Pilot cognitive ability assessment method, system and storage medium
Through multi-channel data acquisition and machine learning models, combined with expert experience, the problem of inaccurate pilot cognitive ability assessment in the existing technology is solved, and a comprehensive, accurate and objective assessment of pilot cognitive ability is achieved, which is suitable for pilot selection and Task rating.
Patent Information
- Application Number
- CN202211055993.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-08-31
- Publication Date
- 2024-11-12
- Estimated Expiration
- 2042-08-31
AI Technical Summary
The prior art cannot effectively comprehensively evaluate the pilot's cognitive ability, and cannot accurately evaluate specific flight mission situations, and the evaluation results are not accurate and objective enough.
Through multi-channel acquisition of pilot interaction, physiological and brain function indicators in performing flight situational tasks and cognitive assessment tasks, combined with the results of the questionnaire scale, the pilot's comprehensive assessment scores were calculated using a pre-trained machine learning model, providing experts-based A dual evaluation method of experience and machine learning.
It achieves a more comprehensive, accurate and objective assessment of pilot cognitive abilities, which is suitable for pilot selection and mission scoring, and improves the ecological validity and pertinence of the assessment.
Smart Images

Figure CN115462755B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of human factors engineering, and in particular to a pilot cognitive ability assessment method, system and storage medium. Background Art
[0002] Flying activities are special, and flying professions require practitioners to have much higher cognitive abilities than ordinary professions. Pilot cognitive ability is a quality that pilots must possess during the flight process, and it is also an indispensable quality for pilots to complete flight missions. Pilot cognitive ability includes spatial memory ability, perception ability, hand-eye coordination ability, conflict handling ability, decision-making ability in risk situations, problem-solving ability, and basic personality traits related to tasks. Therefore, detecting the basic cognitive ability of pilots is of great significance for the selection of flight trainees and the evaluation of flight ability. However, the existing technology only provides behavioral paradigms and frameworks for cognitive ability assessment, and does not have an algorithm for comprehensively assessing cognitive ability to obtain comprehensive cognitive ability assessment results. In addition, although the existing technology has a means of evaluating cognitive ability through physiological assessment, it cannot be combined with specific task situations (situations, also called situations). The existing cognitive ability assessment usually uses a psychological assessment paradigm, that is, the existing cognitive ability assessment technology cannot be targeted at special jobs, and the existing behavioral paradigm for measuring cognitive ability is relatively single, the cognitive task assessment method is weakly combined with specific jobs, and cannot be combined with the concrete working situation of pilots, and the ecological validity needs to be strengthened.
[0003] Therefore, there is an urgent need for a cognitive ability assessment scheme that can more accurately detect the cognitive ability of pilots. Summary of the invention
[0004] In view of this, an embodiment of the present invention provides a method and system for evaluating pilot cognitive ability, which calculates a comprehensive evaluation score of pilot cognitive ability by collecting multi-channel indicator parameters and obtaining two or more pilot cognitive abilities based on multiple evaluation models through the pilot performing multiple tasks such as flight situational tasks and cognitive assessment tasks, thereby enabling a more accurate evaluation of pilot cognitive ability.
[0005] One aspect of the present invention provides a method for evaluating a pilot's cognitive ability, the method comprising the following steps:
[0006] In the process of the pilot performing the flight situation task presented by the flight situation task presentation module, the pilot's index parameters are collected through multiple channels, and the collected multi-channel index parameters include: interaction index, physiological index and brain function index; and / or
[0007] During the pilot's execution of the cognitive assessment task presented by the cognitive presentation module, the pilot's index parameters are collected through multiple channels, and the collected multi-channel index parameters include: interaction index, physiological index and brain function index; and / or
[0008] The flight crew's personality-related cognitive ability assessment results are obtained through the questionnaire response results obtained when the pilot performs the questionnaire scale presented by the flight mission-related cognitive questionnaire scale assessment module;
[0009] A feature vector is obtained based on one or more of the multi-channel indicator parameters of the pilot in performing the flight situation task, the multi-channel indicator parameters in the cognitive assessment task, and the personality-related cognitive ability assessment results, and the feature vector is input into a pre-trained pilot cognitive ability assessment model, and a first assessment score of the pilot cognitive ability is obtained based on the output data;
[0010] Normalizing the multi-channel indicator parameters of the pilot in the flight situation task to obtain a first normalized result, and / or normalizing the multi-channel indicator parameters in the cognitive assessment task to obtain a second normalized result, and performing weighted averaging on one or more of the first normalized result, the second normalized result, and the personality-related cognitive ability assessment result based on preset indicator parameter weights to obtain a second assessment score of the pilot's cognitive ability;
[0011] A comprehensive assessment score of the pilot's cognitive ability is obtained based on the first assessment score of the pilot's cognitive ability and the second assessment score of the pilot's cognitive ability.
[0012] In some embodiments of the present invention, the method further comprises:
[0013] The multi-channel index parameters collected during the pilot's execution of various tasks are the same or partially the same index parameters;
[0014] The multi-channel index parameters collected during the pilot's execution of various tasks include physiological indicators, which include one or more of the following indicators: skin electricity index, electrocardiogram index, myoelectric index, pulse index, breathing index, blood oxygen index and blood pressure index.
[0015] In some embodiments of the present invention, the multi-channel indicator parameters collected during the pilot's performance of various tasks also include interaction indicators, and the interaction indicators include one or more of the following indicators: behavioral indicators and eye movement indicators, the behavioral indicators include reaction time and / or accuracy, and the eye movement indicators include scanning time, fixation time, blinking frequency, pupil diameter change, eyelid closure degree and eye movement trajectory.
[0016] In some embodiments of the present invention, the multi-channel index parameters collected during the pilot's performance of various tasks also include brain function indicators, and the brain function indicators include one or more of the following indicators: EEG time-frequency indicators, nonlinear indicators, event-related brain activity indicators, brain function connection indicators, and blood oxygen level change indicators in various brain regions.
[0017] In some embodiments of the present invention, the questionnaire scale includes at least one of the following: a contextualized questionnaire related to the job, a personality scale, and a cognitive scale. In some embodiments of the present invention, the questionnaire scale includes at least one of the following: a contextualized questionnaire related to the job, a personality scale, and a cognitive scale.
[0018] In some embodiments of the present invention, the pilot cognitive ability assessment model is an XGBoost machine learning model obtained based on a gradient boosting tree algorithm.
[0019] The pilot's second cognitive ability assessment score meets the following requirements:
[0020]
[0021] Among them, i t For each task, k is the multi-channel indicator parameter in each task, Z tk represents the normalized multi-channel index parameters in each task, C tk is the weight coefficient, t represents the number of tasks performed by the pilot, and k represents the number of index parameters of each task.
[0022] In some embodiments of the present invention, the flight scenario task presentation module includes a flight simulator for completing simulations of various flight scenarios and flight environments.
[0023] Another aspect of the present invention provides a pilot cognitive ability assessment system, the system comprising a processor and a memory, the memory storing computer instructions, the processor being used to execute the computer instructions stored in the memory, and the system implementing the steps of the method described above when the computer instructions are executed by the processor.
[0024] Another aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, which implements the steps of the above-mentioned method when executed by a processor.
[0025] The pilot cognitive ability assessment method and system of the present invention can specifically target the pilot's work situation and the cognitive ability required for the corresponding work, combine expert experience to construct multiple cognitive ability assessment dimensions, and comprehensively collect the pilot's objective interactive behavior, eye movement, physiology, EEG and brain imaging multi-channel human factor data in each dimensional task; the system provides two optional pilot cognitive ability calculation methods, expert experience-based and machine learning algorithm-based indicator feature extraction and weight distribution, and the calculation method based on the machine learning algorithm is more suitable for the system to use after accumulating a certain amount of data, so as to continuously optimize the training model and make the system more suitable for flight personnel selection.
[0026] Furthermore, the present invention can also provide a comprehensive cognitive ability assessment of the pilot by combining the evaluation results of the pilot's objective interactive behavior and the subjective questionnaire scale.
[0027] Additional advantages, purposes, and features of the present invention will be described in part in the following description, and will become apparent to those skilled in the art after studying the following, or may be learned from the practice of the present invention. The purposes and other advantages of the present invention may be achieved and obtained by the structures specifically indicated in the specification and the accompanying drawings.
[0028] Those skilled in the art will appreciate that the objectives and advantages that can be achieved with the present invention are not limited to the above specific description, and the above and other objectives that can be achieved by the present invention will be more clearly understood from the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS
[0029] The drawings described herein are used to provide a further understanding of the present invention, constitute a part of this application, and do not constitute a limitation of the present invention.
[0030] Figure 1 Schematic diagram of a pilot cognitive ability assessment method according to an embodiment of the present invention.
[0031] Figure 2 The figure is a flow chart of the pilot cognitive ability assessment in one embodiment of the present invention.
[0032] Figure 3 A schematic diagram of a pilot cognitive ability data collection module in one embodiment of the present invention. DETAILED DESCRIPTION
[0033] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments and the accompanying drawings. Here, the illustrative embodiments of the present invention and their descriptions are used to explain the present invention, but are not intended to limit the present invention.
[0034] It should also be noted that, in order to avoid obscuring the present invention due to unnecessary details, only structures and / or processing steps closely related to the solutions according to the present invention are shown in the accompanying drawings, while other details that are not closely related to the present invention are omitted.
[0035] It should be emphasized that the term “include / comprises” when used herein refers to the presence of features, elements, steps or components, but does not exclude the presence or addition of one or more other features, elements, steps or components.
[0036] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. In the accompanying drawings, the same reference numerals represent the same or similar components, or the same or similar steps.
[0037] In order to overcome the problem that the existing technology is not comprehensive, targeted, and single in form in measuring the cognitive ability of pilots, which leads to inaccurate evaluation results, the embodiment of the present invention specifically targets the pilot's work situation and the cognitive ability required for the corresponding work, and combines expert experience to construct multiple cognitive ability evaluation dimensions, and comprehensively collects the human factor data of the pilot in objective interactive behavior, eye movement, physiology, EEG, brain imaging and other multi-channels in each dimension task. Finally, the system automatically calculates the pilot's cognitive ability score based on the weighted average algorithm based on expert experience and the deep learning algorithm, and obtains the comprehensive evaluation result of the pilot's cognitive ability based on the pilot's cognitive ability score obtained by different algorithms, so that the cognitive ability evaluation is more comprehensive, more objective, more accurate, more targeted, and the measurement form is diversified, so that pilots can be selected according to the pilot's cognitive ability evaluation score. Furthermore, the present invention can also be combined with subjective document scales to conduct a more comprehensive evaluation of the pilot's cognitive ability.
[0038] Figure 1 FIG. 1 is a flow chart of a method for evaluating a pilot's cognitive ability in one embodiment of the present invention. Figure 1 As shown, the method of this embodiment includes the following steps:
[0039] Step S110, when the pilot performs the flight situation task presented by the flight situation task presentation module, the pilot's index parameters are collected through multiple channels, and the collected multi-channel index parameters include: interaction index, physiological index and brain function (or electroencephalogram) index.
[0040] Before executing this step, one or more flight scenario tasks can be constructed in advance, and a flight scenario task presentation module can be used to implement a flight mission simulation based on a real flight scenario. That is, the flight scenario in the constructed flight scenario task can be presented through simulation. The flight scenario task presentation module can include a flight simulator for realizing the simulation of various flight scenarios.
[0041] In the process of the pilot performing the flight situation task presented by the flight situation task presentation module, the pilot's index parameters can be collected through multiple data acquisition channels, and the index parameters may include interaction indexes, physiological indexes and brain function indexes. Among them, the interaction index may include behavioral indexes and / or eye movement indexes, and the behavioral index may include reaction time and / or reaction accuracy; the eye movement index may include some or all of the following indicators: scanning time and fixation time within the eye movement area of interest (AOI), blinking frequency, pupil diameter change, eyelid closure degree and eye movement trajectory; the physiological index collected by multi-channel may include at least one of the following indicators: skin electricity index, electrocardiogram index, myoelectric index, pulse index, breathing index, blood oxygen index and blood pressure index, etc. More specifically, these indicators can be embodied as time domain indicators, frequency domain indicators or nonlinear indicators; brain function indicators may include at least one of the following indicators: EEG time-frequency index, nonlinear index, event-related brain activity index, brain function connection index and blood oxygen level change index of each brain region. These indicators are only examples here, and the present invention is not limited to this. During the test, pilots can perform one flight scenario mission or multiple different flight scenario missions in sequence.
[0042] In an embodiment of the present invention, for example, different collection devices or collection modules in the human-machine environment synchronization platform can be used to collect the pilot's interaction indicators, physiological indicators and brain function indicators during the flight situation task presented by the flight situation task presentation module. For example, the pilot's behavior indicators during the flight situation task execution can be collected through the HMI interactive prototype synchronization module; the basic eye movement indicators of the pilot during the flight situation task execution can be synchronously collected by an eye tracker, and further combined with the eye movement trajectory analysis module (software module) to analyze and calculate other eye movement indicators other than the basic eye movement indicators, so as to obtain the pilot's eye movement data during the flight task execution; the electrodermal activity (EDA) analysis module and the electrodermal activity sensor can be used to synchronously collect and analyze the pilot's electrodermal activity indicators during the task execution; the heart rate variability (HRV) analysis module and / or the electrocardiogram (ECG) sensor can be used to synchronously collect the pilot's electrocardiogram indicators during the task execution; the electromyogram (EMG) analysis module and / or the electromyogram sensor can be used to synchronously collect the pilot's electromyogram indicators during the task execution; the respiratory (Respiratory) The pilot's breathing index during the mission can be synchronously collected and analyzed by the PhotoPlethysmographic (PPG) sensor; the pilot's blood oxygen saturation (SPO2) sensor can be synchronously collected by the functional near-infrared spectroscopy (fNIRS) sensor and fNIRS brain imaging analysis module; the pilot's brain blood oxygen index during the mission can be synchronously collected and analyzed to obtain the pilot's blood oxygen level change data in each brain region; the electroencephalogram (EEG) sensor and EEG / ERP (EventRelated Potential, event-related potential) analysis module can synchronously measure and analyze the EEG indicators of pilots in the process of performing tasks, and obtain EEG time-frequency indicators, nonlinear indicators, event-related brain activity indicators and brain functional connection indicators, etc.; in addition, the present invention can also use a general signal analysis module to obtain biomechanical signals, environmental signals (such as temperature, humidity, noise, atmospheric pressure, light, etc.), skin temperature SKT, electrooculogram and other signals as auxiliary reference information. The above-mentioned collected indicators can be cleaned and feature indicators extracted through the embedded algorithm of the human-machine environment synchronization platform. Of course, the present invention can also use other methods to collect various indicators.
[0043] Step S120, when the pilot performs the cognitive assessment task presented by the cognitive presentation module, the pilot's index parameters are collected through multiple channels, and the collected multi-channel index parameters include: interaction index, physiological index and brain function index.
[0044] In an embodiment of the present invention, various cognitive assessment tasks related to cognitive abilities required for flight can be constructed in advance according to pre-established ecological cognition based on flight situations, so as to measure the pilot's spatial memory ability, alertness, hand-eye coordination ability, conflict handling ability, decision-making ability in risk situations, and problem-solving ability. The created cognitive assessment tasks can be presented using a cognitive presentation module based on flight situations. The pilot executes the cognitive assessment tasks presented by the cognitive presentation module, and during the execution process, multiple data acquisition channels are used to realize multi-channel acquisition of the pilot's index parameters, which may include interaction indicators, physiological indicators, and brain function indicators, which may be the same or partially the same as the indicators collected in step S110.
[0045] During the test, pilots can perform one cognitive assessment task or multiple different cognitive assessment tasks in sequence.
[0046] Step S130, obtaining the personality-related cognitive ability assessment result of the flight personnel through the questionnaire response results obtained when the pilot executes the questionnaire scale presented by the flight mission-related cognitive questionnaire scale assessment module.
[0047] In this step, the flight crew's personality-related cognitive ability assessment results are obtained through the questionnaire response results obtained when the pilot performs the questionnaire scale presented by the flight mission-related cognitive questionnaire scale assessment module; the questionnaire scale may include at least one of the following: task-related contextualized questionnaire answers, personality scales, and cognitive scales. Through the questionnaire scale, information on the pilot's personality, cognition, etc. can be evaluated, so as to obtain the pilot's score in the relevant personality and cognitive questionnaire scale process, and obtain the flight crew's personality-related cognitive ability assessment results.
[0048] Step S140, obtaining a feature vector based on one or more of the multi-channel indicator parameters of the pilot in performing the flight situation task, the multi-channel indicator parameters in the cognitive assessment task, and the personality-related cognitive ability assessment results, inputting the feature vector into a pre-trained pilot cognitive ability assessment model, and obtaining a first assessment score of the pilot's cognitive ability based on the output data.
[0049] In this step, the pilot cognitive ability assessment model used is, for example, an XGBoost machine learning model based on a gradient boosting tree algorithm, which evaluates the pilot's cognitive ability in various tasks (i1, i2, ..., i t) in the multi-channel indicator parameters (j1,j2,...,j k ) is constructed as a feature vector X(x1,x2,...,x t ), corresponding to the actual flight level y of the pilot, the feature vector X in the training set and the actual flight level y as the label are input into the constructed XGBoost machine learning model, and the XGBoost machine learning model can be trained to obtain a pre-trained XGBoost machine learning model. In the embodiment of the present invention, the XGBoost machine learning model is only an example, and the present invention is not limited to this. Other machine learning models can also be used to evaluate the pilot's cognitive ability. Furthermore, by using the pre-trained machine learning model and inputting the feature vector obtained after the test, the first evaluation score f(x) of the pilot's cognitive ability can be obtained.
[0050] In the embodiment of the present invention, the feature vector X=(x1, x2, ..., x t ), indicating that the feature vector X is composed of t features x, and the calculation formula for the tth feature is Represents the weighting of the indicator parameters from 1 to k under the tth task.
[0051] In the embodiment of the present invention, examples of various tasks are as follows: i1 represents a simulated flight task, i2 represents an attention vigilance paradigm PVT task, ..., i t represents the Tower of Hanoi task (a type of cognitive assessment task); examples of multi-channel indicator parameters are: j1 represents the behavioral reaction time (Reaction Time) in the interactive indicators, j2 represents the behavioral accuracy (Accuracy) in the interactive indicators, j3 represents the mean of the skin electrode response (SC mean) in the physiological indicators, j4 represents the low frequency to high frequency (HRV LF / HF) frequency domain indicator of heart rate variability in the physiological indicators, j5 represents the standard deviation of the NN interval (HRV SDNN) in the time domain indicator of heart rate variability in the physiological indicators, j6 represents the α wave to β wave power value (EEG α / β power) in the frequency domain indicator of EEG in the physiological indicators, j7 represents the β wave power value (EEG β power) in the frequency domain indicator of EEG in the physiological indicators, j8 represents the average pupil diameter (Pupildiameter) in the eye movement indicators, j9 represents the total fixation duration in the area of interest (AOI total fixation duration) in the eye movement indicators, j 10 Indicates the eyelid closure degree (PERCLOS) in the eye movement index. These parameters are only examples and the present invention is not limited thereto.
[0052] The pilot cognitive ability assessment method based on machine learning algorithm of the present invention is particularly suitable for use after the system has accumulated a certain amount of data. By continuously optimizing the training model, the system can be made more suitable for flight personnel selection.
[0053] Step S150, normalizing the multi-channel indicator parameters of the pilot in the flight scenario task to obtain a first normalized result, and / or normalizing the multi-channel indicator parameters in the cognitive assessment task to obtain a second normalized result, and performing weighted averaging on one or more of the first normalized result, the second normalized result, and the personality-related cognitive ability assessment result based on pre-set indicator parameter weights to obtain a second assessment score of the pilot's cognitive ability.
[0054] In an embodiment of the present invention, based on the behavioral indicators of the pilot during the mission collected by the interactive prototype synchronization module, the reaction time and / or accuracy of the pilot's operational behavior during the mission can be obtained based on the behavioral data of each operation in the mission; based on the eye movement indicators collected by the eye tracker, the visual attention information of the pilot during the mission can be obtained based on the eye movement data in each area of interest, and the visual attention information includes one or more of the following information: scanning time, fixation time, blinking frequency, pupil diameter change, eyelid closure degree and eye movement trajectory; based on the skin electricity indicators, electrocardiogram indicators, electromyography indicators, pulse indicators, breathing indicators, blood oxygen indicators and blood pressure indicators collected by various physiological indicator measurement sensors, the pilot's physical reaction during the mission is obtained; based on the EEG data collected by the EEG sensor and the brain blood oxygen index collected by the near-infrared brain imaging sensor, the overall reflection of the electrophysiological activity of the pilot's brain nerve cells during the mission on the cerebral cortex or scalp surface is obtained. The multi-channel index parameters collected by the pilot in the process of performing the flight situation task can be normalized to obtain a first normalized result, and the various parameters collected by the pilot in the process of performing the cognitive assessment task can be normalized to obtain a second normalized result.
[0055] In the embodiment of the present invention, the pilot is assigned to each task (i1, i2, ..., i t ) in the multi-channel indicator parameters (j1,j2,...,j k ) is normalized to obtain the normalized evaluation index (Z 11 ,Z 12 ,...,Z tk ), and then the expert group assigns weights to one or more of the normalized evaluation indicators and personality-related cognitive ability evaluation results based on expert experience, and the weighted average can be used to calculate the pilot cognitive ability second evaluation score S. The pilot cognitive ability second evaluation score S satisfies the following formula:
[0056]
[0057] Among them, C tk is the weight coefficient, t represents the number of tasks performed by the pilot, k represents the number of index parameters of each task, and examples of each task are: i1 represents a simulated flight task, i2 represents a PVT task of the attention vigilance paradigm, ..., i t represents the Tower of Hanoi task (a type of cognitive assessment task); examples of multi-channel indicator parameters are: j1 represents the behavioral reaction time (Reaction Time) in the interactive indicator, j2 represents the behavioral accuracy (Accuracy) in the interactive indicator, j3 represents the mean of the electrocutaneous response (SC mean) in the physiological indicator, j4 represents the frequency domain indicator of heart rate variability, low frequency to high frequency (HRV LF / HF) in the physiological indicator, j5 represents the standard deviation of the NN interval (HRV SDNN) in the time domain indicator of heart rate variability in the physiological indicator, j6 represents the α wave to β wave power value (EEG α / β power) in the frequency domain indicator of EEG in the physiological indicator, j7 represents the β wave power value (EEG β power) in the frequency domain indicator of EEG in the physiological indicator, j8 represents the average pupil diameter (Pupil diameter) in the eye movement indicator, j9 represents the total fixation duration in the area of interest (AOI total fixation duration) in the eye movement indicator, and j 10 Indicates the eyelid closure degree (PERCLOS) in the eye movement index. These parameters are only examples and the present invention is not limited thereto.
[0058] In this step, the first normalization result, the second normalization result, and one or more of the personality-related cognitive ability assessment results are weighted averaged based on the indicator parameter weights determined by expert experience to calculate the second assessment score of the pilot's cognitive ability. However, this assessment transition relies on expert experience, is susceptible to expert subjectivity, and is not suitable for scenarios where there are no experts to confirm the weights. Therefore, in order to more accurately assess the pilot's cognitive ability, the present invention further combines the assessment results of the pilot's cognitive ability based on deep learning, as described in the following steps.
[0059] In the training process of the XGBoost machine learning model adopted in step S140, the model input is the feature vectors and corresponding labels of n training samples, T = {(X1, y1), (X2, y2), ..., (X n ,y n )}, where Xn represents the feature vector of the nth sample, y nRepresents a label; the model output is represented by: Gradient Boosting Decision Tree (GBDT); the specific steps of machine learning model training are as follows:
[0060] Step S1, initialize the first round of learning strong learner f0(X) = 0, according to the model input T = {(X1, y1), (X2, y2), ..., (X n ,y n )} is used to train the strong learner f1(X) for the second round of learning, that is, the first training is performed to obtain f1(X), which is also called the first sub-model, and X represents the input feature vector; based on the first sub-model (or the first round of sub-model), the first model representation can be obtained: F1(X) = f0(X) + f1(X).
[0061] Step S2, starting from the second round (m=2), obtain the mth model based on the following formula:
[0062] F m (X) = F m-1 (X)+f m (X);
[0063] This formula indicates that the mth model is the m-1th model F m-1 (X) and the mth sub-model f m (X), where is the m-1th model, f m (X) is the mth sub-model;
[0064] For the m-1th model F m-1 (X) No training is required. For the mth sub-model f m (X) Requires training.
[0065] In the embodiment of the present invention, the adopted loss function can be expressed as:
[0066] L(y,f m (X))=L(y,f m-1 (X)+h m (X)) (2)
[0067] Among them, L(y,f m (X)) is the loss function of the mth round, f m-1 (X) is the strong learner in the m-1th round, h m (X) is the weak learner in round m. In this round of iteration, the decision tree is found to minimize the loss of the sample.
[0068] After m rounds of training, a pre-trained machine learning model is obtained. The feature vector obtained by the test is input into the pre-trained model to obtain the final gradient boosting model:
[0069]
[0070] Among them, M is the final number of rounds.
[0071] Based on the pre-trained machine learning model and the feature vector obtained after measurement, a first evaluation score of the pilot's cognitive ability can be obtained.
[0072] Step S160: obtaining a comprehensive evaluation score of the pilot's cognitive ability based on the first evaluation score of the pilot's cognitive ability and the second evaluation score of the pilot's cognitive ability.
[0073] In this step, a comprehensive evaluation is performed based on the first evaluation score of the pilot's cognitive ability obtained in step S140 and the second evaluation score of the pilot's cognitive ability obtained in step S150, such as a weighted average operation (the present invention is not limited to this), to obtain a comprehensive evaluation score of the pilot's cognitive ability.
[0074] The pilot cognitive ability assessment method of the present invention is as follows: Figure 2 and Figure 3 As shown in the figure, based on expert experience, scenarios, behavioral paradigms and questionnaire scales suitable for evaluating pilot cognitive abilities are constructed. By collecting the physiological indicators and interaction indicators of pilots when completing various scenario simulation tasks, behavioral paradigms and questionnaire scales, parameter indicators of pilots in various dimensions can be obtained, such as interaction indicators, physiological indicators, brain function indicators, cognitive assessment indicators and personality assessment indicators. Based on these indicators, through dual evaluation based on expert experience and machine learning algorithms (such as assigning weights to the scores of physiological indicators and interaction indicators of various scenario tasks, behavioral paradigms and questionnaire scales based on expert experience, and obtaining the pilot cognitive ability assessment score; using decision trees to assign feature weights based on the actual level of pilots, continuously training models, and obtaining the pilot cognitive ability assessment score), and further combining the two evaluation results, the pilot's cognitive ability can be obtained more comprehensively and accurately.
[0075] In summary, the present invention can specifically target the pilot's operating situation and the cognitive ability required for the corresponding operation, combine expert experience to construct multiple cognitive ability assessment dimensions, and comprehensively collect pilots' subjective questionnaire scales and objective interactive behaviors, eye movements, physiology, EEG and brain imaging multi-channel human factors data in tasks of each dimension; the system of the present invention provides two optional pilot cognitive ability calculation methods, indicator feature extraction and weight allocation based on expert experience and based on machine learning algorithms. The calculation method based on machine learning algorithms is more suitable for use after the system accumulates a certain amount of data, so as to continuously optimize the training model and make the system more suitable for flight personnel selection.
[0076] Corresponding to the above method, the present invention also provides a system, which includes a computer device, the computer device includes a processor and a memory, the memory stores computer instructions, the processor is used to execute the computer instructions stored in the memory, and when the computer instructions are executed by the processor, the system implements the steps of the method described above.
[0077] The system of the present invention is an evaluation system suitable for pilots that is obtained by collecting multi-channel subjective and objective human factor data, which is helpful for the selection and evaluation of pilots; and the system of the present invention is based on a dual evaluation of expert experience and machine learning algorithm, making the evaluation more objective and accurate; the present invention comprehensively considers the cognitive level of pilots in various dimensions, and the ecological validity of the evaluation result is high.
[0078] The embodiment of the present invention also provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the steps of the aforementioned edge computing server deployment method are implemented. The computer-readable storage medium can be a tangible storage medium, such as a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a floppy disk, a hard disk, a removable storage disk, a CD-ROM, or any other form of storage medium known in the technical field.
[0079] It should be understood by those skilled in the art that the exemplary components, systems and methods described in conjunction with the embodiments disclosed herein can be implemented in hardware, software or a combination of the two. Whether it is performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention. When implemented in hardware, it can be, for example, an electronic circuit, an application specific integrated circuit (ASIC), appropriate firmware, a plug-in, a function card, etc. When implemented in software, the elements of the present invention are programs or code segments used to perform the required tasks. The program or code segment can be stored in a machine-readable medium, or transmitted on a transmission medium or a communication link via a data signal carried in a carrier.
[0080] It should be clear that the present invention is not limited to the specific configuration and processing described above and shown in the figures. For the sake of simplicity, a detailed description of the known method is omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present invention is not limited to the specific steps described and shown, and those skilled in the art can make various changes, modifications and additions, or change the order between the steps after understanding the spirit of the present invention.
[0081] In the present invention, features described and / or illustrated for one embodiment may be used in the same or similar manner in one or more other embodiments, and / or combined with features of other embodiments or replace features of other embodiments.
[0082] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the embodiments of the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.
Claims
1. A method for evaluating pilot cognitive ability, characterized in that: The method comprises the following steps: In the process of the pilot performing the flight situation task presented by the flight situation task presentation module, the pilot's index parameters are collected through multiple channels, and the collected multi-channel index parameters include: interaction index, physiological index and brain function index; When the pilot performs the cognitive assessment task presented by the cognitive presentation module, the pilot's index parameters are collected through multiple channels, and the collected multi-channel index parameters include: interaction index, physiological index and brain function index; The personality-related cognitive ability assessment results of the flight personnel are obtained by the questionnaire response results obtained when the pilot performs the questionnaire scale presented by the flight mission-related cognitive questionnaire scale assessment module; wherein the interactive indicators include: behavioral indicators and eye movement indicators, the behavioral indicators include reaction time and / or accuracy, and the eye movement indicators include scanning time, fixation time, blinking frequency, pupil diameter change, eyelid closure degree and eye movement trajectory; A feature vector is obtained based on the multi-channel indicator parameters of the pilot in the flight situation task, the multi-channel indicator parameters in the cognitive assessment task, and the personality-related cognitive ability assessment results, and the feature vector is input into a pre-trained pilot cognitive ability assessment model, and a first assessment score of the pilot cognitive ability is obtained based on the output data; wherein the pilot cognitive ability assessment model is an XGBoost machine learning model obtained based on the gradient boosting tree algorithm; each feature in the feature vector is obtained by weighting the indicator parameters collected under the corresponding task; Normalizing the multi-channel indicator parameters of the pilot in the flight situation task to obtain a first normalized result, normalizing the multi-channel indicator parameters in the cognitive assessment task to obtain a second normalized result, and performing weighted averaging based on multiple preset indicator parameter weights in the first normalized result, the second normalized result, and the personality-related cognitive ability assessment result to obtain a second assessment score of the pilot's cognitive ability; and A comprehensive assessment score of the pilot's cognitive ability is obtained based on the first assessment score of the pilot's cognitive ability and the second assessment score of the pilot's cognitive ability.
2. The method according to claim 1, characterized in that The multi-channel index parameters collected during the pilot's execution of various tasks are the same or partially the same index parameters; The multi-channel index parameters collected during the pilot's execution of various tasks include physiological indicators, which include one or more of the following indicators: skin electricity index, electrocardiogram index, myoelectric index, pulse index, breathing index, blood oxygen index and blood pressure index.
3. The method according to claim 2, characterized in that The multi-channel index parameters collected during the pilot's execution of each task also include brain function indicators, which include one or more of the following indicators: EEG time-frequency indicators, nonlinear indicators, event-related brain activity indicators, brain function connection indicators, and blood oxygen level change indicators in various brain regions.
4. The method according to claim 1, characterized in that: The questionnaire scale includes at least one of the following: a task-related contextualized questionnaire, a personality scale, and a cognitive scale.
5. The method according to claim 1, characterized in that The pilot's second cognitive ability assessment score meets the following requirements: Among them, i t For each task, k is the multi-channel indicator parameter in each task, Z tk represents the normalized multi-channel index parameters in each task, C tk is the weight coefficient, t represents the number of tasks performed by the pilot, and k represents the number of index parameters of each task.
6. The method according to claim 1, characterized in that The flight scenario task presentation module includes a flight simulator for completing simulations of various flight scenarios and flight environments.
7. A pilot cognitive ability assessment system, comprising a processor and a memory, characterized in that: The memory stores computer instructions, and the processor is used to execute the computer instructions stored in the memory. When the computer instructions are executed by the processor, the system implements the steps of the method as claimed in any one of claims 1 to 6.
8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.
