Pilot situational consciousness state evaluation method and device based on physiological data

Abstract

The invention provides a pilot situation consciousness state evaluation method and device based on physiological data, and is applied to the technical field of artificial intelligence. The method comprises the following steps: acquiring physiological data of a pilot in a preset duration in real time, wherein the physiological data comprises electrocardio data and eye movement data; determining an electrocardio characteristic index according to the electrocardio data, and determining an eye movement characteristic index according to the eye movement data; inputting the electrocardio characteristic indexes and the eye movement characteristic indexes into a state evaluation model as physiological characteristic indexes to obtain a situation awareness state of the pilot output by the state evaluation model; wherein the state evaluation model is obtained by training a physiological feature index sample and a situational consciousness state sample corresponding to the pilot sample. According to the method, on-line evaluation of the situational awareness state is realized through the state evaluation model based on the physiological characteristic indexes corresponding to the physiological data acquired in real time, and human participation is effectively avoided in the whole process, so that the accuracy of the finally determined situational awareness state is relatively high.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence technology, and in particular, to a method and device for evaluating the situation awareness state of pilots based on physiological data. Background Art

[0002] The situation awareness (SA) state of a pilot refers to the pilot's perception of various elements in the environment, understanding of the meaning of these elements, and prediction of the subsequent state of these elements within a specific time and space. Achieving real-time evaluation of the pilot's situation awareness state is the key condition for reflecting and grasping the pilot's cognitive status of various factors and conditions affecting the operation during a specific time period and specific situation, and is of great significance for reducing flight accidents caused by errors in the situation awareness state.

[0003] Traditional methods for evaluating the situation awareness state of pilots usually include subjective evaluation methods, memory probe (freeze / real-time) measurement methods, and task performance measurement methods, etc. Among them, the subjective evaluation method and the memory probe measurement method belong to the direct evaluation measurement methods of the situation awareness state, and the task performance measurement method belongs to the indirect evaluation measurement method of the situation awareness state. However, whether it is the subjective evaluation method and the memory probe measurement method, or the task performance measurement method, there will be human errors due to human participation in the evaluation measurement, and thus the accuracy of the finally determined evaluation result (i.e., the situation awareness state of the pilot) is relatively low. Summary of the Invention

[0004] This application provides a method and device for evaluating the situation awareness state of pilots based on physiological data. Since there is a very close correlation between physiological data and the situation awareness state, after the electronic device obtains the physiological characteristic indexes corresponding to the physiological data in real time, the physiological characteristic indexes can be used as the input of the state evaluation model, and the situation awareness state of the pilot can be evaluated online through the state evaluation model. The whole process effectively avoids human participation, making the accuracy of the finally determined situation awareness state relatively high, and thus effectively ensuring the flight safety of the pilot during the flight process.

[0005] This application provides a method for evaluating the situation awareness state of pilots based on physiological data, including: obtaining the physiological data of the pilot within a preset time period in real time, where the physiological data includes electrocardiogram data and eye movement data; determining electrocardiogram characteristic indexes according to the electrocardiogram data, and determining eye movement characteristic indexes according to the eye movement data; inputting the electrocardiogram characteristic indexes and the eye movement characteristic indexes as physiological characteristic indexes into a state evaluation model to obtain the situation awareness state of the pilot output by the state evaluation model; where the state evaluation model is trained based on physiological characteristic index samples and situation awareness state samples corresponding to pilot samples.

[0006] A method for evaluating the situation awareness state of a pilot based on physiological data according to an embodiment of the present application, the state evaluation model is trained based on the following steps: determining a plurality of groups of data sample sets according to the physiological characteristic index samples and the situation awareness state samples, each group of data sample sets includes a training set and a test set, and each group of data sample sets is different; for each group of data sample sets, updating the model parameters in the first state evaluation model according to the training set in the data sample set to obtain a second state evaluation model; determining the model accuracy of the second state evaluation model according to the test set in the data sample set; determining the second state evaluation model with the highest model accuracy among the multiple second state evaluation models as the state evaluation model.

[0007] A method for evaluating the situation awareness state of a pilot based on physiological data according to an embodiment of the present application, the updating the model parameters in the first state evaluation model according to the training set in the data sample set to obtain a second state evaluation model includes: inputting the physiological characteristic index samples of the training set in the data sample set into the first state evaluation models corresponding to multiple model parameters respectively to obtain the predicted situation awareness states of the pilot samples output by the multiple first state evaluation models respectively; determining the similarity between each predicted situation awareness state and the situation awareness state sample of the training set in the data sample set; determining the second state evaluation model according to the model parameters of the first state evaluation model corresponding to the maximum similarity among the multiple similarities.

[0008] A method for evaluating the situation awareness state of a pilot based on physiological data according to an embodiment of the present application, the electrocardiogram data includes: electrocardiogram waveform data; the electrocardiogram characteristic indexes include: average heart rate, average RR interval and standardized target frequency band power; the determining the electrocardiogram characteristic indexes according to the electrocardiogram data includes: according to the RR intervals and the electrocardiogram sampling frequency in the electrocardiogram waveform data, determining the average heart rate, where is an integer greater than or equal to 2, and each RR interval is used to represent the time interval between two adjacent R waves corresponding in the electrocardiogram waveform data; according to the RR intervals, determining the average RR interval; determining the standardized target frequency band power according to the target frequency band power of the RR interval sequence in the electrocardiogram waveform data.

[0009] A method for evaluating the situation awareness state of a pilot based on physiological data according to an embodiment of the present application, the determining the average heart rate according to the RR intervals and the electrocardiogram sampling frequency in the electrocardiogram waveform data includes: using a first formula to determine the average heart rate; where the first formula is: ; represents the average heart rate; represents the ECG sampling frequency; represents the th RR interval among the RR intervals; determining the average RR interval according to the RR intervals includes: using a second formula to determine the average RR interval; wherein, the second formula is: ; represents the average RR interval.

[0010] According to a method for evaluating the pilot's situation awareness state based on physiological data provided by an embodiment of the present application, the target frequency band power includes a first frequency band power and a second frequency band power greater than the first frequency band power; the standardized target frequency band power includes a standardized first frequency band power and a standardized second frequency band power; determining the standardized target frequency band power according to the target frequency band power of the RR interval sequence in the ECG waveform data includes: using a third formula to determine the standardized first frequency band power; and using a fourth formula to determine the standardized second frequency band power; wherein, the third formula is: ; the fourth formula is: ; represents the standardized first frequency band power; represents the standardized second frequency band power; represents the first frequency band power; represents the second frequency band power; represents the first frequency band power and the second frequency band power sum; represents a third frequency band power less than the first frequency band power.

[0011] According to a method for evaluating the pilot's situation awareness state based on physiological data provided by an embodiment of the present application, the eye movement data includes: fixation points, saccade points and pupil diameters at valid sampling points, , and are all integers greater than or equal to 2; the eye movement characteristic indexes include: fixation rate, saccade rate, pupil diameter and average fixation duration; determining the eye movement characteristic indexes according to the eye movement data includes: determining the fixation rate according to the fixation points and the preset duration; determining the saccade rate according to the saccade points and the preset duration; determining according to the The pupil diameters of valid sampling points are used to determine the pupil diameter; according to the fixation duration of each fixation point among the fixation points, the average fixation duration is determined.

[0012] An embodiment of the present application further provides a pilot situation awareness state evaluation device based on physiological data, including: A data acquisition module, configured to acquire physiological data of a pilot in a preset duration in real time, where the physiological data includes electrocardiogram data and eye movement data; A data interface module, configured to determine electrocardiogram characteristic indexes according to the electrocardiogram data, and determine eye movement characteristic indexes according to the eye movement data; An online evaluation module, configured to input the electrocardiogram characteristic indexes and the eye movement characteristic indexes as physiological characteristic indexes into a state evaluation model, and obtain the situation awareness state of the pilot output by the state evaluation model; wherein, the state evaluation model is trained based on physiological characteristic index samples and situation awareness state samples corresponding to pilot samples.

[0013] The present application further provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, where when the processor executes the computer program, the method for evaluating the situation awareness state of a pilot based on physiological data as described in any one of the above is implemented.

[0014] The present application further provides a non-transitory computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the method for evaluating the situation awareness state of a pilot based on physiological data as described in any one of the above is implemented.

[0015] The present application further provides a computer program product, including a computer program, and when the computer program is executed by a processor, the method for evaluating the situation awareness state of a pilot based on physiological data as described in any one of the above is implemented.

[0016] The pilot situation awareness state evaluation method and device based on physiological data provided by the embodiments of the present application obtain the physiological data of the pilot within a preset time period in real time, where the physiological data includes electrocardiogram data and eye movement data; determine electrocardiogram characteristic indexes according to the electrocardiogram data, and determine eye movement characteristic indexes according to the eye movement data; input the electrocardiogram characteristic indexes and the eye movement characteristic indexes as physiological characteristic indexes into a state evaluation model to obtain the situation awareness state of the pilot output by the state evaluation model; wherein, the state evaluation model is trained based on physiological characteristic index samples and situation awareness state samples corresponding to pilot samples. In this method, since there is a very close correlation between physiological data and situation awareness state, after the electronic device obtains the physiological characteristic indexes corresponding to the physiological data in real time, the physiological characteristic indexes can be used as the input of the state evaluation model, and the situation awareness state of the pilot can be evaluated online through the state evaluation model. The whole process effectively avoids human participation, making the accuracy of the finally determined situation awareness state relatively high, and thus effectively ensuring the flight safety of the pilot during the flight process. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the present application or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0018] Figure 1 is a schematic flowchart of the pilot situation awareness state evaluation method based on physiological data provided by the embodiments of the present application; Figure 2 is a schematic diagram of the scenario for dynamically evaluating the situation awareness state of the pilot provided by the embodiments of the present application; Figure 3 is a schematic structural diagram of the pilot situation awareness state evaluation system based on physiological data provided by the embodiments of the present application; Figure 4 is a schematic diagram of the pilot information collection interface provided by the embodiments of the present application; Figure 5 is a schematic diagram of the parameter setting window provided by the embodiments of the present application; Figure 6 is a schematic diagram of the online evaluation main interface provided by the embodiments of the present application; Figure 7 is a schematic diagram of the output result interface provided by the embodiments of the present application; Figure 8 is a schematic diagram of the output report interface provided by the embodiments of the present application; Figure 9 FIG. Figure 9 is a schematic structural diagram of a pilot situation awareness state evaluation device based on physiological data provided by an embodiment of the present application; Figure 10 FIG. Figure 10 is a schematic structural diagram of an electronic device provided by an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0019] In order to make the objectives, technical solutions, and advantages of the present application clearer, the technical solutions in the present application will be clearly and completely described below with reference to the accompanying drawings in the present application. Apparently, the described embodiments are some, but not all, of the embodiments of the present application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present application without making creative efforts shall fall within the protection scope of the present application.

[0020] It should be noted that the pilot situation awareness state evaluation method based on physiological data provided by the embodiments of the present application is a physiological measurement method. Compared with traditional pilot situation awareness state evaluation methods, namely subjective evaluation methods, memory probe measurement methods, and task performance measurement methods, this physiological measurement method is considered an effective means for online evaluation of the pilot's operation state (i.e., situation awareness state) because of its good objectivity, real-time performance, and limited task invasiveness.

[0021] The following takes an electronic device as an example to elaborate in detail on the pilot situation awareness state evaluation method based on physiological data provided by the embodiments of the present application: Figure 1 FIG. is a schematic flowchart of the pilot situation awareness state evaluation method based on physiological data provided by the embodiments of the present application. As Figure 1 shown, the method includes the following steps 101 to 104.

[0022] Step 101: Real-time acquire physiological data of a pilot within a preset duration, where the physiological data includes electrocardiogram data and eye movement data.

[0023] Among them, a pilot refers to a professional who operates an aircraft (such as an airplane, helicopter, etc.) to perform flight tasks and needs to have corresponding flight qualifications and skills.

[0024] The preset duration refers to a pre-set time interval used to limit the duration of data acquisition. Exemplarily, the preset duration is 30 seconds (s).

[0025] Physiological data refers to physiological signals collected by physiological sensors and used to reflect the physiological state of the human body.

[0026] Electrocardiogram data refers to the cardiac electrical activity signals recorded by an electrocardiogram device (such as an electrocardiograph) and is used to evaluate cardiac function and rhythm.

[0027] Eye movement data refers to the eye movement trajectory and fixation characteristics recorded by an eye movement device (such as an eye tracker), which is used to reflect visual attention and cognitive states. It should be noted that due to the certain asynchrony and non-real-time nature in the calculation of eye movement characteristics, the electrocardiogram sampling frequency of the eye tracker is 60Hz, that is, 60 eye movement data are collected per second.

[0028] It should be noted that there is no limit on the time sequence for the electronic device to obtain electrocardiogram data and eye movement data.

[0029] Optionally, before step 101, the method may further include: the electronic device collects personal information of the pilot.

[0030] Optionally, the above personal information at least includes: recorded information such as name, age, height, occupation, number, gender, vision, education level, experimental conditions, experimental date, number of experiments, experimental site, and experimental duration, etc.

[0031] Step 102: Determine electrocardiogram characteristic indexes according to the electrocardiogram data, and determine eye movement characteristic indexes according to the eye movement data.

[0032] Among them, the electrocardiogram characteristic index refers to the quantitative parameter extracted by analyzing the cardiac electrical activity signal.

[0033] The eye movement characteristic index refers to the quantitative parameter extracted by analyzing the eye movement data.

[0034] Due to the certain non-real-time nature in the calculation of eye movement data, the electronic device realizes the index calculation conversion of the eye movement data at preset time intervals, that is, determines the eye movement characteristic index according to the eye movement data. At the same time, the electronic device also performs index calculation conversion on the cardiac movement data within the preset time duration to obtain the electrocardiogram characteristic index.

[0035] It should be noted that there is no limit on the time sequence for the electronic device to determine the electrocardiogram characteristic index and the eye movement characteristic index.

[0036] The following elaborates in detail on how the electronic device determines the electrocardiogram characteristic index according to the electrocardiogram data: In some embodiments, the electrocardiogram data may include: electrocardiogram waveform data; the electrocardiogram characteristic indexes may include: Mean Heart Rate (MHR), Mean RR Interval, and normalized target frequency band power.

[0037] Among them, the electrocardiogram waveform data refers to the record of cardiac electrical activity on the body surface, which is a continuous waveform formed after being collected and amplified by electrodes. For example, the electrocardiogram waveform data can be an electrocardiogram.

[0038] The average heart rate refers to the number of heartbeats per unit time. For example, when the unit time is one minute, the unit of the average heart rate is beats per minute.

[0039] The average RR interval refers to the average value of all RR intervals within a preset duration, and the unit of the average RR interval is millisecond (ms).

[0040] The normalized target frequency band power refers to the target frequency band spectrum of the RR interval sequence in the electrocardiogram waveform data.

[0041] In some embodiments, the electronic device determines electrocardiogram characteristic indicators according to the electrocardiogram data, which may include: the electronic device determines the average heart rate according to the RR intervals and the electrocardiogram sampling frequency in the electrocardiogram waveform data. is an integer greater than or equal to 2, and each RR interval is used to represent the time interval between two adjacent R waves corresponding in the electrocardiogram waveform data; the electronic device determines the average RR interval according to RR intervals; the electronic device determines the normalized target frequency band power according to the target frequency band power of the RR interval sequence in the electrocardiogram waveform data.

[0042] Among them, RR intervals are the core waveforms in the above electrocardiogram waveform data. It should be noted that each R wave represents a depolarization of the ventricle, and the RR interval represents a complete cardiac cycle, that is, the time interval from one ventricular contraction to the next ventricular contraction. The length of the RR interval is closely related to the cardiac rhythm and is an important indicator for judging whether the heart rate and rhythm are normal.

[0043] The RR interval sequence is a set of RR intervals on the electrocardiogram within a period of time (i.e., the preset duration), and is used to evaluate the cardiac rhythm and function.

[0044] During the process of the electronic device determining the electrocardiogram characteristic indicators, it can first determine the RR intervals in the electrocardiogram waveform data. Then, the electronic device calculates the average heart rate by calculating the RR intervals and the electrocardiogram sampling frequency. Next, the electronic device calculates the average value of these RR intervals to obtain the average RR interval. Finally, the electronic device performs a normalization calculation on the target frequency band power of the RR interval sequence in the electrocardiogram waveform data to obtain the normalized target frequency band power. These electrocardiogram characteristic indicators can be used as input data for subsequent state evaluation models.

[0045] It should be noted that the timing of the electronic device determining the average heart rate, the average RR interval, and the normalized target frequency band power is not limited.

[0046] In some embodiments, the electronic device determines the average heart rate according to the RR intervals and the electrocardiogram sampling frequency in the electrocardiogram waveform data, which may include: the electronic device uses a first formula to determine the average heart rate.

[0047] Wherein, the first formula is: ; represents the average heart rate; represents the electrocardiogram sampling frequency; represents the th RR interval among the

[0048] Exemplarily, when the unit time is one minute, the value of is 60. At this time, the above first formula is:

[0049] In some embodiments, the electronic device determines the average RR interval according to RR intervals, which may include: the electronic device uses a second formula to determine the average RR interval.

[0050] Wherein, the second formula is: ; represents the average RR interval.

[0051] In some embodiments, the target band power may include a first band power and a second band power greater than the first band power; the normalized target band power may include a normalized first band power and a normalized second band power.

[0052] Among them, the first band power is a low frequency (LF) band power. Exemplarily, the value range of the first band power is 0.04 Hz to 0.15 Hz.

[0053] The second band power is a high frequency (HF) band power. Exemplarily, the value range of the second band power is 0.15 Hz to 0.4 Hz.

[0054] The normalized first band power is a normalized low frequency power, which is used to reflect the low frequency spectrum of the RR interval sequence.

[0055] The normalized second band power is a normalized high frequency power, which is used to reflect the high frequency spectrum of the RR interval sequence.

[0056] In some embodiments, for the electronic device to determine the normalized target frequency band power based on the target frequency band power of the RR interval sequence in the electrocardiogram waveform data, it may include: the electronic device uses the third formula to determine the normalized first frequency band power; and uses the fourth formula to determine the normalized second frequency band power.

[0057] Among them, the third formula is: ; The fourth formula is: ; represents the normalized first frequency band power; represents the normalized second frequency band power; represents the first frequency band power; represents the second frequency band power; represents the first frequency band power and the second frequency band power sum, which is a total power; represents the third frequency band power less than the first frequency band power. This third frequency band power is an extremely low frequency power. It should be noted that the maximum value in the value range of this third frequency band power is less than the minimum value of the above-mentioned first frequency band power.

[0058] It should be noted that the timing for the electronic device to determine the normalized first frequency band power and the normalized second frequency band power is not limited.

[0059] Optionally, for the electronic device to determine the electrocardiogram feature index based on the electrocardiogram data, it may include: the electronic device determines the initial electrocardiogram feature index according to the electrocardiogram data; the electronic device performs normalization processing and preprocessing operations on the initial electrocardiogram feature index to obtain the electrocardiogram feature index.

[0060] Optionally, the normalization processing may include: removing null values and / or outliers, etc.

[0061] Optionally, the preprocessing operation may include: normalization processing, etc.

[0062] After the electronic device analyzes the electrocardiogram data and extracts quantization parameters to obtain the initial electrocardiogram feature index, for the convenience of subsequent unified processing of data, the electronic device may further perform normalization processing and preprocessing operations on the initial electrocardiogram feature index to obtain the electrocardiogram feature index.

[0063] The following elaborates in detail on how the electronic device determines the eye movement feature index based on the eye movement data: In some embodiments, the eye movement data may include: number of fixation points, number of saccade points and number of valid sampling point pupil diameters, , and All are integers greater than or equal to 2; eye movement feature indicators may include: fixation rate (FixationRate, FR), saccade rate (Saccade Rate, SR), pupil diameter (Pupil Diameter, PD) and mean fixation duration (Mean Fixation Duration, MFD).

[0064] Among them, the fixation point refers to a short-term stay point where the pilot's eyes remain relatively still (without obvious movement) at a certain position, usually lasting for a first preset duration (such as 100-300ms) to reflect attention to a specific visual target or information processing.

[0065] A saccade point refers to a trajectory point where the eyeball moves rapidly between fixation points, usually lasting for a second preset duration (such as 20-50ms) at a speed of 300-700 degrees per second (° / s), and is used to switch visual focus targets.

[0066] The pupil diameter of a valid sampling point refers to the pupil diameter value corresponding to a sampling point that is determined to be valid (such as no blinking, occlusion, or signal loss) during the eye movement data collection process. The unit of the pupil diameter of the valid sampling point is millimeter (mm).

[0067] The fixation rate refers to the number of fixations per unit time.

[0068] The scanning rate refers to the number of scanning points per unit time.

[0069] Pupil diameter is The average pupil diameter of the effective sampling points. The unit of the pupil diameter is (mm).

[0070] The average fixation duration refers to the ratio of the total fixation duration of the eye movement to the number of fixation points, and the unit of the average fixation duration is ms.

[0071] In some embodiments, the electronic device determines the eye movement characteristic index according to the eye movement data, which may include: the electronic device determines the eye movement characteristic index according to the eye movement data. The electronic device determines the gaze rate according to the preset gaze point and duration. scanning points and preset duration to determine the scanning rate; the electronic device The pupil diameter is determined by the pupil diameter of the effective sampling points; the electronic device The fixation duration of each fixation point among the fixation points is calculated to determine the average fixation duration.

[0072] It should be noted that there is no limit to the timing of the electronic device determining the gaze rate, the scan rate, the pupil diameter and the average gaze duration.

[0073] Optionally, the electronic device is based on Based on a fixation point and a preset duration, determine the fixation rate, which may include: The electronic device uses the fifth formula to determine the fixation rate.

[0074] Among them, the fifth formula is: ; represents the fixation rate; represents the preset duration.

[0075] Optionally, the electronic device determines the saccade rate according to the number of saccade points and the preset duration, which may include: The electronic device uses the sixth formula to determine the saccade rate.

[0076] Among them, the sixth formula is: ; represents the saccade rate.

[0077] Optionally, the electronic device determines the pupil diameter according to the pupil diameters of a number of valid sampling points, which may include: The electronic device uses the seventh formula to determine the pupil diameter.

[0078] Among them, the seventh formula is: ; represents the pupil diameter; represents the th pupil diameter among the pupil diameters of a number of valid sampling points.

[0079] Optionally, the electronic device determines the average fixation duration according to the fixation duration of each fixation point among a number of fixation points, which may include: The electronic device uses the eighth formula to determine the average fixation duration.

[0080] Among them, the eighth formula is: ; represents the average fixation duration; represents the th fixation duration of the

[0081] Optionally, the electronic device determines an eye movement feature index according to the eye movement data, which may include: The electronic device determines an initial eye movement feature index according to the eye movement data; The electronic device performs a normalization process and a preprocessing operation on the initial eye movement feature index to obtain the eye movement feature index.

[0082] After the electronic device analyzes the eye movement data and extracts quantitative parameters to obtain the initial eye movement feature index, for the convenience of subsequent unified processing of the data, the electronic device can further perform normalization processing and preprocessing operations on the initial eye movement feature index to obtain the eye movement feature index.

[0083] Step 103: Use the electrocardiogram feature index and the eye movement feature index as physiological feature indexes and input them into the state evaluation model to obtain the situational awareness state of the pilot output by the state evaluation model.

[0084] Among them, the state evaluation model is trained based on the physiological feature index samples and situational awareness state samples corresponding to the pilot samples.

[0085] It should be noted that the above state evaluation model can be three "One-Against-All" (OAA) classifiers, namely the first classifier, the second classifier, and the third classifier. Among them, the first classifier is used to realize the discrimination of the high situational awareness state; the second classifier is used to realize the discrimination of the medium situational awareness state; the third classifier is used to realize the discrimination of the low situational awareness state.

[0086] Based on this, after the electronic device obtains the electrocardiogram feature index and the eye movement feature index, it can input these two feature indexes as physiological feature indexes into these three classifiers, so that these three classifiers can respectively calculate the physiological feature indexes.

[0087] Specifically, during the process of the first classifier analyzing and calculating the physiological feature index, the vector corresponding to the first classifier is used as the positive set +1, and the vectors corresponding to the second classifier and the third classifier are used as the negative set -1. During the process of the second classifier analyzing and calculating the physiological feature index, the vector corresponding to the second classifier is used as the positive set +1, and the vectors corresponding to the first classifier and the third classifier are used as the negative set -1. During the process of the third classifier analyzing and calculating the physiological feature index, the vector corresponding to the third classifier is used as the positive set +1, and the vectors corresponding to the first classifier and the second classifier are used as the negative set -1. In this way, each classifier will output a prediction result, that is, the first classifier outputs the first prediction result F(1), the second classifier outputs the second prediction result F(2), and the third classifier outputs the third prediction result F(3).

[0088] Then, the electronic device takes the maximum value of the above three prediction results as the situational awareness state of the pilot output by the state evaluation model, that is, max{F(1), F2(2), F(3)}.

[0089] It should be noted that if the situation awareness state is F(1), then this situation awareness state is a high situation awareness state; if the situation awareness state is F(2), then this situation awareness state is a medium situation awareness state; if the situation awareness state is F(3), then this situation awareness state is a low situation awareness state, so as to realize the online real-time evaluation of different situation awareness state levels.

[0090] It should be noted that during the flight of a pilot, the electronic device can obtain the physiological data of the pilot in the current preset time period in real time, determine the corresponding physiological characteristic indexes, and then output the situation awareness state corresponding to the pilot in the current preset time period by means of the state evaluation model; then, the electronic device obtains the physiological data in a new preset time period in real time, determines the corresponding physiological characteristic indexes, and then outputs the situation awareness state corresponding to the pilot in the new preset time period by means of the above state evaluation model. In this way, the electronic device can determine the situation awareness state corresponding to the pilot in each preset time period during the flight, and realize the dynamic evaluation of the situation awareness state of the pilot during the flight by following the change of the pilot's physiological state.

[0091] Exemplarily, Figure 2 is a schematic diagram of a scenario for dynamically evaluating the situation awareness state of a pilot provided by an embodiment of the present application. As Figure 2 shown, each preset time period is 30s. Based on this, the electronic device can determine the situation awareness state corresponding to the pilot within each 30s.

[0092] It should be noted that the execution subject during the training of the state evaluation model can be an electronic device or other devices. If it is other devices, after the state evaluation model is trained by the other devices, the state evaluation model can be transmitted to the electronic device for subsequent use.

[0093] Taking the electronic device as an example below, the training of the state evaluation model will be elaborated in detail: In some embodiments, the state evaluation model is trained based on the following steps: The electronic device determines multiple groups of data sample sets according to the physiological characteristic index samples and situation awareness state samples. Each group of data sample sets can include a group of training sets and a group of test sets, and each group of data sample sets is different; for each group of data sample sets, the electronic device updates the model parameters in the first state evaluation model according to the training sets in the data sample sets to obtain a second state evaluation model; according to the test sets in the data sample sets, the model accuracy of the second state evaluation model is determined; the electronic device determines the second state evaluation model with the highest model accuracy among the multiple second state evaluation models as the state evaluation model.

[0094] During the process of training the state evaluation model, the electronic device can first obtain the electrocardiogram data samples, eye movement data samples, and situation awareness state samples corresponding to the pilot samples, and determine the electrocardiogram feature index samples corresponding to the electrocardiogram data samples and the eye movement feature index samples corresponding to the eye movement data samples. Then, the electronic device uses the electrocardiogram feature index samples and the eye movement feature index samples as physiological feature index samples, and uses the above situation awareness state samples as labels to determine multiple groups of data sample sets.

[0095] During the process of determining multiple groups of data set samples, taking three groups of data set samples as an example, the first group of data set samples includes a training set composed of 90% of the physiological feature index samples and 90% of the situation awareness state samples, and a test set composed of 10% of the physiological feature index samples and 10% of the situation awareness state samples; the second group of data set samples includes a training set composed of 80% of the physiological feature index samples and 80% of the situation awareness state samples, and a test set composed of 20% of the physiological feature index samples and 20% of the situation awareness state samples; the first group of data set samples includes a training set composed of 70% of the physiological feature index samples and 70% of the situation awareness state samples, and a test set composed of 30% of the physiological feature index samples and 30% of the situation awareness state samples.

[0096] Since the processing process of each group of data set samples is the same, for a group of data set samples, the electronic device can update the model parameters in the first state evaluation model according to the training set in this group of data sample sets to obtain the second state evaluation model, and then determine the model accuracy of the second state evaluation model according to the test set in this group of data sample sets. In this way, the electronic device can obtain the model accuracies of multiple second state evaluation models.

[0097] Finally, the electronic device determines the second state evaluation model with the highest model accuracy among these multiple second state evaluation models as the state evaluation model.

[0098] It should be noted that by using multiple different data sample sets (each containing a training set and a test set) for model training and evaluation, the model can be exposed to more diverse data. Different data sample sets cover different changes and distributions of physiological feature indicators and situation awareness states, which helps the model learn more general and essential feature laws, rather than simply adapting to the features of a specific data set. This enables the finally determined state evaluation model to have better generalization ability to be able to determine the situation awareness state of pilots with higher accuracy subsequently.

[0099] In some embodiments, the electronic device updates the model parameters in the first state evaluation model according to the training set in the data sample set to obtain a second state evaluation model, which may include: the electronic device inputs the physiological characteristic index samples of the training set in the data sample set into the first state evaluation models corresponding to multiple model parameters respectively, and obtains the predicted situation awareness states of the pilot samples respectively output by the multiple first state evaluation models; the electronic device determines the similarity between each predicted situation awareness state and the situation awareness state sample of the training set in the data sample set; the electronic device determines the second state evaluation model according to the model parameters of the first state evaluation model corresponding to the maximum similarity among the multiple similarities.

[0100] It should be noted that the first state evaluation model corresponds to multiple model parameters. The above process is to determine the optimal model parameters from these multiple model parameters, and use the first state evaluation model corresponding to the optimal model parameters as the second state evaluation model.

[0101] Specifically, for a set of data set samples and the first state evaluation model corresponding to a model parameter, the electronic device can input the physiological characteristic index samples of the training set in this set of data sample set into the first state evaluation model, and obtain the predicted situation awareness state output by the first state evaluation model. In this way, for a set of data set samples and the first state evaluation models corresponding to multiple model parameters, the electronic device can finally determine the predicted situation awareness states respectively output by the multiple first state evaluation models.

[0102] Then, the electronic device respectively determines the similarity between the above multiple predicted situation awareness states and the situation awareness state samples of the training set in this set of data set samples, determines the maximum similarity from these multiple similarities, and further determines the model parameters of the first state evaluation model corresponding to the maximum similarity as the optimal model parameters, and determines the first state evaluation model corresponding to the optimal model parameters as the second state evaluation model.

[0103] In summary, the whole process inputs the physiological characteristic index samples of the training set into the first state evaluation models with multiple different model parameters, obtains multiple predicted situation awareness states, and calculates the similarity between these multiple predicted situation awareness states and the real situation awareness state samples, which can intuitively measure the closeness of each model's prediction result to the actual situation. Determining the second state evaluation model with the maximum similarity as the standard is equivalent to accurately selecting the model that can most accurately reflect the real situation awareness state of the pilot from multiple first state evaluation models, avoiding subjective judgment and blind selection, and improving the accuracy of model selection.

[0104] Optionally, after step 103, the method may further include at least one of the following implementation manners: Implementation Mode 1: The electronic device visually outputs the status evaluation result in a preset form, and the status evaluation result is a status evaluation graph of the pilot's situation awareness state that changes over time.

[0105] Optionally, the preset form may include: a data list form and / or an image form, etc.

[0106] Optionally, the image form may include formats such as Portable Network Graphics (PNG) or Plain Text (TXT) files, etc.

[0107] This enables the pilot to intuitively determine the pilot's situation awareness state during the flight.

[0108] Implementation Mode 2: The electronic device outputs the pilot's personal information.

[0109] This enables the pilot to intuitively determine whether the pilot's personal information is incorrect.

[0110] It should be noted that whether it is the output of the status evaluation result or the output of the personal information, the data can be exported in a preset image form by specifying a path.

[0111] In the embodiment of the present application, physiological data of the pilot within a preset duration is obtained in real time, and the physiological data includes electrocardiogram data and eye movement data; according to the electrocardiogram data, an electrocardiogram characteristic index is determined, and according to the eye movement data, an eye movement characteristic index is determined; the electrocardiogram characteristic index and the eye movement characteristic index are used as physiological characteristic indexes and input into the status evaluation model to obtain the situation awareness state of the pilot output by the status evaluation model; wherein, the status evaluation model is trained based on physiological characteristic index samples and situation awareness state samples corresponding to pilot samples. In this method, since there is a very close correlation between physiological data and the situation awareness state, after the electronic device obtains the physiological characteristic indexes corresponding to the physiological data in real time, the physiological characteristic indexes can be used as the input of the status evaluation model, and the online evaluation of the pilot's situation awareness state can be realized through the status evaluation model. The whole process effectively avoids human participation, making the accuracy of the finally determined situation awareness state relatively high, and thus effectively ensuring the flight safety of the pilot during the flight.

[0112] To better understand the method for evaluating the pilot's situation awareness state based on physiological data provided in the embodiment of the present application, the following elaborates in detail on the system for evaluating the pilot's situation awareness state based on physiological data: Figure 3 It is a schematic structural diagram of the system for evaluating the pilot's situation awareness state based on physiological data provided in the embodiment of the present application. As Figure 3As shown, the system may include: a data acquisition module, a data interface module, an online evaluation module, and an output report module.

[0113] Regarding the data acquisition module, specifically, the data acquisition module may include: a personal information acquisition module, an electrocardiogram (ECG) module, and an eye movement module.

[0114] Among them, the personal information acquisition module is used to collect the personal information of the pilot and input it through the user interaction interface.

[0115] The ECG module is used to collect the ECG data read in real time by wearing an ECG device, such as the spatial position coordinates of eye movement, blink markers, fixation markers, and saccade markers, etc.

[0116] The eye movement module is used to collect the eye movement data read in real time by wearing an eye movement device, such as heart rate, ECG waveform data, etc.

[0117] Regarding the data interface module, the data interface module is used to receive the data transmitted by the data acquisition module and convert the index calculation. Specifically, the data interface module may include: a data transceiver interface module, a data storage module, and a data conversion module.

[0118] Among them, the data transceiver interface module is used to read the ECG data and eye movement data. Exemplarily, the data transceiver interface module reads the ECG data at a frequency of 1 Hz and reads the eye movement data at a frequency of 60 Hz.

[0119] The data storage module is used to store the real-time collected ECG data and eye movement data in the memory space.

[0120] The data conversion module is used to determine the ECG characteristic indexes according to the ECG data and determine the eye movement characteristic indexes according to the eye movement data.

[0121] It should be noted that first, set the UDP port number and Host address in the above-mentioned ECG module and the above-mentioned eye movement module respectively. Then, based on the program development platform (such as the C++ platform), establish UDP data communication between the data transceiver interface module and the ECG module and the eye movement module through the data transceiver interface to realize the subsequent transmission of the ECG data and the eye movement data.

[0122] Regarding the online evaluation module, the online evaluation module is used to receive the data transmitted by the data interface module and realize the online evaluation of the pilot's situation awareness state. Specifically, the online evaluation module may include: an ECG characteristic preprocessing module, an eye movement characteristic preprocessing module, and an online evaluation classifier module.

[0123] Among them, the ECG feature preprocessing module is used to determine the initial ECG feature index according to the ECG data, and then perform normalization and preprocessing operations on the initial ECG feature index to obtain the ECG feature index.

[0124] The eye movement feature preprocessing module is used to determine the initial eye movement feature index according to the eye movement data, and then perform normalization and preprocessing operations on the initial eye movement feature index to obtain the eye movement feature index.

[0125] The online evaluation classifier module is used to input the electrocardiogram characteristic index and the eye movement characteristic index as physiological characteristic indexes into the state evaluation model to obtain the pilot's situational awareness state output by the state evaluation model.

[0126] The output report module is used to output the pilot's situational awareness status and also to output the pilot's personal information.

[0127] It should be noted that Figure 3 The system shown does not need to rely on cumbersome questionnaires and post-analysis evaluations. It can measure and evaluate the pilot's situational awareness level in real time during the flight by simply collecting the pilot's electrocardiogram data and eye movement data during the flight. The system is simple and feasible to use, and provides an optimized user interface and usage process. It is suitable for online monitoring and status grasp of the pilot's situational awareness status during the flight, which helps to better complete the flight test.

[0128] Optionally, the user interaction interface of the above system may include: a pilot information collection interface, an online evaluation interface and an output evaluation report interface.

[0129] For the pilot information collection interface, for example, Figure 4 Schematic diagram of the pilot information collection interface provided by the embodiment of the present application. Figure 4 The pilot can input personal information into the pilot information collection interface by selecting a drop-down menu or a text box style. In other words, the pilot information collection interface is used to collect the pilot's personal information.

[0130] For the online evaluation interface, the online evaluation interface is the main interface of the online evaluation observation. Optionally, the online evaluation interface may include: a parameter setting window and an online evaluation main interface. Exemplarily, Figure 5 is a schematic diagram of a parameter setting window provided in an embodiment of the present application; Figure 6 It is a schematic diagram of the online evaluation main interface provided in the embodiment of the present application.

[0131] Combination Figure 5, the pilot can enter the parameter setting window by clicking the "Parameter Setting" button in the online evaluation interface. Among them, the parameter setting window includes parameter setting contents such as device address, UDP port, Transmission Control Protocol (TCP) port, port selection, and selection mode.

[0132] Combined with Figure 6 , the pilot can enter the device address of the computer where the system is located (such as the Internet Protocol (IP) address) at the "Device Address" in the above parameter setting window; and enter the port numbers of the electrocardiogram module and the eye movement module through the "Port Selection" to establish a connection with the electrocardiogram module and the eye movement module; then, the pilot selects the "Number of Eigenvalues" and selects the number of electrocardiogram feature indicators and the number of eye movement feature indicators that need to be read during the online evaluation of the above system according to actual needs; finally, after the pilot clicks the "OK" button, at this time, the pilot has completed the parameter setting. After that, the pilot continues to click the "Run" button to start the online evaluation of the pilot's situation awareness state. The pilot's situation awareness state changing over time can be observed through the graphic area, and the situation awareness state of one cell is updated every 30s. Figure 6 Among them, different colors represent different situation awareness states. Specifically, green represents "high situation awareness state", yellow represents "medium situation awareness state", and red represents "low situation awareness state".

[0133] Regarding the output evaluation report interface, optionally, the output evaluation report interface may include: an output result interface and an output report interface. Exemplarily, Figure 7 is a schematic diagram of the output result interface provided by an embodiment of the present application; Figure 8 is a schematic diagram of the output report interface provided by an embodiment of the present application.

[0134] Combined with Figure 7 , the output result interface is used to output personal information and the visualized situation awareness state.

[0135] Combined with Figure 8 , the output report interface can, through the "Output Report" function, be used to output the evaluation result data (i.e., the situation awareness state) and the visualization graph (i.e., the visualized situation awareness state) according to the specified path.

[0136] The pilot situation awareness state evaluation device based on physiological data provided by the embodiments of the present application will be described below. The pilot situation awareness state evaluation device based on physiological data described below can be mutually referred to corresponding to the pilot situation awareness state evaluation method based on physiological data described above.

[0137] Figure 9 This is a schematic structural diagram of a pilot situation awareness state evaluation device based on physiological data provided by an embodiment of the present application. As Figure 9 shown, the device includes: a data acquisition module 901, a data interface module 902, and an online evaluation module 903.

[0138] The data acquisition module 901 is configured to obtain the physiological data of the pilot in a preset time period in real time, and the physiological data includes electrocardiogram data and eye movement data; The data interface module 902 is configured to determine electrocardiogram feature indexes according to the electrocardiogram data, and determine eye movement feature indexes according to the eye movement data; The online evaluation module 903 is configured to input the electrocardiogram feature indexes and the eye movement feature indexes as physiological feature indexes into a state evaluation model, and obtain the situation awareness state of the pilot output by the state evaluation model; wherein, the state evaluation model is trained based on physiological feature index samples and situation awareness state samples corresponding to pilot samples.

[0139] Optionally, the state evaluation model is trained based on the following steps: determining multiple groups of data sample sets according to the physiological feature index samples and the situation awareness state samples, each group of data sample sets includes a group of training sets and a group of test sets, and each group of data sample sets is different; for each group of data sample sets, updating the model parameters in the first state evaluation model according to the training sets in the data sample sets to obtain a second state evaluation model; determining the model accuracy of the second state evaluation model according to the test sets in the data sample sets; and determining the second state evaluation model with the highest model accuracy among multiple second state evaluation models as the state evaluation model.

[0140] Optionally, inputting the physiological feature index samples of the training set in the data sample set into the first state evaluation models corresponding to multiple model parameters respectively, and obtaining the predicted situation awareness states of the pilot samples respectively output by the multiple first state evaluation models; determining the similarity between each predicted situation awareness state and the situation awareness state samples of the training set in the data sample set; and determining the second state evaluation model according to the model parameters of the first state evaluation model corresponding to the maximum similarity among multiple similarities.

[0141] Optionally, the electrocardiogram data includes: electrocardiogram waveform data; the electrocardiogram feature indexes include: average heart rate, average RR interval, and standardized target frequency band power; the data interface module 902 is specifically configured to determine the average heart rate according to the RR intervals and the electrocardiogram sampling frequency in the electrocardiogram waveform data, where is an integer greater than or equal to 2, and each RR interval is used to represent the time interval between two adjacent R waves corresponding in the electrocardiogram waveform data; according to the For each RR interval, determine the average RR interval; based on the power of the target frequency band in the RR interval sequence in the electrocardiogram waveform data, determine the standardized target frequency band power.

[0142] Optionally, the data interface module 902 is specifically configured to use the first formula to determine the average heart rate; where the first formula is: ; represents the average heart rate; represents the electrocardiogram sampling frequency; represents the th RR interval among the

[0143] Optionally, the data interface module 902 is specifically configured to use the second formula to determine the average RR interval; where the second formula is: ; represents the average RR interval.

[0144] Optionally, the target frequency band power includes a first frequency band power and a second frequency band power greater than the first frequency band power; the standardized target frequency band power includes a standardized first frequency band power and a standardized second frequency band power; the data interface module 902 is specifically configured to use the third formula to determine the standardized first frequency band power; and use the fourth formula to determine the standardized second frequency band power; where the third formula is: ; the fourth formula is: ; represents the standardized first frequency band power; represents the standardized second frequency band power; represents the first frequency band power; represents the second frequency band power; represents the first frequency band power and the second frequency band power sum; represents a third frequency band power less than the first frequency band power.

[0145] Optionally, the eye movement data includes: fixation points, saccade points, and pupil diameters at valid sampling points, , and are all integers greater than or equal to 2; the eye movement feature indicators include: fixation rate, saccade rate, pupil diameter, and average fixation duration; the data interface module 902 is specifically configured to determine the fixation rate according to the fixation points and the preset duration; determine the saccade rate according to the The saccade rate is determined based on the number of saccade points and the preset duration; according to the pupil diameters of the valid sampling points, the pupil diameter is determined; according to the fixation duration of each fixation point among the fixation points, the average fixation duration is determined.

[0146] Figure 10 is a schematic structural diagram of an electronic device provided by an embodiment of the present application. As Figure 10 shown, the electronic device may include: a processor 1010, a communication interface 1020, a memory 1030, and a communication bus 1040. Among them, the processor 1010, the communication interface 1020, and the memory 1030 communicate with each other through the communication bus 1040. The processor 1010 may call logical instructions in the memory 1030 to execute a method for evaluating the pilot's situation awareness state based on physiological data. The method includes: obtaining the physiological data of the pilot in a preset duration in real time, where the physiological data includes electrocardiogram data and eye movement data; determining an electrocardiogram feature index according to the electrocardiogram data, and determining an eye movement feature index according to the eye movement data; inputting the electrocardiogram feature index and the eye movement feature index as physiological feature indexes into a state evaluation model to obtain the situation awareness state of the pilot output by the state evaluation model; where the state evaluation model is trained based on physiological feature index samples and situation awareness state samples corresponding to pilot samples.

[0147] In addition, when the logical instructions in the above-mentioned memory 1030 are implemented in the form of software functional units and sold or used as an independent product, they may be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or a part of this technical solution, may be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present application. The foregoing storage medium includes: various media such as a USB flash drive, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk, or an optical disc that can store program codes.

[0148] On the other hand, the present application also provides a computer program product, which includes a computer program. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the method for evaluating the pilot's situation awareness state based on physiological data provided by the above-mentioned various methods. The method includes: acquiring in real time the physiological data of the pilot within a preset time period, where the physiological data includes electrocardiogram data and eye movement data; determining electrocardiogram characteristic indexes according to the electrocardiogram data, and determining eye movement characteristic indexes according to the eye movement data; inputting the electrocardiogram characteristic indexes and the eye movement characteristic indexes as physiological characteristic indexes into a state evaluation model to obtain the situation awareness state of the pilot output by the state evaluation model; wherein the state evaluation model is trained based on physiological characteristic index samples and situation awareness state samples corresponding to pilot samples.

[0149] On another aspect, the present application also provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it is implemented to execute the method for evaluating the pilot's situation awareness state based on physiological data provided by the above-mentioned various methods. The method includes: acquiring in real time the physiological data of the pilot within a preset time period, where the physiological data includes electrocardiogram data and eye movement data; determining electrocardiogram characteristic indexes according to the electrocardiogram data, and determining eye movement characteristic indexes according to the eye movement data; inputting the electrocardiogram characteristic indexes and the eye movement characteristic indexes as physiological characteristic indexes into a state evaluation model to obtain the situation awareness state of the pilot output by the state evaluation model; wherein the state evaluation model is trained based on physiological characteristic index samples and situation awareness state samples corresponding to pilot samples.

[0150] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. A person of ordinary skill in the art can understand and implement it without creative labor.

[0151] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on such an understanding, the essence of the above technical solution, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0152] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, and are not intended to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for evaluating the situation awareness state of pilots based on physiological data, characterized in that, Including: Obtaining the physiological data of the pilot in a preset duration in real time, where the physiological data includes electrocardiogram data and eye movement data; Determining electrocardiogram feature indexes according to the electrocardiogram data, and determining eye movement feature indexes according to the eye movement data; Taking the electrocardiogram feature indexes and the eye movement feature indexes as physiological feature indexes and inputting them into a state evaluation model to obtain the situation awareness state of the pilot output by the state evaluation model; Wherein, the state evaluation model is trained based on physiological feature index samples and situation awareness state samples corresponding to pilot samples.

2. The method for evaluating the pilot situation awareness state based on physiological data according to claim 1, wherein The state evaluation model is trained based on the following steps: Determining multiple groups of data sample sets according to the physiological feature index samples and the situation awareness state samples, each group of data sample sets includes a group of training sets and a group of test sets, and each group of data sample sets is different; For each group of data sample sets, updating the model parameters in the first state evaluation model according to the training sets in the data sample sets to obtain a second state evaluation model; determining the model accuracy of the second state evaluation model according to the test sets in the data sample sets; Determining the second state evaluation model with the highest model accuracy among multiple second state evaluation models as the state evaluation model.

3. The method for evaluating the pilot's situation awareness state based on physiological data according to claim 2, wherein, The step of updating the model parameters in the first state evaluation model according to the training sets in the data sample sets to obtain a second state evaluation model includes: Inputting the physiological feature index samples of the training sets in the data sample sets into the first state evaluation models corresponding to multiple model parameters respectively to obtain the predicted situation awareness states of the pilot samples respectively output by the multiple first state evaluation models; Determining the similarity between each predicted situation awareness state and the situation awareness state samples of the training sets in the data sample sets; Determining the second state evaluation model according to the model parameters of the first state evaluation model corresponding to the maximum similarity among multiple similarities.

4. The method for evaluating the pilot situation awareness state based on physiological data according to any one of claims 1-3, characterized in that, The electrocardiogram data includes: electrocardiogram waveform data; the electrocardiogram feature indexes include: average heart rate, average RR interval and standardized target frequency band power; The step of determining electrocardiogram feature indexes according to the electrocardiogram data includes: Based on the RR intervals and the electrocardiogram sampling frequency in the electrocardiogram waveform data, determine the average heart rate, is an integer greater than or equal to 2, and each RR interval is used to represent the time interval between two adjacent R waves corresponding in the electrocardiogram waveform data; Based on the described RR intervals, determine the average RR interval; Determining the standardized target frequency band power according to the target frequency band power of the RR interval sequence in the electrocardiogram waveform data.

5. The method for evaluating the pilot situation awareness state based on physiological data according to claim 4, wherein Said determining the average heart rate according to the RR intervals and the electrocardiogram sampling frequency in the electrocardiogram waveform data includes: Determining the average heart rate by using a first formula; Among them, the first formula is: ; represents the average heart rate; represents the electrocardiogram sampling frequency; represents the th RR interval among the th RR interval; Said according to the said RR intervals, determining the average RR interval, comprising: Determining the average RR interval by using a second formula; Among them, the second formula is: ; represents the average RR interval.

6. The method for evaluating the pilot's situation awareness state based on physiological data according to claim 4, characterized in that The target frequency band power includes a first frequency band power and a second frequency band power greater than the first frequency band power; the standardized target frequency band power includes standardized first frequency band power and standardized second frequency band power; The step of determining the standardized target frequency band power according to the target frequency band power of the RR interval sequence in the electrocardiogram waveform data includes: Determining the standardized first frequency band power by using a third formula; and determining the standardized second frequency band power by using a fourth formula; Among them, the third formula is as follows: ; The fourth formula is as follows: ; Represents the standardized first band power; Represents the standardized second band power; Represents the first band power; Represents the second band power; Represents the first band power And the second band power Sum; Represents the third band power less than the first band power.

7. The method for evaluating the pilot's situation awareness state based on physiological data according to any one of claims 1-3, characterized in that, The eye movement data includes: fixation points, saccade points, and pupil diameters of valid sampling points, , and are all integers greater than or equal to 2; the eye movement feature indexes include: fixation rate, saccade rate, pupil diameter, and average fixation duration; The step of determining eye movement feature indexes according to the eye movement data includes: Based on the number of fixation points and the preset duration, determine the fixation rate; Determine the saccade rate according to the number of saccade points and the preset duration. Based on the pupil diameters of the valid sampling points, determine the pupil diameter; Based on the fixation duration of each fixation point among the fixation points, determine the average fixation duration.

8. An assessment device for the situation awareness state of a pilot based on physiological data, characterized in that, Including: A data acquisition module for obtaining the physiological data of the pilot in a preset duration in real time, where the physiological data includes electrocardiogram data and eye movement data; A data interface module, configured to determine electrocardiogram feature indexes according to the electrocardiogram data, and determine eye movement feature indexes according to the eye movement data; An online evaluation module, configured to input the electrocardiogram feature indexes and the eye movement feature indexes as physiological feature indexes into a state evaluation model, and obtain the situation awareness state of the pilot output by the state evaluation model; wherein, the state evaluation model is trained based on physiological feature index samples and situation awareness state samples corresponding to pilot samples.

9. An electronic device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the processor executes the computer program, the method for evaluating the situation awareness state of a pilot based on physiological data according to any one of claims 1 to 7 is implemented.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, the method for evaluating the situation awareness state of a pilot based on physiological data according to any one of claims 1 to 7 is implemented.

Images