.

Description

Technical Field

[0001] This invention relates to the field of vehicle testing and verification, and in particular to a digital vehicle testing and verification system. Background Technology

[0002] Vehicle testing is a crucial step in verifying the design performance of general-purpose ground mobility platforms, identifying potential defects, and ensuring their reliability. However, existing technologies suffer from the following core shortcomings: Open-loop verification is inefficient: traditional methods rely on engineers' experience to set operating conditions and manually analyze results. When the test fails to meet the standards, the test plan needs to be adjusted repeatedly, which is time-consuming, costly, and difficult to automate. Multiphysics data fragmentation: Existing systems only collect kinematic parameters (such as velocity and acceleration) or electrical signals, and cannot simultaneously acquire multiphysics coupled response data such as structural strain, heat flow distribution, and electromagnetic environment, which makes it impossible to truly reflect the comprehensive performance of general ground mobile platforms under extreme working conditions; Reliance on historical data leads to the failure of verification of newly developed general-purpose ground mobility platforms: Mainstream data-driven methods (such as statistical models and machine learning) rely on massive amounts of historical fault or usage frequency data. For newly developed general-purpose ground mobility platforms and the first set of general-purpose ground mobility platforms, the lack of such data has caused them to fail, becoming a bottleneck for innovative research and development. The verification conclusions lack quantitative basis: Traditional methods only provide a binary judgment of "pass / fail", which cannot quantify the credibility of the verification conclusions and is difficult to support high-risk engineering decisions.

[0003] Although existing digital twin technology is used in parking lot management, its application is limited to visual monitoring and post-event analysis, and it has failed to build a closed-loop verification system based on physical laws, with real-time observation as input and uncertainty quantification as the driving force.

[0004] Therefore, there is an urgent need for a digital vehicle testing and verification system that does not require historical data, can automatically perform closed-loop optimization, and whose credibility can be quantitatively verified. Summary of the Invention

[0005] The present invention aims to solve at least one of the technical problems existing in the prior art, and provides a digital vehicle testing and verification system.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: In a first aspect, the present invention provides a digital vehicle testing and verification system, comprising a test task injection module, a multi-physics field synchronous acquisition module, a physical information fusion modeling module, an uncertainty quantification evaluation module, and a verification result feedback module that are sequentially connected in communication and form a closed-loop feedback loop; The test task injection module is used to receive the performance verification target of the general ground mobility platform from external input, and automatically extract key performance boundary conditions based on the digital mock-up of the general ground mobility platform, convert them into a test constraint set, determine the corresponding test task type, and generate the corresponding executable test parameter set based on the test task type. The multi-physics synchronous acquisition module is used to synchronously acquire multi-physics measured data corresponding to the performance verification target of the general ground mobility platform during the execution of the operating conditions defined by the executable test parameter set on the general ground mobility platform. The physical information fusion modeling module is used to construct a general ground mobility platform behavior proxy model that satisfies the corresponding physical law constraints based on the multi-physics field measured data and the prior design parameters characterizing the inherent physical properties and control logic of the general ground mobility platform, and according to the control equations that match the test task type. The uncertainty quantification and evaluation module is used to perform credibility analysis on the output of the general ground mobility platform behavior agent model and generate credibility indexes for performance verification. The verification result feedback module sends a parameter adjustment instruction to the test task injection module based on the comparison result of the credibility index and the preset threshold, so as to trigger a new round of testing process and realize closed-loop verification.

[0007] In some possible embodiments, the multiphysics measured data includes at least two of the following: structural strain, heat flux distribution, electromagnetic environment, and motion response.

[0008] In some possible embodiments, the physical information fusion modeling module employs a physical information neural network (PINN), whose loss function includes a data fitting term and a partial differential equation residual term: ,in, To measure the physical field output, This is the network prediction value. For sampling points Network parameters The derived partial differential equation residuals, To balance the weights.

[0009] In some possible embodiments, the uncertainty quantification assessment module employs the Monte Carlo Dropout method, performing multiple forward propagations on the same input during the inference phase to obtain the distribution of predicted results, and calculating the 95% confidence interval width based on the distribution of predicted results. The credibility metric used for performance verification is as follows: ,in, and These are the upper and lower 2.5 percentiles of the predicted results.

[0010] In some possible embodiments, when the credibility index is lower than the preset threshold, the verification result feedback module is configured to: send a parameter adjustment instruction to the test task injection module to trigger test reconfiguration; and call the sensitivity analysis submodule to calculate the contribution of each test parameter in the executable test parameter set to the credibility index, identify the test parameter with the largest contribution, and prioritize adjusting the value range of the test parameter.

[0011] In some possible embodiments, when constructing the behavior agent model of the general ground mobility platform, the physical information fusion modeling module does not rely on historical fault records or usage frequency statistics, but only on the physical structure parameters, control logic, and real-time collected multiphysics field observation data of the general ground mobility platform, and models the control equations derived from first principles, thereby supporting the performance verification of newly developed general ground mobility platforms or the first set of general ground mobility platforms without historical operating data.

[0012] In some possible embodiments, the control equations include at least one of the longitudinal dynamics equations of a general ground mobility platform, the thermal diffusion equation of a power battery, or the electromagnetic field equation of a motor, and the type of the control equations is dynamically loaded by the type of the test mission to satisfy the corresponding physical constraints.

[0013] In a second aspect, the present invention provides an electronic device, comprising: One or more processors; A storage unit is used to store one or more programs, which, when executed by one or more processors, enable the one or more processors to implement the digital vehicle testing and verification system described above.

[0014] Thirdly, the present invention provides a computer-readable storage medium having a computer program stored thereon, characterized in that the computer program, when executed by a processor, can realize the digital vehicle testing and verification system described above.

[0015] The digital vehicle testing and verification system of this invention has the following advantages: 1. This invention enables highly reliable verification in scenarios without historical data: The physical information fusion modeling module relies solely on the physical structure parameters, control logic, and real-time multiphysics field measured data of the general-purpose ground mobility platform. Based on the control equations derived from first principles (such as the thermal diffusion equation of the power battery), it constructs a behavioral proxy model of the general-purpose ground mobility platform. It does not require historical fault or usage frequency statistics, thus providing a feasible verification path for newly developed general-purpose ground mobility platforms and the first set of general-purpose ground mobility platforms without historical operating data. 2. This invention can quantitatively verify the conclusions to support engineering decisions: The uncertainty quantification assessment module generates a 95% confidence interval width as a credibility index through Monte Carlo Dropout, transforming the fuzzy "pass / fail" judgment into an objective quantitative assessment, which is conducive to improving the reliability of high-risk engineering decisions; 3. This invention enables efficient closed-loop iteration of the verification process: the verification result feedback module automatically triggers sensitivity analysis based on the credibility index, identifies key test parameters and prioritizes adjusting their value range, avoiding blind trial and error and improving verification efficiency; 4. This invention enables precise integration of design and testing: The test task injection module automatically extracts key performance boundary conditions based on a digital prototype of a general ground mobility platform, converts them into a set of test constraints, and generates a set of executable test parameters, accurately covering the design objectives. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the structure of an example electronic device for the digital vehicle testing and verification system of the present invention; Figure 2 This is a schematic diagram of the structure of the digital vehicle testing and verification system of the present invention. Detailed Implementation

[0017] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] Figure 1 This is a schematic diagram of an example electronic device for implementing a digital vehicle testing and verification system according to the present invention. Figure 1 As shown, the electronic device 100 includes one or more processors 110, one or more storage devices 120, one or more input devices 130, one or more output devices 140, etc., and these components are interconnected via a bus system 150 and / or other forms of connection mechanisms. It should be noted that... Figure 1The components and structures of the electronic devices shown are merely illustrative descriptions and not restrictive requirements. Depending on actual deployment needs, the electronic devices may also have other components and structures. For example, the electronic devices may take the form of edge computing servers, in-vehicle embedded systems, or cloud platforms.

[0019] The processor 110 may be a central processing unit (CPU), or may be a processing unit consisting of multiple processing cores, or other forms of processing unit with data processing capabilities and / or instruction execution capabilities, and may control other components in the electronic device 100 to perform desired functions.

[0020] Storage device 120 may include one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may include, for example, random access memory (RAM) and / or cache memory. The non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc. One or more computer program instructions may be stored on the computer-readable storage medium, which a processor may execute to implement the client functions (implemented by the processor) in the embodiments of this disclosure described below, and / or other desired functions. Various applications and various data may also be stored in the computer-readable storage medium, such as various data used and / or generated by the applications.

[0021] The input device 130 may be a device used by a user to input commands, and may include one or more of a keyboard, mouse, microphone, and touch screen.

[0022] The output device 140 can output various information (such as images or sounds) to the outside (e.g., a user) and may include one or more of a display, a speaker, etc.

[0023] Specifically, the processor 110 in this invention is configured to execute the computer program stored in the storage device 120 to realize the core functions of the digital vehicle test verification system described in this invention. When the computer program stored in the storage device 120 is executed by the processor 110, it can realize all the functions of the test task injection module 201, the multi-physics synchronous acquisition module 202, the physical information fusion modeling module 203, the uncertainty quantification evaluation module 204, and the verification result feedback module 205. The input device 130 (such as a touch screen or voice input device) is used to receive the target input for the performance verification of the general ground mobility platform. The output device 140 (such as a display screen or data interface) is used to output the verification report and parameter adjustment instructions.

[0024] Figure 2This is a schematic diagram of the structure of the digital vehicle testing and verification system of the present invention. Figure 2 As shown, a digital vehicle testing and verification system includes a test task injection module 201, a multi-physics field synchronous acquisition module 202, a physical information fusion modeling module 203, an uncertainty quantification and evaluation module 204, and a verification result feedback module 205, which are sequentially connected and form a closed-loop feedback loop. The test task injection module 201 receives externally input performance verification targets for a general-purpose ground mobility platform (e.g., "verifying the power heat source management performance of a certain type of general-purpose ground mobility platform in a -20℃ environment"), and, based on a digital prototype of the general-purpose ground mobility platform... The Mock-up module automatically extracts key performance boundary conditions (such as power heat source heat generation rate of 3.2kW and heat dissipation capacity of 2.8kW), converts them into test constraint sets (such as ambient temperature of -20℃ and power charge / discharge rate of 1C-3C), determines the corresponding test task type, and generates corresponding executable test parameter sets based on the test task type (such as the general ground mobility platform driving sequence: 0-80km / h uniform acceleration, 80km / h uniform speed for 30 minutes, power replenishment strategy: constant current replenishment to 80% capacity); the multi-physics field synchronous acquisition module 202 is used to synchronously acquire data with the general ground mobility platform during the execution of the operating conditions defined in the executable test parameter set (such as 0-80km / h uniform acceleration, 80km / h uniform speed for 30 minutes). The system uses multi-physics field measured data corresponding to the performance verification target; the physical information fusion modeling module 203 is used to construct a general ground mobility platform behavior proxy model that satisfies the corresponding physical law constraints based on the multi-physics field measured data and prior design parameters (including physical structural parameters such as the centroid position of the general ground mobility platform, control logic such as power heat source management strategy, and real-time observation data) characterizing the inherent physical properties and control logic of the general ground mobility platform; the uncertainty quantification evaluation module 204 is used to perform credibility analysis on the output of the general ground mobility platform behavior proxy model (such as the predicted maximum temperature of the power heat source) and generate a credibility index for performance verification; the verification result feedback module 205 is used to perform credibility index and preset threshold (such as...) The comparison results are used to send parameter adjustment instructions (such as "extend the power charge and discharge rate range to 1C-4C") to the test task injection module 205 to trigger a new round of test process and achieve closed-loop verification.

[0025] Specifically, the digital vehicle testing and verification system is deployed in an edge-cloud collaborative architecture. The cloud is responsible for the initial training of the Physical Information Neural Network (PINN), while the edge nodes (deployed in the test site control center) perform real-time inference, uncertainty quantification, and closed-loop feedback control to ensure that the test feedback latency is ≤200ms. The edge node hardware configuration is an NVIDIA Jetson AGX Xavier edge server (8-core CPU / 32-core GPU / 16GB memory), which meets the real-time requirements.

[0026] Specifically, the multi-physics synchronous acquisition module 202 is equipped with a timestamp alignment unit, which uses the IEEE 1588 Precision Time Protocol (PTP) to achieve microsecond-level synchronization, ensuring that sensor data from different physical domains [structural strain (1kHz), heat flow distribution (100Hz), electromagnetic environment (500Hz), and motion response (10Hz) data] are accurately aligned in a unified spatiotemporal coordinate system.

[0027] Specifically, the verification result feedback module 205 generates a digital thread record during the closed-loop process, storing the input parameters, collected data, model output, credibility indicators and reconfiguration decisions of each experiment into the blockchain ledger in an immutable manner, supporting full lifecycle traceable verification and auditing.

[0028] In some embodiments, the measured multiphysics data include at least two of the following: structural strain, heat flux distribution, electromagnetic environment, and motion response (e.g., heat flux distribution of the power heat source and strain of the general ground mobile platform frame structure) to ensure comprehensive capture of the multiphysics coupled response.

[0029] In some embodiments, the physical information fusion modeling module 203 employs a physical information neural network (PINN), whose loss function includes a data fitting term and a partial differential equation residual term: ,in, To measure the physical field output, This is the network prediction value. For sampling points Network parameters The residuals of the derived partial differential equations (such as the residuals of the thermal diffusion equation for power batteries). To balance the weights (such as) =0.7, =0.3).

[0030] In some embodiments, the uncertainty quantification assessment module employs the Monte Carlo Dropout method, performing multiple (preferably 100) forward propagations on the same input during the inference phase to obtain the distribution of predicted results, and calculating the 95% confidence interval width based on the distribution of predicted results. As a credibility indicator for performance verification: ,in, and These are the upper and lower 2.5 percentiles of the predicted results.

[0031] Specifically, in each forward propagation, the Dropout layer of the neural network is randomly activated with a retention rate r (0.5≤r≤0.9, preferably r=0.8) to obtain the distribution of the prediction results.

[0032] In some embodiments, when the credibility index is lower than a preset threshold (e.g., 3°C), the verification result feedback module 205 is configured to: send a parameter adjustment instruction to the test task injection module 201 to trigger test reconfiguration; and call the sensitivity analysis submodule (e.g., Sobol global sensitivity analysis) to calculate the contribution of each test parameter (e.g., charge / discharge rate, ambient temperature) in the executable test parameter set to the credibility index, identify the test parameter with the largest contribution (e.g., power charge / discharge rate contribution of 58%), and prioritize adjusting the value range of the test parameter.

[0033] In some embodiments, when the physical information fusion modeling module 203 constructs a behavior agent model for a general-purpose ground mobility platform, it does not rely on historical fault records or usage frequency statistics. Instead, it models the general-purpose ground mobility platform based solely on its physical structural parameters (such as the heat capacity of the power heat source of 4.18 kJ / (kg*K)), control logic (such as starting the cooling unit when the temperature is >45℃), and real-time collected multi-physics field observation data, according to the control equations derived from first principles (such as the power battery thermal diffusion equation). This supports the performance verification of newly developed general-purpose ground mobility platforms or the first set of general-purpose ground mobility platforms without historical operating data.

[0034] To visually demonstrate the system's effectiveness in verifying platforms without historical data, a specific implementation verification is conducted based on the functions of the aforementioned modules: This embodiment verifies the core performance of a general-purpose ground mobility platform without historical operational data. This platform features a completely new design, and the absence of historical fault data and operational statistics can be used for model training. Traditional verification methods require manual setting of operational conditions based on experience with similar platforms, with parameter iterations and adjustments no less than three times. Only kinematic parameters are collected, and the complete verification cycle takes 25-35 days, providing only a qualitative "pass / fail" conclusion.

[0035] When performing verification using the system of this invention, the specific process is as follows: The test task injection module extracts the heat generation rate of the power heat source (referring to the heat generated by the power heat source per unit time) of 3.2 kW based on the digital prototype of the general ground mobility platform and converts it into a test constraint set; The multi-physics synchronous acquisition module synchronously acquires heat flow data from the power heat source and strain data from the platform frame structure. The physical information fusion modeling module loads the thermal diffusion equation of the power battery. Complete the modeling; The uncertainty quantification assessment module calculated a credibility index (used to measure the degree of conformity between the verification results and the actual situation) of 2.8℃, which is lower than the preset threshold of 3℃. The verification result feedback module triggers test reconfiguration and prioritizes adjusting the power charge / discharge rate parameter range through sensitivity analysis.

[0036] The verification cycle was only 12 days, which is 50%-70% shorter than the traditional solution. It achieved highly reliable quantitative verification of the newly developed general-purpose ground mobility platform and the first set of general-purpose ground mobility platforms without historical data.

[0037] In some embodiments, the governing equations include at least one of the following: the longitudinal dynamics equation of the general-purpose ground mobility platform, the thermal diffusion equation of the power battery, or the electromagnetic field equation of the motor. The type of the governing equation is dynamically loaded according to the test task type to satisfy the corresponding physical constraints. For example, when the verification target is "power heat source management performance of the general-purpose ground mobility platform", the system automatically loads the thermal diffusion equation of the power battery as the governing equation to reflect the physical constraints of the thermal field constituted by the boundary conditions of heat conduction, heat generation, and heat dissipation. When the verification target is "braking performance of the general-purpose ground mobility platform", the system automatically loads the braking dynamics equation of the general-purpose ground mobility platform (including braking force distribution, friction coefficient, and inertial response) as the governing equation to reflect the physical constraints of the mechanical field. When the verification target is "electromagnetic compatibility of the motor of the general-purpose ground mobility platform", the system automatically loads the electromagnetic field equation of the motor (such as the simplified form of the Maxwell equations) to satisfy the physical constraints of the electromagnetic field.

[0038] The digital vehicle testing and verification system of this invention has the following advantages: 1. This invention enables highly reliable verification in scenarios without historical data: the physical information fusion modeling module relies only on the physical structure parameters, control logic, and real-time multiphysics field measured data of the general ground mobility platform. Based on the control equations derived from first principles (such as the battery thermal diffusion equation), it constructs a behavioral proxy model of the general ground mobility platform. It does not require historical fault or usage frequency statistics, providing a feasible verification path for newly developed general ground mobility platforms and the first set of general ground mobility platforms. 2. This invention can quantitatively verify the conclusions to support engineering decisions: The uncertainty quantification assessment module generates a 95% confidence interval width as a credibility index through Monte Carlo Dropout, transforming the fuzzy "pass / fail" judgment into an objective quantitative assessment, which is conducive to improving the reliability of high-risk engineering decisions; 3. This invention enables efficient closed-loop iteration of the verification process: the verification result feedback module automatically triggers sensitivity analysis based on the credibility index, identifies key test parameters and prioritizes adjusting their value range, avoiding blind trial and error and improving verification efficiency; 4. This invention enables precise integration of design and testing: The test task injection module automatically extracts key performance boundary conditions based on a digital prototype of a general ground mobility platform, converts them into a set of test constraints, and generates a set of executable test parameters, accurately covering the design objectives.

[0039] In another aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, enables the implementation of the method described above.

[0040] The computer-readable medium may be included in the apparatus, device, or system disclosed herein, or it may exist independently.

[0041] The computer-readable storage medium can be any tangible medium that contains or stores a program, and can be an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device. More specific examples include, but are not limited to, electrical connections having one or more wires, portable computer disks, hard disks, optical fibers, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0042] The computer-readable storage medium may also include data signals propagated in baseband or as part of a carrier wave, carrying computer-readable program code, specific examples of which include, but are not limited to, electromagnetic signals, optical signals, or any suitable combination thereof.

[0043] Parameter description: All parameter values ​​mentioned in this manual are not subjective assumptions, but rather a comprehensive result of the application scenario's security / efficiency requirements, industry standards and specifications, and industry practice experience thresholds. In actual applications, the parameters will be fine-tuned according to the relevant scenarios.

[0044] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A digitized proving ground test verification system, characterized by, It includes an experimental task injection module that is sequentially connected to form a closed-loop feedback loop, a multi-physics field synchronous acquisition module, a physical information fusion modeling module, an uncertainty quantification assessment module, and a verification result feedback module; The test task injection module is used to receive the performance verification target of the general ground mobility platform from external input, and automatically extract key performance boundary conditions based on the digital mock-up of the general ground mobility platform, convert them into a test constraint set, determine the corresponding test task type, and generate the corresponding executable test parameter set based on the test task type. The multiphysics synchronous acquisition module is used to synchronously acquire multiphysics measured data corresponding to the performance verification target of the general ground mobility platform during the execution of the operating conditions defined by the executable test parameter set on the general ground mobility platform. The physical information fusion modeling module is used to construct a general ground mobility platform behavior proxy model that satisfies the corresponding physical law constraints based on the multi-physics field measured data and the prior design parameters characterizing the inherent physical properties and control logic of the general ground mobility platform, and according to the control equations that match the test task type. The uncertainty quantification and evaluation module is used to perform credibility analysis on the output of the general ground mobility platform behavior agent model and generate credibility indexes for performance verification. The verification result feedback module sends a parameter adjustment instruction to the test task injection module based on the comparison result of the credibility index and the preset threshold, so as to trigger a new round of testing process and realize closed-loop verification.

2. The digital yard test verification system of claim 1, wherein, The measured multiphysics data include at least two of the following: structural strain, heat flow distribution, electromagnetic environment, and motion response.

3. The digital yard test verification system of claim 1, wherein, The physical information fusion modeling module employs a physical information neural network (PINN), whose loss function includes a data fitting term and a partial differential equation residual term: wherein, is the measured physical field output, is the network prediction, is the partial differential equation residual derived from the network parameters at the sampling point , is the balancing weight.

4. The digital yard test verification system of claim 1, wherein, The uncertainty quantification evaluation module adopts a Monte Carlo Dropout method, performs multiple forward propagations on the same input in an inference stage to obtain a prediction result distribution, and calculates a 95% confidence interval width based on the prediction result distribution , as the credibility index for performance verification: wherein, and are the upper and lower 2.5% quantiles of the prediction, respectively.

5. The digital yard test verification system of claim 1, wherein, When the credibility index is lower than the preset threshold, the verification result feedback module is configured to send a parameter adjustment instruction to the test task injection module to trigger test reconfiguration. The system also calls the sensitivity analysis submodule to calculate the contribution of each test parameter in the executable test parameter set to the credibility index, identifies the test parameter with the largest contribution, and prioritizes adjusting the value range of that test parameter.

6. The digital yard test verification system of claim 1, wherein, When constructing the behavior agent model of the general-purpose ground mobility platform, the physical information fusion modeling module does not rely on historical fault records or usage frequency statistics. Instead, it models the general-purpose ground mobility platform based solely on its physical structure parameters, control logic, and real-time multi-physics field observation data, according to the control equations derived from first principles. This supports performance verification of newly developed general-purpose ground mobility platforms or the first set of general-purpose ground mobility platforms without historical operating data.

7. The digital yard test verification system of claim 1, wherein, The control equations include at least one of the following: the longitudinal dynamics equation of a general ground mobility platform, the thermal diffusion equation of a power battery, or the electromagnetic field equation of a motor. The type of the control equations is dynamically loaded by the type of the test mission to meet the corresponding physical constraints.

8. An electronic device, characterized in that, include: One or more processors; a memory unit configured to store one or more programs that, when executed by the one or more processors, enable the one or more processors to implement the digital yard test validation system according to any one of claims 1 to 7.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, The computer program, when executed by a processor, enables the digital yard test validation system according to any one of claims 1 to 7.

Images