A Civil Aviation Pilot Psychological Quality Evaluation and Selection System

Abstract

The present invention discloses a civil aviation pilot psychological quality evaluation and selection system, including: a personnel information recording module: used to record the physiological data of the test personnel before the test and the data under the normal pupil shape, and record the physiological data of the test personnel and the pupil fluctuation data during the test; a simulated flight device: used to simulate and output the images of the civil aviation aircraft working, and at the same time, an eye tracker is integrated inside to collect pupil data. By determining the data of the left and right pupils of the test personnel, the subsequent detection is calibrated. At the same time, the pupil oscillation value Z is introduced. By judging the size of Z, the psychological fluctuation value of the test personnel within every 2 s is fed back, and the oscillation values of the left and right pupils are calculated separately. Furthermore, when the oscillation value Z is abnormal, the left pupil oscillation value and the right pupil oscillation value are used for judgment, so that the pupil oscillation data is more accurate and can better reflect the psychological fluctuation of the test personnel.

Description

Technical Field

[0001] The present invention relates to the technical field of pilot psychological quality assessment, and particularly to a civil aviation pilot psychological quality assessment and selection system. Background Art

[0002] Pilot psychological quality assessment is an important link in pilot selection, training and continuous airworthiness assessment, aiming to ensure that pilots possess the psychological qualities and abilities required to complete flight tasks;

[0003] The main purpose of pilot psychological quality assessment is to determine whether a pilot has the psychological qualities required for flight work, including positive flight motivation, good character, stable emotions, excellent attention and decision-making abilities, etc.

[0004] After retrieval, the invention patent with the Chinese patent number CN106814850A discloses a simulated flight operation test system and method based on line-of-sight trajectory. The method includes: when entering a certain simulated flight operation mode, receiving a test instruction of a simulated aircraft; retrieving a first line-of-sight trajectory corresponding to the test instruction according to the test instruction, and the first line-of-sight trajectory is preset and stored locally; collecting a second pupil image of a pilot, and determining a second line-of-sight trajectory of the pilot according to the second pupil image; comparing the deviation rate between the first line-of-sight trajectory and the second line-of-sight trajectory to assess the operation of the pilot.

[0005] Compared with the prior art, the invention patent with the Chinese patent number CN106814850A can, through the comparison of line-of-sight trajectories, form a standardized line-of-sight trajectory range by means of big data analysis of the line-of-sight trajectories of qualified pilots in a certain driving mode or scenario, and compare it with the line-of-sight trajectory of the tested pilot, and use the coincidence degree as one of the indexes for assessing flight trainees.

[0006] However, in the actual use process of the above system, it is difficult to accurately reflect the problem of abnormal pupils of the tested person during the actual test only through the comparison of the coincidence degree between the first line-of-sight trajectory and the second line-of-sight trajectory. At the same time, there is also a lack of accurate judgment on the reaction when the pupil changes in the simulated picture. Therefore, a civil aviation pilot psychological quality assessment and selection system is needed. Summary of the Invention

[0007] The purpose of the present invention is to solve the problems in the prior art that it is difficult to accurately reflect the problem of abnormal pupils of the tested person during the actual test, and at the same time, there is also a lack of accurate judgment on the reaction when the pupil changes in the simulated picture, and a civil aviation pilot psychological quality assessment and selection system is proposed.

[0008] In order to achieve the above purpose, the present invention adopts the following technical solutions:

[0009] A civil aviation pilot psychological quality evaluation and selection system, including a personnel information recording module: used to record the physiological data of the test personnel before the test and the data under normal pupil morphology, and record the physiological data of the test personnel and the pupil fluctuation data during the test;

[0010] A simulated flight device: including an external flight cockpit and an internal head instrument. The head instrument has real-time video monitoring of the pupils of the test personnel. The eye tracker inside the head instrument will move following the pupils, ensuring the data accuracy during pupil testing. The head instrument integrates an image input unit inside, which is used to collect the pupils and faces of the test personnel, and is used to collect the real-time brightness of the simulated flight images in the flight cockpit, simulate and output images of civil aviation aircraft work. At the same time, it integrates an eye tracker inside to collect pupil data;

[0011] A pupil recognition module: processes the pupil data collected by the eye tracker, and at the same time revises the data when the pupils deflect. Denote the average data of the left pupil from 0.3s to 0.5s before the test as M1, the average data of the right pupil as M2, and record the screen brightness K1 at this time;

[0012] Denote the average data of the left pupil from 0.25s to 0.35s after the screen appears as N1, the average data of the right pupil as N2, and record the screen brightness K2 at this time. The brightness of K1 and K2 is the same;

[0013] At the same time, record the data of the left and right pupils respectively;

[0014] A data processing module: determine the pupil baseline: S1 = (M1 + M2) / 2, S2 = (N1 + N2) / 2, where S1 is the left pupil baseline value and S2 is the right pupil baseline value;

[0015] Calculate the left pupil oscillation value within every 2s: (X is the left pupil oscillation value, L i is the i-th left pupil value);

[0016] Calculate the right pupil oscillation value within every 2s: (Y is the right pupil oscillation value, R i is the i-th right pupil value);

[0017] At the same time, calculate the pupil oscillation value Z within every 2s:

[0018] A data comparison module: when Z > 1.3, compare the X value and the Y value in this time period. If one of the X value and the Y value exceeds 15.7, then the psychology of the test personnel shows strong fluctuations;

[0019] When Z > 1.3, the contraction or dilation value of the left and right pupils within this time period exceeds 5.4%, and it is determined that the fluctuation value of the tester's pupils is relatively large. When one of the X value and the Y value exceeds 15.7, at this time, the deviation between the pupil value and the baseline value exceeds 10.4%. The instantaneous changes in the left and right pupils do not cause fluctuations in the pupil oscillation value Z, effectively processing the normal pupil data fluctuations caused by light stimulation and sudden changes in the simulated image, avoiding the evaluation of the tester due to normal pupil data fluctuations;

[0020] When Z < 1.3, the psychological fluctuations of the tester are gentle, meeting the evaluation and selection criteria.

[0021] The above technical solution further includes:

[0022] The eye tracker collects the pupil data of both eyes every 0.1 s, and at the same time records the data of the left and right pupils separately. Among them, the data of the left pupil is separately packaged into a data set A, and the data of the right pupil is also separately packaged into a data set B. The image input unit inside the simulated flight device takes real-time pictures of the tester's pupils.

[0023] During the test, when it is detected that the tester blinks, interpolation processing is performed on the pupil measurement values 150 ms before and 100 ms after the blink.

[0024] Based on the real-time captured pupil images, the first line-of-sight trajectory of the left eye and the second line-of-sight trajectory of the right eye are established. The data processing module compares and analyzes the coincidence degree of the first line-of-sight trajectory and the second line-of-sight trajectory by the least squares method, and records the time points with a deviation degree exceeding 3.4% at the same time, and transmits the recorded time points and deviation values to the data comparison module.

[0025] Record the time points when both eyes' pupils are deflected in the real-time captured pupil images, and obtain the data of the left and right pupils at this time according to the time points. Through deep learning using the pupil deflection angle obtained before the test, the pupil data collected by the eye movement instrument, and the actual pupil data, a pupil deflection correction model is established. Input the pupil data collected by the eye movement instrument during deflection and the deflection angle recognized by image recognition into the pupil deflection correction model to obtain the actual pupil size.

[0026] Compare the consistency of the left and right pupils when the pupils are deflected. When the abnormal pupil deflection time is between 35 ms and 60 ms, it is judged that the binocular following degree of the tester is average;

[0027] When the abnormal pupil deflection time is between 60 ms and 120 ms, it is judged that the binocular following degree of the tester is low;

[0028] When the time of abnormal pupil deflection exceeds 120 ms, it is determined that the binocular tracking ability of the tester is poor;

[0029] The coincidence degree between the first line-of-sight trajectory and the second line-of-sight trajectory represents the synchronous tracking degree of the tester's binoculars. At the same time, the first line-of-sight trajectory and the second line-of-sight trajectory also need to be compared with the tester's historical line-of-sight trajectory inside the system. Select the time period with a lower coincidence degree, and process the tester's current test data in combination with the X, Y, and Z values. And it is necessary to comprehensively evaluate the current test in combination with the graphic changes and brightness in this time period.

[0030] The brightness of the simulation demonstration in the flight simulation device is synchronously transmitted to the data processing module. The data processing module selects a brightness change interval where the brightness change exceeds 200 nits within 1 s. The brightness change interval is 6 s after the brightness change;

[0031] The data processing module judges the time for the pupil to return to the stable value within the brightness change interval, and transmits the judgment result to the data comparison module.

[0032] The present invention has the following beneficial effects:

[0033] In the present invention, by determining the data of the left and right pupils of the tester to calibrate the subsequent detection, at the same time, the pupil oscillation value Z is introduced. By judging the size of Z, it is used to feedback the psychological fluctuation value of the tester within every 2 s, and the oscillation values of the left and right pupils are calculated separately. Furthermore, when the oscillation value Z is abnormal, it can be judged through the left pupil oscillation value and the right pupil oscillation value, making the pupil oscillation data more accurate and better reflecting the psychological fluctuation of the tester. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 It is a flowchart of a civil aviation pilot psychological quality evaluation and selection system proposed by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0035] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0036] A civil aviation pilot psychological quality evaluation and selection system includes:

[0037] Personnel Information Recording Module: Used to record the physiological data of the test personnel before the test, as well as the data under normal pupil morphology, and record the physiological data of the test personnel and the pupil fluctuation data during the test;

[0038] Among them, the physiological data of the test personnel recorded before the test include heart rate, blood pressure, respiration, and blood oxygen parameters;

[0039] The heart rate requirements are that the heart rates of the pilots during sitting still and simulated flight are 69±7 beats per minute and 75±6 beats per minute respectively;

[0040] The respiration requirement is that the pulmonary ventilation volume during sitting still is 6.8±0.77 liters per minute per meter 2 (BTPS), and during simulated flight it is 7.0±0.74 liters per minute per meter 2 (BTPS);

[0041] The physiological data during the test is as described above;

[0042] Simulated Flight Equipment: Used to simulate and output the images of civil aviation aircraft working, and at the same time an eye tracker is integrated inside it to collect pupil data;

[0043] Among them, this equipment includes an external flight cockpit and an internal head instrument. The head instrument has real-time video monitoring of the pupils of the test personnel. The eye tracker inside the head instrument will move following the pupils, ensuring the data accuracy during pupil testing. The head instrument integrates an image input unit inside it. This image input unit is used to collect the pupils and faces of the test personnel, and in the flight cockpit, it is used to collect the real-time brightness of the simulated flight images.

[0044] Pupil Recognition Module: Processes the pupil data collected by the eye tracker, and at the same time revises the data when the pupil deflects. The average data of the left pupil from 0.3s to 0.5s before the test is denoted as M1, the average data of the right pupil is denoted as M2, and the screen brightness K1 at this time is recorded;

[0045] Among them, the time points when both pupils of both eyes deflect are recorded through the real-time captured pupil images, and the data of the left and right pupils at this time are obtained according to the time points. Through deep learning on the pupil deflection angles obtained before the test, the pupil data collected by the eye movement instrument, and the actual pupil data, a pupil deflection correction model is established. The pupil data collected by the eye movement instrument during deflection and the deflection angles recognized by the image are input into the pupil deflection correction model to obtain the actual pupil size. Subsequently, all pupil data is processed through this pupil deflection correction model. In practice, due to pupil deflection, it is easy to cause errors in the measurement results. The pupil deflection correction model reduces the generation of errors, making the subsequent data reaction based on pupil changes more real and reliable;

[0046] The average data of the left pupil from 0.25 s to 0.35 s after the screen appears is denoted as N1, and the average data of the right pupil is denoted as N2. Record the screen brightness K2 at this time. K1 and K2 have the same brightness. The purpose of setting the same brightness for K1 and K2 in this step is to detect the pupil response data of the tester generated by the screen under the same brightness. The fluctuation of the data at this time also reflects the psychological quality of the tester. Based on this, setting the data recording from 0.25 s to 0.35 s after the screen appears is also to reduce the deviation in subsequent detections and make the pupil baseline more accurate.

[0047] At the same time, the data of the left and right pupils are recorded separately. The data of the left pupil is separately packaged into a data set A, and the data of the right pupil is also separately packaged into a data set B;

[0048] Data processing module: Determine the pupil baseline: S1 = (M1 + M2) / 2, S2 = (N1 + N2) / 2. Here, S1 is the left pupil baseline value, and S2 is the right pupil baseline value. By separately selecting the baselines for the left and right pupils, the abnormal pupil oscillation value caused by the pilot's own physical condition can be reduced;

[0049] Calculate the left pupil oscillation value within every 2 s: X is the left pupil oscillation value, L i is the i-th left pupil value;

[0050] Calculate the right pupil oscillation value within every 2 s: Y is the right pupil oscillation value, R i is the i-th right pupil value;

[0051] The pupil recording data is recorded once every 0.1 s. The selected 2 s pupil oscillation value is the calculation and processing of the 20 pupil data collected.

[0052] At the same time, calculate the pupil oscillation value Z within every 2 s:

[0053] Among them, the value of the pupil oscillation value Z reflects the fluctuation value of the left and right eyes in the same time period. The larger the value, the more inconsistent the left and right pupils of the tester in this time period. The reasons are inattentiveness or large psychological fluctuations, and the pupils do not simultaneously and accurately recognize the screen changes.

[0054] Data comparison module: When Z > 1.3, compare the X value and the Y value in this time period. If one of the X value and the Y value exceeds 15.7, then the psychology of the tester shows strong fluctuations;

[0055] When Z > 1.3, the contraction or dilation value of the left and right pupils within this time period exceeds 5.4%, and it is determined that the fluctuation value of the tester's pupils is relatively large. When one of the X value and the Y value exceeds 15.7, at this time, the deviation of the pupil value from the baseline value exceeds 10.4%. The instantaneous changes in the left and right pupils do not cause fluctuations in the pupil oscillation value Z. The normal pupil data fluctuations caused by light stimulation and sudden changes in the simulated picture are effectively processed to avoid the evaluation of the tester due to normal pupil data fluctuations;

[0056] When Z < 1.3, the psychological fluctuations of the tester are gentle, meeting the evaluation and selection criteria.

[0057] The eye tracker collects and records the pupil data of both eyes every 0.1 s. However, in actual use, the pupil data recording interval of the eye tracker is 20 ms. The eye tracker selects the data every 0.1 s as the average data. At the same time, the image input unit inside the simulated flight device takes real-time pictures of the tester's pupils.

[0058] During the test, when it is detected that the tester blinks, interpolation processing is performed on the pupil measurement values 150 ms before and 100 ms after the blink;

[0059] Regarding the influence caused by blinking, there are two aspects. First, the pupil data recording before and after blinking is inaccurate. Second, the pupil light intake before and after blinking is uneven, which will cause deviations in the measured pupil data. However, the pupil data before and after blinking shows a linear change. Therefore, interpolation is used for the pupil measurement values 150 ms before and 100 ms after the blink to supplement the data lost during the blinking process, and the pupil data before and after blinking is compared to determine how long it takes for the pupil to return to the normal value after blinking.

[0060] Based on the real-time captured pupil images, the first line-of-sight trajectory of the left eye and the second line-of-sight trajectory of the right eye are established. The data processing module compares and analyzes the coincidence degree of the first line-of-sight trajectory and the second line-of-sight trajectory by the least squares method (the method of processing data by the least squares method belongs to the prior art and will not be elaborated here), and records the time points with a deviation degree exceeding 3.4% at the same time, and transmits the recorded time points and deviation values to the data comparison module;

[0061] The consistency of the left pupil and the right pupil when the pupil deflects is compared. When the abnormal pupil deflection time is between 35 ms and 60 ms, it is determined that the binocular tracking ability of the tester is average;

[0062] When the abnormal pupil deflection time is between 60 ms and 120 ms, it is determined that the binocular tracking ability of the tester is low;

[0063] When the time of abnormal pupil deflection exceeds 120 ms, it is determined that the binocular tracking ability of the tester is poor;

[0064] The coincidence degree between the first line-of-sight trajectory and the second line-of-sight trajectory represents the synchronous tracking degree of the tester's binoculars. At the same time, the first line-of-sight trajectory and the second line-of-sight trajectory also need to be compared with the tester's historical line-of-sight trajectory inside the system. The time period with a lower coincidence degree is selected, and the X, Y, and Z values are combined to process the tester's current test data. And it is necessary to combine the changes in the drawing surface and brightness during this time period to comprehensively evaluate the current test.

[0065] The brightness of the simulation demonstration in the flight simulation device is synchronously transmitted to the data processing module. The data processing module selects a light change interval where the brightness change exceeds 200 nits within 1 s. The light change interval is 6 s after the brightness change. The data processing module judges the time for the pupil to return to the stable value within the light change interval and transmits the judgment result to the data comparison module, reducing the abnormal values in pupil data collection caused by light stimulation.

[0066] Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A civil aviation pilot psychological quality assessment and selection system, characterized by: include: Personnel information recording module: used to record the tester's physiological data and pupil normal morphology data before the test, and to record the tester's physiological data and pupil fluctuation data during the test; Flight simulator: used to simulate and output images of civil aircraft operations, and has an integrated eye tracker for collecting pupil data; Pupil recognition module: processes pupil data collected by the eye tracker and modifies the data during pupil deflection. The average data of the left pupil from 0.3s to 0.5s before the test is recorded as M1, the average data of the right pupil is recorded as M2, and the screen brightness at this time is recorded as K1. The average data of the left pupil from 0.25s to 0.35s after the image appears is recorded as N1, and the average data of the right pupil is recorded as N2. The brightness of the image at this time is also recorded as K2. K1 and K2 have the same brightness. At the same time, the data of the left and right pupils are recorded separately; Data processing module: Determine pupil baseline: S1 = (M1 + M2) / 2, S2 = (N1 + N2) / 2, where S1 is the left pupil baseline value and S2 is the right pupil baseline value; Calculate the left pupil oscillation value every 2 seconds: X is the left pupil oscillation value, L i is the value of the left pupil at the i-th time; Calculate the right pupil oscillation value every 2 seconds: Y is the right pupil oscillation value, R i is the right pupil value at the i-th time; At the same time, the pupil oscillation value Z is calculated every 2 seconds: Data comparison module: When Z>1.3, compare the X value and Y value of the time period. If either the X value or the Y value exceeds 15.7, the tester's psychology has fluctuated strongly; When Z<1.3, the tester's psychological fluctuations are smooth, which is consistent with psychological quality assessment and selection.

2. A civil aviation pilot psychological quality assessment and selection system according to claim 1, characterized in that: The eye tracker collects pupil data of both eyes every 0.1s, and the image input unit inside the simulated flight equipment takes real-time photos of the test person's pupils.

3. The civil aviation pilot psychological quality assessment and selection system according to claim 1, characterized in that: When the test person is detected to have blinked during the test, the pupil measurement values 150ms before and 100ms after the blink are processed.

4. A civil aviation pilot psychological quality assessment and selection system according to claim 2, characterized in that: The first sight track of the left eye and the second sight track of the right eye are established through the pupil images captured in real time. The data processing module compares and analyzes the overlap between the first sight track and the second sight track, and records the time points where the deviation exceeds 3.4% at the same time. The recorded time points and deviation values are transmitted to the data comparison module.

5. The civil aviation pilot psychological quality assessment and selection system according to claim 2, characterized in that: The time points at which both pupils of the eyes deflect are recorded through real-time pupil images, and the data of the left and right pupils at that time are obtained based on the time points. A pupil deflection correction model is established by performing deep learning on the pupil deflection angle obtained before the test, the pupil data collected by the eye movement instrument, and the actual pupil data. The pupil data collected by the eye movement instrument during deflection and the deflection angle recognized by the image are input into the pupil deflection correction model to obtain the actual pupil size.

6. A civil aviation pilot psychological quality assessment and selection system according to claim 5, characterized in that: Compare the consistency of the left and right pupils when the pupils deflect. If the abnormal pupil deflection time is between 35ms and 60ms, the tester's binocular tracking ability is judged to be average. When the duration of pupil deflection abnormality is between 60ms and 120ms, it is judged that the tester's binocular tracking degree is low; When the duration of abnormal pupil deflection exceeds 120ms, the tester is judged to have poor binocular tracking.

7. The civil aviation pilot psychological quality assessment and selection system according to claim 2, characterized in that: The brightness of the simulation demonstration in the simulated flight equipment is synchronously transmitted to the data processing module. The data processing module selects the brightness change interval in which the brightness change exceeds 200 nits within 1 second. The brightness change interval is 6 seconds after the brightness change.

8. A civil aviation pilot psychological quality assessment and selection system according to claim 7, characterized in that: The data processing module determines the time it takes for the pupil to return to a stable value within the light change interval, and transmits the determination result to the data comparison module.

Images