Virtual practical training method for automobile vocational education

Abstract

The invention relates to the technical field of automobile education virtual practical training, in particular to a virtual practical training method for automobile vocational education. According to the technical scheme, the method comprises the following steps: initializing a multi-modal interaction system, specifically comprising the following steps: starting a virtual reality module, and creating an automobile practical training scene, including whole automobile internal structure display, engine operation simulation and automobile driving environment simulation; starting an augmented reality module, and preparing to superpose virtual information on a real automobile model; and starting the mixed reality module. Through the steps of multi-modal interaction system initialization, trainee practical training operation, intelligent cooperative tutoring network analysis, ability evaluation and course adjustment, dynamic updating and continuous learning and virtual competition module setting, a vivid practical training scene, personalized tutoring and learning paths are provided for trainees, learning enthusiasm and competition awareness are stimulated, and the training efficiency is improved. The ability of trainees is accurately evaluated, the teaching content is ensured to closely follow the industry development, and the effect of automobile vocational education is significantly improved.

Description

Technical Field

[0001] The invention relates to the technical field of virtual training for automobile education, and in particular to a virtual training method for automobile vocational education. Background Art

[0002] Virtual training in automotive vocational education can create realistic automotive training scenes, such as the display of the internal structure of the whole vehicle, the simulation of engine operation, and the simulation of the driving environment of the car. Participants can observe the various components and systems of the car from all directions and angles to understand their location, structure, and working principle. For example, in a VR environment, participants can view the structure of the engine and understand the movement status of each component during its operation, just like in a real car. Participants can perform various operations in a virtual environment, such as disassembly and assembly of auto parts, fault diagnosis and troubleshooting, etc. This interactivity enables trainees to have a deeper understanding of the process and technical points of automobile maintenance and improve their practical operation ability. For example, through devices such as VR handles, trainees can simulate the disassembly of the car engine, and the system will give feedback based on the correctness of the operation. Superimposing virtual information on real car entities or models provides trainees with a more intuitive learning experience. For example, trainees can view the name, function, disassembly and assembly steps of auto parts through AR devices, and operate them in accordance with the actual parts to deepen their understanding and memory of knowledge.

[0003] Although some virtual training software has begun to use intelligent algorithms for evaluation and analysis, there are still deficiencies in intelligent tutoring. For example, when trainees encounter difficulties during operation, the system cannot provide personalized tutoring and suggestions based on the specific situation of the trainees, and manual intervention by teachers is still required.

[0004] The virtual training system is not intelligent enough and cannot automatically adjust the teaching content and difficulty according to the participants' learning progress and abilities. Participants lack opportunities for independent exploration and learning during the learning process, which affects learning outcomes.

[0005] In summary, this application proposes a virtual training method for automotive vocational education. Summary of the invention

[0006] The purpose of the present invention is to propose a virtual training method for automobile vocational education in view of the problem that virtual training for automobile vocational education in the background technology is still insufficient in intelligent guidance.

[0007] The technical solution of the present invention: a virtual training method for automotive vocational education, comprising the following method steps:

[0008] Step 1: Initialize the multimodal interaction system, including:

[0009] Start the virtual reality module to create a car training scene, including the internal structure of the vehicle, engine operation simulation, and car driving environment simulation;

[0010] Start the augmented reality module and prepare to overlay virtual information on the real car model;

[0011] Start the mixed reality module to enhance the integration of reality and virtuality;

[0012] Connect tactile feedback equipment to provide trainees with tactile feedback during operation;

[0013] Step 2: Participants conduct practical training, which includes the following steps:

[0014] Trainees use the multimodal fusion interactive system to conduct practical vehicle training operations, including component disassembly and assembly, fault diagnosis and troubleshooting. During the operation, the VR module provides trainees with a full range of observation perspectives, while the AR and MR modules provide auxiliary information and guidance.

[0015] The tactile feedback device provides realistic tactile feedback based on the trainee’s operation, including the weight, texture and operating resistance of the components;

[0016] Step 3: Intelligent collaborative tutoring network analysis, specifically including the following steps:

[0017] The deep learning algorithm analyzes the operation data of the trainees. Suppose the operation data of the trainees is O = {o1, o2, ...o n}, where o i Represents an operation action, the deep learning model is f(O;θ), where θ is the model parameter, by minimizing the loss function To train the model, where m is the number of training samples, y i For the real operation results;

[0018] The knowledge graph integrates automotive expertise and practical training experience. When trainees encounter a fault, the system searches the knowledge graph to find nodes and paths related to the fault phenomenon and gives possible causes and solutions.

[0019] Step 4: Capacity assessment and curriculum adjustment, including the following steps:

[0020] The system evaluates the trainees' ability level and knowledge mastery by analyzing the trainees' learning progress, operation performance, and test scores in multiple dimensions. The evaluation algorithm adopts a comprehensive weighted approach. The specific formula is as follows:

[0021]

[0022] Among them, A represents the comprehensive ability evaluation value of the trainees, n represents the number of evaluation indicators, and w i represents the weight of the i-th evaluation indicator, x i Represents the score of the trainee on the i-th evaluation indicator;

[0023] Based on the evaluation results, the system automatically adjusts the teaching content and difficulty, and generates a personalized learning path for the trainees. For trainees with weak foundations, the system will give priority to recommending basic knowledge and simple operation training courses; for trainees with stronger abilities, the system will provide more challenging fault diagnosis and complex maintenance tasks;

[0024] Step 5: Dynamic update and continuous learning, including the following steps:

[0025] The system will update teaching content and cases in a timely manner according to the technological development and market demand of the automotive industry;

[0026] Participants continue to learn according to personalized learning paths.

[0027] Optionally, the virtual reality module uses high-resolution image rendering technology to make the automobile training scene more realistic. The definition calculation formula of the image rendering is: Where C represents clarity, k is the adjustment coefficient, P is the total number of image pixels, and A is the scene area.

[0028] Optionally, the augmented reality module can automatically adjust the display position and size of the virtual information according to the trainee's line of sight focus. The focus and information adjustment formula is: S=f(G), where S represents the display parameters of the virtual information, G is the trainee's line of sight focus parameter, and f is the adjustment function.

[0029] Optionally, the mixed reality module introduces dynamic light and shadow effect simulation technology to enhance the realism of the fusion of reality and virtuality. The light and shadow effect simulation parameter calculation formula is: L=g(T,E), where L represents the light and shadow effect parameter, T is the time variable, E is the environment variable, and g is the simulation function.

[0030] Optionally, the tactile feedback device can simulate the tactile difference of automobile parts made of different materials, and the tactile difference simulation formula is: F=h(M), where F represents the tactile parameter, M is the material parameter of the part, and h is the simulation function.

[0031] Optionally, when analyzing the participant's operation data, the deep learning algorithm is combined with a reinforcement learning strategy, and the reinforcement learning reward calculation formula is: R=u(O,A), where R represents the reward value, O is the participant's operation data, A is the operation action, and u is the reward function.

[0032] Optionally, the knowledge graph can be personalized according to the learning history of the trainees, and the personalized adjustment formula is: K=v(H), where K represents the adjusted knowledge graph parameters, H is the learning history parameters of the trainees, and v is the adjustment function.

[0033] Optionally, the system introduces fuzzy logic judgment when evaluating the ability level and knowledge mastery of the trainees to more accurately reflect the actual situation of the trainees. The fuzzy logic judgment formula is: B=w(X), where B represents the fuzzy judgment result, X is the evaluation data, and w is the fuzzy logic function.

[0034] Optionally, the system adopts an intelligent screening algorithm when updating teaching content and cases, and the screening algorithm formula is: S=z(C, F), where S represents the screening result, C is the automotive industry technology development parameter, F is the market demand parameter, and z is the screening function.

[0035] Optionally, the system sets up a virtual competition module, and the trainees participate in an automobile maintenance skills competition in a virtual environment. The competition results adopt a comprehensive scoring mechanism, and the scoring formula is: Score = a*P+b*T+c*Q, where Score represents the competition score, P is the operation accuracy score, T is the operation time score, Q is the innovation method score, and a, b, c are corresponding weight coefficients.

[0036] Compared with the prior art, the present invention has at least one of the following beneficial technical effects:

[0037] This application provides realistic scenes through a high-resolution virtual reality module to improve the intuitiveness of learning, the augmented reality module automatically adjusts the information display to improve learning efficiency, the light and shadow effects of the mixed reality module enhance the sense of reality and enrich the training scenes; the tactile feedback device simulates the touch of different materials to improve practical operation capabilities.

[0038] Comprehensive skills are further cultivated through multimodal fusion interaction, real tactile feedback helps to master operating strength and techniques, deep learning is combined with reinforcement learning, personalized guidance is provided, and personalized knowledge graphs are provided to quickly solve problems and meet learning needs.

[0039] Further, through comprehensive weighted evaluation and fuzzy logic judgment, the trainees are accurately evaluated, personalized learning paths are generated to meet the needs of different trainees, and teaching content is intelligently screened and updated to ensure the practicality of learning. Personalized continuous learning improves the enthusiasm and autonomy of the trainees.

[0040] Furthermore, through multimodal interactive system initialization, participant training operations, intelligent collaborative tutoring network analysis, capability assessment and curriculum adjustment, dynamic update and continuous learning, and setting up virtual competition module steps, trainees are provided with realistic training scenarios, personalized tutoring and learning paths, which stimulate learning enthusiasm and competitive awareness, accurately assess the abilities of trainees, ensure that the teaching content keeps pace with industry development, and significantly improve the effectiveness of automotive vocational education. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 The flowchart of the virtual training method used for automotive vocational education. DETAILED DESCRIPTION

[0042] In order to make the objectives, technical solutions and advantages of the present invention more clear, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

[0043] like Figure 1 As shown in the figure, the virtual training method used for automotive vocational education is

[0044] The method comprises the following steps:

[0045] Step 1: Initialize the multimodal interaction system, including the following steps:

[0046] Start the virtual reality module and use the advanced graphics rendering engine to create realistic car training scenes. In the display of the internal structure of the whole vehicle, each component of the car is modeled with high precision to ensure that the trainees can clearly see every detail. Including the pointers on the dashboard and the texture of the seat can be presented in high resolution. In the engine operation simulation, the working process of the engine is accurately simulated, including the movement trajectory of the piston, the fuel injection method and the feedback signals of various sensors. The car driving environment simulation takes into account different road conditions, weather conditions and traffic conditions to provide trainees with a rich and diverse training scene. The clarity calculation formula for image rendering is: Where C represents clarity, k is the adjustment coefficient, P is the total number of image pixels, and A is the scene area. By continuously adjusting the value of k and optimizing the ratio between the total number of image pixels and the scene area, the clarity of the training scene is ensured to be optimal. For complex engine structure display scenes, the total number of image pixels is appropriately increased, and the scene area is adjusted at the same time to improve clarity. For wide-area car driving environment simulation scenes, the clarity is dynamically adjusted according to different field of view to ensure that trainees can get a clear visual experience when observing distant and near objects.

[0047] Start the augmented reality module. When the trainee places the real car model in the training area, the augmented reality module recognizes the model through a high-precision camera and accurately superimposes the virtual information on the model. When the trainee observes the car engine, the augmented reality module can display virtual annotation information on various parts of the engine, including the name, function and working principle of the parts. At the same time, the display position and size of the virtual information are automatically adjusted according to the trainee's line of sight focus to ensure that the trainee can see the relevant information in the most appropriate position. The focus and information adjustment formula is: S = f (G), where S represents the display parameter of the virtual information, G is the trainee's line of sight focus parameter, and f is the adjustment function. By continuously optimizing the adjustment function f, the display of virtual information can more accurately follow the changes in the trainee's line of sight focus. When the trainee's line of sight focuses on a specific part of the engine, the augmented reality module automatically enlarges the virtual annotation information of the part and provides a more detailed description.

[0048] Start the mixed reality module, which further integrates reality and virtual elements to provide trainees with a more realistic training experience. In automobile maintenance training, when trainees perform actual operations, the mixed reality module can display virtual maintenance guides and fault diagnosis information in real maintenance scenes. At the same time, the dynamic light and shadow effect simulation technology is introduced to make the integration of virtual elements and the real environment more natural and realistic. The calculation formula for light and shadow effect simulation parameters is: L = g(T, E), where L represents the light and shadow effect parameters, T is the time variable, E is the environment variable, and g is the simulation function;

[0049] Connect the tactile feedback device, and when the trainees are performing operations such as disassembling and assembling auto parts, the tactile feedback device provides real tactile feedback according to the operation action. When the trainees pick up a virtual auto part, the tactile feedback device can simulate the weight, texture and operating resistance of the part, allowing the trainees to feel the same as operating a real part. The simulation formula for tactile difference is: F = h(M), where F represents the tactile parameter, M is the material parameter of the part, and h is the simulation function; by continuously optimizing the simulation function h, the tactile feedback device can more accurately simulate the tactile difference of auto parts of different materials. For example, for metal parts, its hard texture and heavy weight can be simulated; for plastic parts, its light weight and relatively soft texture can be simulated.

[0050] Step 2: Participants conduct practical training, which includes the following steps:

[0051] Automobile training operation:

[0052] Trainees use the multimodal fusion interactive system to conduct practical automotive training operations, including parts disassembly and assembly, fault diagnosis and troubleshooting. In the parts disassembly and assembly training, trainees first use the VR module to observe the position and structure of automotive parts in all directions, and then use the auxiliary information and guidance provided by the AR and MR modules to perform actual operations. The AR module displays the steps for disassembly and assembly of parts, and the MR module superimposes virtual operation prompts on the real scene. In the fault diagnosis and troubleshooting training, trainees use the VR module to observe the operating status of the automotive system, use deep learning algorithms to analyze operating data, and find the cause of the fault.

[0053] During the operation, the VR module provides trainees with a full range of observation perspectives to ensure that they can operate accurately. Trainees can freely rotate their perspectives through the VR helmet to observe various parts of the car and understand the specific manifestations of the fault. At the same time, the auxiliary information and guidance provided by the AR and MR modules can help trainees better understand the operation steps and principles.

[0054] Haptic Feedback:

[0055] The tactile feedback device provides real-time tactile feedback during the trainee's operation, including the weight, texture and operating resistance of the parts. When the trainee tightens a screw, the tactile feedback device simulates the tightening force of the screw, allowing the trainee to feel the realism of the operation. At the same time, the VR module provides the trainee with a full range of observation perspectives to ensure that the trainee can operate accurately.

[0056] Step 3: Intelligent collaborative tutoring network analysis, specifically including the following steps:

[0057] The deep learning algorithm analyzes the operation data of the trainees. Suppose the operation data of the trainees is O = {o1, o2, ...o n}, where o i Represents an operation action, the deep learning model is f(O;θ), where θ is the model parameter, by minimizing the loss function To train the model, where m is the number of training samples, y i is the actual operation result; when analyzing the operation data of the trainees, the deep learning algorithm is combined with the reinforcement learning strategy, and the reinforcement learning reward calculation formula is: R=u(O,A), where R represents the reward value, O is the operation data of the trainees, A is the operation action, and u is the reward function.

[0058] The knowledge graph integrates automotive expertise and practical training experience. When trainees encounter a fault, the system searches the knowledge graph to find nodes and paths related to the fault phenomenon and gives possible causes and solutions. The knowledge graph can be personalized based on the trainees' learning history. The personalized adjustment formula is: K=v(H), where K represents the adjusted knowledge graph parameters, H is the trainee's learning history parameters, and v is the adjustment function.

[0059] Step 4: Capacity assessment and curriculum adjustment, including the following steps:

[0060] The system evaluates the trainees' ability level and knowledge mastery by analyzing the trainees' learning progress, operation performance, and test scores in multiple dimensions. The evaluation algorithm adopts a comprehensive weighted approach. The specific formula is as follows:

[0061]

[0062] Among them, A represents the comprehensive ability evaluation value of the trainees, n represents the number of evaluation indicators, and w i represents the weight of the i-th evaluation indicator, x i Represents the score of the trainee on the i-th evaluation indicator;

[0063] When evaluating the ability level and knowledge mastery of trainees, the system introduces fuzzy logic judgment to more accurately reflect the actual situation of the trainees. The fuzzy logic judgment formula is: B=w(X), where B represents the fuzzy judgment result, X is the evaluation data, and w is the fuzzy logic function.

[0064] Based on the evaluation results, the system automatically adjusts the teaching content and difficulty, and generates a personalized learning path for the trainees. For trainees with weak foundations, the system will give priority to recommending basic knowledge and simple operation training courses; for trainees with stronger abilities, the system will provide more challenging fault diagnosis and complex maintenance tasks;

[0065] Through the above steps, the system analyzes multi-dimensional data such as the participants' learning progress, operating performance, test scores, etc., and adopts a comprehensive weighted evaluation algorithm to comprehensively and accurately evaluate the participants' ability level and knowledge mastery.

[0066] The introduction of fuzzy logic judgment can more accurately reflect the actual situation of the trainees, avoid the limitations of a single evaluation indicator, and provide trainees with a more objective and fair evaluation.

[0067] Based on the evaluation results, the system automatically adjusts the teaching content and difficulty, and generates a personalized learning path for the trainees. This can meet the learning needs of different trainees and improve learning outcomes. For trainees with weak foundations, we provide basic knowledge and simple operation training courses to help them gradually improve; for trainees with stronger abilities, we provide more challenging fault diagnosis and complex maintenance tasks to stimulate their learning potential.

[0068] Step 5: Dynamic update and continuous learning, including the following steps:

[0069] The system will update teaching content and cases in a timely manner according to the technological development and market demand of the automotive industry;

[0070] Trainees continue to learn according to personalized learning paths. When updating teaching content and cases, the system adopts an intelligent screening algorithm. The screening algorithm formula is: S=z(C, F), where S represents the screening result, C is the technical development parameter of the automotive industry, F is the market demand parameter, and z is the screening function. By continuously optimizing the screening function z, the system can more accurately screen out teaching content and cases that meet the development trend of the automotive industry and market demand. Trainees continue to learn according to personalized learning paths. The system generates personalized learning paths for trainees based on their ability assessment results and learning history. Trainees choose practical training courses and learning resources that suit them according to their learning progress and needs, and continuously improve their automobile maintenance skills.

[0071] Intelligent screening algorithm and timely update of teaching content:

[0072] The system uses intelligent screening algorithms to timely update teaching content and cases based on the technological development and market demand of the automotive industry. This ensures that trainees learn the latest automotive maintenance knowledge and skills and improve their employment competitiveness.

[0073] The screening algorithm formula can accurately select teaching content and cases that are in line with the development trends and market demands of the automotive industry, providing participants with more practical learning resources.

[0074] Personalized Continuous Learning:

[0075] Participants can continue to learn according to their personalized learning path and choose practical training courses and learning resources that suit them according to their learning progress and needs. This personalized learning method can improve the learning enthusiasm and initiative of participants and enable them to master automobile maintenance skills more independently.

[0076] Example 2

[0077] Based on Example 1, this embodiment sets up a virtual competition module in the system, and trainees compete in automobile maintenance skills in a virtual environment to further improve the learning enthusiasm and competitive awareness of the trainees.

[0078] Competition Settings:

[0079] The system sets up a virtual competition module, where trainees can compete in automobile maintenance skills in a virtual environment. The competition content includes parts disassembly and assembly, fault diagnosis and troubleshooting, covering all aspects of automotive vocational education. The competition can set different difficulty levels and time limits to meet the needs of different trainees.

[0080] For junior trainees, a simple parts disassembly and assembly competition is set up with a relatively long time limit; for advanced trainees, a complex fault diagnosis and troubleshooting competition is set up with a relatively short time limit. The competition results adopt a comprehensive scoring mechanism, and the scoring formula is: Score = a*P+b*T+c*Q, where Score represents the competition score, P is the operation accuracy score, T is the operation time score, Q is the innovation method score, and a, b, c are the corresponding weight coefficients.

[0081] Competition Incentives:

[0082] In order to motivate participants to actively participate in the competition, generous incentives are set up. Participants with excellent performance in the competition can be given material rewards, such as car repair tools and teaching materials; honorary awards, certificates, and commendations within the school.

[0083] At the same time, the competition results also serve as an important reference for the evaluation of the trainees' abilities. In the comprehensive evaluation of the trainees' abilities, the trainees' performance in the competition is considered and corresponding bonus points are given. The system sets up a virtual competition module, and the trainees compete in the automobile maintenance skills competition in the virtual environment, which can stimulate their learning enthusiasm and competitive awareness. The trainees will work harder to improve their skills in the competition and strive to get better results.

[0084] The competitive atmosphere can make participants more devoted to learning, while also cultivating their team spirit and competitive awareness, laying a good foundation for their future career development.

[0085] Competition Feedback and Learning:

[0086] After the competition, the system provides participants with detailed competition feedback, including error analysis and improvement suggestions during operation. Based on the competition feedback, participants can understand their own shortcomings and conduct targeted learning and improvement.

[0087] The system analyzes the participants' operation data in the competition, finds out the wrong steps that the participants made during the disassembly and assembly of parts, and provides the correct operation methods and demonstration videos. By watching the demonstration videos, the participants learn the correct operation techniques and improve their maintenance skills. The competition results adopt a comprehensive scoring mechanism, which takes into account multiple factors such as operation accuracy, operation time and innovative methods, and can comprehensively evaluate the skill level of the participants. This can guide the participants to focus on improving operation speed and innovation ability while improving operation accuracy. After the competition, the system provides detailed competition feedback to the participants, including error analysis and improvement suggestions. The participants can conduct targeted learning and improvement based on the feedback to further improve their skills.

[0088] The above specific embodiments are only several optional embodiments of the present invention. Based on the technical solutions of the present invention and the relevant inspirations of the above embodiments, those skilled in the art can make various alternative improvements and combinations to the above specific embodiments.

Claims

1. A virtual training method for automotive vocational education, characterized in that: The method comprises the following steps: Step 1: Initialize the multimodal interaction system, including: Start the virtual reality module to create a car training scene, including the internal structure of the vehicle, engine operation simulation, and car driving environment simulation; Start the augmented reality module and prepare to overlay virtual information on the real car model; Start the mixed reality module to enhance the integration of reality and virtuality; Connecting a tactile feedback device to provide tactile feedback to the operator; Step 2: Participants’ practical training includes: Trainees use the multimodal fusion interactive system to conduct practical vehicle training operations, including component disassembly and assembly, fault diagnosis and troubleshooting. During the operation, the VR module provides trainees with a full range of observation perspectives, while the AR and MR modules provide auxiliary information and guidance. The tactile feedback device provides realistic tactile feedback based on the trainee’s operation, including the weight, texture and operating resistance of the components; Step 3: Intelligent collaborative tutoring network analysis, including: The deep learning algorithm analyzes the operation data of the trainees. Suppose the operation data of the trainees is O = {o1, o2, ...o n }, where o i Represents an operation action, the deep learning model is f(O;θ), where θ is the model parameter, by minimizing the loss function To train the model, where m is the number of training samples, y i For the real operation results; The knowledge graph integrates automotive expertise and practical training experience. When trainees encounter a fault, the system searches the knowledge graph to find nodes and paths related to the fault phenomenon and gives possible causes and solutions. Step 4: Capacity assessment and curriculum adjustment, including: The system evaluates the trainees' ability level and knowledge mastery by analyzing the trainees' learning progress, operation performance, and test scores in multiple dimensions. The evaluation algorithm adopts a comprehensive weighted approach. The specific formula is as follows: Among them, A represents the comprehensive ability evaluation value of the trainees, n represents the number of evaluation indicators, and w i represents the weight of the i-th evaluation indicator, x i Represents the score of the trainee on the i-th evaluation indicator; Based on the evaluation results, the system automatically adjusts the teaching content and difficulty, and generates a personalized learning path for the trainees. For trainees with weak foundations, the system will give priority to recommending basic knowledge and simple operation training courses; for trainees with stronger abilities, the system will provide more challenging fault diagnosis and complex maintenance tasks; Step 5: Dynamic update and continuous learning.

2. The virtual training method for automotive vocational education according to claim 1, characterized in that: The virtual reality module uses high-resolution image rendering technology to make the automobile training scene more realistic. The clarity calculation formula of the image rendering is: Where C represents clarity, k is the adjustment coefficient, P is the total number of image pixels, and A is the scene area.

3. The virtual training method for automotive vocational education according to claim 1, characterized in that: The augmented reality module can automatically adjust the display position and size of virtual information according to the trainee's line of sight focus. The focus and information adjustment formula is: S=f(G), where S represents the display parameter of virtual information, G is the trainee's line of sight focus parameter, and f is the adjustment function.

4. The virtual training method for automotive vocational education according to claim 1, characterized in that: The mixed reality module introduces dynamic light and shadow effect simulation technology to enhance the realism of the fusion of reality and virtuality. The light and shadow effect simulation parameter calculation formula is: L=g(T,E), where L represents the light and shadow effect parameter, T is the time variable, E is the environment variable, and g is the simulation function.

5. The virtual training method for automotive vocational education according to claim 1, characterized in that: The tactile feedback device can simulate the tactile differences of automobile parts made of different materials. The tactile difference simulation formula is: F=h(M), where F represents the tactile parameter, M is the material parameter of the part, and h is the simulation function.

6. The virtual training method for automotive vocational education according to claim 1, characterized in that: When analyzing the participant's operation data, the deep learning algorithm is combined with the reinforcement learning strategy, and the reinforcement learning reward calculation formula is: R=u(O,A), where R represents the reward value, O is the participant's operation data, A is the operation action, and u is the reward function.

7. The virtual training method for automotive vocational education according to claim 1, characterized in that: The knowledge graph can be personalized according to the learning history of the trainees. The personalized adjustment formula is: K=v(H), where K represents the adjusted knowledge graph parameters, H is the learning history parameters of the trainees, and v is the adjustment function.

8. The virtual training method for automotive vocational education according to claim 1, characterized in that: When evaluating the ability level and knowledge mastery of trainees, the system introduces fuzzy logic judgment to more accurately reflect the actual situation of the trainees. The fuzzy logic judgment formula is: B=w(X), where B represents the fuzzy judgment result, X is the evaluation data, and w is the fuzzy logic function.

9. The virtual training method for automotive vocational education according to claim 1, characterized in that: When updating teaching content and cases, the system adopts an intelligent screening algorithm, and the screening algorithm formula is: S=z(C, F), where S represents the screening result, C is the automotive industry technology development parameter, F is the market demand parameter, and z is the screening function.

10. The virtual training method for automotive vocational education according to claim 1, characterized in that: The system sets up a virtual competition module, and trainees participate in an automobile maintenance skills competition in a virtual environment. The competition results adopt a comprehensive scoring mechanism. The scoring formula is: Score = a*P+b*T+c*Q, where Score represents the competition score, P is the operation accuracy score, T is the operation time score, Q is the innovation method score, and a, b, and c are the corresponding weight coefficients.

Images