In-vehicle changeable environment intelligent simulation system for driving training

Abstract

The invention discloses an in-vehicle changeable environment intelligent simulation system for driving training, and relates to the technical field of driving simulation, and the system comprises the steps: obtaining a plurality of driving sample scenes, and recognizing an out-vehicle environment sample data set and an in-vehicle environment sample data set; constructing a mapping data set inside and outside the vehicle; training based on the in-vehicle and out-vehicle mapping data set, and outputting an in-vehicle environment mapping model; acquiring a first outside-vehicle simulation scene of the current training user for driving training; inputting an in-vehicle environment mapping model to obtain a first in-vehicle simulation parameter; and a simulation device is connected, and the simulation device simulates the in-vehicle environment according to the first in-vehicle simulation parameter. The technical problems that in the prior art, driving training environment simulation is single, the sense of reality is lacked, and a comprehensive, real and changeable driving training environment is difficult to provide are solved, and the technical effects that the changeable environment in the vehicle is accurately simulated according to driving training requirements, and the authenticity and diversity of driving training are improved are achieved.

Description

Technical Field

[0001] The present application relates to the technical field of driving simulation, and in particular to an intelligent simulation system for a vehicle's variable environment for driving training. Background Art

[0002] With the continuous increase in the number of cars, road traffic safety issues have become increasingly prominent, and improving drivers' driving skills and ability to cope with complex environments has become the key. Driving training is crucial for the growth of novice drivers and the improvement of skills of experienced drivers. Traditional driving training methods mainly rely on actual road driving and simple field training, which have many limitations. In actual road driving training, although a real traffic environment can be provided, it is limited by time, location, weather and other conditions, making it difficult for drivers to experience various road conditions and environmental changes in a short period of time. For example, in certain seasons or regions, it is difficult for drivers to frequently encounter severe weather scenes such as heavy rain, heavy snow, and dense fog for training; at the same time, actual road training needs to consider traffic flow and safety factors, and cannot arbitrarily simulate special road conditions, such as road construction and sudden accident scenes.

[0003] The existing technology has the technical problem that the driving training environment simulation is single and lacks realism, and it is difficult to provide a comprehensive, real and changeable driving training environment. Summary of the invention

[0004] The present application provides an intelligent simulation system for a variable in-vehicle environment for driving training, which is used to solve the technical problems in the prior art that the driving training environment simulation is single, lacks realism, and is difficult to provide a comprehensive, realistic and variable driving training environment.

[0005] The present application provides an intelligent simulation system for a vehicle with a variable environment for driving training, including:

[0006] A sample data set identification module, the sample data set identification module is used to obtain multiple driving sample scenes, and identify the external vehicle environment sample data sets and the internal vehicle environment sample data sets corresponding to the multiple driving sample scenes; an internal and external vehicle mapping data set construction module, the internal and external vehicle mapping data set construction module is used to perform time-series synchronization processing according to the external vehicle environment sample data set and the internal vehicle environment sample data set, and construct an internal and external vehicle mapping data set; an internal vehicle environment mapping model output module, the internal vehicle environment mapping model output module is trained based on the internal and external vehicle mapping data sets, and outputs an internal vehicle environment mapping model; a first external vehicle simulation scene acquisition module, the first external vehicle simulation scene acquisition module is used to obtain a first external vehicle simulation scene for driving training by a current training user; a first internal vehicle simulation parameter acquisition module, the first internal vehicle simulation parameter acquisition module is used to input the first external vehicle simulation scene into the internal vehicle environment mapping model, and obtain a first internal vehicle simulation parameter corresponding to the first external vehicle simulation scene; an internal vehicle environment simulation module, the internal vehicle environment simulation module is used to connect a simulation device, and the simulation device simulates the internal vehicle environment according to the first internal vehicle simulation parameter.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0008] The sample data set identification module is used to obtain multiple driving sample scenes and identify the external vehicle environment sample data sets and the internal vehicle environment sample data sets corresponding to the multiple driving sample scenes; the internal and external vehicle mapping data set construction module performs time series synchronization processing and constructs the internal and external vehicle mapping data sets; the internal vehicle environment mapping model output module performs training based on the internal and external vehicle mapping data sets and outputs the internal vehicle environment mapping model; the first external vehicle simulation scene acquisition module obtains the first external vehicle simulation scene used by the current training user for driving training; the first internal vehicle simulation parameter acquisition module is used to input the first external vehicle simulation scene into the internal vehicle environment mapping model and obtain the first internal vehicle simulation parameter corresponding to the first external vehicle simulation scene; the internal vehicle environment simulation module is used to connect the simulation device, and the simulation device simulates the internal vehicle environment according to the first internal vehicle simulation parameter. The technical effect of accurately simulating the changing internal vehicle environment according to the driving training requirements and improving the authenticity and diversity of driving training is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] 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 the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0010] Figure 1A structural schematic diagram of an in-vehicle multi-changeable environment intelligent simulation system for driving training is provided for an embodiment of the present application;

[0011] Figure 2 A schematic flow chart of outputting an in-vehicle environment mapping model of an in-vehicle variable environment intelligent simulation system for driving training is provided for an embodiment of the present application.

[0012] Explanation of the reference numerals: sample data set identification module 10 , in-vehicle and out-vehicle mapping data set construction module 20 , in-vehicle environment mapping model output module 30 , first in-vehicle simulation scene acquisition module 40 , first in-vehicle simulation parameter acquisition module 50 , in-vehicle environment simulation module 60 . DETAILED DESCRIPTION

[0013] The present application provides an intelligent simulation system for a variable in-vehicle environment for driving training, aiming to solve the technical problems in the prior art that the driving training environment simulation is single, lacks realism, and is difficult to provide a comprehensive, realistic and variable driving training environment.

[0014] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0015] Examples, such as Figure 1 As shown, the present application provides an in-vehicle variable environment intelligent simulation system for driving training, the system comprising:

[0016] The sample data set identification module 10 is used to obtain a plurality of driving sample scenes and identify the vehicle exterior environment sample data sets and vehicle interior environment sample data sets corresponding to the plurality of driving sample scenes.

[0017] Specifically, the sample data set identification module 10 plays the role of data collection and preliminary processing in the driving training in-vehicle variable environment intelligent simulation system. First, multiple driving sample scenes are obtained through various channels, including but not limited to field driving data collection, scene libraries in driving simulation software, and typical scene cases provided by professional driving training institutions. The driving sample scenes obtained are rich and diverse, covering various road conditions, such as urban congested roads, highways, mountain bends, etc.; different weather conditions, such as sunny days, rainy days, snowy days, foggy days, etc.; and various traffic conditions, such as busy traffic during peak traffic hours, driving at night with low visibility, etc. For each driving sample scene, various sensors installed at different positions of the vehicle are used to identify the corresponding sample data set of the vehicle's external environment. The collection of the sample data set of the vehicle's external environment involves many aspects, such as installing a high-precision meteorological sensor on the top of the vehicle to obtain real-time temperature, humidity, air pressure and weather conditions (such as raindrop size, snowfall intensity, fog concentration, etc.); multiple cameras and radars are equipped around the vehicle body to capture information such as traffic flow, vehicle type, road signs, and obstacle locations; at the same time, the vehicle's geographical location and road type (such as urban main roads, rural roads, etc.) are determined through the vehicle's onboard GPS and map data. At the same time, it is also responsible for identifying the sample data set of the vehicle's internal environment. A variety of sensors are arranged in the car, such as pressure sensors and position sensors installed on the seats to obtain the driver and passengers' sitting posture and seat usage; light sensors are set near the dashboard to detect the light intensity in the car; temperature and humidity sensors and noise sensors are installed in the car space to monitor the temperature, humidity and noise level in the car respectively. Through the collaborative work of these sensors, the sample data set recognition module 10 can comprehensively and accurately collect the sample data set of the external environment and the sample data set of the internal environment corresponding to each driving sample scene, providing a solid data foundation for the subsequent construction of the internal and external mapping data set and the training of the internal environment mapping model.

[0018] The vehicle interior and exterior mapping data set construction module 20 is used to perform time-series synchronization processing according to the vehicle exterior environment sample data set and the vehicle interior environment sample data set to construct the vehicle interior and exterior mapping data set.

[0019] Specifically, the in-vehicle and out-vehicle mapping data set construction module 20 undertakes the key data integration task in the entire driving training simulation system. The module first obtains the out-vehicle environment sample data set and the in-vehicle environment sample data set collected by the sample data set recognition module. These data sets contain rich environmental information, but due to the time difference in the collection process, time series synchronization processing is required. High-precision timestamp marking technology is used to mark the precise collection time for each data in the out-vehicle environment sample data set (such as out-vehicle temperature, humidity, traffic conditions, etc.) and the corresponding data in the in-vehicle environment sample data set (such as in-vehicle temperature, noise, seat status, etc.). Then, through the data alignment algorithm, the out-vehicle environment data and the in-vehicle environment data at the same time are matched one by one according to the timestamp to form a series of ordered data pairs. These data can clearly reflect the relationship between the out-vehicle environmental factors and the in-vehicle environmental status at a specific time point. For example, when the out-vehicle temperature suddenly rises at a certain moment, the corresponding in-vehicle temperature changes and other related environmental factors (such as the operating status of the in-vehicle air conditioner, whether the driver opens the window, etc.) After such time-series synchronization processing, a vehicle interior and exterior mapping dataset was successfully constructed. This dataset provides an accurate and time-series logical data basis for the subsequent training of the vehicle interior environment mapping model, enabling the model to learn the dynamic relationship between changes in the vehicle interior and exterior environments, thereby more accurately simulating and predicting the changing trends of the vehicle interior environment under different vehicle exterior environment conditions.

[0020] The in-vehicle environment mapping model output module 30 is trained based on the in-vehicle and out-vehicle environment mapping data set to output an in-vehicle environment mapping model.

[0021] Specifically, the in-vehicle environment mapping model output module 30 undertakes the important task of building a high-precision in-vehicle environment mapping model in the driving training in-vehicle variable environment intelligent simulation system, and optimizes the training process with the help of an adversarial network. First, the module obtains a trained adversarial network, which is composed of a generator and a discriminator that cooperate with each other, and establishes a connection with the input end of the in-vehicle environment mapping model to form an organic whole. At the beginning of training, the in-vehicle and out-vehicle mapping data set is input into the generator in the adversarial network. The generator uses its powerful neural network architecture and advanced learning algorithm to deeply analyze the complex relationships and features contained in the in-vehicle and out-vehicle mapping data set. It continuously adjusts its own parameters and strives to learn how to generate realistic in-vehicle and out-vehicle environment data pairs, thereby outputting in-vehicle and out-vehicle mapping generation data sets. These generated data can simulate, to a certain extent, the changes in the in-vehicle environment that may occur in real scenes but are not directly presented in the original data set and their corresponding in-vehicle environment states. Subsequently, the generated in-vehicle and out-vehicle mapping generation data set is used to expand the original in-vehicle and out-vehicle mapping data set, and the virtual generated data is combined with the actual collected real data to construct an expanded data set with a larger data volume, wider data distribution and more diversity. This expansion operation greatly enriches the sample space of training data, allowing the in-vehicle environment mapping model to learn more diverse in-vehicle and out-vehicle environment mapping relationships. Based on the expanded data set and the identification information of the identification learning accuracy, the in-vehicle environment mapping model is trained. During the training process, the in-vehicle environment mapping model takes the expanded data set as input and continuously adjusts the internal parameters to minimize the error between the model prediction result and the actual expected output (indicated by the identification information). At the same time, the discriminator in the adversarial network continuously discriminates the data generated by the generator and the data in the original data set, and its discrimination accuracy is transmitted back to the generator as a key feedback signal. Based on the feedback of the discrimination accuracy, the generator continuously optimizes its own generation ability and strives to generate data that can confuse the discriminator more, that is, data that is closer to the real data distribution. Through this repeated adversarial interactive training between the generator and the discriminator, and the in-vehicle environment mapping model's in-depth learning of the expanded data set, a vehicle environment mapping model with excellent performance and high accuracy is finally output. The model can accurately calculate and output highly matching in-vehicle simulation parameters based on any input out-of-vehicle simulation scene information, providing solid core model support for the driving training simulation system, ensuring the authenticity and reliability of the simulated in-vehicle environment, and significantly improving the effectiveness and quality of driving training.

[0022] The first vehicle exterior simulation scene acquisition module 40 is used to acquire a first vehicle exterior simulation scene for driving training of a current training user.

[0023] Specifically, the first vehicle exterior simulation scene acquisition module 40 is responsible for accurately acquiring the first vehicle exterior simulation scene for driving training for the training user in the driving training in-vehicle variable environment intelligent simulation system. First, interact with the system's training course setting module or the user-defined training demand interface to obtain the current training user's training goals, skill levels, and personalized training requirements. If the user is a novice driver, the training goal may focus on mastering basic driving skills, such as vehicle control under normal road conditions and weather conditions; while for experienced drivers, they are more inclined to respond to training under complex road conditions and bad weather scenarios. Based on this information, select appropriate scenes from the rich scene library built into the system. The scene library contains various types of vehicle exterior simulation scenes, such as busy traffic intersection scenes in the city center, with complex traffic light rules, dense pedestrian and vehicle flow; highway scenes, involving different speed limits, lane changes, and vehicle spacing requirements; and various special weather scenes, such as obstructed vision and road waterlogging in heavy rain, slippery roads and low visibility in heavy snow, etc. The module will select the scene with the highest matching degree as the first vehicle exterior simulation scene according to the user's training needs. For example, if the user needs to improve their driving skills in rainy weather, the module will give priority to a rainy day scenario that includes different rainfall intensities, road slipperiness, and traffic conditions. In addition, the module also has the ability to dynamically adjust the simulation scenario based on real-time user feedback and training progress. If the user proposes to modify the difficulty or type of the current scenario during training, the module will immediately re-evaluate the user's needs and select a more appropriate first off-vehicle simulation scenario from the scenario library to ensure that driving training always meets the user's expectations and training goals, providing users with the most targeted and effective driving training experience.

[0024] The first in-vehicle simulation parameter acquisition module 50 is used to input the first external vehicle simulation scene into the in-vehicle environment mapping model to acquire the first in-vehicle simulation parameter corresponding to the first external vehicle simulation scene.

[0025] Specifically, the first in-vehicle simulation parameter acquisition module 50 plays a key role in parameter conversion and generation in the driving training in-vehicle variable environment intelligent simulation system. First, it receives the first in-vehicle simulation scene information determined by the first in-vehicle simulation scene acquisition module. The scene information covers a wealth of in-vehicle environment data, such as weather conditions (sunny, rainy, snowy, foggy, etc.), temperature, humidity, light intensity, traffic flow, road type (urban roads, highways, rural roads, etc.) and surrounding terrain features. Subsequently, these detailed first in-vehicle simulation scene data are input into the pre-trained in-vehicle environment mapping model. The in-vehicle environment mapping model has established a mapping relationship model between complex in-vehicle environment factors and in-vehicle environment parameters based on the learning of a large number of in-vehicle and out-vehicle mapping data sets. After receiving the in-vehicle simulation scene input, the model quickly processes and analyzes these data through its internal neural network structure or other algorithm mechanisms. The model calculates the first in-vehicle simulation parameters corresponding to the current in-vehicle simulation scene based on the learned mapping rules. These parameters include, but are not limited to, the specific value to which the in-car temperature should be adjusted to simulate the effect of the outside temperature on the car; the humidity setting value, which reflects the change trend of the in-car humidity under the change of the outside humidity; the noise level parameter, which simulates the effect of the outside traffic noise, wind and rain noise, etc. entering the car; the light intensity parameter, which creates a lighting atmosphere in the car that matches the weather and time outside the car; the seat position adjustment parameter, which provides appropriate driving posture suggestions according to the driving scene requirements (such as high-speed driving or complex road conditions); and the simulated display status of the instrument panel display information (such as the speedometer, tachometer, warning lights, etc.), which provides the driver with realistic driving visual feedback. Through this process, the first in-car simulation parameter acquisition module 50 successfully obtains the first in-car simulation parameters that are highly matched with the first outside car simulation scene, providing accurate data basis for the subsequent in-car environment simulation module to create a real in-car environment.

[0026] The in-vehicle environment simulation module 60 is used to connect to a simulation device, and the simulation device simulates the in-vehicle environment according to the first in-vehicle simulation parameters.

[0027] Specifically, the in-car environment simulation module 60 is a key execution part for realizing the realistic reproduction of the in-car environment in the driving training in-car variable environment intelligent simulation system. The module first establishes a close connection with various professional simulation devices, including but not limited to high-precision temperature adjustment system, humidity control system, advanced noise simulation generator, adjustable lighting equipment, multi-functional seat adjustment device and highly simulated instrument panel display system. After receiving the first in-car simulation parameter from the first in-car simulation parameter acquisition module, the module quickly distributes these parameters to the corresponding simulation device. The temperature adjustment system accurately controls the heating or cooling device in the car according to the set temperature parameters, so that the temperature in the car quickly and stably reaches the simulation requirements. For example, when simulating a cold weather scene, the temperature in the car is reduced to a corresponding low temperature state, so that the driver can actually feel the impact of the cold environment on the driving operation. The humidity control system adjusts the humidity level in the car according to the humidity parameter to create a dry or humid in-car environment atmosphere, just like driving under different actual climatic conditions. The noise simulation generator generates realistic traffic noise, wind and rain, engine roar, etc. outside the car according to the noise level parameter, so that the driver can immerse himself in the real driving scene in hearing. The adjustable lighting device simulates the lighting effects in the car at different times (such as early morning, noon, evening) and weather (sunny, cloudy, rainy) according to the light intensity parameters, including direct sunlight, shadows, and different degrees of dim light, which affect the driver's visual perception and operation judgment. The multifunctional seat adjustment device automatically adjusts the front and rear position, backrest angle, height, etc. of the seat according to the seat position adjustment parameters, providing the driver with a comfortable and reasonable sitting posture that meets the current driving scene. The highly simulated instrument panel display system accurately displays information such as vehicle speed, speed, fuel volume, various warning lights, etc. according to the instrument panel display information parameters, matches the simulated scene outside the car, and provides visual feedback to the driver, so that he can obtain vehicle operation status information like in real driving. Through the collaborative work of these simulation devices, the in-car environment simulation module 60 successfully creates a highly realistic in-car environment according to the first in-car simulation parameters, greatly improving the authenticity and effectiveness of driving training, allowing the driver to fully experience and respond to various changing driving scenes in a safe simulation environment.

[0028] In one possible implementation, Figure 2 As shown, the in-vehicle environment mapping model output module 30 includes:

[0029] An adversarial network acquisition unit, wherein the adversarial network acquisition unit is used to acquire a trained adversarial network, wherein the adversarial network is connected to an input end of the in-vehicle environment mapping model.

[0030] A data learning unit is used to perform data learning on the vehicle interior and exterior mapping data set according to the adversarial network, and output a vehicle interior and exterior mapping generation data set.

[0031] The in-vehicle environment mapping model output unit is used to expand the in-vehicle and out-vehicle mapping data set according to the in-vehicle and out-vehicle mapping data set, train according to the expanded data set and identification information of the identification learning accuracy, and output the in-vehicle environment mapping model.

[0032] Specifically, in the intelligent simulation system of the in-vehicle variable environment for driving training, in order to improve the accuracy and generalization ability of the in-vehicle environment mapping model, the adversarial network technology is used for optimization. First, the trained adversarial network is obtained. The adversarial network consists of two parts, the generator and the discriminator, and is connected to the input end of the in-vehicle environment mapping model to form a collaborative architecture.

[0033] In the data learning phase, the vehicle interior and exterior mapping dataset is input into the generator in the adversarial network. The generator uses its internal complex neural network structure and learning algorithm to deeply mine and learn the data features in the vehicle interior and exterior mapping dataset. By continuously adjusting its own parameters, the generator attempts to generate new data that is similar to the original dataset but has certain changes, thereby outputting the vehicle interior and exterior mapping generated dataset. This generated dataset contains virtual vehicle interior and exterior environment data pairs created by the generator based on the learned patterns and rules. These data pairs can simulate, to a certain extent, scene combinations that may appear in real situations but are not explicitly included in the original dataset.

[0034] The in-vehicle and out-vehicle mapping dataset is used to expand the in-vehicle and out-vehicle mapping dataset, and the virtual data in the generated dataset is combined with the real data in the original dataset to form a larger and more diverse expanded dataset. The expanded dataset not only contains the actual collected in-vehicle and out-vehicle environment data relationships, but also covers the potential relationships inferred by the generator, thereby enriching the data sample space. Finally, training is carried out based on the expanded dataset and the identification information of the identification learning accuracy. The in-vehicle environment mapping model takes the expanded dataset as input, and by continuously adjusting the parameters inside the model, the error between the output result of the model and the expected output represented by the identification information is gradually reduced. In this process, the discriminator in the adversarial network also plays an important role. The discriminator is responsible for distinguishing and judging the data generated by the generator and the data in the original dataset. Its discrimination accuracy is transmitted to the generator as a feedback signal, prompting the generator to continuously optimize its generation ability to generate high-quality data that is more difficult to be distinguished by the discriminator. Through this adversarial training between the generator and the discriminator, and the learning of the in-vehicle environment mapping model on the expanded dataset, a better-performing and more accurate in-vehicle environment mapping model is finally output. This optimized model can more accurately predict the in-vehicle environmental parameters based on the simulated scenes outside the vehicle, providing a more realistic and practical in-vehicle environment simulation effect for driving training.

[0035] In a possible implementation manner, the adversarial network acquisition unit further includes:

[0036] A generated data set acquisition unit is used to input the vehicle interior and exterior mapping data set into the generator, perform data learning according to the generator, and acquire a generated data set after multiple learnings.

[0037] The discrimination accuracy acquisition unit is used to input the generated data set after multiple learning into the discriminator for discrimination and obtain the discrimination accuracy.

[0038] A feedback optimization unit, wherein the feedback optimization unit is used to perform feedback optimization on the generator according to the discrimination accuracy if the discrimination accuracy is less than a preset threshold, until the discrimination accuracy of the discriminator is greater than or equal to the preset threshold, thereby obtaining a trained adversarial network.

[0039] Specifically, first, the vehicle interior and exterior mapping dataset is input into the generator, and the generator begins to learn data based on its complex neural network structure and learning algorithm. It attempts to extract key data features and potential mapping relationship patterns from the vehicle interior and exterior mapping dataset. During the learning process, the generator continuously adjusts its internal parameters to generate new, simulated vehicle interior and exterior environment data pairs based on the learned patterns. These data pairs constitute the generated dataset after multiple learning. The goal of the generator is to generate data with a distribution similar to that of the real vehicle interior and exterior mapping dataset in order to deceive the discriminator.

[0040] In the adversarial network training process of the intelligent simulation system for driving training in-vehicle variable environment, it is a key step to input the generated data set after multiple learning into the discriminator for discrimination to obtain the discrimination accuracy. After receiving the generated data set, the discriminator will start its complex discrimination mechanism. It first performs a detailed feature analysis on each in-vehicle and out-vehicle environment data in the generated data set. These features include but are not limited to the temperature change trend, humidity fluctuation range, light intensity distribution pattern, traffic flow change law in the out-vehicle environment data, and the temperature adjustment amplitude, humidity response characteristics, noise simulation level, and light simulation effect in the corresponding in-vehicle environment data. The discriminator uses pre-set discrimination rules and algorithms, which are formed based on the learning and summary of a large amount of real in-vehicle and out-vehicle mapping data. By comparing the data features in the generated data set with these rules, the discriminator calculates the similarity of each data pair with the real data distribution. For example, for the response change of the in-vehicle temperature when the outside temperature rises, there may be a certain range and law of change in the real data. The discriminator will check whether the corresponding change in the generated data conforms to this law. After discriminating the entire generated data set one by one, the discriminator calculates the discrimination accuracy based on the discrimination results of all data pairs. The discrimination accuracy is usually calculated by dividing the number of correctly discriminated data pairs by the total number of data pairs to obtain a ratio value. This ratio value intuitively reflects the discriminator's ability to discriminate the generated data set. If the discrimination accuracy is high, it means that the data generated by the generator is similar to the characteristics and distribution of the real data, and the performance of the generator is good; conversely, if the discrimination accuracy is low, it means that the data generated by the generator has many discrepancies with the real data and needs further optimization. By obtaining the discrimination accuracy, the current generation effect of the generator can be accurately evaluated, thereby providing a key basis for subsequent feedback optimization, ensuring that the adversarial network can continuously improve its ability to generate high-quality simulated data, and thus provide more reliable data support for the training of the in-vehicle environment mapping model.

[0041] If the discrimination accuracy is less than the preset threshold, this indicates that the data generated by the generator is still quite different from the real data and is easy to be identified by the discriminator. At this point, according to the problem reflected by the discrimination accuracy, the relevant information is passed to the generator as feedback. Based on these feedbacks, the generator adjusts its internal parameters in a targeted manner to improve the way and quality of data generation. Then, the updated generated data set is input into the discriminator for discrimination again, and the above process is repeated to continuously optimize the generator's generation ability. In this way, the generator continues to improve, and the discriminator continues to discriminate and feedback until the discrimination accuracy of the discriminator is greater than or equal to the preset threshold. At this point, it shows that the generator has been able to generate sufficiently realistic data, which is highly similar to the distribution of real data and difficult to be distinguished by the discriminator. So far, the trained adversarial network has been successfully obtained, and its generator can effectively generate high-quality simulated data, which provides a solid foundation for the subsequent expansion of the in-vehicle and out-of-vehicle mapping data set and the optimization training of the in-vehicle environment mapping model, thereby improving the accuracy and reliability of the entire driving training simulation system for in-vehicle environment simulation.

[0042] In a possible implementation, the sample data set identification module 10 includes:

[0043] A scene type identification unit is used to identify scene types of a plurality of driving sample scenes.

[0044] A feature parameter label set construction unit is used to identify scene feature parameters according to the scene type and construct a feature parameter label set corresponding to each driving sample scene.

[0045] An external vehicle environment sample data set acquisition unit is used to acquire external vehicle environment sample data sets corresponding to the multiple driving sample scenes according to the feature parameter label set.

[0046] Specifically, in the intelligent simulation system of the variable environment in the driving training vehicle, identifying the scene types of multiple driving sample scenes is the first step to build a high-quality simulation environment. This goal is achieved through a variety of data collection methods and analysis methods. First, the on-board camera is used to capture the visual information of the driving sample scene, including road conditions, sky conditions, traffic signs, and the conditions of surrounding vehicles and pedestrians. For example, the sky is blue, the sun is shining, and the road is dry in the picture captured by the camera, which initially suggests a sunny scene; if there is rain, water on the road, and the wipers are in working condition in the picture, it tends to be judged as a rainy scene. At the same time, the meteorological sensors equipped on the vehicle also play an important role. The environmental data collected by the temperature sensor, humidity sensor, and air pressure sensor provide a key basis for the judgment of the scene type. For example, when the temperature is high, the humidity is high, and the air pressure is relatively low, combined with the possible signs of dark clouds in the visual information, it can assist in judging the thunderstorm weather scene. In addition, the GPS positioning system, combined with map data, can determine the geographical location and road type of the vehicle, further assisting in the recognition of the scene type. If the vehicle is on a mountain road and the surrounding environment shows many bends and large terrain undulations, combined with the vegetation and terrain features in the camera image, it helps to judge it as a mountain driving scene. Through the comprehensive analysis and processing of these multi-source data, the type of each driving sample scene can be accurately identified, laying the foundation for the subsequent feature parameter identification and sample data set collection for different scene types, thereby building a more realistic and comprehensive driving training simulation environment.

[0047] In the intelligent simulation system of the variable environment in the driving training car, identifying the scene feature parameters according to the identified scene type and constructing the corresponding feature parameter label set is the key link to accurately simulate the in-car environment. For sunny scenes, illumination is the most critical feature parameter. The intensity, angle and distribution of sunlight are accurately measured by the illumination sensor. For example, at noon, the sunlight is directly on the ground, and the light intensity may reach the maximum value. At this time, the sun angle is close to vertical to the ground, and the light distribution is relatively uniform. These parameters are crucial for simulating the lighting environment in the car. At the same time, the color and transparency of the sky on sunny days will also affect the light atmosphere in the car. The camera collects images and analyzes their color characteristics and clarity to obtain relevant parameters. These parameters related to sunny day illumination are integrated to construct feature parameter labels such as "sunny day-high light intensity-specific sun angle-uniform light distribution-blue sky", forming a feature parameter label set corresponding to the sunny scene. In rainy scenes, humidity and wind speed become the focus of feature parameters. The humidity sensor monitors the humidity of the outside environment in real time and accurately records the range of humidity changes under different rainfall intensities, because high humidity environments not only affect the visibility of the car windows, but also affect the humidity control system in the car. The anemometer measures the wind speed and direction. Different wind speeds will change the movement trajectory and impact force of raindrops, which in turn affects the vehicle's driving stability and the driver's line of sight. In addition, raindrop size and density are also important parameters, which are obtained through special raindrop sensors or camera image analysis technology. These rainy day related parameters are summarized to construct feature parameter labels such as "rainy day-high humidity range-different wind speeds-specific raindrop size and density" to form a feature parameter label set for rainy day scenes. For night scenes, brightness and light distribution are the core feature parameters. The light sensor measures the ambient light brightness, including the comprehensive brightness level of light sources such as street lights, other vehicle headlights, and moonlight. The camera is used to analyze the light distribution, such as the spacing and lighting range of street lights, the illumination angle of vehicle headlights, and the distribution of high and low beams. At the same time, the characteristics of light changes in different time periods at night are taken into consideration, such as the gradual dimming of light in the evening, the dim ambient light and more uniform light distribution in the middle of the night, etc. These night light-related parameters are sorted out to construct characteristic parameter labels such as "night-low brightness range-specific light distribution-light changes in different time periods" to form a characteristic parameter label set for night scenes. Through such detailed characteristic parameter identification and label set construction for different scene types, precise guidance is provided for the subsequent accurate collection of external environment sample data sets, ensuring that the simulation system can truly restore the environmental characteristics of various driving scenarios.

[0048] According to the feature parameter label set, the vehicle exterior environment sample data sets corresponding to multiple driving sample scenarios are collected. For the feature parameters represented by each label, the corresponding sensors are used for data collection. For example, for light intensity, a light sensor is used; for humidity, a humidity sensor is used; for wind speed, an anemometer is used, etc. By deploying these sensors in different scenarios, data of various environmental factors can be accurately collected, thereby constructing a comprehensive and accurate vehicle exterior environment sample data set, providing a solid data foundation for the subsequent construction of the vehicle interior and exterior mapping data set and the training of the vehicle interior environment mapping model, ensuring that the simulation system can truly restore the vehicle exterior environment conditions under different driving scenarios, and improving the effectiveness and authenticity of driving training.

[0049] In a possible implementation, the in-vehicle environment mapping model output module 30 further includes:

[0050] A scene switching mode determination unit is used to determine whether the scene currently used by the training user for driving training is a scene switching mode.

[0051] A scene switching node acquisition unit, wherein the scene switching node acquisition unit is used to acquire a scene switching node if the current driving training scene of the training user is in scene switching mode, wherein the scene switching node is a node for switching the first external vehicle simulation scene to the second external vehicle simulation scene.

[0052] The second in-vehicle simulation parameter acquisition unit is used to acquire second in-vehicle simulation parameters corresponding to the second out-of-vehicle simulation scene according to the in-vehicle environment mapping model.

[0053] A switching simulation unit is used for the simulation device to simulate the in-vehicle environment according to the first in-vehicle simulation parameter and then perform switching simulation on the in-vehicle environment with the second in-vehicle simulation parameter.

[0054] Specifically, in the intelligent simulation system of the changing environment in the driving training car, the scene switching simulation function provides drivers with a richer and more challenging training experience. First, it is necessary to determine whether the scene currently used by the training user for driving training is in the scene switching mode. This judgment is based on the system's pre-set training course logic and the user's operation instructions. For example, if the user selects a comprehensive training course that includes scene switching, or manually triggers a scene switching instruction during training, it is determined that the current scene switching mode is in progress.

[0055] When the scene switching mode is determined, the scene switching node will be obtained immediately. The scene switching node refers to the precise moment or position point when the first simulated scene outside the vehicle switches to the second simulated scene outside the vehicle. The determination of this node is crucial to ensure the continuity and authenticity of the scene switching. The scene switching node is accurately located by monitoring the time progress of the simulated scene and tracking the position of the vehicle in the virtual environment. For example, when a city road driving scene switches to a highway driving scene, the scene switching node is set to the point where the vehicle drives to the junction of the city road and the highway entrance, or when the simulation time reaches the predetermined switching moment.

[0056] Once the scene switching node is obtained, the second in-car simulation parameters corresponding to the second exterior simulation scene are quickly calculated and obtained based on the trained interior environment mapping model. The interior environment mapping model is trained based on a large number of interior and exterior mapping data sets, and can accurately predict the adjustment requirements of interior environment parameters corresponding to different exterior environment changes. For the second exterior simulation scene, considering its unique environmental factors, such as wind noise changes caused by higher speeds on highways, different lighting conditions, and more stable temperature environments, the corresponding interior temperature, humidity, noise, lighting and other parameters are calculated.

[0057] In terms of simulation equipment, the in-car environment is first simulated according to the first in-car simulation parameters, so that the driver can experience the real driving environment in the initial scenario. When the scene switching node is reached, the simulation equipment switches quickly to simulate the switching of the in-car environment with the second in-car simulation parameters. For example, the in-car temperature adjustment system adjusts from the appropriate temperature in the urban road scene to a slightly lower temperature in the highway scene to simulate the ventilation effect when driving at high speed; the noise simulation equipment increases the intensity of wind noise and tire noise to create a noisy environment on the highway; the lighting system adjusts the brightness and angle of the light according to the time and weather conditions of the new scene. Through such a scene switching simulation process, the driver's adaptability and coping skills when switching between different driving scenarios can be effectively improved, providing them with a training environment that is closer to actual driving conditions.

[0058] In a possible implementation, the switching simulation unit further includes:

[0059] A parameter difference index output unit is used to perform parameter difference analysis on the first in-vehicle simulation parameter and the second in-vehicle simulation parameter, and output a parameter difference index.

[0060] The switching transition module activation unit is used to activate the switching transition module if the parameter difference index is greater than a preset difference index.

[0061] A transition switching simulation unit is used to perform a transition switching simulation on the in-vehicle environment according to the switching transition module.

[0062] Specifically, a detailed parameter difference analysis is carried out on the first in-car simulation parameter and the second in-car simulation parameter. The first in-car simulation parameter corresponds to the in-car environment settings under the initial driving scenario (such as driving on a sunny city road), including in-car temperature, humidity, noise level, light intensity, etc. The second in-car simulation parameter is the ideal in-car environment parameter corresponding to the switched scenario (such as driving on a rainy night highway). During the analysis process, the differences of each parameter are compared one by one. For example, in terms of temperature, the in-car temperature may be more comfortable when driving on a sunny city road, while the in-car temperature needs to be appropriately lowered when driving on a rainy night highway due to the high speed and low temperature outside the car; in terms of humidity, the humidity of the environment on a rainy night is extremely high, which is significantly different from the humidity on a sunny city road; the noise level will increase significantly in the rainy night highway scenario due to the sound of wind and rain and the noise generated by high-speed driving; the light intensity will also be significantly weakened due to the dual effects of night and rainy days. By accurately calculating the differences in these parameters in terms of numerical values, change trends, and the overall degree of impact on the in-car environment, a comprehensive parameter difference index is output.

[0063] The parameter difference index is compared with the preset difference index, which is a threshold set based on a large amount of actual driving scene switching data and the human body's sensitivity to environmental changes. If the calculated parameter difference index is greater than the preset difference index, it means that the scene switching brings about a large change in the in-car environment. Direct switching may give the driver a sense of abruptness, affecting the authenticity and effect of the simulation training. At this time, the system activates the switching transition module.

[0064] The switching transition module simulates the transition switching of the in-car environment according to the parameter difference index and the preset transition strategy. Taking the rainy night scene as an example, assuming that the in-car temperature is switched from a sunny city road, it will not drop to the low temperature required for the rainy night highway instantly, but will gradually decrease at a certain rate, simulating the gradual change process of the in-car temperature affected by the external environment; the humidity will gradually increase from a low level, and considering factors such as window fogging, the humidity in the car will be gradually adjusted through the simulation equipment and the defog effect will be simulated; the noise level will gradually increase from the relatively quiet environment of the city road to the noisy environment of the rainy night highway, simulating the gradual increase of the wind, rain and the friction between the tires and the slippery road surface; the light intensity will also gradually dim from the bright state of the sunny day, simulating the effect of the gradually dimming light at night and rainy days. Through such a transition switching simulation, the in-car environment can naturally transition from the initial state to the scene state after the switch, allowing the driver to more realistically feel the environmental changes brought about by the scene switch during the simulated driving process, improve their ability to cope with complex and changeable driving environments, and make driving training closer to the actual situation.

[0065] In a possible implementation, the transition switching simulation unit further includes:

[0066] A preset transition step length configuration unit, wherein the preset transition step length configuration unit is used to configure a preset transition step length.

[0067] An in-vehicle simulation parameter acquisition unit, wherein the in-vehicle simulation parameter acquisition unit is used to input the preset transition step length into the switching transition module after activating the activation switching transition module, and acquire transition simulation parameters of the first in-vehicle simulation parameter and the second in-vehicle simulation parameter.

[0068] The in-vehicle environment transition switching unit is used to perform a transition switching simulation on the in-vehicle environment using the transition simulation parameters.

[0069] Specifically, in the intelligent simulation system of the variable environment in the driving training car, in order to achieve a smooth transition of the in-car environment when switching scenes, the gradual change process of the environmental parameters is accurately controlled by configuring the preset transition step. First, the preset transition step is configured according to the characteristics of the driving scene and the human body's perception of environmental changes. The preset transition step is a series of values ​​used to control the gradual change of the in-car environmental parameters. The setting of these values ​​is based on a reasonable estimate of the speed of change of different environmental factors and the consideration of ensuring that the driver can naturally adapt to environmental changes. For example, for temperature parameters, the transition step may be set to decrease or increase a certain degree per minute to simulate the rate of temperature change in the real environment; for humidity changes, the transition step can be expressed as an increase or decrease of a certain humidity percentage every several seconds.

[0070] When the switching transition module is activated, the preset transition step is input into the module. The switching transition module calculates the transition simulation parameters based on these transition step lengths and the first in-car simulation parameters and the second in-car simulation parameters. Taking the switch from a sunny driving scene to a rainy night driving scene as an example, if the initial in-car temperature is 25°C, the target temperature after switching is 18°C, and the preset transition step is 1°C lower per minute, then during the transition process, according to the advancement of time, the module will calculate the corresponding transition simulation temperature values ​​at different times, such as 24°C after 1 minute, 23°C after 2 minutes, and so on, to generate the temperature transition simulation parameters for the entire transition period. Similarly, for other environmental parameters such as humidity, noise, and light, the corresponding transition simulation parameters will also be calculated based on their respective preset transition step lengths and initial target parameter values.

[0071] Finally, these transition simulation parameters are used to simulate the transition switching of the in-car environment. The temperature control system in the car gradually adjusts the temperature in the car according to the temperature transition simulation parameters, so that the driver can gradually adapt to the temperature change; the humidity control system simulates the gradual change of humidity in the car according to the humidity transition simulation parameters, and cooperates with the window fogging simulation effect; the noise simulation equipment gradually increases the noise level according to the noise transition simulation parameters, simulating the transition from a relatively quiet environment on a sunny day to a noisy environment on a rainy night; the lighting system gradually dims the light in the car according to the lighting transition simulation parameters, simulating the change from bright to dim light. Through this transition simulation method based on a preset transition step length, the in-car environment can smoothly and naturally transition from the first in-car simulation scene to the second in-car simulation scene during the scene switching process, providing the driver with a more realistic, comfortable and actual driving simulation experience, effectively improving the quality and effect of driving training.

[0072] In a possible implementation, the first vehicle exterior simulation scene acquisition module 40 includes:

[0073] A mixed scene determination unit is used to determine whether the first external simulated scene is a mixed scene.

[0074] A plurality of mixed scene determination units are configured to determine a plurality of mixed scenes if the first external vehicle simulation scene is a mixed scene.

[0075] A plurality of in-vehicle simulation parameter acquisition units are used to acquire a plurality of in-vehicle simulation parameters corresponding to the plurality of mixed scenarios respectively according to the in-vehicle environment mapping model.

[0076] A fused in-vehicle simulation parameter output unit is used to fuse the multiple in-vehicle simulation parameters and output the fused in-vehicle simulation parameters.

[0077] A hybrid simulation unit is used by the simulation device to perform hybrid simulation on the in-vehicle environment according to the fused in-vehicle simulation parameters.

[0078] Specifically, in the intelligent simulation system of the variable environment in the driving training vehicle, judging whether the first simulated scene outside the vehicle is a mixed scene is an important beginning to realize the simulation of complex environment. First, the first simulated scene outside the vehicle is comprehensively extracted and analyzed, and various types of information in the scene are obtained through a variety of data collection equipment and technical means, such as meteorological sensors collecting meteorological data such as temperature, humidity, air pressure, precipitation type, etc., cameras capturing visual information of road conditions, traffic signs, surrounding objects and weather phenomena, GPS positioning combined with map data to determine the geographical location and topographic features, and the vehicle's own sensors obtaining dynamic information such as vehicle speed and driving direction. For these collected data, the preset scene recognition algorithms and rules are used for comprehensive judgment. For example, if the meteorological data shows that there is rainfall and low temperature at the same time, and there are signs of snow or ice on the road in the camera image, and the geographical location information indicates that it is in a mountainous road, it is initially judged that the scene is a mixed scene of rain and snow weather and mountainous road conditions; if the camera captures road construction signs and traffic congestion traffic conditions, and at the same time confirms that it is in a city construction section in combination with map data, it tends to be judged as a mixed scene of construction and congestion. However, to ensure the accuracy of the judgment, the interrelationships and influence of various environmental factors will be further considered. For example, when judging whether it is a mixed scene of severe weather and special road conditions, it is necessary not only to confirm the simultaneous existence of severe weather elements (such as rain, snow, fog, etc.) and special road condition elements (such as curves, steep slopes, construction sections, etc.), but also to analyze whether the impact of severe weather on the difficulty of driving in special road conditions conforms to the characteristics of mixed scenes. For example, water accumulation in heavy rain weather will increase the danger of driving on curves. If the road water accumulation in the scene and data such as the curvature of the curve confirm this influence relationship, it will be more certain that the scene is a mixed scene. Through such detailed and multi-dimensional analysis and judgment, it is possible to accurately identify whether the first off-vehicle simulation scene is a mixed scene, providing a reliable basis for subsequent processing of mixed scenes, thereby providing drivers with a real driving training environment.

[0079] After determining that the first simulated scene outside the vehicle is a mixed scene, it is necessary to further determine the multiple mixed scene elements that constitute the mixed scene, that is, multiple specific sub-scenes. First, the various types of data collected related to the scene are deeply analyzed, including meteorological data, road condition data, geographic information data, and vehicle status data. For example, in an urban driving scene determined to be a mixed scene, the meteorological data shows that there is rain and the temperature is low, the road condition data indicates that there are waterlogged sections and traffic congestion, the geographic information data determines that it is located in the downtown commercial district, and the vehicle status data shows that the vehicle speed is low and the vehicle starts and stops frequently. Based on these data, multiple mixed scene elements are determined according to the preset scene classification rules and feature recognition algorithms. For the above example, rainy weather is regarded as a mixed scene element, which is characterized by increased humidity, obstructed vision, slippery road surface, etc.; traffic congestion is another mixed scene element, which is characterized by dense vehicles, slow driving, and the driver's attention is highly focused on keeping the distance between vehicles and frequent start-stop operations; the waterlogged section is also an important mixed scene element, which will affect the vehicle's handling and increase the risk of vehicle skidding; and the low temperature environment, as a separate scene element, affects the temperature adjustment in the car and the driver's somatosensory comfort. When determining these mixed scene elements, not only their respective independent characteristics are considered, but also the interactions and influence relationships between them are analyzed. For example, rainy weather and flooded road sections are interrelated. Rainfall causes water accumulation, which further affects driving safety in rainy weather; traffic congestion and low temperature environment work together to make the engine heat up slowly during low-speed driving, which may affect the heating supply effect in the car, and then affect the driver's comfort and attention. Through such a comprehensive and detailed analysis, multiple mixed scene elements can be accurately determined, laying the foundation for the subsequent acquisition of corresponding in-car simulation parameters and ultimately achieving realistic mixed scene simulation, so that drivers can fully experience the complex and changeable driving environment under mixed scenes in simulation training.

[0080] According to the in-vehicle environment mapping model, the corresponding in-vehicle simulation parameters are obtained for each determined mixed scene element. The in-vehicle environment mapping model is based on a large number of previous in-vehicle and out-vehicle mapping data sets, and can accurately predict the in-vehicle environment parameters corresponding to changes in different single environmental factors. For rainstorm weather elements, the model will consider the impact of factors such as raindrop impact, increased humidity, and obstructed vision on parameters such as noise, humidity, and light in the car, and calculate the corresponding in-vehicle simulation parameters; for dense fog environment elements, the model will focus on the impact of reduced visibility on the light in the car and the driver's visual perception, and determine the corresponding in-vehicle simulation parameters such as light intensity and contrast.

[0081] These multiple in-vehicle simulation parameters are fused. During the fusion process, the weight distribution of the impact of each mixed scene element on the in-vehicle environment is considered comprehensively, and the interaction relationship between different parameters is comprehensively considered to perform weighted summation or other reasonable fusion calculations on each in-vehicle simulation parameter. For example, in a mixed scene of heavy rain and dense fog, if the weight of the impact of heavy rain on noise is larger, and the weight of the impact of dense fog on illumination is larger, then when fusing the in-vehicle simulation parameters, the parameter changes of these two aspects will be highlighted accordingly, and a fused in-vehicle simulation parameter that comprehensively considers the combined impact of heavy rain and dense fog will be obtained.

[0082] The simulation equipment performs mixed simulation of the in-car environment based on the fused in-car simulation parameters. The noise simulation equipment in the car emits a sound that includes the sound of heavy rain and a relatively quiet sound with special sound effects in a dense fog environment based on the fused noise parameters. The humidity control system adjusts to the fused humidity level to simulate the high humidity brought by heavy rain and the possible humidity changes in a dense fog environment. The lighting system creates an in-car lighting atmosphere that meets both the dark light of a heavy rain day and the low visibility in dense fog based on the fused lighting parameters. Through this mixed simulation process, the driver can experience the real changes in the in-car environment under complex mixed scenarios in simulated driving, and improve his ability to cope with complex and changing environments in actual driving.

[0083] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0084] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.

[0085] This specification and the drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. An intelligent simulation system for in-vehicle variable environment for driving training, characterized in that: The system comprises: A sample data set identification module, the sample data set identification module is used to obtain a plurality of driving sample scenes, and identify an external vehicle environment sample data set and an internal vehicle environment sample data set corresponding to the plurality of driving sample scenes; A vehicle interior and exterior mapping data set construction module, wherein the vehicle interior and exterior mapping data set construction module is used to perform time-series synchronization processing according to the vehicle exterior environment sample data set and the vehicle interior environment sample data set to construct a vehicle interior and exterior mapping data set; An in-vehicle environment mapping model output module, which is trained based on the in-vehicle and out-vehicle mapping data set and outputs an in-vehicle environment mapping model; A first vehicle exterior simulation scene acquisition module, the first vehicle exterior simulation scene acquisition module is used to acquire a first vehicle exterior simulation scene for driving training of a current training user; A first in-vehicle simulation parameter acquisition module, the first in-vehicle simulation parameter acquisition module is used to input the first external vehicle simulation scene into the in-vehicle environment mapping model to acquire a first in-vehicle simulation parameter corresponding to the first external vehicle simulation scene; An in-vehicle environment simulation module, wherein the in-vehicle environment simulation module is used to connect to a simulation device, and the simulation device simulates the in-vehicle environment according to the first in-vehicle simulation parameters.

2. The in-vehicle variable environment intelligent simulation system for driving training as claimed in claim 1, characterized in that: The in-vehicle environment mapping model output module includes: An adversarial network acquisition unit, the adversarial network acquisition unit is used to acquire a trained adversarial network, wherein the adversarial network is connected to an input end of the in-vehicle environment mapping model; A data learning unit, the data learning unit is used to perform data learning on the vehicle interior and exterior mapping data set according to the adversarial network, and output a vehicle interior and exterior mapping generation data set; The in-vehicle environment mapping model output unit is used to expand the in-vehicle and out-vehicle environment mapping data set according to the in-vehicle and out-vehicle mapping data set, train according to the expanded data set and identification information of the identification learning accuracy, and output the in-vehicle environment mapping model.

3. The in-vehicle variable environment intelligent simulation system for driving training as claimed in claim 2, characterized in that: The adversarial network includes a generator and a discriminator, and the generator and the discriminator are trained alternately, and the adversarial network acquisition unit includes: A generated data set acquisition unit, the generated data set acquisition unit is used to input the vehicle interior and exterior mapping data set into the generator, perform data learning according to the generator, and acquire a generated data set after multiple learnings; A discrimination accuracy acquisition unit, the discrimination accuracy acquisition unit is used to input the generated data set after multiple learning into the discriminator for discrimination, and obtain the discrimination accuracy; A feedback optimization unit, wherein the feedback optimization unit is used to perform feedback optimization on the generator according to the discrimination accuracy if the discrimination accuracy is less than a preset threshold, until the discrimination accuracy of the discriminator is greater than or equal to the preset threshold, thereby obtaining a trained adversarial network.

4. The in-vehicle variable environment intelligent simulation system for driving training as claimed in claim 1, characterized in that: The sample data set identification module includes: A scene type identification unit, the scene type identification unit is used to identify the scene types of a plurality of driving sample scenes; A feature parameter label set construction unit, the feature parameter label set construction unit is used to identify scene feature parameters according to the scene type, and construct a feature parameter label set corresponding to each driving sample scene; An external vehicle environment sample data set acquisition unit is used to acquire external vehicle environment sample data sets corresponding to the multiple driving sample scenes according to the feature parameter label set.

5. The in-vehicle variable environment intelligent simulation system for driving training as claimed in claim 1, characterized in that: The in-vehicle environment mapping model output module also includes: A scene switching mode determination unit, the scene switching mode determination unit is used to determine whether the scene currently used by the training user for driving training is a scene switching mode; A scene switching node acquisition unit, wherein the scene switching node acquisition unit is used to acquire a scene switching node if the current driving training scene of the training user is a scene switching mode, wherein the scene switching node is a node at which the first external vehicle simulation scene switches to a second external vehicle simulation scene; a second in-vehicle simulation parameter acquisition unit, the second in-vehicle simulation parameter acquisition unit being used to acquire second in-vehicle simulation parameters corresponding to the second out-of-vehicle simulation scene according to the in-vehicle environment mapping model; A switching simulation unit is used for the simulation device to simulate the in-vehicle environment according to the first in-vehicle simulation parameter and then perform switching simulation on the in-vehicle environment with the second in-vehicle simulation parameter.

6. The in-vehicle multi-changeable environment intelligent simulation system for driving training as claimed in claim 5, characterized in that: The switching simulation unit also includes: a parameter difference index output unit, the parameter difference index output unit being used to perform parameter difference analysis on the first in-vehicle simulation parameter and the second in-vehicle simulation parameter, and output a parameter difference index; A switching transition module activation unit, the switching transition module activation unit is used to activate the switching transition module if the parameter difference index is greater than a preset difference index; A transition switching simulation unit is used to perform a transition switching simulation on the in-vehicle environment according to the switching transition module.

7. The in-vehicle multi-changeable environment intelligent simulation system for driving training as claimed in claim 6, characterized in that: The transition switching simulation unit comprises: A preset transition step length configuration unit, wherein the preset transition step length configuration unit is used to configure a preset transition step length; an in-vehicle simulation parameter acquisition unit, the in-vehicle simulation parameter acquisition unit being used to input the preset transition step length into the switching transition module after activating the activation switching transition module, and to acquire transition simulation parameters of the first in-vehicle simulation parameter and the second in-vehicle simulation parameter; The in-vehicle environment transition switching unit is used to perform a transition switching simulation on the in-vehicle environment using the transition simulation parameters.

8. The in-vehicle multi-changeable environment intelligent simulation system for driving training as claimed in claim 1, characterized in that: The first vehicle exterior simulation scene acquisition module further includes: A mixed scene determination unit, the mixed scene determination unit is used to determine whether the first vehicle exterior simulation scene is a mixed scene; a plurality of mixed scene determination units, the plurality of mixed scene determination units being used to determine a plurality of mixed scenes if the first vehicle exterior simulation scene is a mixed scene; A plurality of in-vehicle simulation parameter acquisition units, wherein the plurality of in-vehicle simulation parameter acquisition units are used to acquire a plurality of in-vehicle simulation parameters corresponding to the plurality of mixed scenarios respectively according to the in-vehicle environment mapping model; A fused in-vehicle simulation parameter output unit, wherein the fused in-vehicle simulation parameter output unit is used to perform parameter fusion on the multiple in-vehicle simulation parameters and output the fused in-vehicle simulation parameters; A hybrid simulation unit is used by the simulation device to perform hybrid simulation on the in-vehicle environment according to the fused in-vehicle simulation parameters.

Images