Remote training method and system based on virtual reality

Abstract

The invention provides a remote training method and system based on virtual reality. The method comprises the following steps: performing personalized customization processing on a training environment by utilizing a professional background, a skill level and a learning target of a student to obtain a simulation operation scene; according to the simulation operation scene, carrying out natural language processing algorithm analysis processing on the obtained language instruction, generating an operation prompt, superposing the operation prompt into a visual interface of a student through an augmented reality technology, and carrying out real-time monitoring and automatic correction on an operation process of the student to obtain a personalized guidance scheme; and based on the personalized guidance scheme, performing decision tree algorithm classification analysis processing on the behavior data of the trainees, and comprehensively evaluating the performance of the trainees from multiple dimensions in combination with an educational psychology theory to obtain a multi-dimensional performance evaluation report. According to the technical scheme provided by the invention, the pertinence and effectiveness of training are remarkably improved.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of virtual reality technology, and in particular to a remote training method and system based on virtual reality. Background Art

[0002] In modern vocational education and skills training, the construction of a personalized training environment has become increasingly important. With the continuous improvement of professional skills requirements in various industries, the traditional unified training model has been unable to meet the needs of different trainees. In order to improve the training effect, there is an urgent need for a method that can be customized according to the trainees' professional background, skill level and learning goals. In addition, real-time guidance and feedback mechanisms are crucial to improving trainees' operational proficiency and reducing the error rate, but existing training methods are still insufficient in this regard.

[0003] At present, most training institutions use standardized training environments and fixed teaching content, supplemented by simple multimedia teaching tools or virtual simulation software. Although these methods can simulate actual work scenarios to a certain extent, they lack customized designs for individual differences. At the same time, some existing augmented reality (AR) applications are mainly focused on the display level, and fail to deeply combine natural language processing (NLP) technology and real-time monitoring functions to provide trainees with personalized operation guidance and instant feedback.

[0004] The existing solutions have the following major defects: the training environment cannot be customized according to the specific circumstances of the trainees, resulting in a disconnect between the training content and the actual needs of the trainees, affecting the learning effect; there is a lack of real-time operation prompts and automatic correction mechanisms based on natural language processing, and trainees cannot get timely and effective help when they encounter problems during operation; they mostly rely on static performance evaluations and fail to comprehensively evaluate the performance of trainees from multiple dimensions, especially in terms of psychological state and learning effects, making it difficult to provide scientific improvement suggestions. Summary of the invention

[0005] The embodiments of the present application provide a remote training method and system based on virtual reality to solve the problems of lack of personalization, insufficient real-time guidance and incomplete evaluation in the prior art.

[0006] In a first aspect, an embodiment of the present application provides a remote training method based on virtual reality, comprising:

[0007] The training environment is customized based on the trainees’ professional background, skill level and learning goals to obtain simulated operation scenarios.

[0008] According to the simulated operation scenario, the acquired language instructions are parsed by a natural language processing algorithm to generate operation prompts, and the operation prompts are superimposed on the trainee's visual interface through augmented reality technology, so that the trainee's operation process is monitored and automatically corrected in real time to obtain a personalized guidance plan;

[0009] Based on the personalized guidance program, the behavioral data of the trainees are classified and analyzed using a decision tree algorithm, and combined with educational psychology theory, the performance of the trainees is comprehensively evaluated from multiple dimensions to obtain a multi-dimensional performance evaluation report.

[0010] Optionally, according to the simulated operation scenario, the acquired language instructions are parsed by a natural language processing algorithm to generate operation prompts, and the operation prompts are superimposed on the trainee's visual interface through augmented reality technology, and the trainee's operation process is monitored and automatically corrected in real time to obtain a personalized guidance plan, including:

[0011] Utilizing the simulated operation scenario, the acquired language instructions are parsed using a natural language processing algorithm to obtain operation prompts;

[0012] According to the operation prompt, the operation prompt is superimposed on the trainee's visual interface through augmented reality technology to generate an augmented reality guidance interface;

[0013] Based on the augmented reality guidance interface, the trainee's operation process is monitored in real time and automatically corrected to obtain real-time monitoring data;

[0014] The real-time monitoring data is used in combination with the specific behavior patterns and error types of the trainees to perform optimization algorithm analysis and processing on the personalized learning path to generate a personalized guidance plan.

[0015] Optionally, the step of superimposing the operation prompt onto the trainee's visual interface through augmented reality technology according to the operation prompt to generate an augmented reality guidance interface includes:

[0016] According to the operation prompt, configure the image recognition accuracy, transparency of the virtual object and interaction mode parameters in the augmented reality technology to obtain display parameters;

[0017] Based on the display parameters, the trainee's visual interface is embedded with augmented reality elements, and the operation prompts are accurately superimposed on the corresponding real objects or operation points in the trainee's field of vision in the form of virtual objects to generate a preliminary augmented reality guidance interface;

[0018] Using the preliminary augmented reality guidance interface, visually annotate key positions and actions in the trainee's operating environment to obtain real-time annotation information;

[0019] According to the real-time annotation information and in combination with the progress and changes of the trainees in actual operation, the preliminary augmented reality guidance interface is dynamically updated to generate an augmented reality guidance interface.

[0020] Optionally, based on the display parameters, the visual interface of the trainee is embedded with augmented reality elements, and the operation prompt is accurately superimposed on the corresponding real object or operation point in the trainee's field of vision in the form of a virtual object to generate a preliminary augmented reality guidance interface, including:

[0021] Based on the display parameters, the trainee's visual interface is embedded with augmented reality elements, and the relative position of the virtual object in the real world is determined by spatial positioning technology to obtain position mapping information of the virtual object;

[0022] According to the position mapping information, the operation prompt is accurately superimposed on the corresponding real object or operation point in the trainee's field of view in the form of a virtual object, ensuring that the virtual object is accurately aligned with the real object, and generating a preliminary virtual object layout;

[0023] Using the preliminary virtual object layout, interactively designing the augmented reality elements to obtain interactive augmented reality elements;

[0024] Based on the interactive augmented reality elements and in combination with the trainees' operating habits and environmental conditions, the preliminary augmented reality guidance interface is personalized and adjusted to generate a preliminary augmented reality guidance interface.

[0025] Optionally, based on the personalized guidance program, the behavioral data of the trainees are classified and analyzed by a decision tree algorithm, and the performance of the trainees is comprehensively evaluated from multiple dimensions in combination with educational psychology theory to obtain a multi-dimensional performance evaluation report, including:

[0026] Based on the personalized guidance program, the behavioral data of trainees during the training period are systematically collected and structured to obtain a standardized behavioral data set;

[0027] Based on the standardized behavior data set, a decision tree algorithm is applied to perform classification analysis and processing, different behavior patterns and their corresponding performance levels are identified, the trainees' performance is evaluated from multiple dimensions such as operation accuracy, task completion, and time efficiency, and a preliminary performance evaluation result is generated;

[0028] Using the preliminary performance evaluation results and combining them with educational psychology theory, we conduct an in-depth analysis of the students' psychological state and learning effects, and obtain a psychological and learning effect analysis report;

[0029] Based on the psychological and learning effect analysis report, multiple dimensions of evaluation indicators are integrated, and factors such as the students' learning progress, error types and corrections are taken into consideration to conduct a comprehensive and integrated evaluation of the students' overall performance and generate a multi-dimensional performance evaluation report.

[0030] Optionally, the preliminary performance evaluation results are used in combination with educational psychology theory to conduct in-depth analysis and processing of the trainees' psychological state and learning effect to obtain a psychological and learning effect analysis report, including:

[0031] Using the preliminary performance evaluation results, the trainees' behavior patterns during the training period are carefully analyzed to identify the psychological factors that may affect the trainees' performance and obtain psychological state characteristics;

[0032] According to the characteristics of the psychological state, combined with the model trained by educational psychology theory, the psychological state of the trainees is quantitatively analyzed and processed, and supplementary data is collected through questionnaires and interviews to generate psychological state assessment data;

[0033] Based on the psychological state assessment data, correlation analysis is performed on the learning effect of the trainees to assess the trainees' knowledge mastery, skill improvement speed, error types and their correction status, and obtain learning effect assessment data;

[0034] The learning effect evaluation data is used in combination with the psychological state evaluation data to conduct a comprehensive and integrated analysis and processing, and a combination of qualitative and quantitative methods is used to generate a psychological and learning effect analysis report.

[0035] Optionally, the mental state assessment data is used to perform correlation analysis on the student's learning effect, assess the student's knowledge mastery, skill improvement speed, error type and its correction status, and obtain learning effect assessment data, including:

[0036] Based on the psychological state assessment data, correlation analysis is performed on the learning effect of the trainees to assess the trainees' knowledge mastery, skill improvement speed, error types and their correction status, and obtain learning effect assessment data;

[0037] Utilizing the psychological state assessment data, quantitatively analyzing and processing the trainees' knowledge mastery during the training period, measuring the trainees' knowledge level through objective data such as test scores and homework completion, and obtaining a knowledge mastery score;

[0038] Based on the knowledge mastery score and the trainee's historical operation records, the skill improvement speed is dynamically evaluated and processed, the time difference and quality change of each task completion are calculated, and a skill improvement curve is generated;

[0039] Based on the skill improvement curve, the types of students' errors and their corrections are classified and statistically processed to identify common error patterns and the effectiveness of improvement measures, and obtain an error improvement report;

[0040] The error improvement report is used in combination with the psychological state assessment data to conduct a comprehensive and integrated analysis and processing, and a combination of qualitative and quantitative methods is used to evaluate the learning effects of the trainees and generate learning effect assessment data.

[0041] In a second aspect, an embodiment of the present application provides a remote training system based on virtual reality, including:

[0042] The customization module is used to customize the training environment based on the trainees’ professional background, skill level and learning goals to obtain simulation operation scenarios;

[0043] The parsing module is used to perform natural language processing algorithm parsing on the acquired language instructions according to the simulated operation scenario, generate operation prompts, and superimpose the operation prompts on the trainee's visual interface through augmented reality technology, so as to monitor and automatically correct the trainee's operation process in real time and obtain a personalized guidance plan;

[0044] The evaluation module is used to perform decision tree algorithm classification analysis on the trainees' behavior data based on the personalized guidance plan, and to comprehensively evaluate the trainees' performance from multiple dimensions in combination with educational psychology theory to obtain a multi-dimensional performance evaluation report.

[0045] In a third aspect, an embodiment of the present application provides a computing device, comprising a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a virtual reality-based remote training method as described in the first aspect.

[0046] In a fourth aspect, an embodiment of the present application provides a computer storage medium storing a computer program. When the computer program is executed by a computer, a remote training method based on virtual reality as described in the first aspect is implemented.

[0047] In an embodiment of the present application, the training environment is personalized and customized using the professional background, skill level and learning goals of the trainees to obtain a simulated operation scenario; based on the simulated operation scenario, the acquired language instructions are parsed by a natural language processing algorithm to generate operation prompts, and the operation prompts are superimposed on the trainee's visual interface through augmented reality technology, and the trainee's operation process is monitored and automatically corrected in real time to obtain a personalized guidance plan; based on the personalized guidance plan, the trainee's behavior data is classified and analyzed by a decision tree algorithm, and combined with educational psychology theory, the trainee's performance is comprehensively evaluated from multiple dimensions to obtain a multi-dimensional performance evaluation report.

[0048] The technical solution of this application has the following beneficial effects:

[0049] The training environment is customized based on the trainees' professional background, skill level and learning goals, ensuring that the simulation operation scenarios are highly consistent with the trainees' actual working environment and personal development goals, thus improving the relevance and pertinence of the training; personalized simulation operation scenarios can stimulate the trainees' interest, enhance their sense of participation and immersion, and thus significantly improve their learning efficiency; the instructor's language instructions are parsed through natural language processing algorithms to generate precise operation prompts, and these prompts are superimposed on the trainees' visual interface with the help of augmented reality technology, realizing real-time monitoring and automatic correction of the trainees' operation process. This not only reduces the occurrence of errors, but also improves the trainees' operational proficiency and self-confidence; based on the trainees' performance and feedback, the system can dynamically adjust the guidance plan and provide more personalized support to ensure that each trainee can obtain the training path that best suits them; combined with educational psychology theory, the trainees' performance is comprehensively evaluated from multiple dimensions such as operation accuracy, task completion, and time efficiency, and a multi-dimensional performance evaluation report is obtained. This comprehensive evaluation method helps to more accurately identify the strengths and weaknesses of students and provide a scientific basis for subsequent training. By classifying and analyzing students' behavioral data through the decision tree algorithm, the system can discover potential problems and make suggestions for improvement, helping students to continuously optimize their operational skills and form a virtuous cycle of learning. The personalized training environment and real-time guidance mechanism make training more flexible and adaptable to the needs of different students, thereby improving the overall training effect and quality. By collecting and analyzing students' behavioral data, the system can continuously optimize training content and methods to ensure that training is always in the best condition and meet the ever-changing industry needs and student expectations.

[0050] Furthermore, this method achieves a high degree of personalization and intelligence in the training environment, ensures the accuracy and immediacy of operation prompts, and greatly improves the learning efficiency and operation proficiency of trainees. The guidance interface generated by augmented reality technology not only enhances the trainees' immersion and participation, but also provides intuitive operation guidance and reduces the learning curve. The real-time monitoring and automatic correction mechanism can quickly detect and correct errors to ensure that trainees make rapid progress on the right path. More importantly, based on real-time monitoring data and the trainees' specific behavior patterns and error types, the system can optimize personalized learning paths and provide tailored guidance plans, thereby significantly improving the training effect and quality and promoting the continuous improvement and development of trainees' skills.

[0051] Furthermore, by systematically collecting and structuring the behavioral data of trainees during the training, a standardized behavioral data set is obtained, which ensures the consistency and reliability of the data and lays a solid foundation for subsequent precise analysis; the decision tree algorithm is used to classify and analyze the standardized behavioral data set, which can accurately identify different behavioral patterns and their corresponding performance levels, comprehensively evaluate trainees' performance from multiple dimensions such as operation accuracy, task completion, and time efficiency, and generate preliminary performance evaluation results. This method not only improves the accuracy of the evaluation, but also provides detailed feedback to help trainees clarify the direction of improvement; combined with educational psychology theory, the preliminary performance evaluation results are used to conduct in-depth analysis of the trainees' psychological state and learning effects, and generate a psychological and learning effect analysis report. This helps to more comprehensively understand the trainees' learning motivation, emotional state, and cognitive development, and provide a scientific basis for personalized tutoring; based on the psychological and learning effect analysis report, the evaluation indicators of multiple dimensions are integrated, and factors such as the trainees' learning progress, error types, and corrections are considered to conduct a comprehensive and comprehensive evaluation of the overall performance of the trainees, and generate a multi-dimensional performance evaluation report. This comprehensive evaluation method not only reflects the trainees' current ability level, but also reveals their potential for development, providing strong support for future training plans. Through multi-dimensional performance evaluation reports, trainers can adjust training strategies in a timely manner, optimize course content and teaching methods, and ensure that each trainee can get the most suitable training experience for themselves. At the same time, trainees can also improve their own shortcomings in a targeted manner based on the evaluation results, so as to achieve continuous improvement of skills and maximize personal development.

[0052] These and other aspects of the present application will become more clearly understood in the description of the following embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0054] Figure 1 A flowchart of a remote training method based on virtual reality provided in an embodiment of the present application;

[0055] Figure 2 A schematic diagram of the structure of a remote training system based on virtual reality provided in an embodiment of the present application;

[0056] Figure 3 A schematic diagram of the structure of a computing device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0057] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.

[0058] In some of the processes described in the specification and claims of this application and the above-mentioned figures, multiple operations that appear in a specific order are included, but it should be clearly understood that these operations may not be executed in the order in which they appear in this article or executed in parallel. The serial numbers of the operations, such as 101, 102, etc., are only used to distinguish between different operations, and the serial numbers themselves do not represent any execution order. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions of "first", "second", etc. in this article are used to distinguish different messages, devices, modules, etc., do not represent the order of precedence, and do not limit the "first" and "second" to be different types.

[0059] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions 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 those skilled in the art without creative work are within the scope of protection of this application.

[0060] Figure 1 A flowchart of a remote training method based on virtual reality is provided for an embodiment of the present application, such as Figure 1 As shown, the method includes:

[0061] 101. Use the trainees’ professional background, skill level and learning goals to personalize the training environment and obtain simulated operation scenarios;

[0062] In this step, the professional background, skill level and learning goals of the trainees are used for personalized customization, including collecting data such as the trainees' personal information, professional experience, current skill level and specific learning goals, to build a simulation operation scenario that is highly tailored to the trainees' needs. This information not only helps to create a more realistic training environment, but also ensures that the training content is closely related to the trainees' career development.

[0063] In the embodiment of the present application, the system first obtains the basic information and needs of the trainees through questionnaires, interviews or data analysis, and then adjusts the parameters in the virtual environment based on this information, such as equipment type, task difficulty, workflow, etc., and finally generates a simulated operation scenario that can reflect the actual working situation of the trainees, providing a basis for subsequent training.

[0064] For example, if a mechanical engineer is receiving training on how to operate a new automated production line, the system will customize a simulation scenario that includes the latest automated equipment and complex production processes based on his professional background (mechanical engineering), skill level (intermediate), and learning goal (mastering the operation of a new production line). This makes the situations he faces in training highly similar to those he will encounter in his future work, thereby improving the training effect.

[0065] 102. According to the simulated operation scenario, the acquired language instructions are parsed by a natural language processing algorithm to generate operation prompts, and the operation prompts are superimposed on the trainee's visual interface through augmented reality technology, so as to monitor and automatically correct the trainee's operation process in real time and obtain a personalized guidance plan;

[0066] In this step, according to the simulated operation scenario, the acquired language instructions are parsed by a natural language processing algorithm to generate operation prompts, and these prompts are superimposed on the trainee's visual interface through augmented reality technology. The language instructions can come from the instructor or other sources, and after parsing, they are converted into specific operation instructions to help trainees perform tasks correctly. At the same time, the system will monitor and automatically correct the trainee's operation process in real time to reduce errors and improve operation proficiency.

[0067] In the embodiment of the present application, the system receives the instructor's verbal instructions, converts them into specific operation prompts through natural language processing algorithms, and then uses augmented reality technology to intuitively display these prompts in the trainee's field of vision. At the same time, the system continuously monitors the trainee's actions, and once an error or irregular operation is detected, it immediately provides instant feedback and correction suggestions to ensure that the trainee's operation meets the standard requirements.

[0068] Assume that in the training of the mechanical engineer mentioned above, the instructor issued a voice command "Start the production line and check the safety lock". After analysis, the system generates corresponding operation prompts, such as "Press the green start button and then check the safety lock status on the screen". These prompts are projected directly into the trainee's field of vision through augmented reality glasses. At the same time, the system monitors their actions in real time. If any deviation is found, corrective instructions are given immediately, such as "Please confirm again that the safety lock is activated."

[0069] 103. Based on the personalized guidance program, the behavioral data of the trainees are classified and analyzed using a decision tree algorithm, and combined with educational psychology theory, the performance of the trainees is comprehensively evaluated from multiple dimensions to obtain a multi-dimensional performance evaluation report.

[0070] In this step, based on the personalized guidance plan, the decision tree algorithm is used to classify and analyze the trainee's behavior data, and combined with educational psychology theory, the trainee's performance is comprehensively evaluated from multiple dimensions to obtain a multi-dimensional performance evaluation report. The behavior data includes information on operation accuracy, task completion, time efficiency, etc., while educational psychology theory helps understand the trainee's psychological state and learning effect, thereby providing a comprehensive evaluation result.

[0071] In the embodiment of the present application, the system collects all behavioral data of trainees during training, applies decision tree algorithm to classify and analyze it, and identifies different behavioral patterns and their corresponding performance levels. Subsequently, the psychological state and learning effect of the trainees are deeply analyzed in combination with educational psychology theory, and a detailed performance evaluation report is generated. This report not only reflects the current ability level of the trainees, but also reveals their potential development space, providing strong support for future training plans.

[0072] Assume that after the mechanical engineer's training, the system collects all his operation records, including data such as the completion time and accuracy of each task. Through decision tree algorithm analysis, the system identifies his characteristics of being good at responding quickly but occasionally neglecting details. Combined with educational psychology theory, further evaluation shows that he is prone to anxiety in high-pressure environments, which affects his operation accuracy. The final performance evaluation report describes his strengths and weaknesses in detail and puts forward targeted improvement suggestions, such as strengthening psychological adjustment training and detail management skills.

[0073] In summary, steps 101 to 103 cover the entire process from the construction of personalized simulation operation scenarios, real-time operation guidance and automatic correction, to multi-dimensional performance evaluation, aiming to provide a scientific, efficient and personalized training system to meet the specific needs and development goals of different trainees. This approach not only improves the pertinence and effectiveness of training, but also promotes the continuous improvement of trainees' skills and the maximization of their personal development.

[0074] In order to meet the needs of real-time guidance and personalized learning path optimization, in some embodiments, according to the simulation operation scenario, the acquired language instructions are parsed by a natural language processing algorithm in step 102 to generate operation prompts, and the operation prompts are superimposed on the trainee's visual interface through augmented reality technology, and the trainee's operation process is monitored and automatically corrected in real time to obtain a personalized guidance plan, including:

[0075] Using the simulated operation scenario, the acquired language instructions are parsed by a natural language processing algorithm to obtain operation prompts; based on the operation prompts, the operation prompts are superimposed on the trainee's visual interface through augmented reality technology to generate an augmented reality guidance interface; based on the augmented reality guidance interface, the trainee's operation process is monitored in real time and automatically corrected to obtain real-time monitoring data; using the real-time monitoring data, combined with the trainee's specific behavior patterns and error types, the personalized learning path is analyzed and processed by an optimization algorithm to generate a personalized guidance plan.

[0076] In this embodiment, the simulated operation scenario is used to perform natural language processing algorithm analysis on the acquired language instructions to obtain operation prompts; based on the operation prompts, the operation prompts are superimposed on the trainee's visual interface through augmented reality technology to generate an augmented reality guidance interface; based on the augmented reality guidance interface, the trainee's operation process is monitored in real time and automatically corrected to obtain real-time monitoring data; using the real-time monitoring data, combined with the trainee's specific behavior patterns and error types, the personalized learning path is analyzed and processed by optimization algorithm to generate a personalized guidance plan.

[0077] In the embodiment of the present application, first, the system receives and parses the instructor's language instructions and converts them into specific operation prompts; second, these prompts are intuitively displayed in the trainee's field of vision through augmented reality technology to form an augmented reality guidance interface; third, the system monitors the trainee's operation process in real time, records any deviations or errors, and generates real-time monitoring data; finally, based on the real-time monitoring data, combined with the trainee's behavior patterns and error types, an optimization analysis of the personalized learning path is performed, and finally a personalized guidance plan is generated.

[0078] Here is a specific example:

[0079] Suppose an aviation engineer is receiving aircraft maintenance training; first, the system receives the voice command from the instructor, "Check the engine's oil pressure sensor", and interprets it into a specific operation prompt, "Find the oil pressure sensor and confirm its status"; second, these prompts are projected directly into the trainee's field of vision through augmented reality glasses to help him find the sensor accurately; third, the system monitors the trainee's operation in real time, and once it finds that he has missed a step during the inspection, it immediately gives a corrective suggestion, "Please recheck whether the sensor connection is tight"; finally, based on the collected real-time monitoring data, the system identifies the characteristics of the trainee who tends to ignore details in a high-pressure environment, and optimizes his learning path, adding more exercises on detail management. Through the above steps, the system not only ensures the accuracy of the trainee in each operation, but also provides a more targeted learning plan for his future training.

[0080] In order to solve the accuracy and real-time problems of the augmented reality guidance interface, in some embodiments, the operation prompt is superimposed on the trainee's visual interface through augmented reality technology according to the operation prompt to generate the augmented reality guidance interface, including:

[0081] According to the operation prompts, the image recognition accuracy, transparency of virtual objects and interaction mode parameters in the augmented reality technology are configured to obtain display parameters; based on the display parameters, the trainee's visual interface is embedded with augmented reality elements, and the operation prompts are accurately superimposed on the corresponding real objects or operation points in the trainee's field of view in the form of virtual objects to generate a preliminary augmented reality guidance interface; using the preliminary augmented reality guidance interface, key positions and actions in the trainee's operating environment are visually annotated to obtain real-time annotation information; according to the real-time annotation information, combined with the progress and changes of the trainee in actual operation, the preliminary augmented reality guidance interface is dynamically updated to generate an augmented reality guidance interface.

[0082] In this embodiment, according to the operation prompts, the image recognition accuracy, transparency of virtual objects and interaction mode parameters in the augmented reality technology are configured to obtain display parameters; based on the display parameters, the trainee's visual interface is embedded with augmented reality elements, and the operation prompts are accurately superimposed in the form of virtual objects on the corresponding real objects or operation points in the trainee's field of view to generate a preliminary augmented reality guidance interface; using the preliminary augmented reality guidance interface, key positions and actions in the trainee's operating environment are visually annotated to obtain real-time annotation information; according to the real-time annotation information, combined with the progress and changes of the trainee in actual operation, the preliminary augmented reality guidance interface is dynamically updated to generate an augmented reality guidance interface.

[0083] In the embodiment of the present application, first, the system adjusts various parameters of the augmented reality technology, such as image recognition accuracy, virtual object transparency and interaction mode, according to the operation prompts to ensure the best display effect; secondly, based on these parameters, the augmented reality elements are embedded in the trainee's visual interface, so that the operation prompts can be accurately superimposed on the corresponding real objects or operation points to form a preliminary augmented reality guidance interface; thirdly, the system visually marks the key positions and actions in the trainee's operating environment and provides real-time annotation information; finally, the system dynamically updates the preliminary augmented reality guidance interface according to the trainee's actual operation progress and changes to ensure that the guidance always fits the current situation.

[0084] Here is a specific example:

[0085] Suppose an electronic engineer is receiving training on circuit board repair; first, based on the operation prompt "check the specific resistor on the circuit board", the system adjusts the image recognition accuracy of the augmented reality glasses to ensure that the target resistor can be accurately located, and sets the transparency and interaction mode of the virtual objects (such as arrows or highlight boxes) so that the trainee can clearly see and interact; secondly, based on these parameters, the system accurately superimposes the virtual objects on the corresponding resistor positions in the trainee's field of view, and generates a preliminary augmented reality guidance interface; thirdly, the system visually annotates the key positions of the circuit board (such as welding points and connecting lines), and provides annotation information in real time to help the trainee confirm the operation position; finally, as the trainee gradually completes each step, the system dynamically updates the guidance interface according to its actual operation progress. For example, when the trainee starts to measure the resistance, the interface will display the measurement results in real time and prompt the next operation. Through the above steps, the system not only ensures the accuracy of the operation prompts, but also provides real-time and dynamic guidance support, which significantly improves the trainee's operation efficiency and accuracy.

[0086] In order to solve the problem of accurate alignment between virtual objects and real objects in the augmented reality guidance interface, in some embodiments, based on the display parameters, the trainee's visual interface is embedded with augmented reality elements, and the operation prompt is accurately superimposed on the corresponding real object or operation point in the trainee's field of view in the form of a virtual object to generate a preliminary augmented reality guidance interface, including:

[0087] Based on the display parameters, the trainee's visual interface is embedded with augmented reality elements, and the relative position of the virtual object in the real world is determined by spatial positioning technology to obtain position mapping information of the virtual object; based on the position mapping information, the operation prompt is accurately superimposed on the corresponding real object or operation point in the trainee's field of view in the form of a virtual object to ensure that the virtual object is accurately aligned with the actual object, and a preliminary virtual object layout is generated; using the preliminary virtual object layout, the augmented reality elements are interactively designed to obtain interactive augmented reality elements; based on the interactive augmented reality elements, combined with the trainee's operating habits and environmental conditions, the preliminary augmented reality guidance interface is personalized and adjusted to generate a preliminary augmented reality guidance interface.

[0088] In this embodiment, based on the display parameters, the trainee's visual interface is embedded with augmented reality elements, and the relative position of the virtual object in the real world is determined by spatial positioning technology to obtain position mapping information of the virtual object; according to the position mapping information, the operation prompt is accurately superimposed on the corresponding real object or operation point in the trainee's field of view in the form of a virtual object to ensure that the virtual object is accurately aligned with the actual object, and a preliminary virtual object layout is generated; using the preliminary virtual object layout, the augmented reality elements are interactively designed to obtain interactive augmented reality elements; based on the interactive augmented reality elements, combined with the trainee's operating habits and environmental conditions, the preliminary augmented reality guidance interface is personalized and adjusted to generate a preliminary augmented reality guidance interface.

[0089] In the embodiment of the present application, first, the system uses spatial positioning technology to determine the relative position of the virtual object in the real world according to the display parameters, and obtains position mapping information; secondly, based on these position mapping information, the operation prompts are accurately superimposed on the corresponding real objects or operation points in the trainee's field of view in the form of virtual objects, ensuring that the virtual objects are accurately aligned with the actual objects; thirdly, based on the preliminary generated virtual object layout, the system performs interactive design on the augmented reality elements, such as adding interactive functions such as clicking and dragging, to form interactive augmented reality elements; finally, the system personalizes the preliminary augmented reality guidance interface according to the trainees' operating habits and specific environmental conditions to ensure that it meets the actual needs of the trainees, and generates the final preliminary augmented reality guidance interface.

[0090] Here is a specific example:

[0091] Suppose an automobile repair technician is receiving engine disassembly training; first, the system uses spatial positioning technology to determine the specific position of the virtual arrow on the engine based on the pre-configured display parameters (such as image recognition accuracy, transparency, etc.), ensuring that the virtual arrow can accurately point to the screw that needs to be removed; second, based on these position mapping information, the system accurately superimposes the virtual arrow and highlight box on the corresponding screw position in the trainee's field of view to ensure that the virtual object is accurately aligned with the actual object; third, based on the preliminary generated virtual object layout, the system adds a click function to the virtual arrow. When the trainee clicks, the system will provide detailed disassembly steps and precautions to form an interactive augmented reality element; finally, the system personalizes the preliminary augmented reality guidance interface according to the trainee's operating habits (such as preference for left-handed operation) and environmental conditions (such as light intensity), such as adjusting the color and size of the virtual object to ensure that the interface is clear and easy to use. Through the above steps, the system not only ensures the precise alignment of virtual objects with actual objects, but also provides rich interactive functions and a personalized guidance interface, which significantly improves the trainee's learning experience and operating efficiency.

[0092] The present application takes into account that, in the prior art, due to the lack of accurate parsing of language instructions and the ability to perceive context, traditional training systems have difficulty providing personalized and accurate operation prompts. In addition, the existing system fails to fully consider the historical behavior patterns of trainees and the specific circumstances of the current operation scenario, resulting in the lack of targeted guidance provided, which affects the training effect. Therefore, an embodiment of the present invention proposes a method for parsing a natural language processing algorithm using a simulated operation scenario, which aims to solve the above problems and enhance the intelligence level and personalized guidance capabilities of the training system.

[0093] Optionally, the utilizing the simulated operation scenario to perform natural language processing algorithm analysis on the acquired language instructions to obtain operation prompts includes:

[0094] By using the simulated operation scenario, the acquired language instructions are preprocessed and context-aware word embedding analysis is performed. By calculating the semantic vector of the operation prompt and combining the historical behavior pattern of the trainees, a foundation is laid for the subsequent generation of the operation prompt. The semantic vector of the operation prompt is calculated by the following formula:

[0095] V op =σ(W sem ·(E cmd +α·E ctx )+b sem +β·H hist )

[0096] Among them, V op The semantic vector representing the operation prompt; W semis the semantic transformation matrix, which is used to adjust the mapping from word embedding to action prompt space; E cmd is the word embedding vector of the tutor's language instruction, obtained through the pre-training model; E ctx is the word embedding vector of the current operation scene context, reflecting the environmental information; α is the context weight coefficient, which controls the influence of the context on the final semantic vector; b sem is a bias term used to calibrate the converted semantic vector; β is the historical behavior weight coefficient, which affects the impact of historical operation records on the current prompt; H hist is the student's historical operation record feature vector; σ is a nonlinear activation function;

[0097] According to the semantic vector V of the operation prompt op , the operation prompts are integrated with the features of the current operation scenario, the relevance is evaluated using the context weight matrix, and the scoring mechanism is adjusted to generate the context relevance score S by comparing with the ideal operation conditions. ctx , where the context relevance score S ctx Calculated by the following formula:

[0098] S ctx =exp(-||W ctx ·[V op ; C sc ]+b ctx || 2 )×max(0,1-||C sc -C ref ||)+γ·F feed

[0099] Among them, S ctx represents the context relevance score, which measures the relevance of the operation prompt to the current operation scenario; W ctx is the context weight matrix, which is used to evaluate the association strength between the operation prompt and the scene context; V op is the semantic vector of the action hint obtained from the first formula; C sc is the context feature vector of the simulation operation scenario; C ref is the reference context feature vector, representing the ideal operating scenario conditions; b ctx is a bias term used to adjust the scoring benchmark; exp is an exponential function used to compress the scoring range to a positive value close to zero; ∥·∥ represents the Euclidean norm, which is used to calculate the distance between vectors; the max function ensures that the score will not be lower than zero, and adjusts the score by comparing the distance between the current context and the reference context; γ is the feedback weight coefficient, which affects the impact of the tutor's real-time feedback on the score; F feed It is the instant feedback feature vector provided by the instructor, reflecting the instructor’s evaluation of the student’s current operation;

[0100] Based on the context relevance score S ctx , intelligently filter and optimize the most relevant prompt options, fine-tune the content based on the tutor's immediate feedback, and verify the effect through simulation tests to generate operation prompts.

[0101] This formula is designed to achieve more accurate and personalized operation prompt generation. The system calculates the semantic vector V of the operation prompt through preprocessing and context-aware word embedding analysis. op The system then evaluates the relevance of the operation prompt to the current operation scenario and generates a context relevance score S ctx , to ensure the effectiveness and pertinence of the prompts. Finally, the system generates the most appropriate operation prompts through intelligent screening and optimization processing, combined with the instructor's feedback.

[0102] The following is a brief introduction to the design reasons of each sub-item of the formula:

[0103] V op =σ(W sem ·(E cmd +α·E ctx )+b sem +β·H hist )

[0104] W sem ·(E cmd +α·E ctx ) Overall design reason: This sub-item aims to combine the word embedding vector E of the tutor's language instructions cmd and the word embedding vector E of the current operation scene context ctx , and through the semantic transformation matrix W sem Adjust the mapping to generate an operation prompt semantic vector that comprehensively considers language instructions and environmental information. The context weight coefficient α controls the degree of influence of the context on the final semantic vector, ensuring that the generated operation prompt not only accurately reflects the instructor's intention, but also fully considers the specific circumstances of the current operation scenario, thereby improving the relevance and pertinence of the prompt.

[0105] β·H hist Overall design reason: Taking into account the trainee’s past behavior patterns and common mistakes, more personalized guidance is provided. The integration of historical data helps the system to dynamically adjust according to the trainee’s actual performance, improving the effectiveness of training and personalized development.

[0106] The following is a brief introduction to how to obtain the parameters of the formula:

[0107] Semantic transformation matrix W sem Obtained by training the model; word embedding vector E cmdExtracted by pre-trained language model (such as BERT); context weight coefficient α is obtained through experimental setting or machine learning optimization; context word embedding vector E ctx Extracted through text description in the current simulation operation scene; bias term b sem Automatically adjusted during model training; historical behavior weight coefficient β is obtained based on the student's historical operation data statistics; historical operation record feature vector H hist Extract the trainees' historical operation records from the database;

[0108] The following is a brief introduction to the design reasons of each sub-item of the formula:

[0109] S ctx =exp(-||W ctx ·[V op ; C sc ]+b ctx || 2 )×max(0,1-||C sc -C ref ||)+γ·F feed

[0110] exp(-||W ctx ·[V op ; C sc ]+b ctx || 2 ) Overall design reason: This sub-item calculates the semantic vector V of the operation prompt op and the current operation scene context feature vector C sc The association strength is calculated by using the context weight matrix W ctx The evaluation is performed to generate a context relevance score. The exponential function exp compresses the score range to a positive value close to zero to ensure the rationality and stability of the score. The core purpose of this sub-item is to quantify the match between the operation prompt and the current operation scenario to ensure the effectiveness and pertinence of the prompt.

[0111] max(0,1-||C sc -C ref ||) Overall design reason: This sub-item compares the current operation scene context feature vector C sc and the reference context feature vector C under ideal operating conditions ref , and use the maximum function max to ensure that the score will not be lower than zero. The main function of this sub-item is to measure the gap between the current operation scenario and the ideal conditions, and adjust the scoring mechanism to ensure that the generated operation prompts are not only in line with the current situation, but also can guide students to gradually approach the ideal operation state.

[0112] γ·F feedOverall design reason: Combined with the real-time evaluation of the tutor, ensure that the prompt content is more in line with the actual needs and tutor expectations. The integration of tutor feedback enhances the interactivity and flexibility of the system and improves the pertinence and effectiveness of training.

[0113] The following is a brief introduction to how to obtain the parameters of the formula:

[0114] Context weight matrix W ctx Obtained by training the model; semantic vector V of the operation prompt op Calculated by formula 1; the context feature vector C of the simulation operation scenario sc Extracted from the simulation environment; reference context feature vector C ref Set according to industry standards or best practices; bias term b ctx Automatically adjusted during model training; feedback weight coefficient γ is obtained through experimental setting or machine learning optimization; instant feedback feature vector F feed collected through real-time feedback provided by the tutor;

[0115] Suppose a mechanical engineer is receiving training on the operation of an automated production line. First, the system receives a voice command from the instructor: "Start the production line and check the safety lock." The system performs preprocessing and context-aware word embedding analysis on this command, using the formula:

[0116] V op =σ(W sem ·(E cmd +0.7·E ctx )+b sem +0.5 H hist )

[0117] Where W sem is the semantic transformation matrix obtained by training the model, E cmd is the tutor language instruction word embedding vector extracted by the pre-trained model, the context weight coefficient α = 0.7, and the context word embedding vector E ctx Extracted from the current simulation operation scenario, the bias term b sem Automatic adjustment, historical behavior weight coefficient β = 0.5, historical operation record feature vector H hist Extract the trainee's historical operation records from the database. The calculation result is:

[0118] V op =σ(specific value)

[0119] Next, the system is based on V op Evaluate the relevance of the operation prompt to the current operation scenario using the formula:

[0120] S ctx=exp(-||W ctx ·[V op ; C sc ]+b ctx || 2 )×max(0,1-||C sc -C ref ||)+0.8·F feed

[0121] Where W ctx is the context weight matrix obtained by training the model, C sc Extracted from the simulation environment, the reference context feature vector C ref According to industry standard settings, the bias term b ctx Automatic adjustment, feedback weight coefficient γ = 0.8, instant feedback feature vector F feed Collected through real-time feedback provided by the tutor. The calculation result is: S ctx =Specific value based on S ctx The system screens out the most relevant prompt options, fine-tunes the content based on the instructor's immediate feedback, verifies the effect through simulation tests, and finally generates an operation prompt "Press the green start button and then check the safety lock status on the screen."

[0122] Through the above steps, the system not only provides accurate and personalized operation prompts, but also ensures that these prompts are highly matched with the current operation scenario, significantly improving the training effect and personalized development. Assuming that the threshold is set to 0.8, due to the context relevance score S ctx It is greater than the set threshold, which indicates that the operation prompt highly matches the current operation scenario and can effectively guide students to complete the task.

[0123] In order to meet the multi-dimensional evaluation needs of trainees' performance, in some embodiments, based on the personalized guidance plan described in step 103, the trainees' behavior data is classified and analyzed by a decision tree algorithm, and the trainees' performance is comprehensively evaluated from multiple dimensions in combination with educational psychology theory to obtain a multi-dimensional performance evaluation report, including:

[0124] Based on the personalized guidance program, the behavioral data of the trainees during the training period are systematically collected and structured to obtain a standardized behavioral data set; based on the standardized behavioral data set, a decision tree algorithm is applied to perform classification and analysis to identify different behavioral patterns and their corresponding performance levels, and the performance of the trainees is evaluated from multiple dimensions such as operational accuracy, task completion, and time efficiency to generate preliminary performance evaluation results; using the preliminary performance evaluation results and combining them with educational psychology theory, the psychological state and learning effects of the trainees are deeply analyzed to obtain a psychological and learning effect analysis report; based on the psychological and learning effect analysis report, evaluation indicators of multiple dimensions are integrated to take into account the trainees' learning progress, error types, and correction factors, and a comprehensive and integrated evaluation of the overall performance of the trainees is performed to generate a multi-dimensional performance evaluation report.

[0125] In this embodiment, based on the personalized guidance program, the behavioral data of the trainees during the training period are systematically collected and structured to obtain a standardized behavioral data set; based on the standardized behavioral data set, a decision tree algorithm is applied to perform classification and analysis to identify different behavioral patterns and their corresponding performance levels, and the performance of the trainees is evaluated from multiple dimensions such as operation accuracy, task completion, and time efficiency to generate preliminary performance evaluation results; using the preliminary performance evaluation results, combined with educational psychology theory, the psychological state and learning effects of the trainees are deeply analyzed and processed to obtain a psychological and learning effect analysis report; based on the psychological and learning effect analysis report, evaluation indicators of multiple dimensions are integrated, and factors such as the trainees' learning progress, error types, and correction situations are taken into consideration to conduct a comprehensive and integrated evaluation of the overall performance of the trainees to generate a multi-dimensional performance evaluation report.

[0126] In the embodiment of the present application, first, the system systematically collects all behavioral data of trainees during the training period, and organizes them in a structured manner to ensure the consistency and reliability of the data, thereby forming a standardized behavioral data set; secondly, the system applies a decision tree algorithm to classify and analyze the standardized behavioral data set, identifies different behavioral patterns and their corresponding performance levels, evaluates trainee performance from dimensions such as operational accuracy, task completion, and time efficiency, and generates preliminary performance evaluation results; thirdly, the system combines educational psychology theory and uses preliminary performance evaluation results to deeply analyze the psychological state and learning effects of trainees, and generates a psychological and learning effect analysis report; finally, the system integrates evaluation indicators of multiple dimensions, takes into account factors such as trainees' learning progress, error types, and correction situations, conducts a comprehensive and integrated evaluation of the overall performance of the trainees, and finally generates a multi-dimensional performance evaluation report.

[0127] Here is a specific example:

[0128] Suppose a software development engineer is receiving code review skills training; first, the system systematically collects all his behavioral data during the training (such as code submission records, problem solving speed, team collaboration) and organizes them in a structured manner to ensure that the data format is unified and form a standardized behavioral data set; secondly, the system applies a decision tree algorithm to classify and analyze these data, identify different behavioral patterns (such as efficient coding, frequent debugging, and relying on others for help), and evaluates his performance from multiple dimensions such as operation accuracy, task completion, and time efficiency, and generates preliminary performance evaluation results; thirdly, the system combines educational psychology theory and uses preliminary performance evaluation results to deeply analyze the trainees' psychological state (such as stress level, self-confidence) and learning effects (such as knowledge mastery and skill improvement speed), and generates a psychological and learning effect analysis report; finally, the system integrates evaluation indicators from multiple dimensions, considers the trainees' learning progress (such as improvement in code quality), error types and corrections (such as reduction in common programming errors), and conducts a comprehensive and integrated evaluation of the trainees' overall performance, and finally generates a multi-dimensional performance evaluation report.

[0129] In order to meet the needs of comprehensive evaluation of trainees' psychological state and learning effects, in some embodiments, the preliminary performance evaluation results are used in combination with educational psychology theory to conduct in-depth analysis and processing of trainees' psychological state and learning effects to obtain a psychological and learning effect analysis report, including: using the preliminary performance evaluation results to conduct a detailed analysis of trainees' behavioral patterns during training, identify psychological factors that may affect trainees' performance, and obtain psychological state characteristics; based on the psychological state characteristics, combined with the model trained by educational psychology theory, quantitative analysis and processing of trainees' psychological states are performed, and supplementary data are collected through questionnaires and interviews to generate psychological state evaluation data; based on the psychological state evaluation data, correlation analysis and processing of trainees' learning effects are performed to evaluate trainees' knowledge mastery, skill improvement speed, error types and their correction learning indicators to obtain learning effect evaluation data; using the learning effect evaluation data in combination with the psychological state evaluation data, a comprehensive and integrated analysis and processing is performed, and a combination of qualitative and quantitative methods is used to generate a psychological and learning effect analysis report.

[0130] In this embodiment, the preliminary performance evaluation results are used to conduct a detailed analysis of the behavioral patterns of trainees during training, identify psychological factors that may affect the performance of trainees, and obtain psychological state characteristics; based on the psychological state characteristics, combined with the model trained by educational psychology theory, the psychological state of the trainees is quantitatively analyzed and processed, and supplementary data is collected through questionnaires and interviews to generate psychological state evaluation data; based on the psychological state evaluation data, the learning effects of the trainees are analyzed and processed for correlation, and the learning indicators of the trainees' knowledge mastery, skill improvement speed, error types and their correction are evaluated to obtain learning effect evaluation data; the learning effect evaluation data is used, combined with the psychological state evaluation data, to conduct a comprehensive and integrated analysis and processing, and a combination of qualitative and quantitative methods is used to generate a psychological and learning effect analysis report.

[0131] In the embodiment of the present application, first, the system uses the preliminary performance evaluation results to conduct a detailed analysis of the behavioral patterns of trainees during training, identify psychological factors that may affect their performance, such as anxiety levels, self-confidence, etc., and form psychological state characteristics; secondly, based on these psychological state characteristics, the system combines the model trained by educational psychology theory, collects supplementary data through questionnaires and interviews, etc., conducts quantitative analysis of the trainees' psychological states, and generates psychological state evaluation data; thirdly, based on the psychological state evaluation data, the system conducts a correlation analysis on the trainees' learning effects, evaluates their knowledge mastery, skill improvement speed, and error types and corrections, and obtains learning effect evaluation data; finally, the system combines the psychological state evaluation data and the learning effect evaluation data, adopts a combination of qualitative and quantitative methods, conducts a comprehensive and integrated analysis of the trainees' overall performance, and finally generates a psychological and learning effect analysis report.

[0132] Here is a specific example:

[0133] Suppose a marketing specialist is receiving digital marketing skills training. First, the system uses the preliminary performance evaluation results to conduct a detailed analysis of his behavior patterns during the training period and finds that he shows a high level of anxiety when facing difficult tasks, and this anxiety may affect his task completion. Secondly, based on this psychological state characteristic, the system combines the model trained by educational psychology theory to collect more data about his psychological state through questionnaires and one-on-one interviews, and conducts quantitative analysis to generate psychological state evaluation data. The results show that he is prone to self-doubt under high-pressure environments. Thirdly, based on these psychological state evaluation data, the system conducts a correlation analysis on his learning effect, evaluates his knowledge mastery (such as understanding of social media advertising strategies), skill improvement speed (such as improvement in data analysis capabilities), and error types and corrections (such as repeating the same error many times), and obtains learning effect evaluation data, which shows that he has deficiencies in time management. Finally, the system combines the psychological state evaluation data and the learning effect evaluation data, adopts a combination of qualitative and quantitative methods, conducts a comprehensive and integrated analysis of his overall performance, and finally generates a psychological and learning effect analysis report, pointing out that he needs to strengthen his stress management and time management skills.

[0134] In order to meet the needs of comprehensive evaluation of students' learning effects, in some embodiments, the correlation analysis of students' learning effects is performed based on the psychological state evaluation data to evaluate students' knowledge mastery, skill improvement speed, error types and their correction status, and obtain learning effect evaluation data, including:

[0135] Based on the psychological state assessment data, correlation analysis is performed on the learning effects of the trainees, and the learning indicators of the trainees' knowledge mastery, skill improvement speed, error types and their corrections are evaluated to obtain learning effect evaluation data; using the psychological state assessment data, quantitative analysis is performed on the trainees' knowledge mastery during the training period, and the trainees' knowledge level is measured by objective data such as test scores and homework completion to obtain knowledge mastery scores; based on the knowledge mastery score and in combination with the trainees' historical operation records, dynamic evaluation is performed on the skill improvement speed, and the time difference and quality change for each task completion are calculated to generate a skill improvement curve; based on the skill improvement curve, the trainees' error types and their corrections are classified and statistically processed to identify common error patterns and the effectiveness of improvement measures to obtain an error improvement report; using the error improvement report and in combination with the psychological state assessment data, a comprehensive and integrated analysis is performed to evaluate the trainees' learning effects and generate learning effect evaluation data.

[0136] In this embodiment, based on the psychological state assessment data, the learning effect of the trainees is subjected to correlation analysis and processing, and the learning indicators of the trainees' knowledge mastery, skill improvement speed, error types and their correction are evaluated to obtain learning effect evaluation data; using the psychological state assessment data, the trainees' knowledge mastery during the training period is quantitatively analyzed and processed, and the trainees' knowledge level is measured by objective data such as test scores and homework completion to obtain a knowledge mastery score; based on the knowledge mastery score and in combination with the trainees' historical operation records, the skill improvement speed is dynamically evaluated and processed, and the time difference and quality change for each task completion are calculated to generate a skill improvement curve; based on the skill improvement curve, the trainees' error types and their correction are classified and statistically processed to identify common error patterns and the effectiveness of improvement measures to obtain an error improvement report; using the error improvement report and in combination with the psychological state assessment data, a comprehensive and integrated analysis and processing is performed, and a combination of qualitative and quantitative methods is used to evaluate the trainees' learning effect to generate learning effect evaluation data.

[0137] In the embodiment of the present application, first, the system performs a correlation analysis on the student's learning effect based on the psychological state assessment data, evaluates his knowledge mastery, skill improvement speed, error types and correction status, and forms learning effect evaluation data; secondly, the system uses the psychological state assessment data to conduct a quantitative analysis of the student's knowledge mastery, measures his knowledge level through objective data such as test scores and homework completion, and generates a knowledge mastery score; thirdly, based on the knowledge mastery score and in combination with the student's historical operation records, the system dynamically evaluates the skill improvement speed, calculates the time difference and quality change for each task completion, and generates a skill improvement curve; finally, based on the skill improvement curve, the system classifies and counts the student's error types and their correction status, identifies common error patterns and the effectiveness of improvement measures, generates an error improvement report, and combines the psychological state assessment data to adopt a combination of qualitative and quantitative methods to comprehensively evaluate the student's learning effect, and finally generates learning effect evaluation data.

[0138] Here is a specific example:

[0139] Suppose a financial analyst is receiving risk management skills training; first, based on the psychological state assessment data, the system finds that the student shows a high level of anxiety in a high-pressure environment, which may affect his learning effect; the system conducts a correlation analysis on his learning effect, evaluates his knowledge mastery (such as understanding of risk models), skill improvement speed (such as improvement in risk assessment ability), and error types and corrections (such as frequently making the same calculation errors), and forms preliminary learning effect evaluation data; secondly, the system uses the psychological state assessment data to conduct a quantitative analysis of his knowledge mastery through objective data such as test scores and homework completion. The results show that he has a solid theoretical knowledge, but there are deficiencies in practical application, and the system generates knowledge Mastery score; again, based on the knowledge mastery score and combined with his past historical operation records, the system dynamically evaluates his skill improvement speed, calculates the time difference and quality change for each task completion, and generates a skill improvement curve that shows his low efficiency under high pressure; finally, based on this skill improvement curve, the system classifies and counts his error types and their corrections, and finds that he often makes calculation errors when time is tight, and the improvement measures (such as practicing time management skills) are not effective, so an error improvement report is generated; the system combines these data and adopts a combination of qualitative and quantitative methods to comprehensively evaluate his learning effect, and finally generates detailed learning effect evaluation data, pointing out that he needs to strengthen psychological adjustment training and time management skills.

[0140] This application takes into account that in the prior art, due to the lack of accurate identification of trainees' behavior patterns and the ability to conduct multi-dimensional comprehensive evaluation, traditional performance evaluation methods can often only provide static, single-dimensional evaluation results, which are difficult to fully reflect the actual performance and potential problems of trainees. Therefore, the embodiment of the present invention proposes a method that combines nonlinear mapping functions and decision tree algorithms to solve the above problems. By identifying different behavior patterns and their corresponding performance levels, the performance of trainees is comprehensively evaluated from multiple dimensions (operation accuracy, task completion, time efficiency), and a preliminary performance evaluation result is generated.

[0141] Optionally, the standardized behavior data set is classified and analyzed by applying a decision tree algorithm to identify different behavior patterns and their corresponding performance levels, evaluate the trainee's performance from multiple dimensions such as operation accuracy, task completion, and time efficiency, and generate preliminary performance evaluation results, including:

[0142] The data expression ability is improved by nonlinear mapping functions, and the behavioral characteristics of trainees in different tasks are captured by analyzing the identified behavioral pattern vectors in combination with the trainees' historical operation records, providing a basis for generating preliminary performance evaluation results. pat Calculated by the following formula:

[0143] M pat =tanh(U pat ·F(X std )+V hist ·H hist +c pat )

[0144] Among them, M pat Represents the identified behavior pattern vector; U pat is the behavior feature transformation matrix, which is used to adjust the standardized behavior dataset X std The importance of each feature in; F(·) is a nonlinear feature mapping function used to enhance the expressiveness of the original data; V hist is the historical behavior impact matrix, which is used to quantify the impact of the trainee's historical operation records on the current behavior pattern; H hist is the characteristic vector of the student's historical operation records; c pat is a bias term used to calibrate the results of behavioral pattern recognition; tanh is a nonlinear activation function;

[0145] According to the behavior pattern vector M pat The system builds a comprehensive evaluation framework, integrating the quantitative indicators of behavior patterns and operation accuracy, task completion, and time efficiency. It accurately measures the impact of each dimension by introducing weight matrices and bias terms, and evaluates the matching degree between the behavior pattern and the current operation scenario context to generate a performance level score P. lev , where the performance level score P lev Calculated by the following formula:

[0146]

[0147] Among them, P lev Indicates the student's performance level score; W acc is the operation accuracy weight, which is used to evaluate the accuracy of the trainees in performing the task; acc It is a quantitative indicator of operation accuracy; W cmp is the task completion weight, which measures whether the trainees complete the task as required; T cmp is a quantitative indicator of task completion; W tim is the time efficiency weight, which indicates the time required for the trainee to complete the task; E tim It is a quantitative indicator of time efficiency; b lev is a bias term used to adjust the scoring benchmark; exp is an exponential function used to implement nonlinear transformations in the form of logistic regression; η is a context-related weight coefficient that affects the impact of context on the final score; G(·,·) is a context adaptation function used to evaluate the behavior pattern M pat and the current operation scene context C ctxThe matching degree;

[0148] Based on the performance level score P lev ,The system classifies the scoring results, and fine-tunes the evaluation content based on the ,immediate feedback provided by the tutor, verifies the effectiveness of the evaluation ,criteria through simulation tests, and generates a preliminary performance ,evaluation report.

[0149] The purpose of this formula is to achieve more accurate behavior pattern recognition and multi-dimensional performance evaluation. The system first improves the data expression ability through nonlinear mapping functions, and combines the trainees' historical operation records to identify the behavior pattern vector M. pat Then, the system constructs a comprehensive evaluation framework, integrating the behavior pattern with quantitative indicators of multiple dimensions such as operation accuracy, task completion, and time efficiency, and generates a performance level score P lev , ensuring that the evaluation results are both comprehensive and personalized. Finally, the system classifies the results according to the scoring, fine-tunes the evaluation content based on the immediate feedback provided by the tutor, verifies the effect through simulation tests, and generates a preliminary performance evaluation report.

[0150] The following is a brief introduction to the design reasons of each sub-item of the formula:

[0151] M pat =tanh(U pat ·F(X std )+V hist ·H hist +c pat )

[0152] U pat ·F(X std ) Overall design reason: This sub-item aims to transform the matrix U through the behavior characteristics pat and nonlinear feature mapping function F(·), improve the standardized behavior dataset X std This processing method not only enhances the complexity and expressiveness of the original data, but also captures the subtle behavioral differences of students in different tasks, which is helpful for generating accurate behavioral pattern vectors M. pat This provides a solid foundation. In this way, the system can more accurately identify the behavioral characteristics of trainees and provide more targeted performance evaluations.

[0153] V hist ·H hist Overall design reason: This sub-item introduces the historical behavior impact matrix V hist and the student's historical operation record feature vector H hist , to quantify the impact of the student’s past behavior on the current behavior pattern. This design takes into account the student’s historical operating habits and common mistakes, so that the generated behavior pattern vector M patIt not only reflects current performance, but also takes into account the trainee's growth trajectory and personal characteristics. This helps the system provide more personalized and dynamic performance evaluation and improve training effectiveness.

[0154] The following is a brief introduction to how to obtain the parameters of the formula:

[0155] Behavior feature conversion matrix U pat Obtained through training model; nonlinear feature mapping function F(·) is implemented by pre-trained or customized neural network layer; historical behavior influence matrix V hist The student's historical operation record feature vector H is obtained through machine learning optimization; hist Extract the student's historical operation records from the database; bias item c pat Automatically adjust during model training;

[0156] The following is a brief introduction to the design reasons of each sub-item of the formula:

[0157]

[0158] Overall design reason: This sub-item uses nonlinear transformation in the form of logistic regression to comprehensively evaluate the students' performance in three dimensions: operation accuracy, task completion, and time efficiency. acc ,W cmp ,W tim The importance of each dimension is measured separately, and the quantitative indicator A acc ,T cmp ,E tim It specifically reflects the actual performance of the students in these three aspects. lev Used to adjust the scoring benchmark. The core purpose of this sub-item is to quantify multi-dimensional performance into a comprehensive performance level score P lev , ensuring that the evaluation results are both comprehensive and comparable.

[0159] η·G(M pat ,C ctx ) Overall design reason: This sub-item evaluates the behavior pattern vector M through the context association weight coefficient η and the context adaptation function G(·,·) pat and the current operation scene context C ctx This design ensures that the generated performance level score P lev It not only considers the students’ behavioral performance, but also fully combines the specific conditions of the current operating environment. The context adaptation function G(·,·) is used to quantify the fit between the behavioral pattern and the operating scenario, making the evaluation results more in line with the actual situation and improving the system’s intelligence level and personalized guidance capabilities.

[0160] The following is a brief introduction to how to obtain the parameters of the formula:

[0161] Operation accuracy weight W acc Obtained through expert setting or machine learning optimization; quantitative indicator of operation accuracy A acc Calculated through objective data such as test scores or homework completion; Task completion weight W cmp Obtained through expert setting or machine learning optimization; quantitative indicator T of task completion cmp Through task completion statistics; time efficiency weight W tim Obtained through expert setting or machine learning optimization; quantitative index E of time efficiency tim By task completion time statistics; bias term b lev Automatically adjusted during model training; context association weight coefficient η is obtained through experimental setting or machine learning optimization; context adaptation function G(·,·) is implemented through pre-trained model or custom algorithm;

[0162] Example Assume that a software development engineer is receiving code review skills training; first, the system improves data expression capabilities through nonlinear mapping functions, and combines the trainee's historical operation records to identify the behavior pattern vector M pat , through the formula:

[0163] M pat =tanh(U pat ·F(X std )+V hist ·H hist +c pat )

[0164] Among them U pat is the behavioral feature conversion matrix obtained by training the model, F(·) is the nonlinear feature mapping function implemented by the pre-trained neural network layer, and V hist is the historical behavior influence matrix obtained through machine learning optimization, H hist Extract the trainee's historical operation records from the database, c pat Automatically adjusted during model training. The calculation result is: M pat = specific value Next, the system will calculate the behavior pattern vector M pat , build a comprehensive evaluation framework and generate a performance level score P lev , through the formula:

[0165]

[0166] The operation accuracy weight W acc =0.6, quantitative index of operation accuracy A acc =0.9, task completion weight Wcmp =0.3, quantitative indicator of task completion T cmp =0.8, time efficiency weight W tim =0.1, quantitative index of time efficiency E tim =0.7, bias term b lev =0.5, context association weight coefficient η = 0.4, and context adaptation function G(·,·) is implemented by pre-training model. The calculation result is: P lev =Specific value based on P lev ,The system classifies the scoring results, and fine-tunes the evaluation content based on the ,immediate feedback provided by the tutor, verifies the effect through simulation ,tests, and finally generates a preliminary performance evaluation report.

[0167] Through the above steps, the system not only provides a comprehensive and dynamic performance evaluation, but also reveals the performance characteristics of trainees in different dimensions, significantly improving the training effect and personal development. lev It is greater than the set threshold, so it shows that the student has performed well in multiple dimensions and can effectively guide the student to further improve his skills.

[0168] Figure 2 A structural diagram of a remote training system based on virtual reality is provided for the embodiment of the present application, such as Figure 2 As shown, the device comprises:

[0169] A customization module 21 is used to customize the training environment based on the trainees’ professional background, skill level and learning goals to obtain a simulation operation scenario;

[0170] The parsing module 22 is used to perform natural language processing algorithm parsing on the acquired language instructions according to the simulated operation scenario, generate operation prompts, and superimpose the operation prompts on the trainee's visual interface through augmented reality technology, so as to monitor and automatically correct the trainee's operation process in real time and obtain a personalized guidance plan;

[0171] The evaluation module 23 is used to perform decision tree algorithm classification analysis on the trainee's behavior data based on the personalized guidance plan, and to comprehensively evaluate the trainee's performance from multiple dimensions in combination with educational psychology theory to obtain a multi-dimensional performance evaluation report.

[0172] Figure 2 The virtual reality-based remote training system can be implemented Figure 1The implementation principle and technical effect of the remote training method based on virtual reality described in the embodiment shown are not repeated here. The specific way in which each module and unit performs operations in the remote training system based on virtual reality in the above embodiment has been described in detail in the embodiment of the method, and will not be elaborated here.

[0173] In one possible design, Figure 2 A remote training system based on virtual reality in the embodiment shown can be implemented as a computing device, such as Figure 3 As shown, the computing device may include a storage component 31 and a processing component 32;

[0174] The storage component 31 stores one or more computer instructions, wherein the one or more computer instructions are called and executed by the processing component 32 .

[0175] The processing component 32 is used to: utilize the professional background, skill level and learning goals of the trainees to personalize the training environment and obtain a simulated operation scenario; according to the simulated operation scenario, perform natural language processing algorithm analysis on the acquired language instructions to generate operation prompts, and superimpose the operation prompts on the trainees' visual interface through augmented reality technology, monitor and automatically correct the trainees' operation process in real time, and obtain a personalized guidance plan; based on the personalized guidance plan, perform decision tree algorithm classification and analysis on the trainees' behavior data, combine educational psychology theory, and comprehensively evaluate the trainees' performance from multiple dimensions to obtain a multi-dimensional performance evaluation report.

[0176] The processing component 32 may include one or more processors to execute computer instructions to complete all or part of the steps in the above method. Of course, the processing component may also be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors or other electronic components to perform the above method.

[0177] The storage component 31 is configured to store various types of data to support operations at the terminal. The storage component can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0178] Of course, the computing device may also include other components, such as input / output interfaces, display components, communication components, etc.

[0179] The input / output interface provides an interface between the processing component and the peripheral interface module, which may be an output device, an input device, etc.

[0180] The communication component is configured to facilitate, among other things, wired or wireless communications between the computing device and other devices.

[0181] Among them, the computing device can be a physical device or an elastic computing host provided by a cloud computing platform, etc. In this case, the computing device can refer to a cloud server, and the above-mentioned processing components, storage components, etc. can be basic server resources rented or purchased from the cloud computing platform.

[0182] The present application also provides a computer storage medium storing a computer program, wherein the computer program can achieve the above-mentioned Figure 1 A remote training method based on virtual reality in the illustrated embodiment.

[0183] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0184] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0185] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0186] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A remote training method based on virtual reality, characterized in that: include: The training environment is customized based on the trainees’ professional background, skill level and learning goals to obtain simulated operation scenarios. According to the simulated operation scenario, the acquired language instructions are parsed by a natural language processing algorithm to generate operation prompts, and the operation prompts are superimposed on the trainee's visual interface through augmented reality technology, so that the trainee's operation process is monitored and automatically corrected in real time to obtain a personalized guidance plan; Based on the personalized guidance program, the behavioral data of the trainees are classified and analyzed using a decision tree algorithm, and combined with educational psychology theory, the performance of the trainees is comprehensively evaluated from multiple dimensions to obtain a multi-dimensional performance evaluation report.

2. The method according to claim 1, characterized in that According to the simulation operation scenario, the acquired language instructions are parsed by a natural language processing algorithm to generate operation prompts, and the operation prompts are superimposed on the trainee's visual interface through augmented reality technology, and the trainee's operation process is monitored and automatically corrected in real time to obtain a personalized guidance plan, including: Utilizing the simulated operation scenario, the acquired language instructions are parsed using a natural language processing algorithm to obtain operation prompts; According to the operation prompt, the operation prompt is superimposed on the trainee's visual interface through augmented reality technology to generate an augmented reality guidance interface; Based on the augmented reality guidance interface, the trainee's operation process is monitored in real time and automatically corrected to obtain real-time monitoring data; The real-time monitoring data is used in combination with the trainees' specific behavior patterns and error types to perform optimization algorithm analysis and processing on the personalized learning path to generate a personalized guidance plan.

3. The method according to claim 2, characterized in that According to the operation prompt, the operation prompt is superimposed on the trainee's visual interface through augmented reality technology to generate an augmented reality guidance interface, including: According to the operation prompt, configure the image recognition accuracy, transparency of the virtual object and interaction mode parameters in the augmented reality technology to obtain display parameters; Based on the display parameters, the trainee's visual interface is embedded with augmented reality elements, and the operation prompts are accurately superimposed on the corresponding real objects or operation points in the trainee's field of vision in the form of virtual objects to generate a preliminary augmented reality guidance interface; Using the preliminary augmented reality guidance interface, visually annotate key positions and actions in the trainee's operating environment to obtain real-time annotation information; According to the real-time annotation information and in combination with the progress and changes of the trainees in actual operation, the preliminary augmented reality guidance interface is dynamically updated to generate an augmented reality guidance interface.

4. The method according to claim 3, characterized in that Based on the display parameters, the trainee's visual interface is embedded with augmented reality elements, and the operation prompt is accurately superimposed on the corresponding real object or operation point in the trainee's field of vision in the form of a virtual object to generate a preliminary augmented reality guidance interface, including: Based on the display parameters, the trainee's visual interface is embedded with augmented reality elements, and the relative position of the virtual object in the real world is determined by spatial positioning technology to obtain position mapping information of the virtual object; According to the position mapping information, the operation prompt is accurately superimposed on the corresponding real object or operation point in the trainee's field of view in the form of a virtual object, ensuring that the virtual object is accurately aligned with the real object, and generating a preliminary virtual object layout; Using the preliminary virtual object layout, interactively designing the augmented reality elements to obtain interactive augmented reality elements; Based on the interactive augmented reality elements and in combination with the trainees' operating habits and environmental conditions, the preliminary augmented reality guidance interface is personalized and adjusted to generate a preliminary augmented reality guidance interface.

5. The method according to claim 1, characterized in that: Based on the personalized guidance program, the decision tree algorithm is used to classify and analyze the trainee's behavior data, and combined with educational psychology theory, the trainee's performance is comprehensively evaluated from multiple dimensions to obtain a multi-dimensional performance evaluation report, including: Based on the personalized guidance program, the behavioral data of trainees during the training period are systematically collected and structured to obtain a standardized behavioral data set; Based on the standardized behavior data set, a decision tree algorithm is applied to perform classification analysis and processing, different behavior patterns and their corresponding performance levels are identified, the trainees' performance is evaluated from multiple dimensions such as operation accuracy, task completion, and time efficiency, and a preliminary performance evaluation result is generated; Using the preliminary performance evaluation results and combining them with educational psychology theory, we conduct an in-depth analysis of the students' psychological state and learning effects, and obtain a psychological and learning effect analysis report; Based on the psychological and learning effect analysis report, multiple dimensions of evaluation indicators are integrated, and factors such as the students' learning progress, error types and corrections are taken into consideration to conduct a comprehensive and integrated evaluation of the students' overall performance and generate a multi-dimensional performance evaluation report.

6. The method according to claim 5, characterized in that The preliminary performance evaluation results are used in combination with educational psychology theory to conduct in-depth analysis and processing of the students' psychological state and learning effects, and obtain a psychological and learning effect analysis report, including: Using the preliminary performance evaluation results, the trainees' behavior patterns during the training period are carefully analyzed to identify the psychological factors that may affect the trainees' performance and obtain psychological state characteristics; According to the characteristics of the psychological state, combined with the model trained by educational psychology theory, the psychological state of the trainees is quantitatively analyzed and processed, and supplementary data is collected through questionnaires and interviews to generate psychological state assessment data; Based on the psychological state assessment data, correlation analysis is performed on the learning effect of the trainees to assess the trainees' knowledge mastery, skill improvement speed, error types and their correction status, and obtain learning effect assessment data; The learning effect evaluation data is used in combination with the psychological state evaluation data to conduct a comprehensive and integrated analysis and processing, and a combination of qualitative and quantitative methods is used to generate a psychological and learning effect analysis report.

7. The content according to claim 6, characterized in that: Based on the psychological state evaluation data, correlation analysis is performed on the learning effect of the trainees to evaluate the learning indicators of the trainees' knowledge mastery, skill improvement speed, error types and their correction status, and obtain learning effect evaluation data, including: Based on the psychological state assessment data, correlation analysis is performed on the learning effect of the trainees to assess the trainees' knowledge mastery, skill improvement speed, error types and their correction status, and obtain learning effect assessment data; Utilizing the psychological state assessment data, quantitatively analyzing and processing the trainees' knowledge mastery during the training period, measuring the trainees' knowledge level through objective data such as test scores and homework completion, and obtaining a knowledge mastery score; Based on the knowledge mastery score and the trainee's historical operation records, the skill improvement speed is dynamically evaluated and processed, the time difference and quality change of each task completion are calculated, and a skill improvement curve is generated; Based on the skill improvement curve, the types of students' errors and their corrections are classified and statistically processed to identify common error patterns and the effectiveness of improvement measures, and obtain an error improvement report; The error improvement report is used in combination with the psychological state assessment data to conduct a comprehensive and integrated analysis and processing, and a combination of qualitative and quantitative methods is used to evaluate the learning effects of the trainees and generate learning effect assessment data.

8. A remote training system based on virtual reality, characterized in that: include: The customization module is used to customize the training environment based on the trainees’ professional background, skill level and learning goals to obtain simulation operation scenarios; The parsing module is used to perform natural language processing algorithm parsing on the acquired language instructions according to the simulated operation scenario, generate operation prompts, and superimpose the operation prompts on the trainee's visual interface through augmented reality technology, so as to monitor and automatically correct the trainee's operation process in real time and obtain a personalized guidance plan; The evaluation module is used to perform decision tree algorithm classification analysis on the trainees' behavior data based on the personalized guidance plan, and to comprehensively evaluate the trainees' performance from multiple dimensions in combination with educational psychology theory to obtain a multi-dimensional performance evaluation report.

9. A computing device, characterized in that It comprises a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a remote training method based on virtual reality as described in any one of claims 1 to 7.

10. A computer storage medium, characterized in that: A computer program is stored, and when the computer program is executed by a computer, a remote training method based on virtual reality as claimed in any one of claims 1 to 7 is implemented.

Images