A driving simulation system, a method and a device for calibrating its authenticity

Abstract

The present invention discloses a driving simulation system, a method and device for calibrating its authenticity, comprising: a driving simulation cabin, wherein a driving simulator is provided in the driving simulation cabin; wherein a processor is provided in the driving simulator, wherein the processor is connected with an input module, an output module, a virtual scene projection module, an environmental noise and vibration simulation module, a road driving scene design and operation module, and a vehicle dynamic simulation module; wherein the input module of the driving simulator is also connected with a brake pedal via a first angle sensor, the input module is also connected with an oil port pedal via a second angle sensor, the input module is also connected with a steering wheel via a third angle sensor, and the input module is also connected with a gear position via a displacement sensor; and the output module of the driving simulator is also connected with an instrument panel. The simulation degree of the driving simulation system to the real driving scene is improved, and theoretical and technical support is provided for its service in the field of highway engineering design.

Description

Technical Field

[0001] The present invention relates to the technical field of driving simulation system reality calibration, and in particular to a driving simulation system, and a reality calibration method and device thereof. Background Art

[0002] The statements in this section merely mention background art related to the present invention and do not necessarily constitute prior art.

[0003] In recent years, driving simulation technology has been widely used in the fields of traffic safety and driving training. In the field of traffic safety, the research focus is often on dangerous driving scenarios or dangerous driving behaviors that are difficult to obtain on real roads. It mainly conducts feature analysis, modeling, characterization and prediction, accident reproduction, etc. on the driver behavior data obtained through simulated driving, mostly qualitative analysis or trend analysis. In the field of driving training, it is usually to reduce the safety risks of students learning to drive on the road, and use driving simulation systems to provide students with virtual driving experience, aiming to improve their driving skills.

[0004] In order to expand the application scope of driving simulation systems and improve the driving experience, researchers usually use driving simulation systems in combination with a variety of devices, such as VR glasses / helmets, brain-computer interaction devices, etc., but there is always a lack of relevant basis and methods for calibrating the realism of driving simulation systems. Since the realism of driving simulation systems is difficult to evaluate, there are few cases of applying them to the highway engineering design industry. Summary of the invention

[0005] In order to address the deficiencies in the prior art, the present invention provides a driving simulation system, a method and device for calibrating its reality; this patent designs a new driving simulation system that combines virtual reality technology with driving simulation technology, and on this basis proposes a means for calibrating and correcting the reality of the driving simulation system, aiming to further improve the degree of simulation of the driving simulation system for real driving scenes, and provide theoretical and technical support for its service in the field of highway engineering design.

[0006] In a first aspect, the present invention provides a driving simulation system;

[0007] A driving simulation system comprises: a driving simulation cabin, wherein a driving simulator is arranged in the driving simulation cabin;

[0008] The driving simulator is provided with a processor, which is connected to an input module, an output module, a virtual scene projection module, an environmental noise and vibration simulation module, a road driving scene design and operation module, and a vehicle dynamic simulation module;

[0009] The input module of the driving simulator is also connected to the brake pedal via a first angle sensor, the input module is also connected to the oil port pedal via a second angle sensor, the input module is also connected to the steering wheel via a third angle sensor, and the input module is also connected to the gear position via a displacement sensor;

[0010] The output module of the driving simulator is also connected to the instrument panel.

[0011] In a second aspect, the present invention provides a method for calibrating the realism of a driving simulation system;

[0012] A driving simulation system realism calibration method, comprising:

[0013] Preliminary calibration steps: Collect multi-dimensional driving behavior data; compare the similarities and differences of multi-dimensional behavior indicators of different drivers when using the driving simulation system and when driving on real road scenes, and determine and adjust the setting parameters that affect the realism of the driving simulation system through similarity analysis and calculation of the two behaviors, so as to complete the preliminary calibration of the driving simulation system;

[0014] Recalibration steps: Replace another group of drivers, and calculate the similarity between the two behavior data through comparative analysis of the multi-dimensional driving behavior data of real and virtual scenarios; based on the similarity between the two behavior data, complete the recalibration of the driving simulation system.

[0015] Final calibration steps: Replace a group of drivers again for re-verification until the calibration result is less than the critical threshold and the final calibration is completed.

[0016] In a third aspect, the present invention also provides a device for calibrating the authenticity of a driving simulation system;

[0017] A device for calibrating the authenticity of a driving simulation system, comprising:

[0018] A similarity calculation module is configured to: calculate the index similarity between the real driving scene and the virtual driving scene based on the weights and the corresponding index values, and determine whether the number of parameter optimizations reaches the set number. If not, return to the acquisition module; if yes, it means that the preliminary calibration of the driving simulation system has been completed, and record the current comprehensive index similarity as the critical threshold;

[0019] The judgment module is configured to: calculate the similarity of comprehensive indicators of a new batch of drivers in real driving scenarios and virtual driving scenarios, and judge whether the similarity of comprehensive indicators of the new batch of drivers is less than a critical threshold. If so, the calibration of the driving simulation system is terminated; if not, the parameters affecting the realism of the driving simulator are adjusted and new drivers are continued to be recruited for experiments, thereby further adjusting and optimizing the relevant parameters of the driving simulator.

[0020] The device further comprises:

[0021] An acquisition module is configured to: acquire a number of behavioral and physiological reaction indicators of a driver when performing a set driving task in a real road scene; acquire a number of behavioral and physiological reaction indicators of a driver when performing a set driving task in a simulated scene using a driving simulator;

[0022] The indicator selection module is configured to: set m evaluation optimization dimensions, and for each dimension, select a number of corresponding behavior and physiological response indicators; under the same dimension, the indicators selected in the real road scene are the same as those selected in the simulation scene;

[0023] The weight setting module is configured to set weights for a number of selected behavior and physiological response indicators.

[0024] Compared with the prior art, the present invention has the following beneficial effects:

[0025] The main purpose of the present invention is to improve the simulation realism of the driving simulation system, combine the driving simulator with virtual reality technology, create a virtual driving scene, and build a new driving simulation system. The driving behavior data and psychological and physiological data of the driver when performing driving tasks on the real road and the driving simulation system, as well as the driver's subjective feelings, are obtained, and through comparative analysis and calculation, a reference basis for calibrating the realism of the driving simulation system is formed, and the system parameters are optimized and adjusted accordingly. It aims to improve the simulation realism of the driving simulation system for real driving scenes, and provide technical and theoretical support for the application of driving simulation experiments in the field of traffic engineering design. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The accompanying drawings in the specification, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0027] Figure 1 This is a schematic diagram of hardware connection of Embodiment 1;

[0028] Figure 2 This is a flow chart of the method of embodiment 1. DETAILED DESCRIPTION

[0029] It should be noted that the following detailed descriptions are exemplary and are intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present invention belongs.

[0030] In the absence of conflict, the embodiments of the present invention and the features of the embodiments may be combined with each other.

[0031] In this embodiment, all data is obtained in compliance with laws and regulations and based on the user's consent, and is used legally.

[0032] A module for building virtual driving scenarios with high fidelity and strong immersion that are highly similar to real road scenarios and driving tasks is developed. A digital twin model of real roads is imported into the system, and a virtual driving scenario is built based on the digital twin model of the road. Then, AR technology is used to project the built virtual scenario onto the windshield of the driving simulation cabin to form a three-dimensional visual observation effect. The above-mentioned parts form a new driving simulation system based on the combination of virtual reality technology and driving simulation technology. On this basis, a new driving simulation system realism multi-cycle calibration method is designed based on driving experiments conducted in real and virtual driving scenarios and the driving behavior data obtained.

[0033] Embodiment 1

[0034] This embodiment provides a driving simulation system;

[0035] like Figure 1 As shown, a driving simulation system comprises: a driving simulation cabin, wherein a driving simulator is arranged in the driving simulation cabin;

[0036] The driving simulator is provided with a processor, which is connected to an input module, an output module, a virtual scene projection module, an environmental noise and vibration simulation module, a road driving scene design and operation module, and a vehicle dynamic simulation module;

[0037] The input module of the driving simulator is also connected to the brake pedal via a first angle sensor, the input module is also connected to the oil port pedal via a second angle sensor, the input module is also connected to the steering wheel via a third angle sensor, and the input module is also connected to the gear position via a displacement sensor;

[0038] The output module of the driving simulator is also connected to the instrument panel.

[0039] Furthermore, the driving simulation cabin is designed in a 1:1 ratio with a real car body. The basic vehicle operating components such as the brake pedal, gas pedal, steering wheel, gear position, etc. inside the car body are completely consistent with the real situation. At the same time, the display components such as the instrument panel can display the vehicle speed, mileage and other information during simulated driving.

[0040] Furthermore, the vehicle dynamic simulation module is used to utilize various sensors to make driving operations such as braking, accelerator, and steering wheel steering realistic, while coordinating the interaction between the three-dimensional virtual scene and the vehicle so that the driver can feel that he and the vehicle are moving, thereby achieving a realistic dynamic effect when driving.

[0041] Furthermore, the road driving scenario design and operation module is composed of a road scenario design sub-module, a driving scenario design sub-module and a scenario operation sub-module; wherein the road scenario design sub-module is generated through the design of various fixed modules such as road modules, building modules, sign modules, etc., and also provides a loading function for the road digital twin model; the driving scenario design sub-module is used to design certain specific traffic events and realize certain control functions, such as dynamic scenario controls such as vehicle operation control, traffic light control, and weather control; the scenario operation sub-module is used to run the established virtual driving scenario.

[0042] Furthermore, the road digital twin model is a three-dimensional solid model created based on BIM technology to simulate real road scenes.

[0043] Furthermore, the virtual scene projection module is used to project the virtual scene onto the front windshield and each side window of the cockpit using AR technology, and simultaneously project the virtual rearview mirror view on the left and right side windows, providing the driver with a 360° view and a 3D stereoscopic visual observation effect virtual driving environment. The virtual driving scene comes from the road driving scene constructed in the road driving scene design and operation module.

[0044] Furthermore, the environmental noise and vibration simulation module is used to simulate the sound and vibration feeling of real driving according to the interactive relationship between the vehicle driving operation and the scene, including the different sound quality produced by the vehicle under different deceleration and acceleration forces and the noise produced by other surrounding vehicles, etc., and at the same time combine with the dynamic platform to produce front and back, up and down vibrations to enhance the driver's immersive feeling. The data output module is used to output the operation data of the driving simulator and form a driving behavior data file for storage.

[0045] Embodiment 2

[0046] like Figure 2 As shown, the driving simulation system reality calibration method includes:

[0047] Preliminary calibration steps: collect multi-dimensional driving behavior data; compare the similarities and differences between different drivers' driving behavior data when using the driving simulation system and when driving in real road scenes, determine and adjust the setting parameters that affect the realism of the driving simulation system through similarity analysis and calculation of the two behaviors, and then have the same group of drivers perform the driving simulation task again, and adjust the driving simulation system parameters again based on the collected data. After repeating the above steps many times, the simulated driving behavior characteristics collected in the driving simulation gradually approach the driving behavior characteristics of the real scene, thereby completing the preliminary calibration of the driving simulation system and determining the critical threshold of realism;

[0048] Recalibration step: Replace another group of drivers, and calculate the similarity between the real and virtual scene driving behavior data through comparative analysis. If the result is less than the critical threshold of realism, the final calibration of the driving simulation system is completed; if the result is greater than the critical threshold of realism, repeat the initial calibration steps, adjust the parameters of the driving simulation system, and complete the recalibration;

[0049] Final calibration steps: Replace a group of drivers again for re-verification until the calibration result is less than the critical threshold and the final calibration is completed.

[0050] Recruit multiple drivers of different genders, ages, driving experience, professions and non-professionals to perform driving tasks in real road scenarios, and then use the driving simulation system to perform the same driving tasks in the corresponding digital twin virtual road scenarios, combined with eye movement, EEG or other physiological feedback collection equipment and questionnaires.

[0051] Furthermore, the preliminary calibration step specifically includes:

[0052] S201: Select a real road section, build a digital twin virtual driving scene that is the same as the real road scene, project the virtual driving scene onto the front and side windshields of the driving simulation cabin, and set a driving task;

[0053] S202: setting m dimensions for optimizing the realism of the driving simulation system, evaluating the realism of the driving simulation system and optimizing parameters from the m dimensions, selecting corresponding behavioral indicators and physiological response indicators for each dimension; and setting weights for the selected indicators; m is a positive integer;

[0054] In the same dimension, the indicators selected in the real road scenario are the same as those selected in the simulation scenario;

[0055] S203: obtaining a number of behavioral and physiological reaction indicators of the driver when performing a set driving task in a real road scene; obtaining a number of behavioral and physiological reaction indicators of the driver when performing a set driving task in a simulated scene using a driving simulator; the indicators collected in the real scene and the virtual scene correspond one to one and correspond to the indicators set in S202;

[0056] S204: Calculating the similarity of two types of indicators in the real driving scene and the virtual driving scene in m dimensions based on the weights and corresponding indicator values;

[0057] S205: adjusting and optimizing the relevant parameters of the corresponding dimension of the driving simulation system according to the similarity between the real driving behavior and the virtual driving behavior in each dimension;

[0058] S206: Determine whether the number of parameter optimizations reaches the set number. If not, use the same group of drivers for the experiment and return to S203. If yes, it means that the preliminary calibration of the driving simulation system has been completed, and the current indicator similarity is recorded as the critical threshold.

[0059] Furthermore, the recalibration step specifically includes:

[0060] S207: Replace another group of drivers and repeat steps S203 to S204;

[0061] S208: judging whether the behavior similarity of the driving experiment of the batch of drivers obtained in S207 is less than the critical threshold obtained in S206, if yes, terminating the calibration of the driving simulation system; if no, looping through S203, S204, and S205 in sequence until the calculated similarity is less than the critical threshold;

[0062] Furthermore, the final calibration step specifically includes:

[0063] S209: Continue to recruit a new batch of drivers for the experiment until the behavior similarity of a batch of drivers in the first driving experiment is less than the critical threshold, and the calibration is completed. Otherwise, return to S203.

[0064] Furthermore, in S201, the driving tasks include: free driving along lane lines (straight driving and curved driving), following a vehicle, changing lanes, recognizing signs and billboards, etc.

[0065] Furthermore, the S202: setting m dimensions for optimizing the realism of the driving simulation system, evaluating the realism of the driving simulation system and optimizing parameters from these m dimensions, respectively refer to: virtual scene rendering dimension, AR display setting dimension, cockpit internal operation unit parameter setting dimension and sound and somatosensory parameter setting dimension.

[0066] Among them, the virtual scene rendering dimension improves the visual realism of the virtual scene by adjusting the rendering settings of the virtual scene roads, infrastructure layout, traffic signs and other traffic participants;

[0067] Among them, the AR display setting dimension improves the overall realism and immersion of the virtual scene by adjusting the AR imaging distance, imaging viewing angle and display brightness;

[0068] Among them, the parameter setting dimension of the operating unit inside the cockpit adjusts the sensitivity of the driving simulator steering wheel, brake pedal and accelerator pedal to make it closer to the real driving vehicle;

[0069] Among them, the sound and body perception parameter setting dimension improves the realism of sound and body perception by adjusting the parameters of the ambient noise and vibration simulation module setting.

[0070] Furthermore, the step S202: setting weights for the selected behavior indicators and physiological response indicators specifically includes:

[0071] The analytic hierarchy process is used to set weights for the selected behavioral indicators and physiological response indicators; all indicators in the same layer are compared two by two to establish a judgment matrix, and the weights of each indicator are determined through consistency test to form a weight matrix W. i =[w1,w2,...,w m ], where i = 1, 2, 3, 4, m is the total number of indicators in different dimensions, w m is the weight of each indicator;

[0072] Furthermore, the S202: several behavior indicators and physiological reaction indicators, including: eye movement signal indicators, EEG signal indicators, driving behavior data indicators, physiological feedback indicators and questionnaire score indicators;

[0073] Furthermore, the S203: driver driving behavior data collection is to recruit a number of drivers to complete the same driving task in the same real road scene and virtual road scene. While the drivers are conducting the driving experiment, they need to wear eye movement, EEG and other physiological information feedback instruments to collect relevant physiological behavior feedback data, which is combined with the driving behavior data output by the driving simulator and the driver questionnaire survey data before and after the experiment to support the subsequent data analysis process.

[0074] Furthermore, the step S204: calculating the index similarity between the real driving scene and the virtual driving scene in m dimensions based on the weights and the corresponding index values; specifically includes:

[0075] S2041: using a linear function to normalize each indicator to a value of [0, 1], thereby eliminating the dimensional differences between the indicators;

[0076]

[0077] in, is the normalized value, x is the data, x min and x max are the minimum and maximum values ​​of the original data respectively.

[0078] For each dimension, form its initial feature matrix:

[0079]

[0080]

[0081] in, is the initial feature matrix obtained from the real driving scene, is the initial feature matrix obtained from the virtual driving scene, m is the number of features, and n is the number of samples;

[0082] i, j are both integers, and are the same initial feature indicators from the real driving scene and the virtual driving scene respectively.

[0083] S2042: For each dimension, combine its weight matrix W = [w1, w2, ..., w m ], after matrix multiplication operation, we get the weighted feature matrix U,

[0084]

[0085]

[0086] in, is the feature matrix obtained from the real driving scene,

[0087] is the feature matrix obtained from the virtual driving scene.

[0088] For each dimension U r and U v The similarity calculation of two sets of data is carried out, taking Euclidean distance as an example, according to the formula Where i is the sequence number of the four dimensions.

[0089] The above method is used to obtain the similarity results of driving behavior characteristics of real driving scenarios and virtual driving scenarios in four dimensions, which are used as the basis for authenticity.

[0090] Embodiment 3

[0091] This embodiment also provides a device for calibrating the authenticity of a driving simulation system;

[0092] A device for calibrating the authenticity of a driving simulation system, comprising:

[0093] An acquisition module is configured to: acquire a number of behavioral and physiological reaction indicators of a driver when performing a set driving task in a real road scene; acquire a number of behavioral and physiological reaction indicators of a driver when performing a set driving task in a simulated scene with a driving simulator installed;

[0094] The indicator selection module is configured to: set m evaluation optimization dimensions, and for each dimension, select a number of corresponding behavior and physiological reaction indicators in real scenes and a number of behavior and physiological reaction indicators in virtual scenes; under the same dimension, the behavior and physiological reaction indicators in real scenes and virtual scenes are the same;

[0095] A weight setting module, which is configured to: set weights for the behavior and physiological response indicators in the selected real scenes and virtual scenes;

[0096] A similarity calculation module is configured to: calculate the index similarity between the real driving scene and the virtual driving scene based on the weights and the corresponding index values, and determine whether the number of parameter optimizations reaches the set number. If not, return to the acquisition module; if yes, it means that the preliminary calibration of the driving simulation system has been completed, and record the current comprehensive index similarity as the critical threshold;

[0097] The judgment module is configured to: calculate the similarity of comprehensive indicators of a new batch of drivers, real driving scenarios and virtual driving scenarios, and judge whether the similarity of comprehensive indicators of the new batch of drivers is less than a critical threshold. If so, the calibration of the driving simulation system is terminated; if not, the parameters affecting the realism of the driving simulator are adjusted and new drivers are continued to be recruited for experiments, so as to further adjust and optimize.

[0098] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A driving simulation system realism calibration method, characterized in that: include: Initial calibration steps: Collect multi-dimensional driving behavior data; Compare the similarities and differences of multi-dimensional behavioral indicators of different drivers when using the driving simulation system and when driving on real road scenes. Through the similarity analysis and calculation of the two behaviors, determine the setting parameters that affect the realism of the driving simulation system and adjust them, so as to complete the preliminary calibration of the driving simulation system. Recalibration steps: Replace another group of drivers, and calculate the similarity between the real and virtual scene multi-dimensional driving behavior data through comparative analysis; based on the similarity between the two behavior data, complete the recalibration of the driving simulation system; Final calibration step: Replace a group of drivers again for re-verification until the calibration result is less than the critical threshold, completing the final calibration; The similarity analysis and calculation of the two behaviors includes: setting m dimensions for optimizing the realism of the driving simulation system, evaluating the realism of the driving simulation system and optimizing parameters from the m dimensions, selecting corresponding behavioral indicators and physiological response indicators for each dimension; and setting weights for the selected indicators; m is a positive integer; The m dimensions include: a virtual scene rendering dimension, an AR display setting dimension, a cockpit internal operation unit parameter setting dimension, and a sound and body sensory parameter setting dimension; The calculation process of the similarity between two behaviors also includes: Select a real road section, build a digital twin virtual driving scene that is the same as the real road scene, project the virtual driving scene onto the front and side windshields of the driving simulation cabin, and set the driving task; Obtain several behavioral and physiological reaction indicators of the driver when performing the set driving task in the real road scene; obtain several behavioral and physiological reaction indicators of the driver when performing the set driving task in the simulated scene using the driving simulator; Based on the weights and corresponding index values, the similarity of two types of behaviors in the real driving scene and the virtual driving scene under m dimensions is calculated; The critical threshold calculation process includes: According to the similarity between the real driving behavior and the virtual driving behavior in each dimension, the relevant parameters of the corresponding dimension of the driving simulation system are adjusted and optimized; Determine whether the number of parameter optimizations has reached the set number. If so, it means that the preliminary calibration of the driving simulation system has been completed, and the current indicator similarity is recorded as the critical threshold.

2. The driving simulation system reality calibration method according to claim 1, characterized in that: After the step of determining the setting parameters that affect the realism of the driving simulation system and adjusting them, and before completing the preliminary calibration step of the driving simulation system, the method further includes: The same group of drivers will perform the driving simulation task again, and adjust the driving simulation system parameters again based on the collected data. After repeating the above steps many times, the simulated driving behavior characteristics collected in the driving simulation will gradually approach the driving behavior characteristics of the real scene, thereby completing the preliminary calibration of the driving simulation system and determining the critical threshold of authenticity.

3. The driving simulation system reality calibration method according to claim 1, characterized in that: The recalibration of the driving simulation system is completed according to the similarity between the two behavior data; specifically includes: If the result is less than the critical threshold of realism, the final calibration of the driving simulation system is completed; if the result is greater than the critical threshold of realism, the initial calibration steps are repeated, the parameters of the driving simulation system are adjusted, and the recalibration is completed.

4. The driving simulation system reality calibration method according to claim 1, characterized in that: m dimensions for optimizing the realism of the driving simulation system are set, and the realism of the driving simulation system is evaluated and parameters are optimized from these m dimensions.

5. A driving simulation system, characterized in that: Using the driving simulation system reality calibration method according to any one of claims 1 to 4; The driving simulation system comprises: a driving simulation cabin, in which a driving simulator is arranged; The driving simulator is provided with a processor, which is connected to an input module, an output module, a virtual scene projection module, an environmental noise and vibration simulation module, a road driving scene design and operation module, and a vehicle dynamic simulation module; The input module of the driving simulator is also connected to the brake pedal via a first angle sensor, the input module is also connected to the oil port pedal via a second angle sensor, the input module is also connected to the steering wheel via a third angle sensor, and the input module is also connected to the gear position via a displacement sensor; The output module of the driving simulator is also connected to the instrument panel.

6. A device for calibrating the authenticity of a driving simulation system, characterized in that: include: A similarity calculation module is configured to: calculate the index similarity between the real driving scene and the virtual driving scene based on the weights and the corresponding index values, and determine whether the number of parameter optimizations reaches the set number. If not, return to the acquisition module; if yes, it means that the preliminary calibration of the driving simulation system has been completed, and record the current comprehensive index similarity as the critical threshold; A judgment module is configured to: calculate the similarity of comprehensive indicators of a new batch of drivers in real driving scenarios and virtual driving scenarios, and judge whether the similarity of comprehensive indicators of the new batch of drivers is less than a critical threshold. If so, the calibration of the driving simulation system is terminated; if not, the parameters affecting the realism of the driving simulator are adjusted and new drivers are continuously recruited for experiments, and then the relevant parameters of the driving simulator are further adjusted and optimized until the calibration result is less than the critical threshold, and the final calibration is completed; The similarity calculation includes: setting m dimensions for optimizing the realism of the driving simulation system, evaluating the realism of the driving simulation system and optimizing parameters from the m dimensions, selecting corresponding behavioral indicators and physiological response indicators for each dimension; and setting weights for the selected indicators; m is a positive integer; The m dimensions include: a virtual scene rendering dimension, an AR display setting dimension, a cockpit internal operation unit parameter setting dimension, and a sound and body sensory parameter setting dimension; The similarity calculation process also includes: Select a real road section, build a digital twin virtual driving scene that is the same as the real road scene, project the virtual driving scene onto the front and side windshields of the driving simulation cabin, and set the driving task; Obtain several behavioral and physiological reaction indicators of the driver when performing the set driving task in the real road scene; obtain several behavioral and physiological reaction indicators of the driver when performing the set driving task in the simulated scene using the driving simulator; Based on the weights and corresponding index values, the similarity of two types of behaviors in the real driving scene and the virtual driving scene under m dimensions is calculated; The critical threshold calculation process includes: According to the similarity between the real driving behavior and the virtual driving behavior in each dimension, the relevant parameters of the corresponding dimension of the driving simulation system are adjusted and optimized; Determine whether the number of parameter optimizations has reached the set number. If so, it means that the preliminary calibration of the driving simulation system has been completed, and the current indicator similarity is recorded as the critical threshold.

7. A driving simulation system reality calibration device as claimed in claim 6, characterized in that: The device further comprises: An acquisition module is configured to: acquire a number of behavioral and physiological reaction indicators of a driver when performing a set driving task in a real road scene; acquire a number of behavioral and physiological reaction indicators of a driver when performing a set driving task in a simulated scene using a driving simulator; The indicator selection module is configured to: set m evaluation optimization dimensions, and for each dimension, select a number of corresponding behavior and physiological response indicators; under the same dimension, the indicators selected in the real road scene are the same as those selected in the simulation scene; The weight setting module is configured to set weights for a number of selected behavior and physiological response indicators.

Images