Armored vehicle driving operation simulation training system based on attention detection

Through an armored vehicle driving operation simulation training system based on attention detection, using eye tracking, Lyapunov index and fractal dimension analysis, the problem that the existing system cannot evaluate the driver's attention and reaction ability in real time and accurately, achieving more In-depth training assessment and accident risk reduction.

CN118587961BActive Publication Date: 2024-11-08XUZHOU JIUDING ELECTROMECHANICAL FACTORY
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411073817.1
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-07
Publication Date
2024-11-08
Estimated Expiration
2044-08-07

AI Technical Summary

Technical Problem

The existing armored vehicle driving training system lacks scientific and data-driven attention assessment methods, and cannot identify and feedback drivers' attention status and response capabilities in real time and accurately in a high-risk situation, making it difficult to comprehensively and accurately evaluate drivers' Performance in dynamic and complex environments.

Method used

The armored vehicle driving operation simulation training system based on attention detection, including scene risk division module, time stamping module, risk attention detection module, reminder judgment module, reminder reaction analysis module and training compliance evaluation module, through eye tracking and Liyapunov index and fractal dimension analysis, monitor and feedback driver's attention status in real time, and evaluate their operating stability and coping capabilities under high-risk conditions.

Benefits of technology

The system comprehensively evaluates the driver's attention and operational capabilities through a scientific and data-driven method, providing deeper insights than traditional methods, improving the scientificity and effectiveness of training, and helping to reduce the risk of accidents in real driving scenarios.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118587961B_ABST
    Figure CN118587961B_ABST
Patent Text Reader

Abstract

The present invention discloses an armored vehicle driving operation simulation training system based on attention detection, and specifically relates to the field of training simulation detection technology. Through scene risk division and time marking modules, the system can accurately identify risk-increasing time periods in a dynamic environment, and monitor and feedback the driver's attention status in real time through risk attention detection and reminder judgment modules. The reminder reaction analysis module and the training standard assessment module provide deeper insights than traditional linear analysis through advanced nonlinear dynamic analysis and fractal geometry analysis methods, and can accurately assess the driver's operational complexity and coping ability under high-risk conditions, which helps to significantly reduce accident risks in real driving scenarios.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of training simulation detection, and more specifically, to an armored vehicle driving operation simulation training system based on attention detection. Background Art

[0002] As the complexity of modern warfare and military missions continues to increase, the environment that armored vehicle drivers face in the process of performing their missions is becoming increasingly complex and changeable. Armored vehicle driving operations not only require drivers to have superb driving skills, but also require them to maintain a high level of attention and quick response capabilities in unexpected situations.

[0003] Conventional training systems rely mainly on subjective evaluations and lack scientific, data-driven attention assessment methods. They mainly rely on static simulators and realistic training scenarios, making it difficult to comprehensively and accurately assess drivers’ performance in dynamic and complex environments. In addition, existing training systems are deficient in risk assessment and attention detection, and are unable to accurately identify and provide feedback on drivers’ attention status and reaction capabilities in high-risk situations in real time.

[0004] In order to solve the above problems, a technical solution is now provided. Summary of the invention

[0005] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides an armored vehicle driving operation simulation training system based on attention detection to solve the problems raised in the above-mentioned background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] The armored vehicle driving operation simulation training system based on attention detection includes a scene risk division module, a time marking module, a risk attention detection module, a reminder judgment module, a reminder reaction analysis module, and a training standard assessment module;

[0008] Scenario risk classification module: determines the risk of simulated driving scenarios, which include normal risk and high risk;

[0009] Time marking module: marks the time period when the normal risk of the simulated driving scenario changes to high risk as the risk-increasing time period, and marks the time period outside the risk-increasing time period as the non-risk-increasing time period;

[0010] Risk attention detection module: detects the driver's attention during the risk-increasing period through eye tracking to assess whether the driver's attention is normal during the risk-increasing period;

[0011] Reminder judgment module: reminds the driver when the driver's attention is not normal during the risk-increasing period;

[0012] Reminder reaction analysis module: After reminding the driver, the module measures the driver's dynamic reaction rate to environmental changes to evaluate the operational stability and adaptability under simulated high-risk conditions; and evaluates the driver's ability to respond to emergencies by analyzing the fractal dimension of the driver's operation data after the reminder.

[0013] Training compliance assessment module: Through comprehensive analysis of the operational stability and adaptability under simulated high-risk conditions and the driver's ability to respond to emergencies, it is assessed whether the driver's training has met the standards and whether re-training is necessary.

[0014] In a preferred embodiment, the risk of the simulated driving scenario is determined as follows:

[0015] Identify risk factors, including road conditions, traffic conditions, environmental factors, and vehicle performance parameters;

[0016] Collect specific data on risk factors; assign a weight to each risk factor and calculate the risk value of the risk factor; combine the risk values ​​obtained from all single risk factors to calculate the total risk score ;

[0017] Setting the Threshold , the risks of simulated driving scenarios are divided into normal risks and high risks:

[0018] if , then the risk of the simulated driving scenario is classified as normal risk;

[0019] if , the risk of the simulated driving scenario is classified as high risk.

[0020] In a preferred embodiment, the time period when the normal risk of the simulated driving scenario changes to a high risk is marked as a risk increase time period, and the time period outside the risk increase time period is marked as a non-risk increase time period, specifically:

[0021] Mark the time point when the normal risk of the simulated driving scenario changes to a high risk as the risk change time point, and mark the risk increase time period based on the risk change time point;

[0022] All time outside of the risk-elevated time period is automatically marked as a non-risk-elevated time period.

[0023] In a preferred embodiment, the driver's attention is detected during the risk-increasing time period by eye tracking to assess whether the driver's attention is normal during the risk-increasing time period, specifically:

[0024] The eye tracking device records the eye tracking data, which includes the coordinates of the gaze point, the pupil diameter, and the blinking frequency;

[0025] Calculate the change of the gaze point coordinates over time, the expression is: ;

[0026] in, is the fixation stability, No. The gaze point coordinates of the data points, are the average values ​​of the horizontal and vertical coordinates of the gaze point coordinates, is the total number of data points, is the number of the data point;

[0027] Calculate pupil reactivity, the expression is: ;in, is pupil reactivity, is the change in pupil diameter, The time interval used for the pupil diameter to change;

[0028] when Not less than the predetermined threshold and pupil reactivity is within the predetermined normal range When the driver's attention is normal during the risk-increasing period,

[0029] when Below a predetermined threshold or Out of the normal range , it is determined that the driver's attention is abnormal during the risk-increasing time period.

[0030] In a preferred embodiment, the operating stability and adaptability under simulated high-risk conditions are evaluated by measuring the driver's dynamic reaction rate to environmental changes, specifically:

[0031] Collect the driver's operation data after the reminder, including reaction time, operation type and operation quality;

[0032] Introducing initial perturbations during simulation training;

[0033] In the training scenario, randomly change an environment parameter and quantify the change ;

[0034] Track and record the system's response to initial disturbances; record the process from initial disturbance to driver operation stabilization through high-frequency data acquisition to quantify state changes :

[0035] Calculate the Lyapunov exponent, the expression is: ;in, is the Lyapunov exponent, is the system state deviation at time t, is the initial state deviation.

[0036] In a preferred embodiment, the driver's ability to cope with emergencies is evaluated by analyzing the fractal dimension of the driver's operation data after the reminder, specifically:

[0037] Real-time recording of driver response data, including steering wheel rotation angle, accelerator and brake usage, speed changes, and gaze point movement;

[0038] Convert raw data into a format suitable for fractal dimension calculation;

[0039] Select the fractal dimension calculation method; divide the space where the data is embedded into two parts with side lengths of small box;

[0040] Count the number of boxes containing driving action data points ; gradually decrease , repeat the above steps and record different Corresponding value;

[0041] Calculate the fractal dimension, the expression is: ;in, is the fractal dimension, is the diameter of the ball.

[0042] In a preferred embodiment, by comprehensively analyzing the operational stability and adaptability under simulated high-risk conditions and the driver's ability to respond to emergencies, it is evaluated whether the driver's training meets the standards and whether retraining is required, specifically:

[0043] Normalizing the Lyapunov exponent and the fractal dimension, assigning the normalized Lyapunov exponent and the fractal dimension to preset proportional coefficients, and then calculating the training adaptation evaluation coefficient;

[0044] The training adaptation assessment coefficient is compared with the training adaptation assessment threshold: when the training adaptation assessment coefficient is greater than the training adaptation assessment threshold, it is determined that the driver's training does not meet the standard; when the training adaptation assessment coefficient is less than or equal to the training adaptation assessment threshold, it is determined that the driver's training meets the standard.

[0045] The technical effects and advantages of the armored vehicle driving operation simulation training system based on attention detection of the present invention are as follows:

[0046] The attention detection-based armored vehicle driving operation simulation training system comprehensively evaluates the driver's attention and operating ability through a scientific, data-driven approach. Through the scenario risk division and time marking modules, the system can accurately identify the risk-increasing time period in a dynamic environment, and monitor and feedback the driver's attention status in real time through the risk attention detection and reminder judgment modules. The reminder reaction analysis module and the training standard assessment module provide deeper insights than traditional linear analysis through advanced nonlinear dynamic analysis and fractal geometry analysis methods, and can accurately assess the driver's operational complexity and response capabilities under high-risk conditions. This system not only improves the scientificity and effectiveness of training, but also maintains the foresight and adaptability of the evaluation system through continuous data analysis and optimization, which helps to significantly reduce the risk of accidents in real driving scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 It is a structural schematic diagram of the armored vehicle driving operation simulation training system based on attention detection of the present invention. DETAILED DESCRIPTION

[0048] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. Example

[0049] Figure 1 A structural schematic diagram of an armored vehicle driving operation simulation training system based on attention detection of the present invention is given. The armored vehicle driving operation simulation training system based on attention detection includes a scene risk division module, a time marking module, a risk attention detection module, a reminder judgment module, a reminder reaction analysis module and a training standard assessment module.

[0050] Scenario risk classification module: determines the risks of simulated driving scenarios, which include normal risks and high risks.

[0051] Time marking module: marks the time period when the normal risk of the simulated driving scenario changes to high risk as the risk-increased time period, and marks the time period outside the risk-increased time period as the non-risk-increased time period.

[0052] Risk attention detection module: detects the driver's attention during the risk-heightened period through eye tracking to assess whether the driver's attention is normal during the risk-heightened period.

[0053] Reminder judgment module: reminds the driver when the driver's attention is abnormal during the risk-increasing period.

[0054] Reminder reaction analysis module: After reminding the driver, the module evaluates the operational stability and adaptability under simulated high-risk conditions by measuring the driver's dynamic reaction rate to environmental changes; and evaluates the driver's ability to respond to emergencies by analyzing the fractal dimension of the driver's operation data after the reminder.

[0055] Training compliance assessment module: Through comprehensive analysis of the operational stability and adaptability under simulated high-risk conditions and the driver's ability to respond to emergencies, it is assessed whether the driver's training has met the standards and whether re-training is necessary.

[0056] Identify the risks of the simulated driving scenario, specifically:

[0057] It is necessary to identify various risk factors that may affect driving safety. Risk factors generally include but are not limited to:

[0058] Road conditions: such as slippery, icy, winding roads, etc.

[0059] Traffic conditions: such as traffic density, traffic accidents, pedestrian crossings, etc.

[0060] Environmental factors: such as visibility (fog, rain, night), wind speed, etc.

[0061] Vehicle performance parameters: such as vehicle speed, braking system performance, etc.

[0062] For each driving scenario, collect specific data on the above risk factors. For example:

[0063] Road conditions: obtained from maps or actual scene images through image recognition technology.

[0064] Traffic conditions: data generated using traffic simulation software or obtained from traffic monitoring systems.

[0065] Environmental factors: obtained from weather forecast API or monitored in real time through environmental sensors.

[0066] Vehicle performance parameters: collected through the vehicle's sensor system.

[0067] Each risk factor is assigned a weight, and the risk value is calculated based on its impact on security. The determination of the weight usually relies on the experience of experts and historical data analysis. The risk value of a risk factor can be calculated using the following formula: ;in, It is The risk value of each risk factor is It is The actual measured value of the risk factor, is the weight of the risk factor, is the number of the risk factor.

[0068] The risk values ​​obtained from all single risk factors are combined to calculate the total risk score: ;in, is the total risk score, is the total number of risk factors.

[0069] Based on the comprehensive risk score, the risk of the simulated driving scenario is divided into normal risk and high risk. Usually, this can be achieved by setting a threshold accomplish:

[0070] if , then the risk of the simulated driving scenario is classified as normal risk;

[0071] if , the risk of the simulated driving scenario is classified as high risk.

[0072] Threshold The determination can be based on historical accident data and statistical analysis to ensure that there is an adequate safety margin.

[0073] This method can systematically identify and evaluate the risks in driving scenarios, allowing the training system to adjust the complexity of the training scenarios and the reminder strategy in a targeted manner. This risk assessment mechanism helps improve the targeting and efficiency of training, enhance safety, and optimize the allocation of training resources.

[0074] To ensure that the armored vehicle driving operation simulation training system can effectively assess and simulate real-world driving risks, the following is a detailed description of the actual measurement values ​​of each risk factor and how to obtain them:

[0075] Road conditions:

[0076] Slipperiness: Assessed using a slipperiness sensor or through meteorological data (such as precipitation). In a simulated environment, slipperiness can be set to a value from 0 to 1, where 0 means completely dry and 1 means extremely slippery.

[0077] Ice level: estimated from air temperature and precipitation data. Ice level can be set to a range similar to slipperiness.

[0078] Number and angle of curves: Extract road curvature information from map data or define specific road layout parameters directly in the simulation system.

[0079] Traffic conditions:

[0080] Traffic density: Traffic flow data generated by a simulation system, expressed as the number of vehicles passing through a specific road segment per unit time.

[0081] Traffic accidents: In the simulation system, the probability of an accident can be set or accident events can be randomly generated at a specific time and place.

[0082] Pedestrian Crossing: Measured by simulating pedestrian flow and the frequency of pedestrian crossing events.

[0083] Environmental factors:

[0084] Visibility: It is set by simulating the sight range under different weather conditions, such as clear, foggy, rainy, and nighttime visibility levels. The visibility range is usually expressed in meters.

[0085] Wind speed: In the simulation system, you can set a specific wind speed value, usually in meters per second, or adjust it according to the wind level.

[0086] Vehicle performance parameters:

[0087] Vehicle speed: obtained by simulating the real-time speed data of the vehicle, in kilometers per hour or meters per second.

[0088] Braking system performance: can be evaluated by brake response time and braking distance. In simulation systems, these parameters can be preset or dynamically calculated based on the vehicle model.

[0089] In the virtual training environment, all parameters can be preset through software programming or generated according to algorithms to simulate real-world driving conditions.

[0090] The time period when the normal risk of the simulated driving scenario changes to high risk is marked as the risk increase time period, and the time period outside the risk increase time period is marked as the non-risk increase time period, specifically:

[0091] The time point when the normal risk of the simulated driving scenario changes to a high risk is marked as the risk change time point, and the risk increase time period is marked based on the risk change time point. Specifically, a time window can be set with the risk change time point as the center. Generally, this time window can be extended forward and backward. For example, the window can be set to 2 minutes before and 1 minute after the risk change time point. This means that the time period from 2 minutes before to 1 minute after the risk change time point will be marked as the risk increase time period.

[0092] All times outside of the elevated risk periods are automatically marked as non-elevated risk periods. These periods usually correspond to normal risk levels and are relatively safe and stable periods in training.

[0093] By accurately marking and distinguishing between periods of increased risk and periods of non-increased risk, the training system can more specifically train the driver's reaction and handling capabilities under different risk conditions. This approach ensures that the driver can effectively adjust his attention and reaction strategy when faced with sudden increases in risk. By paying special attention to and training during periods of increased risk, the driver can better understand and learn how to stay calm and make quick decisions under stressful or dangerous situations. This training method improves the practical application value of training and makes the driver more adaptable to various emergency situations that may be encountered in the real world.

[0094] Eye tracking is used to detect the driver's attention during the risk-increasing period to assess whether the driver's attention is normal during the risk-increasing period. Specifically:

[0095] Make sure the eye tracking device is properly installed and calibrated to accurately capture the driver's eye movements. Eye tracking devices usually include infrared cameras and processing software that can capture and interpret changes in eye position and pupils in real time.

[0096] When the pilot enters the simulated cockpit and starts training, the eye tracking device starts recording eye tracking data. Eye tracking data includes:

[0097] Gaze point coordinates: The horizontal and vertical coordinates of the point on the screen where the eyes are looking.

[0098] Pupil diameter: The size of the pupil is usually related to the level of concentration.

[0099] Blink frequency: the number of blinks per unit time.

[0100] By analyzing eye tracking data, the driver's concentration can be assessed. The analysis indicators include gaze stability and pupil reactivity.

[0101] Calculate the change of the gaze point coordinates over time to reflect whether the driver's visual attention is focused. The calculation formula for gaze point stability is: ;

[0102] in, is the fixation stability, No. The gaze point coordinates of the data points, are the average values ​​of the horizontal and vertical coordinates of the gaze point coordinates, is the total number of data points, is the number of the data point.

[0103] The pupil diameter is measured to measure the speed of its reaction to light changes, reflecting the sharpness of attention. The formula for calculating pupil reactivity is: ;in, is pupil reactivity, is the change in pupil diameter, The time interval over which pupil diameter changes.

[0104] Combined with the calculation results of gaze point stability and pupil reactivity, it is possible to assess whether the driver's attention is normal during the risk-increasing period:

[0105] When pupil reactivity is large, it means that the pupil responds quickly and significantly to changes in lighting. This usually indicates that the driver has a low cognitive load, is in a more relaxed mental state, and is able to quickly adjust visual focus to respond to external changes. In some cases, it may also mean that the driver is more sensitive to the current task.

[0106] When pupil reactivity is low, it means the pupil is slow or has a small response to changes in lighting. This may indicate that the driver is under high cognitive load or fatigue and has difficulty adapting to changes in the environment quickly. Prolonged concentration, fatigue or stress may cause this reaction.

[0107] If the gaze stability is high, it means that the driver's gaze is relatively fixed and rarely makes rapid or frequent gaze shifts. This usually indicates that the driver is highly focused on the object or task currently being looked at. In driving scenarios, high gaze stability may indicate that the driver is concentrating on handling complex or high-attention tasks.

[0108] Low gaze stability means that the driver's gaze frequently moves and cannot stay on a specific point. This may indicate driver distraction or inattention. In a driving environment, frequent gaze shifts may be a sign of searching for navigation information, multitasking, or responding to unexpected events, or it may be a sign of inattention.

[0109] when Not less than the predetermined threshold and pupil reactivity is within the predetermined normal range When the driver's attention is normal during the risk-increasing period,

[0110] when Below a predetermined threshold or Out of the normal range , it is determined that the driver's attention is abnormal during the risk-increasing time period.

[0111] Predetermined threshold It is based on historical data and experimental measurements. First, gaze stability data is collected by tracking the eyes of a certain number of drivers under standard driving conditions. Then, the data is analyzed to determine the mean and standard deviation of gaze stability. SthS_{\text{th}}Sth is usually set to the mean plus one to two times the standard deviation. Such a threshold can distinguish between normal and abnormal gaze stability and help identify drivers who may be distracted.

[0112] Predetermined normal range It is set by comparing pupillary response data of healthy adult drivers under different lighting conditions. By collecting a large number of data points, the mean and standard deviation of pupillary reaction speed are calculated. The normal range is generally set based on the mean minus or plus one to two standard deviations. This range is designed to cover the reactivity of most healthy drivers under normal conditions, ensuring that it is only marked as abnormal when it deviates significantly from this range.

[0113] When the driver's attention is not normal during the risk-increasing period, the driver is reminded, specifically:

[0114] Select the appropriate alert method (visual, audible or tactile feedback) for warning:

[0115] Visual Reminders: Displays a warning icon or message on the driver's display to indicate a lack of concentration or the need to adjust focus.

[0116] Auditory reminder: Audio signals are sent through headphones or speakers in the cabin to remind the driver of their attention status.

[0117] Haptic reminders: Vibration in the seat or controls reminds the driver to refocus.

[0118] Driver response monitoring: Monitor the driver's response to reminders and adjust the training scenario or difficulty accordingly to ensure that the driver can effectively improve his performance after receiving feedback.

[0119] System iterative optimization: Based on the driver's performance and feedback, the reminder algorithm and parameter settings are continuously optimized to improve training effects and safety.

[0120] The reminder judgment module can monitor and respond to changes in attention status in real time, helping drivers get necessary reminders at critical moments to prevent potential safety risks. The reminder method and training content are adjusted according to each driver's specific performance and reaction to enhance the pertinence and effectiveness of the training. Through continuous reminders and adjustments, drivers can learn how to better control and allocate their attention resources in a simulated environment, so that they can perform more stably and efficiently in actual driving.

[0121] By measuring the driver's dynamic reaction rate to environmental changes, the operational stability and adaptability under simulated high-risk conditions are evaluated, specifically:

[0122] Collect the driver's operation data after the reminder, including reaction time, operation type and operation quality.

[0123] High-precision sensors and eye-tracking equipment are used to record the driver's response in real time, from the moment the reminder is received until the driver completes a series of actions.

[0124] A small initial disturbance is introduced during the simulation training to test the driver's system reaction.

[0125] In the training scenario, randomly change an environmental parameter (such as road conditions, sudden obstacles, etc.) and quantify this change .

[0126] Track and record the system (driver response) response to the initial disturbance.

[0127] Through high-frequency data acquisition, the process from initial disturbance to driver operation stabilization is recorded, and the state changes in this process are quantified. .

[0128] Calculate the Lyapunov exponent, the expression is: ;in, is the Lyapunov exponent (indicating the sensitivity of the system response), is the system state deviation at time t, is the initial state deviation.

[0129] Analyze the results of the Lyapunov index to determine the stability of the system and the adaptability of the driver. If the Lyapunov index is larger, it means that the system is very sensitive to initial disturbances, the driver's reaction may be excessive or unstable, and the operational stability and adaptability under simulated high-risk conditions are worse; if the Lyapunov index is smaller, it means that the system is stable, the driver can effectively handle the initial disturbance, and the operational stability and adaptability under simulated high-risk conditions are better.

[0130] Using the Lyapunov index as an analysis tool, the driver's adaptability to complex situations and the overall stability of the system can be objectively evaluated. This method provides a means to quantify the dynamics of driver behavior, which helps to identify drivers who may overreact or underreact when faced with unexpected situations. In addition, this analysis can help driving training system designers optimize training procedures and improve the driver's coping ability and the overall performance of the system through targeted training; by applying nonlinear dynamic system theory, this method can not only improve the scientificity and effectiveness of the training system, but also increase the understanding of the complexity of driver behavior, thus playing an important role in improving driving safety and operating efficiency. The innovation of this method lies in that it goes beyond the traditional linear reaction time analysis and provides a more in-depth means of dynamic system behavior analysis.

[0131] By analyzing the fractal dimension of the driver's operation data after the reminder, the driver's ability to respond to emergencies is evaluated, specifically:

[0132] Fractal dimension analysis is a method used to quantify the geometric characteristics of a data set in a complex system. In driving behavior analysis, fractal dimension can reveal the complexity and variability of driver operations.

[0133] Collect the driver's operational data after receiving the reminder: Use high-precision sensors and eye tracking equipment to record the driver's reaction data in real time, including steering wheel rotation angle, accelerator and brake usage, speed changes, gaze movement, etc.

[0134] Convert the raw data into a format suitable for fractal dimension calculation, specifically:

[0135] Data cleaning: Remove noise and irrelevant data to ensure data quality.

[0136] Data normalization: Normalize different types of operational data so that they can be compared on the same scale.

[0137] Time series: Arrange data in chronological order to form a time series.

[0138] Select an appropriate fractal dimension calculation method: In driving behavior analysis, commonly used fractal dimension calculation methods include the Box-Counting Method.

[0139] The data embedding space is divided into two parts with side lengths of Small box.

[0140] Count the number of boxes containing driving action data points .

[0141] Gradually reduce , repeat the above steps and record different Corresponding value.

[0142] Using a log-log plot, right Plotting, the slope is the fractal dimension, and the expression of the fractal dimension is: ;in, is the fractal dimension, is the diameter of the ball, as Reduce, The rate of growth reflects the growth in data complexity.

[0143] The driver's driving execution ability is analyzed through fractal dimension. The higher the fractal dimension, the more complex and changeable the driver's operation is, and the stronger the driver's driving execution ability to deal with emergencies is. This may indicate that the driver has adopted a variety of different operation strategies when dealing with emergencies after reminders, and has strong adaptability. The lower the fractal dimension, the more relatively simple and repetitive the driver's operation mode is, and the weaker the driver's driving execution ability to deal with emergencies is, which may indicate that the driver is insufficient in dealing with emergencies after reminders and has a single operation strategy.

[0144] Fractal dimension provides a means to quantify the complexity of driving operations and can accurately evaluate the driver's reaction and execution ability after being reminded. Fractal dimension can reveal operating modes and behavioral characteristics that traditional methods cannot capture, providing deeper insights. Based on the fractal dimension results, personalized training programs can be provided to drivers to improve their specific weaknesses and improve training results. By identifying the complexity and stability of the driver's operations, potential driving risks can be discovered in advance, targeted training can be carried out, and overall driving safety can be improved.

[0145] Through comprehensive analysis of the operational stability and adaptability under simulated high-risk conditions and the driver's ability to respond to emergencies, it is assessed whether the driver's training meets the standards and whether retraining is necessary. Specifically:

[0146] The Lyapunov exponent and the fractal dimension are normalized, and after the normalized Lyapunov exponent and the fractal dimension are assigned preset proportional coefficients respectively, the training adaptation evaluation coefficient is calculated.

[0147] The expression of the training adaptation evaluation coefficient is: ;in, is the training adaptation evaluation coefficient, are the preset scaling factors for the Lyapunov exponent and fractal dimension, respectively, and are greater than 0, is the base of natural logarithms.

[0148] The larger the training adaptation assessment coefficient, the less up to standard the driver's training is. The larger the training adaptation assessment coefficient, the poorer the driver's operational stability and adaptability under simulated high-risk conditions, and the unstable or insufficient response. This means that the driver may overreact or underreact in emergencies, and thus fail to effectively handle risk scenarios. Therefore, a larger training adaptation assessment coefficient indicates that the driver's training is not effective and requires further training and improvement.

[0149] Compare the training adaptation evaluation coefficient to the training adaptation evaluation threshold:

[0150] When the training adaptation assessment coefficient is greater than the training adaptation assessment threshold, the driver's training is judged to be substandard; the driver's performance in the simulation training does not meet the expected standards, his operation is unstable, his adaptability is weak, and he cannot effectively respond to emergencies.

[0151] According to the judgment result, the driver's training does not meet the standards, and the measures taken include:

[0152] Strengthen training: Increase the frequency and intensity of training, especially targeting areas where drivers are weak, such as practicing complex operations and dealing with emergencies.

[0153] Personalized training: Design a personalized training plan and provide customized training content and feedback based on the driver's specific problems.

[0154] Psychological counseling: Provide psychological counseling and stress management training to help drivers improve their performance in high-pressure environments.

[0155] When the training adaptation assessment coefficient is less than or equal to the training adaptation assessment threshold, the driver's training is judged to have met the standards; the driver's performance in the simulation training has reached the expected standards, the operation is stable, the adaptability is strong, and the driver can effectively respond to emergencies.

[0156] The setting of the preset proportionality coefficient needs to be combined with the importance of Lyapunov index and fractal dimension in reflecting the driver's ability. First, the contribution of each indicator to the driver's training adaptability is determined through expert evaluation and experimental data. Then, the corresponding proportionality coefficient is set according to the contribution to ensure that both are reasonably weighted in the comprehensive evaluation. The specific value of the proportionality coefficient should be determined through experiments and optimization, with the goal of making the comprehensive evaluation coefficient accurately reflect the driver's training adaptation. The common practice is to adjust and optimize the proportionality coefficient based on historical data through regression analysis or machine learning methods to achieve the best evaluation effect.

[0157] The setting of the training adaptation evaluation threshold is based on a large amount of historical data and expert experience. First, a large number of Lyapunov exponents and fractal dimension data of drivers at different training stages are collected and normalized. Then, through statistical analysis, the distribution of these data among drivers who meet the training standards and those who do not meet the standards is determined. Using this distribution, a critical value is selected as the training adaptation evaluation threshold. This critical value should be able to effectively distinguish between drivers who meet the standards and those who do not meet the standards. Usually, the point with the lowest misjudgment rate is selected as the threshold. In addition, the threshold is adjusted and verified regularly to ensure its applicability and accuracy.

[0158] By normalizing the Lyapunov index and fractal dimension and calculating the training adaptation evaluation coefficient in combination with the preset proportional coefficient, the driver's operational stability and adaptability under simulated high-risk conditions can be comprehensively and accurately evaluated. This method utilizes the theory of nonlinear dynamic systems and fractal geometry, providing deeper insights than traditional linear analysis, and can identify the driver's potential weaknesses and deficiencies. By setting scientific and reasonable evaluation thresholds, the training system can accurately determine whether the driver meets the standards and avoid waste of training resources. Personalized feedback and intensive training plans ensure that each driver can be targeted to improve their specific weaknesses, thereby improving the driver's resilience and driving safety as a whole. This method not only enhances the scientific nature and effectiveness of training, but also maintains the foresight and adaptability of the evaluation system through continuous data analysis and optimization, which helps to significantly reduce the risk of accidents in real driving scenarios.

[0159] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters and thresholds in the formula are set by technicians in this field according to actual conditions.

[0160] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium, or may be transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from one website, computer, server or data center to another website, computer, server or data center by wired (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium may be a solid-state hard disk.

[0161] Those of ordinary skill in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0162] 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 modules described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0163] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the modules is only a logical function division. There may be other division methods in actual implementation, such as multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or modules, which can be electrical, mechanical or other forms.

[0164] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical modules, and may be located in one place or distributed on multiple network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0165] In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0166] If the functions are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage media include: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, and other media that can store program codes.

[0167] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0168] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. The armored vehicle driving operation simulation training system based on attention detection is characterized by: It includes scenario risk classification module, time marking module, risk attention detection module, reminder judgment module, reminder reaction analysis module and training achievement assessment module; Scenario risk classification module: determines the risk of simulated driving scenarios, which include normal risk and high risk; Time marking module: marks the time period when the normal risk of the simulated driving scenario changes to high risk as the risk-increasing time period, and marks the time period outside the risk-increasing time period as the non-risk-increasing time period; Risk attention detection module: detects the driver's attention during the risk-increasing period through eye tracking to assess whether the driver's attention is normal during the risk-increasing period; Reminder judgment module: reminds the driver when the driver's attention is not normal during the risk-increasing period; Reminder reaction analysis module: After reminding the driver, the module measures the driver's dynamic reaction rate to environmental changes to evaluate the operational stability and adaptability under simulated high-risk conditions; and evaluates the driver's ability to respond to emergencies by analyzing the fractal dimension of the driver's operation data after the reminder. Training compliance assessment module: Through comprehensive analysis of the operating stability and adaptability under simulated high-risk conditions and the driver's ability to respond to emergencies, it is assessed whether the driver's training has met the standards and whether retraining is necessary; Identify the risks of the simulated driving scenario, specifically: Identify risk factors, including road conditions, traffic conditions, environmental factors, and vehicle performance parameters; Collect specific data on risk factors; assign a weight to each risk factor and calculate the risk value of the risk factor; The total risk score is calculated by combining the risk values ​​obtained from all single risk factors. ; Setting the Threshold , the risks of simulated driving scenarios are divided into normal risks and high risks: if , then the risk of the simulated driving scenario is classified as normal risk; if , then the risk of the simulated driving scenario is classified as high risk; The time period when the normal risk of the simulated driving scenario changes to high risk is marked as the risk increase time period, and the time period outside the risk increase time period is marked as the non-risk increase time period, specifically: Mark the time point when the normal risk of the simulated driving scenario changes to a high risk as the risk change time point, and mark the risk increase time period based on the risk change time point; All time outside the risk-increasing time period is automatically marked as non-risk-increasing time period; Eye tracking is used to detect the driver's attention during the risk-increasing period to assess whether the driver's attention is normal during the risk-increasing period. Specifically: The eye tracking device records the eye tracking data, which includes the coordinates of the gaze point, the pupil diameter, and the blinking frequency; Calculate the change of the gaze point coordinates over time, the expression is: ;in, is the fixation stability, No. The gaze point coordinates of the data points, are the average values ​​of the horizontal and vertical coordinates of the gaze point coordinates, is the total number of data points, is the number of the data point; Calculate pupil reactivity, the expression is: ;in, is pupil reactivity, is the change in pupil diameter, The time interval used for the pupil diameter to change; when Not less than the predetermined threshold and pupil reactivity is within the predetermined normal range When the driver's attention is normal during the risk-increasing period, when Below a predetermined threshold or Out of the normal range When the driver's attention is abnormal during the risk-increasing period, By measuring the driver's dynamic reaction rate to environmental changes, the operational stability and adaptability under simulated high-risk conditions are evaluated, specifically: Collect the driver's operation data after the reminder, including reaction time, operation type and operation quality; Introducing initial perturbations during simulation training; In the training scenario, randomly change an environment parameter and quantify the change ; Track and record the system's response to initial disturbances; record the process from initial disturbance to driver operation stabilization through high-frequency data acquisition to quantify state changes ; Calculate the Lyapunov exponent, the expression is: ;in, is the Lyapunov exponent, It's in time The system state deviation, is the initial state deviation; By analyzing the fractal dimension of the driver's operation data after the reminder, the driver's ability to respond to emergencies is evaluated, specifically: Real-time recording of driver response data, including steering wheel rotation angle, accelerator and brake usage, speed changes, and gaze point movement; Convert raw data into a format suitable for fractal dimension calculation; Select the fractal dimension calculation method; divide the space where the data is embedded into two parts with side lengths of small box; Count the number of boxes containing driving action data points ; gradually decrease , repeat the above steps and record different Corresponding value; Calculate the fractal dimension, the expression is: ;in, is the fractal dimension, is the diameter of the sphere; Through comprehensive analysis of the operational stability and adaptability under simulated high-risk conditions and the driver's ability to respond to emergencies, it is assessed whether the driver's training meets the standards and whether retraining is necessary. Specifically: Normalizing the Lyapunov exponent and the fractal dimension, assigning the normalized Lyapunov exponent and the fractal dimension to preset proportional coefficients, and then calculating the training adaptation evaluation coefficient; The training adaptation assessment coefficient is compared with the training adaptation assessment threshold: when the training adaptation assessment coefficient is greater than the training adaptation assessment threshold, it is determined that the driver's training does not meet the standard; when the training adaptation assessment coefficient is less than or equal to the training adaptation assessment threshold, it is determined that the driver's training meets the standard.