Rail transit training method and system based on AR intelligent head-mounted terminal

Abstract

The invention relates to the technical field of rail traffic safety operation and maintenance, in particular to a rail traffic training method and system based on an AR intelligent head-mounted terminal, and the method comprises the steps: classifying rail traffic equipment into high-risk equipment and common equipment based on a preset fault library; generating a virtual training scene aligned with the physical space coordinates; based on a preset student skill portrait, a preset fault library and a preset operation instruction library, configuring a modular training course in the virtual training scene; acquiring operation data when a student executes the modular training course through an AR head-mounted terminal, wherein the operation data comprises spatio-temporal trajectory data, decision delay and an interaction log; based on the spatio-temporal trajectory data and the decision delay, a compliance score is generated through fusion; and generating a skill short board positioning report. The method has the advantages that effectiveness and safety of operation training of the rail transit equipment are improved, and high-precision positioning and targeted strengthening of short skill of trainees are achieved.

Description

Technical Field

[0001] The present application relates to the technical field of rail transit safe operation and maintenance, and in particular to a rail transit training method and system based on an AR smart head-mounted terminal. Background Art

[0002] In the rail transit sector, equipment operation training typically relies on traditional hands-on drills or desktop simulation systems, which suffer from the following drawbacks:

[0003] Insufficient real-world reproduction: Traditional simulation systems struggle to accurately reproduce the physical details and operating space of high-risk equipment (such as high-voltage circuit breakers and track switches), resulting in insufficient training for trainees in responding to complex fault scenarios.

[0004] Static training content: The course design lacks adaptation to the individual skill differences of trainees, and is unable to dynamically adjust the difficulty and priority of training to target weak areas.

[0005] Single evaluation dimension: Existing systems only record the correctness of operational results, without quantifying the compliance of operational paths, decision-making timeliness, or the accuracy of instruction sequences, making it difficult to identify underlying skill gaps.

[0006] Safety risks are hidden: Equipment damage or accident consequences that may be caused by trainees' operational errors are not explicitly linked, and a risk awareness reinforcement mechanism cannot be formed.

[0007] Therefore, improvements are needed. Summary of the Invention

[0008] In order to improve the effectiveness and safety of rail transit equipment operation training and achieve high-precision positioning and targeted reinforcement of trainees' skill shortcomings, this application provides a rail transit training method and system based on an AR smart head-mounted terminal.

[0009] The above-mentioned invention objective of this application is achieved through the following technical solutions:

[0010] A rail transit training method based on an AR smart head-mounted terminal, comprising:

[0011] Classify rail transit equipment into high-risk equipment and ordinary equipment based on the preset fault library;

[0012] Collect millimeter-level laser scanning data of high-risk equipment, centimeter-level photogrammetry data of ordinary equipment, and physical space coordinate data to generate virtual training scenes aligned with physical space coordinates;

[0013] Based on the preset trainee skill profile, the preset fault library, and the preset operation instruction library, a modular training course is configured in the virtual training scenario;

[0014] Collecting operational data of trainees while they execute the modular training course through an AR head-mounted terminal, wherein the operational data includes spatiotemporal trajectory data, decision delay, and interaction log;

[0015] Based on the spatiotemporal trajectory data and the decision delay, generating a compliance score by fusion;

[0016] The operation data and the compliance score are integrated to generate a skill shortcoming location report.

[0017] In a preferred embodiment, the step of classifying rail transit equipment into high-risk equipment and ordinary equipment based on a preset fault library includes:

[0018] Based on the equipment failure probability and accident consequence severity in the preset fault database, rail transit equipment is classified into high-risk equipment and ordinary equipment;

[0019] The high-risk equipment meets the following conditions: the probability of equipment failure ≥ a preset probability threshold and the severity of the accident consequences ≥ a preset severity threshold.

[0020] In a preferred embodiment, the step of configuring a modular training course in the virtual training scenario based on a preset trainee skill profile, a preset fault library, and a preset operation instruction library includes:

[0021] Analyze the preset student skill profile to obtain the student's operational proficiency score for each fault type in the preset fault library;

[0022] Sort the operation proficiency scores from low to high, and combine them with the fault frequency of the corresponding fault types from high to low to generate a progressive task sequence;

[0023] Among them, priority training tasks are generated for fault types whose operation proficiency score is ≤ a preset first threshold and whose fault occurrence frequency is ≥ a preset second threshold.

[0024] In a preferred embodiment, the step of configuring a modular training course in the virtual training scenario based on the preset trainee skill profile, the preset fault library, and the preset operation instruction library further includes:

[0025] Based on the progressive task sequence, extracting the operation instruction steps corresponding to the task from a preset operation instruction library;

[0026] The operation instruction steps are combined according to the progressive task sequence and associated with the equipment operation coordinates in the virtual training scene to generate a modular training course.

[0027] In a preferred embodiment, the step of generating a compliance score based on the spatiotemporal trajectory data and the decision delay comprises:

[0028] Get the key operation nodes in the preset standard path;

[0029] Calculate the Euclidean distance between the trainee's operating position and the preset standard position at the key operating node;

[0030] The distance deviation of all key operation nodes is weighted averaged to calculate the trajectory deviation degree;

[0031] When the decision delay exceeds a preset decision delay threshold, a timeliness deduction item is generated based on a preset timeout ratio formula;

[0032] The trajectory deviation degree and the time deduction item are combined based on a preset fusion formula to generate a compliance score.

[0033] In a preferred embodiment, the step of integrating the operational data with the compliance score to generate a skill shortcoming location report includes:

[0034] Locating weak operational steps based on the compliance scores;

[0035] Input the weak operation steps into a preset equipment-operation step mapping relationship table to match the associated rail transit equipment module;

[0036] Comparing the decision delay data with a preset standard decision duration threshold for a corresponding operation step to identify decision defects;

[0037] Extract the operation instruction sequence in the interaction log:

[0038] Performing a timing comparison between the operation instruction sequence and the standard instruction sequence in the preset operation instruction library;

[0039] If missing instructions, reversed order, or incorrect parameters are detected, it will be marked as an incorrect operation sequence;

[0040] Associating the erroneous operation sequence with corresponding risk items in a preset fault library, wherein the risk items include equipment damage probability and accident level;

[0041] The equipment modules, decision defect types and risk items are integrated to generate a skill shortcoming location report.

[0042] In a preferred embodiment, the step of comparing the decision delay data with a preset standard decision duration threshold for a corresponding operation step to identify decision defects includes:

[0043] If the decision delay exceeds the standard decision time threshold of the corresponding operation step, it is marked as a "slow response" decision defect;

[0044] If the decision delay is lower than the standard decision time threshold of the corresponding operation step, it is marked as a "hasty decision" decision defect item.

[0045] The second object of the present invention is achieved through the following technical solutions:

[0046] A rail transit training system based on an AR smart head-mounted terminal, comprising:

[0047] Classification module: classifies rail transit equipment into high-risk equipment and ordinary equipment based on the preset fault library;

[0048] Acquisition module: collects millimeter-level laser scanning data of high-risk equipment, centimeter-level photogrammetry data of ordinary equipment, and physical space coordinate data, and generates a virtual training scene aligned with the physical space coordinates;

[0049] Configuration module: Based on the preset trainee skill profile, preset fault library, and preset operation instruction library, configure modular training courses in the virtual training scenario;

[0050] Data module: collects operational data of trainees while they are executing the modular training course through AR head-mounted terminals. The operational data includes spatiotemporal trajectory data, decision delay, and interaction logs.

[0051] Fusion module: Based on the spatiotemporal trajectory data and the decision delay, generating a compliance score through fusion;

[0052] Generation module: integrates the operation data and the compliance score to generate a skill shortcoming positioning report.

[0053] The third objective of this application is achieved through the following technical solutions:

[0054] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the above-mentioned rail transit training method based on an AR smart head-mounted terminal are implemented.

[0055] The fourth objective of this application is achieved through the following technical solutions:

[0056] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the above-mentioned rail transit training method based on an AR smart head-mounted terminal.

[0057] In summary, this application includes at least one of the following beneficial technical effects:

[0058] First, data is collected based on the risk level of the equipment (high-risk equipment uses millimeter-level laser scanning to capture mechanical details, while ordinary equipment uses centimeter-level photogrammetry to balance efficiency). This is then combined with physical space coordinates to construct a virtual scene aligned with the real environment at the millimeter level. Modular courses are then dynamically configured based on the trainee's skill characteristics, with AR terminals recording operation trajectories, response delays, and interactive behaviors in real time. Finally, a compliance score is generated through multi-dimensional data fusion, and specific skill shortcomings are identified. This method forms a closed-loop training system of "precise modeling - intelligent class grouping - real-time evaluation - defect feedback," which improves training efficiency while ensuring practical safety, achieving full coverage of factors from macro-operational processes to micro-mechanical structures, and promoting iterative optimization of trainee skills through quantitative evaluation indicators. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 This is an overall flow chart of a rail transit training method based on an AR smart head-mounted terminal in this application;

[0060] Figure 2 This is a flow chart of an embodiment of step S10 of a rail transit training method based on an AR smart head-mounted terminal in the present application;

[0061] Figure 3 This is a flow chart of an embodiment of step S30 of a rail transit training method based on an AR smart head-mounted terminal in the present application;

[0062] Figure 4 This is a flowchart of another embodiment of step S30 of a rail transit training method based on an AR smart head-mounted terminal in the present application;

[0063] Figure 5 This is a flow chart of an embodiment of step S50 of a rail transit training method based on an AR smart head-mounted terminal in the present application;

[0064] Figure 6 This is a flow chart of an embodiment of step S60 of a rail transit training method based on an AR smart head-mounted terminal in the present application;

[0065] Figure 7 This is a flow chart of an embodiment of step S603 of a rail transit training method based on an AR smart head-mounted terminal in the present application;

[0066] Figure 8 This is a principle block diagram of a computer device of the present application. DETAILED DESCRIPTION

[0067] The following is combined with Figure 1-8 This application is described in further detail.

[0068] In one embodiment, if Figure 1As shown, the present application discloses a rail transit training method based on an AR smart head-mounted terminal, which specifically includes the following steps:

[0069] S10: Classifying rail transit equipment into high-risk equipment and ordinary equipment based on a preset fault library;

[0070] S20: Collect millimeter-level laser scanning data of high-risk equipment, centimeter-level photogrammetry data of ordinary equipment, and physical space coordinate data to generate a virtual training scene aligned with the physical space coordinates;

[0071] S30: Based on the preset trainee skill profile, the preset fault library, and the preset operation instruction library, a modular training course is configured in the virtual training scenario;

[0072] S40: collecting, through an AR head-mounted terminal, operational data of the trainee when executing the modular training course, the operational data including spatiotemporal trajectory data, decision delay, and interaction log;

[0073] S50: Based on the spatiotemporal trajectory data and the decision delay, generate a compliance score by fusing them;

[0074] S60: Integrate the operation data with the compliance score to generate a skill shortcoming location report.

[0075] In this embodiment, data is first collected based on the risk level of the equipment (millimeter-level laser scanning is used to capture mechanical details for high-risk equipment, and centimeter-level photogrammetry is used to balance efficiency for ordinary equipment). Physical space coordinates are then combined to construct a virtual scene aligned with the real environment at the millimeter level. Modular courses are then dynamically configured based on the trainee's skill characteristics, and operation trajectories, response delays, and interactive behaviors are recorded in real time through AR terminals. Finally, a compliance score is generated through multi-dimensional data fusion, and specific skill shortcomings are identified. This method forms a closed-loop training system of "precise modeling - intelligent class grouping - real-time evaluation - defect feedback." This system improves training efficiency while ensuring practical safety, achieving full coverage from macro-operational processes to micro-mechanical structures, and promoting iterative optimization of trainee skills through quantitative evaluation indicators.

[0076] like Figure 2 As shown, step S10 includes:

[0077] S101: Classify rail transit equipment into high-risk equipment and ordinary equipment based on the equipment failure probability and accident consequence severity in a preset fault library;

[0078] S102: The high-risk equipment satisfies the following conditions: equipment failure probability ≥ preset probability threshold and accident consequence severity ≥ preset severity threshold.

[0079] In this embodiment, a dual-factor assessment model of equipment failure probability (P) - accident consequence severity (S) is adopted, and the equipment risk is quantified through the formula Risk = P × S. For example, when the equipment meets both P ≥ 0.05 times / year (preset probability threshold) and S ≥ Level III (preset severity threshold), the system automatically marks it as high-risk equipment. Through this approach, the precise delivery of training resources and pre-risk management are achieved, which not only provides a high-fidelity environment for subsequent practical training, but also focuses most of the training time on key safety nodes through the risk grading mechanism, significantly improving the efficiency of high-risk equipment failure simulation and the risk response capabilities of trainees.

[0080] like Figure 3 As shown, step S30 includes:

[0081] S301: Analyze the preset student skill profile to obtain the student's operation proficiency score for each fault type in the preset fault library;

[0082] S302: Sort the operation proficiency scores from low to high, and generate a progressive task sequence based on the fault occurrence frequencies of the corresponding fault types from high to low;

[0083] S303: For the fault types whose operation proficiency score is less than or equal to a preset first threshold and whose fault occurrence frequency is greater than or equal to a preset second threshold, a priority training task is generated.

[0084] In this embodiment, a dynamic sorting matrix is ​​established based on the operational proficiency score (0-100 points) and the frequency of faults (times / year) of each fault type in the preset trainee skill profile. For example, when a certain fault type satisfies both the trainee proficiency score ≤ 60 points (preset first threshold) and the fault frequency ≥ 2 times / year (preset second threshold), the system automatically marks it as a priority training task. This process enables the training content to break away from the static mode and form a personalized course driven by the two-way "capability shortcoming-risk hotspot", ensuring that trainees prioritize strengthening the most needed practical skills. Through the intelligent generation of progressive task sequences, the system not only shortens the key skill compliance cycle, but also ensures the consistency of the operational logic of the training scene with the real equipment environment through the spatiotemporal binding of operating instructions and virtual scene coordinates, ultimately promoting the improvement of trainees' handling efficiency in high-frequency fault scenarios and the leap in overall operational compliance rate.

[0085] like Figure 4 As shown, step S30 further includes:

[0086] S304: Based on the progressive task sequence, extracting operation instruction steps corresponding to the task from a preset operation instruction library;

[0087] S305: Combining the operation instruction steps according to the progressive task sequence and associating them with the device operation coordinates in the virtual training scene to generate a modular training course.

[0088] In this embodiment, by constructing a spatiotemporal mapping mechanism between operating instructions and virtual scenes, the structured presentation of training content and the precise restoration of practical operations are achieved. Taking the "ATP host board failure" task as an example, the system first extracts a standardized handling process from the preset operating instruction library (as shown in S304), which includes three core steps of "restarting the host, replacing the board, and data verification"; then, through S305, these steps are combined in a progressive task sequence and associated with the physical coordinates of the board slots in the virtual scene to ensure that when students perform operations in the AR environment, their gesture trajectories and component interaction positions are completely consistent with the real equipment. This modular course generation method standardizes the teaching process, significantly improves the compliance of practical operations through the hard constraint mechanism of operating coordinates, and optimizes the training efficiency of key components of the equipment, forming a complete teaching closed loop from task decomposition to scene implementation.

[0089] like Figure 5 As shown, step S50 includes:

[0090] S501: Obtain key operation nodes in a preset standard path;

[0091] S502: Calculating the Euclidean distance between the trainee's operating position and the preset standard position at the key operating node;

[0092] S503: Perform weighted averaging of distance deviations of all key operation nodes to calculate trajectory deviation;

[0093] S504: When the decision delay exceeds a preset decision delay threshold, a timeliness deduction item is generated based on a preset timeout ratio formula;

[0094] S505: Generate a compliance score based on the trajectory deviation and the timeliness deduction item based on a preset fusion formula.

[0095] In this embodiment, a multi-dimensional quantitative analysis of operational compliance is achieved by constructing a composite evaluation model combining spatial trajectory and temporal response. This step first extracts key operational nodes within a preset standard path as evaluation benchmarks. In the spatial dimension, the Euclidean distance deviation between the trainee's actual position and the standard position is calculated. A dynamic weighting coefficient (e.g., high-voltage equipment operation points are weighted higher than indicator light confirmation points) is assigned based on the equipment's safety level to form a trajectory deviation index. Simultaneously, a decision-making delay penalty mechanism is established in the temporal dimension. When the operational response time exceeds a preset threshold, a timeliness penalty based on the timeout ratio is triggered. By nonlinearly integrating the geometric deviation of the spatial trajectory with the timeliness of the temporal response, the system can accurately distinguish between different defect types, such as "a trajectory deviation of 15 cm but a timely decision" and "a trajectory deviation of 5 cm but a timeout of 5 seconds." This composite evaluation approach not only improves the accuracy of operational compliance assessments but also, through a spatiotemporal coupled quantitative feedback mechanism, effectively guides trainees to optimize their operational strategies, improving emergency response efficiency while ensuring safety. It also provides a quantifiable diagnostic basis for subsequent skill gap identification, forming a closed-loop training system of "evaluation-feedback-improvement."

[0096] like Figure 6 As shown, step S60 includes:

[0097] S601: Locate weak operation steps based on the compliance score;

[0098] S602: Input the weak operation steps into a preset equipment-operation step mapping relationship table to match the associated rail transit equipment module;

[0099] S603: Compare the decision delay data with a preset standard decision time threshold for the corresponding operation step to identify decision defects;

[0100] S604: Extracting the operation instruction sequence in the interaction log;

[0101] S605: Performing a timing comparison between the operation instruction sequence and the standard instruction sequence in the preset operation instruction library;

[0102] S606: If it is detected that the instruction is missing, the order is reversed, or the parameters are wrong, it is marked as an incorrect operation sequence;

[0103] S607: Associating the erroneous operation sequence with corresponding risk items in a preset fault database, wherein the risk items include equipment damage probability and accident level;

[0104] S608: Integrate the equipment modules, decision defect types and risk items to generate a skill shortcoming location report.

[0105] In this embodiment, by constructing a multi-dimensional skill diagnosis model, accurate positioning and risk association analysis of trainees' operational defects are achieved. This step first identifies weak operational links based on compliance scores, and traces them back to specific equipment modules in combination with the equipment-operation mapping table, forming a two-way positioning of "step-equipment"; at the same time, through decision delay analysis and operation instruction sequence comparison, the system can identify two typical defect modes: decision hysteresis and operation disorder, and further associate them with risk level parameters in the preset fault library. The final generated shortcoming positioning report not only reveals the spatial distribution (equipment module) and temporal characteristics (decision delay) of skill defects, but also quantifies the probability of equipment damage and accident level that may be caused by incorrect operation. This multi-dimensional diagnostic mechanism significantly improves the targeted nature of training intervention measures. By integrating operational defects, equipment risks and training feedback loops, it optimizes the efficiency of training resource allocation and provides trainees with improvement suggestions with risk warning functions, forming a complete improvement path from skill assessment to safety enhancement.

[0106] like Figure 7 As shown, step S603 includes:

[0107] SE1: If the decision delay exceeds the standard decision time threshold of the corresponding operation step, it is marked as a "slow response" decision defect;

[0108] SE2: If the decision delay is lower than the standard decision time threshold of the corresponding operation step, it is marked as a "hasty decision" decision defect item.

[0109] In this embodiment, a two-dimensional diagnostic mechanism for decision latency is established to accurately characterize operational response quality and provide early warning of risks. This step establishes a dynamic response assessment interval based on a preset standard decision latency threshold. When a trainee's decision latency exceeds the threshold (e.g., a 15-second delay in handling a signal system fault), the system automatically identifies it as a "slow response" defect, indicating insufficient risk assessment or operational hesitation in complex scenarios. When the latency falls below the threshold (e.g., an emergency response for a catenary power outage is initiated 8 seconds early), it is flagged as a "hasty decision" defect, indicating possible step skipping or oversight of risk confirmation. This two-way diagnostic mechanism not only improves the accuracy of identifying decision defects but also provides dual guidance for training interventions through correlation analysis with equipment modules and risk items. For "slow response" defects, emergency response plan drills and risk assessment training are strengthened, while for "hasty decision" issues, standardized processes are reinforced and dual confirmation mechanisms are developed. Ultimately, this improves trainees' decision-making stability in complex rail transit scenarios and reduces the rate of equipment misoperation caused by human factors.

[0110] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0111] In one embodiment, a rail transit training system based on an AR smart head-mounted terminal is provided. The rail transit training system based on an AR smart head-mounted terminal corresponds to the rail transit training method based on an AR smart head-mounted terminal in the above embodiment. The rail transit training system based on an AR smart head-mounted terminal includes:

[0112] Classification module: classifies rail transit equipment into high-risk equipment and ordinary equipment based on the preset fault library;

[0113] Acquisition module: collects millimeter-level laser scanning data of high-risk equipment, centimeter-level photogrammetry data of ordinary equipment, and physical space coordinate data, and generates a virtual training scene aligned with the physical space coordinates;

[0114] Configuration module: Based on the preset trainee skill profile, preset fault library, and preset operation instruction library, configure modular training courses in the virtual training scenario;

[0115] Data module: collects operational data of trainees while they are executing the modular training course through AR head-mounted terminals. The operational data includes spatiotemporal trajectory data, decision delay, and interaction logs.

[0116] Fusion module: Based on the spatiotemporal trajectory data and the decision delay, generating a compliance score through fusion;

[0117] Generation module: integrates the operation data and the compliance score to generate a skill shortcoming positioning report.

[0118] Optionally, also include:

[0119] Module 1: Based on the equipment failure probability and accident consequence severity in the preset fault library, rail transit equipment is classified into high-risk equipment and ordinary equipment;

[0120] Module 2: The high-risk equipment meets the following requirements: the probability of equipment failure ≥ the preset probability threshold and the severity of the accident consequences ≥ the preset severity threshold.

[0121] Optionally, also include:

[0122] The third module: Analyze the preset student skill profile and obtain the student's operation proficiency score for each fault type in the preset fault library;

[0123] Module 4: Sort the operation proficiency scores from low to high, and generate a progressive task sequence based on the fault frequency of the corresponding fault type from high to low;

[0124] The fifth module: Among them, for the fault types whose operation proficiency score is ≤ the preset first threshold and the fault occurrence frequency is ≥ the preset second threshold, priority training tasks are generated.

[0125] Optionally, also include:

[0126] The sixth module: extracting the operation instruction steps of the corresponding task from the preset operation instruction library based on the progressive task sequence;

[0127] Module 7: Combine the operation instruction steps according to the progressive task sequence and associate them with the equipment operation coordinates in the virtual training scene to generate a modular training course.

[0128] Optionally, also include:

[0129] Module 8: Obtain key operation nodes in the preset standard path;

[0130] Module 9: Calculate the Euclidean distance between the trainee's operating position and the preset standard position at the key operating node;

[0131] Module 10: Perform weighted average of the distance deviations of all key operation nodes and calculate the trajectory deviation degree;

[0132] Module 11: When the decision delay exceeds a preset decision delay threshold, a timeliness deduction item is generated based on a preset timeout ratio formula;

[0133] Module 12: Generate a compliance score based on the trajectory deviation and the time deduction items based on a preset fusion formula.

[0134] Optionally, also include:

[0135] Module 13: Locating weak operational steps based on the compliance score;

[0136] Module 14: Input the weak operation steps into the preset equipment-operation step mapping relationship table and match the associated rail transit equipment module;

[0137] Module 15: Compare the decision delay data with the preset standard decision time threshold of the corresponding operation step to identify decision defects;

[0138] Module 16: Extracting the operation instruction sequence in the interaction log:

[0139] Module 17: Perform timing comparison between the operation instruction sequence and the standard instruction sequence in the preset operation instruction library;

[0140] Module 18: If it is detected that the instruction is missing, the order is reversed, or the parameter is wrong, it will be marked as an incorrect operation sequence;

[0141] Module 19: Associating the erroneous operation sequence with the corresponding risk item in the preset fault library, wherein the risk item includes the probability of equipment damage and the accident level;

[0142] Module 20: Integrate the equipment modules, decision defect types and risk items to generate a skill shortcoming positioning report.

[0143] Optionally, also include:

[0144] Module 21: If the decision delay exceeds the standard decision time threshold of the corresponding operation step, it is marked as a "slow response" decision defect item;

[0145] Module 22: If the decision delay is lower than the standard decision time threshold of the corresponding operation step, it is marked as a "hasty decision" decision defect item.

[0146] Regarding the specific definition of a rail transit training system based on an AR smart head-mounted terminal, please refer to the definition of a rail transit training method based on an AR smart head-mounted terminal above, which will not be repeated here. The various modules in the above-mentioned rail transit training system based on an AR smart head-mounted terminal can be implemented in whole or in part through software, hardware and their combination. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0147] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 8 As shown. The computer device includes a processor, a memory, a network interface and a database connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, a rail transit training method based on an AR smart head-mounted terminal is implemented.

[0148] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, a rail transit training method based on an AR smart head-mounted terminal is implemented.

[0149] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored, and the computer program is executed by a processor to provide a rail transit training method based on an AR smart head-mounted terminal.

[0150] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0151] Those skilled in the art will clearly understand that for the sake of convenience and brevity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0152] The above-described embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. 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. 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 various embodiments of the present application, and should all be included in the scope of protection of the present application.

Claims

1. A rail transit training method based on an AR smart head-mounted terminal, characterized in that: include: Classify rail transit equipment into high-risk equipment and ordinary equipment based on the preset fault library; Collect millimeter-level laser scanning data of high-risk equipment, centimeter-level photogrammetry data of ordinary equipment, and physical space coordinate data to generate virtual training scenes aligned with physical space coordinates; Based on the preset trainee skill profile, the preset fault library, and the preset operation instruction library, a modular training course is configured in the virtual training scenario; Collecting operational data of trainees while they execute the modular training course through an AR head-mounted terminal, wherein the operational data includes spatiotemporal trajectory data, decision delay, and interaction log; Based on the spatiotemporal trajectory data and the decision delay, generating a compliance score by fusion; The operation data and the compliance score are integrated to generate a skill shortcoming location report.

2. A rail transit training method based on an AR smart head-mounted terminal according to claim 1, characterized in that: The step of classifying rail transit equipment into high-risk equipment and ordinary equipment based on a preset fault library includes: Based on the equipment failure probability and accident consequence severity in the preset fault database, rail transit equipment is classified into high-risk equipment and ordinary equipment; The high-risk equipment meets the following conditions: the probability of equipment failure ≥ a preset probability threshold and the severity of the accident consequences ≥ a preset severity threshold.

3. A rail transit training method based on an AR smart head-mounted terminal according to claim 1, characterized in that: The step of configuring a modular training course in the virtual training scenario based on the preset trainee skill profile, the preset fault library, and the preset operation instruction library includes: Analyze the preset student skill profile to obtain the student's operational proficiency score for each fault type in the preset fault library; Sort the operation proficiency scores from low to high, and combine them with the fault frequency of the corresponding fault types from high to low to generate a progressive task sequence; Among them, priority training tasks are generated for fault types whose operation proficiency score is ≤ a preset first threshold and whose fault occurrence frequency is ≥ a preset second threshold.

4. A rail transit training method based on an AR smart head-mounted terminal as claimed in claim 3, characterized in that: The step of configuring a modular training course in the virtual training scenario based on the preset trainee skill profile, the preset fault library, and the preset operation instruction library also includes: Based on the progressive task sequence, extracting the operation instruction steps corresponding to the task from a preset operation instruction library; The operation instruction steps are combined according to the progressive task sequence and associated with the equipment operation coordinates in the virtual training scene to generate a modular training course.

5. The rail transit training method based on AR smart head-mounted terminal according to claim 1, characterized in that: The S5 includes: Get the key operation nodes in the preset standard path; Calculate the Euclidean distance between the trainee's operating position and the preset standard position at the key operating node; The distance deviation of all key operation nodes is weighted averaged to calculate the trajectory deviation degree; When the decision delay exceeds a preset decision delay threshold, a timeliness deduction item is generated based on a preset timeout ratio formula; The trajectory deviation degree and the time deduction item are combined based on a preset fusion formula to generate a compliance score.

6. The rail transit training method based on an AR smart head-mounted terminal according to claim 1, characterized in that: The step of integrating the operation data with the compliance score to generate a skill shortcoming location report includes: Locating weak operational steps based on the compliance scores; Input the weak operation steps into a preset equipment-operation step mapping relationship table to match the associated rail transit equipment module; Comparing the decision delay data with a preset standard decision duration threshold for a corresponding operation step to identify decision defects; Extract the operation instruction sequence in the interaction log: Performing a timing comparison between the operation instruction sequence and the standard instruction sequence in the preset operation instruction library; If missing instructions, reversed order, or incorrect parameters are detected, it will be marked as an incorrect operation sequence; Associating the erroneous operation sequence with corresponding risk items in a preset fault library, wherein the risk items include equipment damage probability and accident level; The equipment modules, decision defect types and risk items are integrated to generate a skill shortcoming location report.

7. A rail transit training method based on an AR smart head-mounted terminal according to claim 6, characterized in that: The step of comparing the decision delay data with a preset standard decision duration threshold corresponding to the operation step to identify decision defects further includes: If the decision delay exceeds the standard decision time threshold of the corresponding operation step, it is marked as a "slow response" decision defect; If the decision delay is lower than the standard decision time threshold of the corresponding operation step, it is marked as a "hasty decision" decision defect item.

8. A rail transit training system based on AR smart head-mounted terminal, characterized in that: include: Classification module: classifies rail transit equipment into high-risk equipment and ordinary equipment based on the preset fault library; Acquisition module: collects millimeter-level laser scanning data of high-risk equipment, centimeter-level photogrammetry data of ordinary equipment, and physical space coordinate data, and generates a virtual training scene aligned with the physical space coordinates; Configuration module: Based on the preset trainee skill profile, preset fault library, and preset operation instruction library, configure modular training courses in the virtual training scenario; Data module: collects operational data of trainees while they are executing the modular training course through AR head-mounted terminals. The operational data includes spatiotemporal trajectory data, decision delay, and interaction logs. Fusion module: Based on the spatiotemporal trajectory data and the decision delay, generating a compliance score through fusion; Generation module: integrates the operation data and the compliance score to generate a skill shortcoming positioning report.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the steps of a rail transit training method based on an AR smart head-mounted terminal are implemented as claimed in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the steps of a rail transit training method based on an AR smart head-mounted terminal as claimed in any one of claims 1 to 7.

Images