CN-119088215-B - Locomotive driver operation fatigue evaluation system based on virtual reality

CN119088215BCN 119088215 BCN119088215 BCN 119088215BCN-119088215-B

Abstract

The application belongs to the technical field of locomotive driving man-machine work efficiency. The application provides a locomotive driver operation fatigue evaluation system based on virtual reality. According to the embodiment of the disclosure, by constructing a highly simulated driving scene, simulating a long-time and high-strength driving task, reproducing various conditions possibly causing fatigue of a driver in a risk-free manner, and monitoring and analyzing physiological response, behavior change and psychological state of the driver in the simulation process in real time through high-precision equipment, the problem of lack of accuracy and instantaneity of the existing assessment means is effectively solved, the operation fatigue degree of the locomotive driver in the task execution process can be timely and accurately assessed and found, and safety guarantee is provided for locomotive driving.

Inventors

YU MINGJIU
ZHANG YU
WU ZIRUI
Yuan Quanjingzi
CHEN JING
ZHAO DI
TAO HUIMIN
BAI XIN
GAO SIQI

Assignees

西北工业大学

Dates

Publication Date: 20260508
Application Date: 20240814

Claims (4)

1. A locomotive driver operation fatigue evaluation system based on virtual reality, the system comprising: the virtual man-machine driving operation unit is used for providing a simulated driving scene; The system comprises a man-machine data comprehensive acquisition unit, a control unit and a control unit, wherein the man-machine data comprehensive acquisition unit is used for acquiring operation behaviors and operation tasks of a locomotive driver in a driving process to obtain driving operation data, and the driving operation data comprises operation data, response time, accuracy, eye movement data and limb skeleton data of the locomotive driver; The data processing and feature extraction analysis unit is used for preprocessing the driving operation data and extracting index feature values, wherein the index feature values comprise workload information, task performance information, eye fatigue features and limb fatigue features; the cross-system data time synchronization and each index feature dimensionless processing unit is used for performing time alignment on the operation data and normalizing and dimensionless processing on the index feature values so as to obtain a work load index, a task performance index, an eye movement feature index and a limb feature index; The locomotive driver operation fatigue fuzzy comprehensive evaluation unit is used for fusing the workload, the task performance, the eye movement characteristics and the limb characteristics to obtain a final evaluation result; The locomotive driver operation fatigue comprehensive evaluation result grading unit is used for evaluating the operation fatigue state of the locomotive driver according to the final evaluation result; The man-machine data comprehensive acquisition unit comprises: a virtual locomotive driving recording module, a driving task executing recording module, an eye movement data acquisition module and a driving action acquisition module, The virtual locomotive driving recording module is used for collecting the operation data of the locomotive driver; the driving task execution recording module is used for collecting the response time and the accuracy of the locomotive driver in task execution; The eye movement data acquisition module is used for acquiring the eye movement data of the locomotive driver; The driving action acquisition module is used for acquiring the limb skeleton data of the locomotive driver; the data processing and feature extraction analysis unit includes: a work load calculation module, a task performance calculation module, an eye movement feature extraction module and a limb feature extraction module, wherein, The work load calculation module is used for calculating the work behavior time and the work task demand of the locomotive driver so as to obtain work load information; the task performance calculation module is used for obtaining the average response time according to the response time and the accuracy rate and obtaining the task performance information of the locomotive driver in a quantification mode according to the average response time; The eye movement characteristic extraction module is used for preprocessing the eye movement data and extracting eye fatigue characteristics from the preprocessed eye movement data; the limb feature extraction module is used for processing the limb skeleton data and extracting limb fatigue features from the processed limb skeleton data; The workload calculation module comprises locomotive driver operation behavior time calculation and operation task demand calculation, wherein the operation behavior time comprises a perception time T v , a cognitive time T c and an operation time T m : The perception time T v is calculated by a method-time measurement method; the expression of the cognitive time T c is: Wherein n is the number of selectable items; Time to cognition for the selected item; selecting a frequency of occurrence for the ith; the expression of the operation time T m is: in the formula, The time for executing the operation is D is the manual distance required by the operation, W is the diameter of the operation object; the operation time is as follows: The job task demand is: Wherein, the Representing the perceived task demand at time t, Representing the cognitive task demand at time t, The operation task requirement at the time t is the operation task requirement; The continuous symbol of the ith behavior element is that when the behavior exists at the time t, the behavior element is that =1, Otherwise =0; The category number of the behavior elements; the perceived task demand of the ith behavioral element at time t; the cognitive task demand of the ith behavior element at the time t; The operation task requirement of the ith behavior element at the time t; the average workload for each task is: Each driving task has 4 tasks, and after the operation behavior time and the operation task demand of each task are calculated, the workload of the driving task of the present wheel is calculated: Wherein, the Represent the first The average workload of the individual tasks is determined, Represent the first Job behavior time of the individual tasks; the task performance is as follows: Wherein, the The performance of the task is indicated and, The reaction time is indicated as a function of the reaction time, Representing the accuracy; in eye movement feature extraction, the average pupil diameter and average blink time are extracted: The average pupil diameter is: Wherein, the For the number of sampling points in the sample, Is the pupil diameter; The average blink time is: Wherein, the For the number of blinks, Is the time of one blink; In the cross-system data time synchronization and each index feature dimensionless processing unit, the Unix time stamp is converted into a date and time format which is easy to read by calling the system time and the Unix time stamp, so that most of data can reach millisecond-level time precision in the synchronization process, and experimental data analysis and integration are effectively supported, wherein an integration formula is as follows: Wherein AX represents a Unix timestamp value to be converted, and Date represents the time when the conversion is completed; For the extracted index characteristic values, an extremum standardization method is adopted to realize normalization and dimensionless quantification of quantitative index values, and the specific operation steps are as follows: The forward indexes with larger contribution to the target as the value is larger are as follows: the reverse indexes with larger values and smaller contribution to the target are as follows: Wherein, the An ith observation representing an jth indicator, And Represents the maximum value and the minimum value which can be taken by the j index obtained according to the upper limit and the lower limit of the operation time and the operation condition standard of the locomotive driver respectively.
2. The virtual reality-based locomotive driver work fatigue evaluation system of claim 1, wherein the virtual man-machine driving operation unit comprises: A virtual locomotive cockpit, a virtual driving task, a virtual reality head mounted display and a motion catcher, wherein, The virtual locomotive cockpit comprises a seat, an operation desk, an operation lever and a locomotive cockpit digital model, and is used for simulating the operation environment of a locomotive; The virtual driving task comprises a control lever task, a driving situation gesture task, a road information observation task and an instrument panel information viewing task, and is used for simulating the driving task of the locomotive; the virtual reality head mounted display is used for tracking eyeballs of the locomotive driver; the motion catcher is used for catching the driving motion of the locomotive driver.
3. A locomotive driver operation fatigue evaluation method based on virtual reality, which is characterized by being applied to the system as claimed in any one of claims 1-2, and comprising the following steps: Simulating a driving scene; acquiring operation behaviors and operation tasks of a locomotive driver in a driving process based on the driving scene to obtain driving operation data, wherein the driving operation data comprises operation data, response time, accuracy, eye movement data and limb skeleton data of the locomotive driver; Preprocessing the driving operation data and extracting index characteristic values, wherein the index characteristic values comprise workload information, task performance information, eye fatigue characteristics and limb fatigue characteristics; Performing time alignment on the operation data, and normalizing and dimensionless treatment on the index characteristic values to obtain a work load index, a task performance index, an eye movement characteristic index and a limb characteristic index; the cross-system data time synchronization and each index feature dimensionless processing unit normalizes and dimensionless the work load information, the task performance information, the eye fatigue feature and the limb fatigue feature by using an extremum standardization method to obtain the work load index, the task performance index, the eye movement feature index and the limb feature index; The method comprises the steps of fusing the workload, the task performance, the eye movement characteristic and the limb characteristic to obtain a final evaluation result, wherein the workload index, the task performance index, the eye movement characteristic index and the limb characteristic index are two-level indexes, three-level indexes under the workload index are current workload and accumulated workload, three-level indexes under the task performance index are average reaction time, three-level indexes under the eye movement characteristic index are average pupil diameter and average blink time, three-level indexes under the limb characteristic index are current uncomfortableness and accumulated uncomfortableness, and the method specifically comprises the steps of constructing a membership function according to a triangular fuzzy function, carrying out confidence correction and consistency check on a judgment matrix of the membership function, processing the judgment matrix after confidence correction and consistency check by a defuzzification calculation and root finding method to obtain a maximum characteristic root and a corresponding characteristic vector of the judgment, normalizing the maximum characteristic root and the corresponding characteristic vector to obtain an average reaction time, carrying out weighting on the maximum characteristic root and the maximum characteristic vector, and the three-level indexes under the eye movement characteristic index to obtain a current uncomfortableness and accumulated uncomfortableness, and determining a weighted value according to a fatigue grade, and determining corresponding risk grade; and evaluating the operation fatigue state of the locomotive driver according to the final evaluation result.
4. The virtual reality-based locomotive driver work fatigue evaluation method of claim 3, wherein the final evaluation result comprises: no fatigue, slight fatigue, relatively tired, obvious fatigue and very tired, wherein, The fatigue evaluation value is 0-0.2 and is not tired, the fatigue evaluation value is 0.2-0.4 and is slightly tired, the fatigue evaluation value is 0.4-0.6 and is relatively tired, the fatigue evaluation value is 0.6-0.8 and is obviously tired, and the fatigue evaluation value is 0.8-1.0 and is very tired.

Description

Locomotive driver operation fatigue evaluation system based on virtual reality Technical Field The embodiment of the disclosure relates to the technical field of locomotive driving man-machine efficiency, in particular to a locomotive driver operation fatigue evaluation system based on virtual reality. Background In the driving operation process with high strength and long time, the fatigue problem of a driver is increasingly remarkable, and the fatigue problem becomes a great hidden trouble for influencing the driving safety. Research shows that fatigue of a driver can cause phenomena of reduced reaction speed, distraction, weakened judgment and the like, and traffic accidents can be caused when the fatigue is serious. Particularly in complex railway environments, the continuous and efficient working capacity of locomotive drivers is a key factor for ensuring driving safety and maintaining transportation order. Therefore, the fatigue state of locomotive drivers is studied in depth, and is important for improving the safety and efficiency of railway transportation. Although the hazards of driver fatigue have been widely recognized, the problems and deficiencies present in current research remain prominent. The existing fatigue monitoring technology often depends on physiological indexes, such as a researcher uses a camera and other sensors to monitor physiological indexes such as facial expression, blink frequency of eyes, head posture and the like of a driver, and the researcher uses a surface electromyography to acquire electromyographic signals to realize real-time monitoring and evaluation of operation fatigue. These indicators are difficult to capture accurately in an actual driving environment and the assessment has a certain degree of singleness. Although the fatigue characteristics of the driver can be revealed to a certain extent by the commonly used method at present, the method is limited by technical limitations and the diversity of fatigue characterization in a complex driving environment, and the accuracy and the instantaneity of the evaluation means still need to be further improved, so that the method cannot help to improve the driving safety of the locomotive and has the effect of reducing the possibility of accident occurrence caused by fatigue of the driver. And the objective state of the locomotive driver during operation is difficult to monitor by a crew member, especially when the locomotive driver is tired and driving accidents are possible. Methods using existing equipment and methods such as subjective scale assessment, physiological feature-based work fatigue assessment, and the like have certain drawbacks in objectivity, comprehensive comprehensiveness and accuracy. Disclosure of Invention In order to avoid the defects of the prior art, the invention provides a locomotive driver operation fatigue evaluation system based on virtual reality, which is used for solving the problem that the prior art has a certain defect in objectivity, comprehensive comprehensiveness and accuracy by using the existing equipment and methods such as subjective scale evaluation, physiological characteristic-based operation fatigue evaluation and the like. According to a first aspect of embodiments of the present disclosure, there is provided a virtual reality-based locomotive driver work fatigue evaluation system, the system comprising: the virtual man-machine driving operation unit is used for providing a simulated driving scene; The system comprises a man-machine data comprehensive acquisition unit, a control unit and a control unit, wherein the man-machine data comprehensive acquisition unit is used for acquiring operation behaviors and operation tasks of a locomotive driver in a driving process to obtain driving operation data, and the driving operation data comprises operation data, response time, accuracy, eye movement data and limb skeleton data of the locomotive driver; The data processing and feature extraction analysis unit is used for preprocessing the driving operation data and extracting index feature values, wherein the index feature values comprise workload information, task performance information, eye fatigue features and limb fatigue features; the cross-system data time synchronization and each index feature dimensionless processing unit is used for performing time alignment on the operation data and normalizing and dimensionless processing on the index feature values so as to obtain a work load index, a task performance index, an eye movement feature index and a limb feature index; The locomotive driver operation fatigue fuzzy comprehensive evaluation unit is used for fusing the workload, the task performance, the eye movement characteristics and the limb characteristics to obtain a final evaluation result; and the locomotive driver operation fatigue comprehensive evaluation result grading unit is used for evaluating the operation fatigue state of the locomotive driver according to the