A Platform Construction Method Based on a Driver Training and Teaching Platform

Abstract

This invention discloses a platform construction method based on a driver training teaching platform, belonging to the field of driver training teaching technology. The method includes the following steps: conducting needs analysis and scenario modeling, designing a dynamic training path optimization algorithm, constructing a highly realistic virtual driving environment, designing a real-time driving behavior monitoring and feedback mechanism, identifying typical error patterns of learners in different driving scenarios, constructing a learning map, and designing a driving risk prediction model and embedding it into the platform, thus completing the platform construction. This platform construction method based on a driver training teaching platform overcomes the problem of insufficient response to learners' personalized needs in traditional driver training platforms by introducing a dynamic training path optimization algorithm, a multi-dimensional data fusion model, and a driving risk prediction model. The platform can collect learners' driving operation data, environmental factors, and physiological state information in real time, and combine this with learners' historical behavior data and training trajectories to generate highly personalized training task paths and instant feedback.

Description

Technical Field

[0001] This invention relates to the field of driver training and teaching technology, specifically to a platform construction method based on a driver training and teaching platform. Background Technology

[0002] With advancements in intelligent driving technology and data collection methods, driver training is gradually evolving towards data-driven, personalized, and intelligent approaches. Traditional driver training platforms primarily rely on instructors' intuitive judgment and simple learning modules, which have several shortcomings, such as the uniformity of training content, neglect of individual student needs, and delayed feedback mechanisms. Existing technologies typically focus on static task allocation, data analysis, and evaluation, lacking dynamic monitoring and personalized feedback of students' real-time driving behavior, and failing to fully consider the impact of student operational errors and environmental factors in complex driving scenarios.

[0003] While current driver training platforms offer some operational simulation, they largely rely on basic error statistics and general feedback, lacking in-depth training path optimization and risk prediction capabilities. Existing path optimization methods are mostly based on simple task difficulty or time constraints, ignoring the dynamic changes in learners' abilities and learning progress. Furthermore, although some platforms collect driving data through sensors, most applications are limited to recording basic learner operations, without further in-depth integration of multi-dimensional data, resulting in insufficient personalization and targeting of learner training outcomes. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a platform construction method based on a driver training and teaching platform, thereby resolving the problems mentioned in the background section.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] In a first aspect, embodiments of the present invention provide a platform construction method based on a driver training and teaching platform, comprising the following steps:

[0007] S1. Conduct requirements analysis and scenario modeling;

[0008] S2. Based on the scene modeling results, design a dynamic training path optimization algorithm to provide a personalized learning path;

[0009] S3. Based on personalized learning paths, a highly realistic virtual driving environment is constructed;

[0010] S4. Based on the virtual driving environment, design a real-time driving behavior monitoring and feedback mechanism;

[0011] S5. Based on the data feedback from the mechanism, identify typical error patterns of trainees in different driving scenarios;

[0012] S6. Construct a learning map based on typical error patterns;

[0013] S7. Design a driving risk prediction model and embed it into the platform to complete the platform construction.

[0014] To further optimize this technical solution, in step S1:

[0015] Demand analysis involves analyzing the characteristics of the driver training industry to extract the common and individualized needs of learners' learning paths, instructors' teaching methods, and administrators' platform usage requirements.

[0016] Scenario modeling transforms requirements into technical implementation goals, including simulated driving scenarios, real-time learning feedback, and interactive teaching.

[0017] To further optimize this technical solution, in step S2, a path optimization model is designed to construct a dynamic training path optimization algorithm, which combines the student's real-time performance, ability level, and learning habits to generate a personalized training path.

[0018] In the path optimization model, the training path is defined by multiple training tasks. The path optimization model selects the optimal training task sequence for trainees. To maximize the overall learning benefits for students;

[0019] The path optimization model includes a comprehensive learning benefit function, a dynamic weight adjustment formula, and a task ranking formula.

[0020] To further optimize this technical solution, in the path optimization model:

[0021] The overall learning benefit function is shown below:

[0022] ;

[0023] in,

[0024] :Task The overall benefit value;

[0025] :Task Skill matching degree, that is, the degree to which the trainee's current ability matches the difficulty of the task;

[0026] :Task The weakness repair value is the error frequency of this skill in the student's historical data;

[0027] :Task The training fatigue value is the value of the decline in learning interest caused by the similarity between the task and the previous task.

[0028] Weighting coefficients are dynamically adjusted based on the individual needs of each student.

[0029] The dynamic weight adjustment formula is shown below:

[0030] ;

[0031] in,

[0032] : Vector of the student's current ability level;

[0033] : Vector of student's historical training records;

[0034] : A weight adjustment function based on the student's current status and historical data;

[0035] The task sorting formula is as follows:

[0036] ;

[0037] The task ranking formula selects a set of training tasks. This leads to improved overall learning efficiency. maximize.

[0038] To further optimize this technical solution, the path optimization model, when used, includes:

[0039] Input data;

[0040] Calculation steps: calculate skill matching degree, calculate weakness repair value, calculate training fatigue value, calculate and rank comprehensive benefits;

[0041] Output results;

[0042] Dynamic adjustment.

[0043] To further optimize this technical solution, in step S4, the real-time driving behavior monitoring and feedback mechanism captures the student's driving operation and physiological state through sensors, and designs a multi-dimensional data fusion model to determine whether the performance meets expectations.

[0044] Driving operations include steering wheel angle, braking force, and accelerator pressure;

[0045] Physiological states include heart rate, pupil diameter, and eye movement trajectory.

[0046] To further optimize this technical solution, the multidimensional data fusion model is defined as including the following multidimensional data:

[0047] Driving operation data This includes steering wheel angle, braking force, and accelerator pressure;

[0048] Environmental data This includes road type, weather conditions, and traffic flow;

[0049] Physiological state data This includes heart rate, pupil diameter, and eye movement trajectory.

[0050] To further optimize this technical solution, the multidimensional data fusion model includes a fusion weight calculation formula, a data fusion formula, a performance deviation detection formula, and a real-time feedback generation formula.

[0051] The fusion weighting calculation formula is used to represent the degree of contribution of data from different dimensions to driving behavior;

[0052] The data fusion formula is used to comprehensively measure the driving behavior performance of trainees;

[0053] The performance deviation detection formula is used to measure the degree of deviation between current behavior and ideal behavior;

[0054] A real-time feedback generation formula is used to generate feedback.

[0055] To further optimize this technical solution, step S6 involves constructing a learning map, including:

[0056] The knowledge structure of driver training is dynamically reorganized, and driver knowledge points are divided into three levels: basic operation, scenario application, and complex task.

[0057] The graph structure represents the knowledge acquisition path of students, with nodes representing knowledge points and edges representing logical relationships in learning.

[0058] To further optimize this technical solution, in step S7, the driving risk prediction model uses time series analysis to predict potential risk points in the student's future driving, generates a risk report after each training session, and proposes improvement suggestions.

[0059] In a second aspect, embodiments of the present invention provide a computer device, including a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program instructions are executed by the processor, they implement the steps of a platform construction method based on a driver training and teaching platform as described in the first aspect of the present invention.

[0060] Thirdly, embodiments of the present invention provide a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program instructions are executed by a processor, they implement the steps of a platform construction method based on a driver training teaching platform as described in the first aspect of the present invention.

[0061] Compared with existing technologies, this invention provides a platform construction method based on a driver training and teaching platform, belonging to the technical field of intelligent transportation Internet of Things application services and customer interaction services, and has the following beneficial effects:

[0062] This platform construction method, based on a driver training teaching platform, overcomes the problem of insufficient responsiveness to individual student needs in traditional driver training platforms by introducing dynamic training path optimization algorithms, multi-dimensional data fusion models, and driving risk prediction models. The platform can collect students' driving operation data, environmental factors, and physiological state information in real time, and combine this with students' historical behavior data and training trajectories to generate highly personalized training task paths and instant feedback. By dynamically adjusting training content in real time, the system can accurately identify students' weaknesses and potential risks, thereby providing more targeted training guidance and significantly improving the efficiency and safety of driver training. This data-driven intelligent teaching method not only improves students' operational skills but also enhances their ability to cope with complex driving environments, effectively reducing driving risks and promoting the development of driver training towards a more scientific and intelligent direction. Attached Figure Description

[0063] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0064] Figure 1 This is a flowchart illustrating a platform construction method based on a driver training teaching platform proposed in this invention.

[0065] Figure 2 This is a schematic diagram illustrating the specific components of the multi-dimensional data fusion model in the platform construction method based on a driver training teaching platform proposed in this invention. Detailed Implementation

[0066] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0067] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0068] Secondly, the term "an embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places throughout this specification does not necessarily refer to the same embodiment, nor is it a single embodiment or an embodiment selectively excluded from other embodiments.

[0069] Example 1:

[0070] Reference Figures 1-2 This is the first embodiment of the present invention, which provides a platform construction method based on a driver training and teaching platform, including the following steps:

[0071] S1. Conduct requirements analysis and scenario modeling.

[0072] In this embodiment, the requirements analysis identifies the common and individualized needs of learners, instructors, and administrators in terms of learning paths, teaching methods, and platform usage by analyzing the characteristics of the driver training industry.

[0073] Scenario modeling transforms requirements into technical implementation goals, including simulated driving scenarios, real-time learning feedback, and interactive teaching.

[0074] At the same time, the modeling process also needs to determine the core functional modules of the platform and their interaction logic to ensure that the subsequent design is systematic and forward-looking.

[0075] S2. Based on the scene modeling results, design a dynamic training path optimization algorithm to provide a personalized learning path.

[0076] In this embodiment, a path optimization model is designed to construct a dynamic training path optimization algorithm, which combines the student's real-time performance, ability level and learning habits to generate a personalized training path.

[0077] In the path optimization model, the training path is defined by multiple training tasks. The path optimization model selects the optimal training task sequence for trainees. To maximize the overall learning benefits for students;

[0078] The path optimization model includes a comprehensive learning benefit function, a dynamic weight adjustment formula, and a task ranking formula.

[0079] Furthermore, in the path optimization model:

[0080] The overall learning benefit function is shown below:

[0081] ;

[0082] in,

[0083] :Task The overall benefit value;

[0084] :Task Skill matching degree, that is, the degree to which the trainee's current ability matches the task attributes;

[0085] :Task The weakness repair value is the error frequency of this skill in the student's historical data;

[0086] :Task The training fatigue value is the value of the decline in learning interest caused by the similarity between the task and the previous task.

[0087] Weighting coefficients are dynamically adjusted based on the individual needs of each student.

[0088] The dynamic weight adjustment formula is shown below:

[0089] ;

[0090] in,

[0091] : Vector of the student's current ability level;

[0092] In this embodiment, for example, the following capability level vector can be set:

[0093] Rapid acceleration control ability 80 points (out of 100) indicates that the student performed well in rapid acceleration control;

[0094] Emergency braking control capability A score of 70 indicates that the student has some shortcomings in emergency braking techniques.

[0095] Cornering driving skills A score of 60 indicates a significant technical deficiency in driving on sharp curves.

[0096] Response capability in complex road conditions A score of 85 indicates a relatively quick response to complex environments.

[0097] : Vector of student's historical training records;

[0098] In this embodiment, for example, the following historical training record vector can be set:

[0099] Historical performance of rapid acceleration control : Passing 7 out of 10 training sessions, i.e., 0.7;

[0100] Historical performance of emergency braking control 15 out of 20 training sessions were passed, which is 0.75;

[0101] Historical performance of cornering driving skills 18 out of 30 training sessions were passed, which is 0.6;

[0102] Historical performance of response capability in complex road conditions 22 out of 25 training sessions were successful, which is 0.88.

[0103] A weight adjustment function based on the student's current state and historical data dynamically generates a vector by balancing the current ability level vector and the historical training record vector. , , The weighting coefficients reflect the importance and reliability of skills in different tasks. Weighting adjustment functions typically calculate weights through weighted averaging, linear combination, or nonlinear mapping, allowing them to flexibly adapt to differences in learners' skills and training accumulation, ensuring more accurate and personalized predictions and assessments.

[0104] The task sorting formula is as follows:

[0105] ;

[0106] The task ranking formula selects a set of training tasks. This leads to improved overall learning efficiency. maximize.

[0107] Furthermore, the path optimization model, when used, includes:

[0108] Input data:

[0109] The trainee's current ability level and historical records ;

[0110] Task Collection And its attributes (such as task difficulty, corresponding skill points, etc.);

[0111] Limitations on the student's learning path (such as time constraints and scenario constraints).

[0112] Calculation steps:

[0113] Calculate skill matching degree Based on the students' ability level The degree of matching with task attributes is used to calculate the skill matching degree. For example, if a student is highly skilled at driving on curves, their suitability for this type of task will be lower, and its priority will decrease.

[0114] ;

[0115] in and These are the student and the task in the skill points. The value on, This represents the maximum value in the vector of the student's current ability level, that is, the ability value in which the student performs best across all skill points.

[0116] Calculate the repair value of weak points According to historical records Extract the frequency of errors made by trainees in a specific skill area and assign them high priority to tasks where they are weak:

[0117] ;

[0118] in Representative task set All tasks within. The formula uses... To represent a set of tasks The error frequencies of all tasks are summed. Therefore, Primarily used for computational tasks The proportion of the error rate to the total error rate of all tasks.

[0119] Calculate training fatigue value If the skill points for the current task overlap too much with those for the previous task, it will lead to decreased interest and increased fatigue.

[0120] ;

[0121] in, :Task Compared to the previous task The similarity between tasks. This value reflects the degree of overlap between the current task and the previous task in terms of skill points or operation types. The higher the similarity between tasks, the more fatigued the learner may feel, because the task content is too repetitive.

[0122] The maximum possible similarity value among all tasks in the task set, used as a standardization benchmark to convert similarity values ​​into fatigue values. Typically, It is a preset constant that represents the maximum possible value of task similarity.

[0123] This formula is used to measure the similarity between the current task and the previous task. If the similarity is too high, trainees may feel fatigued due to a lack of novelty.

[0124] Comprehensive benefit calculation and ranking: combined with Calculate the benefit value of each task and sort them to generate the optimal path. .

[0125] Output result:

[0126] The system generates personalized training paths for trainees. It provides real-time recommendations for the next task.

[0127] Dynamic adjustment:

[0128] Update based on real-time feedback from trainees and weight This enables dynamic optimization of the training path. For example, when a trainee improves in a certain skill area, the priority of tasks related to that skill will automatically decrease.

[0129] S3. Based on personalized learning paths, a highly realistic virtual driving environment is constructed.

[0130] In this embodiment, constructing a highly realistic virtual driving environment is key to achieving immersive teaching. By introducing a physics engine and graphics rendering technology, realistic recreations of roads, weather, vehicles, and traffic rules are achieved. Unlike traditional static driving simulation systems, this virtual environment is highly dynamic and adaptable, automatically loading relevant scenarios based on the training path selected by the learner. For example, a simulated scenario of high-speed driving at night in rain can be generated in real time, and algorithms can control vehicle dynamics and the behavior of traffic participants to better match the learner's training objectives.

[0131] S4. Based on the virtual driving environment, design a real-time driving behavior monitoring and feedback mechanism.

[0132] In this embodiment, the real-time driving behavior monitoring and feedback mechanism captures the student's driving operation, environment and physiological state through sensors, and designs a multi-dimensional data fusion model to determine whether the performance meets expectations.

[0133] Furthermore, in the multidimensional data fusion model, the multidimensional data includes:

[0134] Driving operation data This includes steering wheel angle, braking force, and accelerator pressure;

[0135] Environmental data This includes road type, weather conditions, and traffic flow;

[0136] Physiological state data This includes heart rate, pupil diameter, and eye movement trajectory.

[0137] The multidimensional data fusion model includes a fusion weight calculation formula, a data fusion formula, a performance deviation detection formula, and a real-time feedback generation formula.

[0138] Formula for calculating fusion weights: Define fusion weights The contribution of different dimensions of data to driving behavior is represented by the following formula:

[0139] ;

[0140] in, It represents the variance of a data dimension, reflecting the impact of its fluctuations on overall performance.

[0141] Data fusion formula: The fused data is represented as Used to comprehensively measure trainees' performance in time Driving behavior at the time:

[0142] ;

[0143] in:

[0144] It is a data mapping function used to normalize and non-linearly map different data dimensions to a unified standard;

[0145] For each student in time Operational data, environmental data, and physiological state data at the time.

[0146] Performance deviation detection formula: Define the deviation function Measure the degree of deviation between current behavior and ideal behavior:

[0147] ;

[0148] in, It is the expected ideal driving behavior, derived from a preset driving behavior model or high-level driving data.

[0149] Real-time feedback generation formula: Generate feedback Based on deviation value and data fusion :

[0150]

[0151] in, This is the tolerance threshold for deviation.

[0152] When using a multidimensional data fusion model, the following are included:

[0153] Real-time collection of driving operation data Environmental data and physiological state data And complete the normalization preprocessing.

[0154] The weights are dynamically calculated based on the data fluctuations (variance). This reflects the importance of each dimension of data in the current scenario. For example, in a complex traffic environment, the weight of environmental data... It will be higher.

[0155] Generate comprehensive evaluation value This value comprehensively reflects whether the student's current driving behavior is consistent with expectations.

[0156] Compare Compared to ideal performance Measurement deviation. If A large discrepancy indicates that the student's current performance deviates from expectations and needs to be corrected promptly.

[0157] Through feedback formula Generate personalized prompts. For example, when At this time, the system will trigger a voice prompt: "Current speed is too high, please slow down immediately."

[0158] S5. Based on the data feedback from the mechanism, identify typical error patterns of trainees in different driving scenarios.

[0159] In this embodiment, typical error patterns include premature braking and cornering drift. Personalized teaching interventions are implemented by combining error frequency and learning progress data. The intervention mechanism is implemented through augmented reality (AR) and voice guidance. For example, when it is detected that a learner has repeatedly failed to comply with right-of-way rules in simulated city driving, the relevant rule explanation will be automatically played, and the frequency of similar scenario practice will be increased in the next training session, thereby filling in the learner's knowledge gaps.

[0160] S6. Construct a learning map based on typical error patterns.

[0161] In this embodiment, constructing a learning map includes:

[0162] The knowledge structure of driver training is dynamically reorganized, and driver knowledge points are divided into three levels: basic operation, scenario application, and complex task.

[0163] The graph structure represents the knowledge acquisition path of students, with nodes representing knowledge points and edges representing logical relationships in learning.

[0164] When a student performs well in a specific knowledge point, the system can automatically reduce the amount of training for that knowledge point while strengthening training for weaker areas, thereby optimizing resource allocation.

[0165] S7. Design a driving risk prediction model and embed it into the platform to complete the platform construction.

[0166] In this embodiment, the driving risk prediction model uses time series analysis to predict potential risks in the trainee's future driving, generates a risk report after each training session, and proposes improvement suggestions.

[0167] Input the following feature data into the model:

[0168] Student driving habit data (derived from typical error patterns) such as rapid acceleration Emergency braking Number of times you deviate from your lane wait.

[0169] Operational error data (derived from typical error patterns) such as the distance from the lane Precision of steering wheel operation wait.

[0170] Environment variable data Such as reaction time in complex scenarios Weather conditions wait.

[0171] The model generates an input vector by combining these features:

[0172] ;

[0173] in, This represents the time step of a time series, and the input vector contains the student's time step. All relevant data at any given time.

[0174] Deep neural networks (DNNs) can learn complex nonlinear relationships in input data to obtain a driving risk score for a learner. Suppose there is a neural network with multiple hidden layers, whose output is a driving risk score. The calculation process can be expressed as follows:

[0175] ;

[0176] in, This represents the mapping function of a deep neural network. Specifically, the DNN model progressively abstracts input features through a hierarchical structure, ultimately predicting the driving risk score of a learner in a specific situation.

[0177] To quantify the driving risks of trainees, we define a risk score. As a student in time The level of safe driving risk at that time. Based on the model output. The risk index is calculated using the following formula. :

[0178] ;

[0179] in, These represent the various risk components output by the model. These are the corresponding weighting coefficients, reflecting the impact of different risk factors on overall risk.

[0180] To further quantify the degree of behavioral deviation by trainees, a risk deviation metric is introduced. To measure the deviation in the trainee's performance in the current driving situation:

[0181] ;

[0182] in, This represents the risk score under ideal driving scenarios, indicating the expected score under risk-free conditions. It is calculated based on deviation. The system can determine in real time whether a student's driving behavior poses an excessive risk.

[0183] Based on risk deviation and driving risk score Generate personalized feedback:

[0184] ;

[0185] in, This is a set risk threshold. Through this feedback, trainees can receive real-time suggestions on their driving performance to reduce future risks.

[0186] Based on the above steps, the training and teaching structure of the driver training platform is completed.

[0187] Example 2:

[0188] This embodiment also provides a computer device applicable to a platform construction method based on a driver training teaching platform, including a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the platform construction method based on a driver training teaching platform as proposed in the above embodiment.

[0189] This embodiment also provides a storage medium storing a computer program that, when executed by a processor, implements a platform construction method based on a driver training and teaching platform as proposed in the above embodiments.

[0190] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0191] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0192] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-including system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0193] More specific examples (a non-exhaustive list) of computer-readable media include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which programs can be printed, because programs can be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0194] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0195] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A platform construction method based on a driving training teaching platform, characterized in that, Comprise the following steps: S1, carry out requirement analysis and scene modeling; S2, based on the scene modeling results, design a dynamic training path optimization algorithm, provide personalized learning path; S3, based on the personalized learning path, build a highly simulated virtual driving environment; S4, based on the virtual driving environment, design real-time driving behavior monitoring and feedback mechanism; S5, based on the data of mechanism feedback, identify the typical error mode of students in different driving scenes; S6, according to the typical error mode, build learning map; S7, design driving risk prediction model and embed platform, complete the construction of platform; In the step S2, the path optimization model is used to build a dynamic training path optimization algorithm, which combines the real-time performance, ability level and learning habit of the students to generate personalized training path; In the path optimization model, the training path is composed of multiple training tasks The path optimization model selects the optimal training task sequence for the student to maximize the overall learning efficiency of the student. ​ The path optimization model includes comprehensive learning benefit function, dynamic weight adjustment formula and task sorting formula; In the path optimization model: The comprehensive learning benefit function is as follows: ； Wherein, : task : overall benefit value : task skill match degree, i.e. the matching degree of the current ability of the student and the task difficulty; : task weakness repair value, i.e. the frequency of errors in the skill in the student's history data; : task a training fatigue value, i.e. a learning interest decay value caused by the similarity of the task to the previous task; : weight coefficient, dynamically adjusted according to the individual needs of the student; The dynamic weight adjustment formula is as follows: ； Wherein, : current ability level vector of the student; : student historical training record vector; : a weight adjustment function based on the current state of the student and historical data; The task sorting formula is as follows: ； The task ranking formula selects a set of training tasks such that the overall learning benefit is maximized.

2. The platform construction method based on the driving training teaching platform according to claim 1, characterized in that, In the step S1: Requirement analysis, through analyzing the characteristics of driving training industry, the common and individualized demand of students' learning path, coaches' teaching method and managers' platform use demand are refined; Scene modeling, the demand is converted into technical implementation target, including simulation driving scene, real-time learning feedback and teaching interaction.

3. The platform construction method based on the driving training teaching platform according to claim 1, characterized in that, The path optimization model in use includes: Input data; Calculation steps: calculate skill matching degree, calculate weak point repair value, calculate training fatigue value, comprehensive benefit calculation and sorting; Output result; Dynamic adjustment.

4. The platform construction method based on the driving training teaching platform according to claim 1, characterized in that, In the step S4, the real-time driving behavior monitoring and feedback mechanism captures the driving operation and physiological state of the students through sensors, and designs a multi-dimensional data fusion model to judge whether the performance meets the expectation; Driving operation includes steering wheel angle, brake force, throttle pressure; Physiological state includes heart rate, pupil diameter, eye movement trajectory.

5. The platform construction method based on the driving training teaching platform according to claim 4, characterized in that, In the multi-dimensional data fusion model, the set multi-dimensional data includes: Driving operation data including steering wheel angle, brake force, throttle pressure; environmental data including road type, weather conditions, traffic flow; physiological state data including heart rate, pupil diameter, eye movement trajectory.

6. The platform construction method based on the driving training teaching platform according to claim 5, characterized in that, The multi-dimensional data fusion model includes fusion weight calculation formula, data fusion formula, performance deviation detection formula and real-time feedback generation formula; Fusion weight calculation formula is used to represent the contribution degree of different dimension data to driving behavior; Data fusion formula is used to comprehensively measure the driving behavior performance of students; Performance deviation detection formula is used to measure the deviation degree of current behavior from ideal behavior; Real-time feedback generation formula is used to generate feedback.

7. The platform construction method based on the driving training teaching platform according to claim 1, characterized in that, In the step S6, build learning map, including: Dynamic recombination of knowledge point structure of driving training, divide driving knowledge points into three levels of basic operation class, scene application class and complex task class; Through graph structure, the knowledge mastering path of students is represented, node represents knowledge point, edge represents learning logic relationship.

8. The platform construction method based on the driving training teaching platform according to claim 1, characterized in that, In the step S7, the driving risk prediction model uses time series analysis method to predict the risk points that may exist in future driving of students, generates risk report after each training and puts forward improvement suggestions.

Images