An intelligent system for pilot psychological assessment and personalized training

Abstract

The present application relates to the technical field of education training data processing, and more particularly to an intelligent system for pilot psychological assessment and individualized training, the system comprising a psychological data processing module, a psychological assessment module, an individualized matching training scheme module and a team cooperation module, the psychological data processing module performing data fusion on feature extraction results, the psychological assessment module acquiring psychological states, the individualized matching training scheme module acquiring training schemes, and the team cooperation module matching roles for pilots, the present application performing real-time monitoring on training data, real-time matching and adjustment of training schemes for pilots, so as to flexibly match training schemes for individual psychological states of pilots, and reasonably allocate team roles, thereby improving the efficiency of pilot training and team task execution efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of education and training data processing, and in particular to an intelligent system for pilot psychological assessment and personalized training. Background Art

[0002] Traditional pilot psychological assessments mainly rely on questionnaires and expert interviews. When filling out questionnaires, pilots may be influenced by factors such as subjective will and social expectations, resulting in biased answers. Although expert interviews can conduct more in-depth communication, the assessment results largely depend on the experts' personal experience and subjective judgment. The assessment standards of different experts may vary, which affects the accuracy and reliability of the assessment. Traditional pilot psychological training usually adopts a unified training model and does not consider individual differences. However, each pilot has different personalities, experiences, psychological foundations, etc., so the real-time acceptance and effect of the training process will also vary. As a result, the use of a unified training model cannot achieve ideal individual training effects based on real-time training feedback.

[0003] Chinese Patent Publication No. CN114693170A discloses a psychological assessment method for pilot parachute training based on multi-source parameter fusion, which relates to the field of psychological assessment technology. A psychological assessment index system for pilot parachute training is established, including the selection of evaluation indicators and the division of evaluation levels and their principles. A psychological assessment method for pilot parachute training is established by collecting pilot physiological data and combining it with individual characteristic data. This solution still lacks feedback on the individual real-time psychological state of pilots during training, and lacks personalized matching of training programs based on real-time feedback, resulting in low training efficiency. Summary of the Invention

[0004] To this end, the present invention provides an intelligent system for pilot psychological assessment and personalized training, which is used to overcome the problem in the prior art of lacking feedback on the individual real-time psychological state of pilots during training, and performing personalized matching training according to the real-time feedback, thereby leading to low training efficiency.

[0005] To achieve the above objectives, the present invention provides an intelligent system for pilot psychological assessment and personalized training, the system comprising:

[0006] Psychological data collection module, used to collect training data;

[0007] The psychological data processing module is used to extract features from the training data to obtain training data features, and is also used to perform data fusion on the training data features to obtain training data fusion vectors;

[0008] A psychological assessment module is used to obtain a psychological state index based on the training data fusion vector, and to obtain the psychological state based on the psychological state index, and to perform task adjustment on the psychological state, and to perform experience adjustment on the task adjustment process;

[0009] A personalized training program matching module is used to classify pilot types according to their psychological state, obtain pilot type classification results, and obtain training programs based on the pilot type classification results. It is also used to adjust the psychological state and training program based on the comprehensive training index, and to update the training adjustment process;

[0010] The team collaboration module is used to match pilots' roles according to the psychological state index and the comprehensive training index through the role matching method, to make team adjustments to the role matching method, and to update the team during the team adjustment process.

[0011] Furthermore, the psychological data processing module collects training data;

[0012] When the psychological data processing module extracts features from the training data, the training data is input into the data feature extraction model to obtain training data features, wherein the training data features include physiological data features hrv, behavioral data features hry and environmental data features hru;

[0013] When the psychological data processing module performs data fusion on the feature extraction results, it calculates the training data fusion vector Fto based on the physiological data feature hrv, the behavioral data feature hry, the environmental data feature hru, the pilot communication ability index Gton, the physiological data feature weight w1, the behavioral data feature weight w2, the environmental data feature weight w3 and the pilot communication ability index weight wt to obtain the training data fusion vector Fto, and sets Fto=hrv×w1+hry×w2+hru×w3+Gton×wt.

[0014] Furthermore, when the psychological assessment module obtains the psychological state index based on the training data fusion vector, the training data fusion vector is input into the psychological state prediction model to obtain the psychological state index;

[0015] When the psychological assessment module obtains the psychological state according to the psychological state index, it compares the psychological state index Hx with the first preset psychological state index Hx1 and the second preset psychological state index Hx2, sets Hx1=35 and Hx2=65, judges the state of the psychological state index according to the comparison result, and outputs the psychological state according to the judgment result, wherein:

[0016] When Hx<Hx1, the psychological assessment module determines that the psychological state index is low and outputs the relaxation state as the psychological state;

[0017] When Hx2≤Hx<Hx2, the psychological assessment module determines that the state of the psychological state index is moderate and outputs the normal state as the psychological state;

[0018] When Hx≥Hx2, the psychological assessment module determines that the psychological state index is high and outputs the tension state as the psychological state;

[0019] When the psychological assessment module adjusts the psychological state according to the complexity of the flight mission, the flight mission complexity Ff is compared with the preset flight mission complexity Ff0, and 0.53≤Ff0≤0.69 is set. The complexity of the flight mission is judged according to the comparison result, and the task is adjusted according to the first preset psychological state index Hx1 and the second preset psychological state index Hx2 according to the judgment result, wherein:

[0020] When Ff≤Ff0, the psychological assessment module determines that the complexity of the flight mission is not complex, and does not adjust the first preset psychological state index Hx1 and the second preset psychological state index Hx2;

[0021] When Ff>Ff0, the psychological assessment module determines that the complexity of the flight mission is complex, and adjusts the first preset psychological state index Hx1 and the second preset psychological state index Hx2 according to the complexity coefficient α, setting α=0.38+0.3×e -0.7 ×（Ff-Ff0） , obtain the adjusted first preset mental state index Hx1` and the adjusted second preset mental state index Hx2`, set Hx1`=Hx1×α, Hx2`=Hx2×α, replace the first preset mental state index Hx1 with the adjusted first preset mental state index Hx1`, replace the second preset mental state index Hx2 with the adjusted second preset mental state index Hx2`, and re-compare the mental state index Hx with the adjusted first preset mental state index Hx1` and the adjusted second preset mental state index Hx2`.

[0022] Furthermore, when the psychological assessment module makes empirical adjustments to the task adjustment process based on the number of flights, the number of flights Cf is compared with the preset number of flights Cf0, and 15 times ≤ Cf0 ≤ 23 times is set. The number of flights is judged based on the comparison result, and the preset flight task complexity Ff0 is empirically adjusted based on the judgment result, wherein:

[0023] When Cf ≥ Cf0, the psychological assessment module determines that the number of flights is sufficient and does not perform empirical adjustments to the preset flight mission complexity Ff0;

[0024] When Cf<Cf0, the psychological assessment module determines that the number of flights is insufficient, and empirically adjusts the preset flight mission complexity Ff0 according to the empirical adjustment coefficient γ, setting γ=0.95-0.4×e -0.7 ×（Cf / cf0） , where e is the base of the natural logarithm. The adjusted preset flight mission complexity Ff0` is obtained, and Ff0`=Ff0×γ is set. The preset flight mission complexity Ff0 is replaced with the adjusted preset flight mission complexity Ff0`, and the flight mission complexity Ff is recompared with the adjusted preset flight mission complexity Ff0`.

[0025] Furthermore, the personalized matching training program module classifies pilot types according to psychological states using a pilot type classification method, wherein the pilot type classification method includes:

[0026] Step B01, classifying the pilots according to their psychological states to obtain classification results, wherein the classification results include high stress sensitivity and emotional stability;

[0027] Step B02: Initially select the cluster centers to obtain the initial cluster centers cj`={cj1,cj2,cj3,...,cjh};

[0028] Step B03, calculate the target distance set d = {d1, d2, d3, ..., dn} based on the initial cluster center cj` = {cj1, cj2, cj3, ..., cjh} and the training data features xi = {xi1, xi2, xi3, ..., xin}, and set , where nj is the total number of data points in the training data features;

[0029] Step B04: assign each data point in the training data feature to the category classification result that is closest to the data point, to obtain an updated category classification result;

[0030] Step B05, clustering and updating the initial cluster centers to obtain updated cluster centers;

[0031] Step B06: Replace the initial cluster center with the updated cluster center, and repeat steps B01 to B06 according to the number of iterations Dd, setting Dd=1;

[0032] Step B07: Compare the number of iterations Dd with the preset number of iterations Dd0, setting 3 ≤ Dd0 ≤ 5. Based on the comparison result, determine whether the number of iterations meets the standard, and output the pilot type classification result based on the determination result, where:

[0033] When Dd<Dd0, the personalized matching training program module determines that the number of iterations does not meet the standard, and continues to repeat steps B01 to B06 until Dd≥Dd0;

[0034] When Dd≥Dd0, the personalized matching training program module determines that the number of iterations has met the standard and outputs the updated category division result as the pilot type classification result.

[0035] Furthermore, when the personalized matching training program module obtains the training program according to the pilot type classification result, if the pilot type classification result is high stress sensitivity, low-intensity flight is output as the training program;

[0036] When the pilot type classification result is emotionally stable, high-intensity flight is output as the training plan;

[0037] When the personalized matching training program module adjusts the psychological state and training program according to the comprehensive training index, the comprehensive training index Zx is calculated according to the performance assessment score Jk, the psychological state index Hx, and the physiological index evaluation value Sp to obtain the comprehensive training index Zx, and Zx=0.3×Jk+0.4×Hx0+0.3×Sp is set. The comprehensive training index Zx is compared with the preset comprehensive training index Zx0, and 0.61≤Zx0≤0.74 is set. The compliance of the comprehensive training index is judged based on the comparison result, and the preset number of flights Cf0 and the preset number of iterations Dd0 are adjusted according to the judgment result.

[0038] When Zx≥Zx0, the personalized matching training program module determines that the comprehensive training index is up to standard, and adjusts the preset number of flights Cf0 according to the first training adjustment coefficient. Adjust the preset flight times Cf0 and set =1.77-0.23×e -0.38×（Zx-Zx0） , where e is the base of the natural logarithm, and the preset number of flights after training adjustment is Cf01, set Cf01=Cf0× , replacing the preset number of flights Cf0 with the preset number of flights Cf01 after training adjustment, and re-comparing the number of flights Cf with the preset number of flights Cf01 after training adjustment;

[0039] When Zx<Zx0, the personalized matching training program module determines that the comprehensive training index is not up to standard, and adjusts the preset number of iterations Dd0 according to the second training adjustment coefficient. Adjust the preset number of iterations Dd0 for training, set =1.27-0.14×e -0.7×（Zx0-Zx） , where e is the base of the natural logarithm, and the preset number of iterations after training adjustment is obtained, Dd0`, set Dd0`=Dd0× , replace the preset number of iterations Dd0 with the preset number of iterations Dd0' after training adjustment, and re-compare the number of iterations Dd with the preset number of iterations Dd0' after training adjustment.

[0040] Furthermore, when the personalized matching training program module performs training updates on the training adjustment process according to the number of training times, the number of training times Cx is compared with the preset number of training times Cx0, and 9 times ≤ Cx0 ≤ 12 times is set. The compliance of the number of training times is judged based on the comparison result, and the preset comprehensive training index Zx0 is updated based on the judgment result, wherein:

[0041] When Cx≤Cx0, the personalized matching training program module determines that the number of training times does not meet the standard, and does not perform training updates on the preset comprehensive training index Zx0;

[0042] When Cx>Cx0, the personalized matching training program module determines that the number of training times has reached the standard, and updates the preset comprehensive training index Zx0 according to the training update coefficient £, setting £=1.38-0.21×e -0.7×（Cx-Cx0） , where e is the base of the natural logarithm, and the preset comprehensive training index Zx01 after training update is obtained, and Zx01=Zx0×£ is set. The preset comprehensive training index Zx0 is replaced with the preset comprehensive training index Zx01 after training update, and the comprehensive training index Zx is re-compared with the preset comprehensive training index Zx01 after training update.

[0043] Furthermore, the team collaboration module performs role matching for pilots according to the psychological state index and the comprehensive training index through a role matching method, and the role matching method includes:

[0044] Step C01: Obtain the pilot's characteristic vector P based on the psychological state index and the comprehensive training index, and set P = {P1, P2, P3, ..., Pf};

[0045] Step C02, set the role demand vector G = {G1, G2, G3, ..., Gv};

[0046] Step C03, calculate the role matching degree Mp according to the pilot feature vector P = {P1, P2, P3, ..., Pf} and the role requirement vector G = {G1, G2, G3, ..., Gv}, and obtain the role matching degree Mp, and set , where nu is the total number of features of the role requirement vector, and ny is the total number of features of the pilot feature vector;

[0047] Step C04: compare the role matching degree Mp with the preset role matching degree Mp0, set 0.68≤Mp0≤0.71, judge the matching status of the role matching degree according to the comparison result, and output the role matching result according to the judgment result, wherein:

[0048] When Mp<Mp0, the team collaboration module determines that the role matching degree is mismatched, and outputs the mismatch between the pilot and the role as a role matching result;

[0049] When Mp≥Mp0, the team collaboration module determines that the role matching degree is a match, and outputs the matching between the pilot and the role as a role matching result.

[0050] Furthermore, when the team collaboration module performs team adjustment on the role matching method according to the team status value, the team status index is obtained according to the team status index obtaining method, and the team status index obtaining method includes:

[0051] Step K01: Calculate the pilot's current pressure value Ps based on heart rate HR, resting heart rate HRrest, maximum heart rate HRmax, galvanic skin response GSR, resting galvanic skin response GSRrest, maximum galvanic skin response GSRmax, and EEG pressure value EEGs to obtain the pilot's current pressure value Ps and set ;

[0052] Step K02: Calculate the pressure average value Pt based on the pressure mean value Pp and the pressure standard deviation σp to obtain the pressure average value Pt, and set Pt=Pp+1.23×σp;

[0053] Step K03, calculate the team status index S according to the pilot's current pressure value Ps and the pressure average value Pt, and obtain the team status index S, set .

[0054] Furthermore, when the team collaboration module adjusts the role matching method according to the team status value, the team status index S is compared with the preset team status index S0, and 0.82≤S0≤0.94 is set. The team status is judged according to the comparison result, and the preset role matching degree Mp0 is adjusted according to the judgment result, wherein:

[0055] When S≥S0, the team collaboration module determines that the team status is up to standard and does not adjust the team to the preset role matching degree Mp0;

[0056] When S<S0, the team collaboration module determines that the team status is not up to standard, and adjusts the preset role matching degree Mp0 according to the team adjustment index eko, setting eko=1.77-0.4×e -0.7 ×（S0-S） , where e is the base of the natural logarithm, and the preset role matching degree after team adjustment Mp0` is obtained, Mp0`=Mp0×eko is set, the preset role matching degree Mp0 is replaced with the preset role matching degree after team adjustment Mp0`, and the role matching degree Mp is re-compared with the preset role matching degree after team adjustment Mp0`;

[0057] The team collaboration module updates the team adjustment process according to the changes in team members. If there are no changes in team members, the team adjustment process will not be updated.

[0058] When there are changes in team members, the team adjustment process is updated according to the team update coefficient. Update the team status index S and set it to 0.71≤ ≤0.77, get the team status index S` after team update, set S`=S× , compare the updated team status index S` with the preset team status index S0, and re-judge whether the team status meets the standards.

[0059] Compared with the existing technology, the beneficial effect of the present invention lies in that it monitors the training data to obtain the pilot's psychological state, and personalizes the training plan for the pilot according to the pilot's psychological state, so as to flexibly match the training plan according to the individual psychological state of the pilot. At the same time, the team roles are allocated and adjusted according to the pilot's psychological state, so as to match the appropriate team roles according to the differences in the pilot's psychological state, thereby improving the efficiency of pilot training and the efficiency of team task execution. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 Schematic diagram of the structure of the intelligent system for pilot psychological assessment and personalized training in this embodiment. DETAILED DESCRIPTION

[0061] In order to make the objects and advantages of the present invention more clearly understood, the present invention is further described below in conjunction with embodiments; it should be understood that the specific embodiments described herein are merely used to explain the present invention and are not intended to limit the present invention.

[0062] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only used to explain the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0063] It should be noted that, in the description of the present invention, terms such as "up", "down", "left", "right", "inside", and "outside" indicating directions or positional relationships are based on the directions or positional relationships shown in the accompanying drawings. This is only for the convenience of description and does not indicate or imply that the device or element must have a specific orientation, be constructed and operated in a specific orientation. Therefore, it cannot be understood as a limitation on the present invention.

[0064] Furthermore, it should be noted that, in the description of the present invention, unless otherwise expressly specified or limited, the terms "mounted," "connected," and "connected" should be understood in a broad sense. For example, they may refer to fixed connections, detachable connections, or integral connections; mechanical connections or electrical connections; direct connections or indirect connections through an intermediate medium; and internal communication between two components. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on specific circumstances.

[0065] See also Figure 1 As shown in FIG. 1 , an intelligent system for pilot psychological assessment and personalized training according to this embodiment is shown. The system includes:

[0066] Psychological data collection module, used to collect training data;

[0067] The psychological data processing module is used to extract features from the training data to obtain training data features, and is also used to perform data fusion on the training data features to obtain training data fusion vectors;

[0068] A psychological assessment module is used to obtain a psychological state index based on the training data fusion vector, and to obtain the psychological state based on the psychological state index, and is also used to perform task adjustment on the psychological state, and is also used to perform experience adjustment on the task adjustment process. The psychological assessment module is connected to the psychological data processing module;

[0069] A personalized training program matching module is used to classify pilot types according to their psychological state, obtain pilot type classification results, and acquire training programs based on the pilot type classification results. It is also used to adjust the psychological state and training program based on the comprehensive training index and to update the training adjustment process. The personalized training program matching module is connected to the psychological assessment module;

[0070] The team collaboration module is used to match pilots' roles according to the psychological state index and the comprehensive training index through the role matching method, to make team adjustments to the role matching method, and to update the team during the team adjustment process. The team collaboration module is connected to the personalized matching training program module.

[0071] Specifically, the intelligent system for pilot psychological assessment and personalized training is applied to the pilot training terminal, which monitors the training data and matches and adjusts the pilot's training plan according to the training data, so as to flexibly match the training plan according to the individual psychological state of the pilot, and reasonably allocate the team roles, thereby improving the efficiency of pilot training and the efficiency of team task execution. The intelligent system for pilot psychological assessment and personalized training collects the training data in real time through the psychological data collection module, so as to match the pilot with the training plan in real time according to the training data, thereby improving the accuracy and matching degree of training. The intelligent system for pilot psychological assessment and personalized training also fuses the training data features through the psychological data processing module, so as to integrate the data into the training data. The system classifies the pilot's psychological state in real time based on the characteristics of the pilot, so as to facilitate the matching of subsequent training plans and improve the training efficiency. The intelligent system for pilot psychological assessment and personalized training also uses the psychological assessment module to analyze the pilot's psychological state in real time, so as to facilitate the evaluation of the pilot's psychological state based on real-time training feedback and improve the real-time training efficiency. The intelligent system for pilot psychological assessment and personalized training also obtains the pilot type classification results through the personalized matching training plan module, so as to reasonably match the training plans based on the pilot type classification results, thereby improving the real-time training efficiency. The intelligent system for pilot psychological assessment and personalized training also uses the team collaboration module to match appropriate team roles according to the differences in the pilot's psychological state, so as to efficiently complete the team task, thereby improving the efficiency of team task execution.

[0072] Specifically, the psychological data processing module collects training data.

[0073] Specifically, the training data includes physiological data, behavioral data and environmental data. The physiological data includes heart rate variability data and electroencephalogram data. The behavioral data includes gaze point, eye saccades, pupil radius and facial micro-expressions. The environmental data refers to relevant environmental parameters in the flight environment. This embodiment does not limit the environmental data. Relevant technical personnel in this field can freely choose according to actual conditions, such as flight altitude and cabin temperature. This embodiment does not limit the specific method of collecting environmental data. Relevant technical personnel in this field can freely choose according to actual needs, such as obtaining it through a flight recording system. The heart rate variability data refers to a numerical value used to reflect the degree of variation between the pilot's successive heartbeat cycles. The electroencephalogram data refers to a numerical value used to reflect the electrical activity of the pilot's cerebral cortex neurons. This implementation does not limit the specific method of collecting heart rate variability data and electroencephalogram data. Relevant technicians in this field can freely choose according to actual conditions, such as collecting heart rate variability data through dynamic electrocardiogram and collecting electroencephalogram data through scalp electrodes. The gaze point refers to the specific position where the pilot's line of sight is focused, and the eye saccade refers to data used to reflect the pilot's switching of line of sight between different gaze points. The pupil radius refers to the radius length of the pilot's pupil. This embodiment does not limit the specific collection method of the gaze point, eye saccade and pupil radius. Relevant technicians in this field can freely choose according to actual conditions, such as collecting the gaze point, eye saccade and pupil radius through an eye tracker. The facial micro-expression refers to data used to reflect subtle changes in the pilot's facial muscles. This embodiment does not limit the specific collection method of facial micro-expressions. Relevant technicians in this field can freely choose according to actual conditions, such as collecting facial micro-expressions through electromyography sensors.

[0074] Specifically, the psychological assessment module collects training data in real time to facilitate the subsequent real-time matching of training plans for pilots based on the training data, thereby improving the accuracy and matching of training.

[0075] Specifically, when the psychological data processing module extracts features from the training data, the training data is input into a data feature extraction model to obtain training data features, where the training data features include physiological data features hrv, behavioral data features hry, and environmental data features hru.

[0076] Specifically, the data feature extraction model refers to a convolutional neural network model that takes training data as input data and training data features as output data. The psychological assessment module constructs the data feature extraction model through a data feature extraction model construction method. This embodiment does not limit the data feature extraction model construction method. Relevant technical personnel in this field can freely choose according to actual needs, such as using historical training data and historical training data features corresponding to historical training data as a data set for training the association between historical training data and historical training data features to train the data feature extraction model. The physiological data features refer to a collection of digital vectors reflecting information features in physiological data obtained according to the data feature extraction model. The behavioral data features refer to a collection of digital vectors reflecting information features in behavioral data obtained according to the data feature extraction model. The environmental data features refer to a collection of digital vectors reflecting information features in environmental data obtained according to the data feature extraction model.

[0077] Specifically, the psychological assessment module obtains the characteristics of the training data and classifies the data to facilitate the matching of subsequent training plans and improve training efficiency.

[0078] Specifically, when the psychological data processing module performs data fusion on the feature extraction results, it calculates the training data fusion vector Fto based on the physiological data feature hrv, the behavioral data feature hry, the environmental data feature hru, the pilot communication ability index Gton, the physiological data feature weight w1, the behavioral data feature weight w2, the environmental data feature weight w3 and the pilot communication ability index weight wt to obtain the training data fusion vector Fto, and sets Fto=hrv×w1+hry×w2+hru×w3+Gton×wt.

[0079] Specifically, the pilot communication ability index refers to a numerical value for evaluating the pilot's communication ability. This embodiment does not limit the specific method for obtaining the pilot communication ability index. Relevant technicians in this field can freely choose according to actual needs, such as team assessment. The physiological data feature weight refers to a coefficient that measures the importance of physiological data features in the training data fusion vector. The behavioral data feature weight refers to a coefficient that measures the importance of behavioral data features in the training data fusion vector. The environmental data feature weight refers to a coefficient that measures the importance of environmental data features in the training data fusion vector. The pilot communication ability index weight refers to a coefficient that measures the importance of the pilot communication ability index in the training data fusion vector. This embodiment does not limit the physiological data feature weight, behavioral data feature weight, environmental data feature weight, and pilot communication ability index weight. Relevant technicians in this field can freely choose according to actual needs, and only need to meet the requirement of w1+w2+w3+wt=1, such as setting w1=0.2, w2=0.2, w3=0.3, and wt=0.3.

[0080] Specifically, the psychological assessment module obtains the training data fusion vector so as to intuitively present the information contained in the training data in a fused digital form, thereby improving the efficiency of real-time training analysis.

[0081] Specifically, when the psychological assessment module obtains the psychological state index according to the training data fusion vector, the training data fusion vector is input into the psychological state prediction model to obtain the psychological state index.

[0082] Specifically, the mental state prediction model refers to a dynamic Bayesian network model that takes the training data fusion vector as input data and the mental state index as output data. The psychological assessment module constructs the mental state prediction model through the mental state prediction model construction method. This embodiment does not limit the mental state prediction model construction method. Relevant technical personnel in this field can freely choose according to actual needs, such as using the historical training data fusion vector and the historical mental state index corresponding to the training data fusion vector as a data set for training the association between the training data fusion vector and the mental state index to train the mental state prediction model.

[0083] Specifically, the psychological assessment module obtains the psychological state index in real time and intuitively presents the pilot's psychological state in numerical form, so as to evaluate the pilot's psychological state based on real-time training feedback and improve real-time training efficiency.

[0084] Specifically, when the psychological assessment module obtains the psychological state according to the psychological state index, it compares the psychological state index Hx with the first preset psychological state index Hx1 and the second preset psychological state index Hx2, sets Hx1=35 and Hx2=65, judges the state of the psychological state index according to the comparison result, and outputs the psychological state according to the judgment result, wherein:

[0085] When Hx<Hx1, the psychological assessment module determines that the psychological state index is low and outputs the relaxation state as the psychological state;

[0086] When Hx2≤Hx<Hx2, the psychological assessment module determines that the state of the psychological state index is moderate and outputs the normal state as the psychological state;

[0087] When Hx≥Hx2, the psychological assessment module determines that the state of the psychological state index is high, and outputs the tense state as the psychological state.

[0088] Specifically, the first preset mental state index refers to a preset lower limit value for judging the state of the mental state index, the second preset mental state index refers to a preset upper limit value for judging the state of the mental state index, the state of the mental state index refers to the degree of relaxation of the pilot's mental state judged based on the mental state index and the first preset mental state index and the second preset mental state index, and the states of the mental state index include a low mental state index state, a moderate mental state index state, and a high mental state index state.

[0089] Specifically, the psychological assessment module obtains the psychological state and intuitively presents the pilot's real-time psychological state to facilitate subsequent matching of training plans, thereby improving the targeted nature of training.

[0090] Specifically, when the psychological assessment module adjusts the psychological state according to the complexity of the flight mission, it compares the flight mission complexity Ff with the preset flight mission complexity Ff0, sets 0.53≤Ff0≤0.69, judges the complexity of the flight mission based on the comparison result, and adjusts the first preset psychological state index Hx1 and the second preset psychological state index Hx2 based on the judgment result, where:

[0091] When Ff≤Ff0, the psychological assessment module determines that the complexity of the flight mission is not complex, and does not adjust the first preset psychological state index Hx1 and the second preset psychological state index Hx2;

[0092] When Ff>Ff0, the psychological assessment module determines that the complexity of the flight mission is complex, and adjusts the first preset psychological state index Hx1 and the second preset psychological state index Hx2 according to the complexity coefficient α, setting α=0.38+0.3×e -0.7 ×（Ff-Ff0） , obtain the adjusted first preset mental state index Hx1` and the adjusted second preset mental state index Hx2`, set Hx1`=Hx1×α, Hx2`=Hx2×α, replace the first preset mental state index Hx1 with the adjusted first preset mental state index Hx1`, replace the second preset mental state index Hx2 with the adjusted second preset mental state index Hx2`, and re-compare the mental state index Hx with the adjusted first preset mental state index Hx1` and the adjusted second preset mental state index Hx2`.

[0093] Specifically, the flight mission complexity refers to a numerical value for evaluating the complexity of the flight mission. This embodiment does not limit the specific method for obtaining the complexity of the flight mission. Relevant technical personnel in this field can freely choose according to actual needs, such as the expert evaluation method. The expert evaluation method refers to the expert user who has the ability to set the flight mission complexity to set the numerical value of the flight mission complexity. The preset flight mission complexity refers to a preset value for judging the complexity of the flight mission. The complexity of the flight mission refers to the complexity of the flight mission judged based on the flight mission complexity and the preset flight mission complexity. The complexity of the flight mission includes the complexity of the flight mission being complex and the complexity of the flight mission being not complex.

[0094] Specifically, the psychological assessment module judges the complexity of the flight mission. When the complexity of the flight mission is high, the pilot needs to remain alert for a long time. At this time, the first preset psychological state index Hx1 and the second preset psychological state index Hx2 are reduced to avoid the impact of the pilot's psychological state changes under high-pressure conditions on the flight mission and improve flight safety.

[0095] Specifically, when the psychological assessment module makes empirical adjustments to the task adjustment process based on the number of flights, the number of flights Cf is compared with the preset number of flights Cf0, and 15 times ≤ Cf0 ≤ 23 times is set. The number of flights is judged based on the comparison result, and the preset flight task complexity Ff0 is empirically adjusted based on the judgment result, where:

[0096] When Cf ≥ Cf0, the psychological assessment module determines that the number of flights is sufficient and does not perform empirical adjustments to the preset flight mission complexity Ff0;

[0097] When Cf -0.7 ×（Cf / cf0） wherein e is the base of natural logarithm, to obtain the adjusted preset flight task complexity Ff0`, set Ff0`= Ff0× γ, replace the preset flight task complexity Ff0with the adjusted preset flight task complexity Ff0`, and re-compare the flight task complexity Ff with the adjusted preset flight task complexity Ff0`.

[0098] Specifically, the flight frequency refers to the number of times that the pilot performs the same flight task, and is used to measure the flight experience of the pilot. The specific acquisition method of the flight frequency is not limited in the embodiment, and can be freely selected by the person skilled in the art according to the actual needs, such as software recording. The preset flight frequency refers to a preset value for judging the condition of the flight frequency. The condition of the flight frequency refers to the sufficiency of the flight experience of the pilot, which is judged according to the flight frequency and the preset flight frequency. The condition of the flight frequency includes the condition of the flight frequency being sufficient and the condition of the flight frequency being insufficient.

[0099] Specifically, the psychological evaluation module judges the condition of the flight frequency. When the pilot's flight experience is insufficient, the preset flight task complexity Ff0is reduced with the decrease of the flight frequency, so as to reduce the psychological influence of the flight task complexity on the pilot, thereby improving the accuracy of judging the psychological state of the pilot.

[0100] Specifically, the individualized matching training scheme module classifies the pilot types according to the psychological state by the pilot type classification method, and the pilot type classification method includes:

[0101] Step B01, classifying the pilot according to the psychological state to obtain a classifying result, and the classifying result includes a high-pressure sensitive type and an emotional stability type.

[0102] Step B02, initially selecting a clustering center to obtain an initial clustering center cj`={cj1, cj2, cj3,..., cjh};

[0103] Step B03, calculating a target distance set d={d1, d2, d3,..., dn} according to the initial clustering center cj`={cj1, cj2, cj3,..., cjh} and the training data feature xi={xi1, xi2, xi3,..., xin}, and setting wherein nj is the total number of data points in the training data feature;

[0104] Step B04: assign each data point in the training data feature to the category classification result that is closest to the data point, to obtain an updated category classification result;

[0105] Step B05, clustering and updating the initial cluster centers to obtain updated cluster centers;

[0106] Step B06: Replace the initial cluster center with the updated cluster center, and repeat steps B01 to B06 according to the number of iterations Dd, setting Dd=1;

[0107] Step B07: Compare the number of iterations Dd with the preset number of iterations Dd0, setting 3 ≤ Dd0 ≤ 5. Based on the comparison result, determine whether the number of iterations meets the standard, and output the pilot type classification result based on the determination result, where:

[0108] When Dd<Dd0, the personalized matching training program module determines that the number of iterations does not meet the standard, and continues to repeat steps B01 to B06 until Dd≥Dd0;

[0109] When Dd≥Dd0, the personalized matching training program module determines that the number of iterations has met the standard and outputs the updated category division result as the pilot type classification result.

[0110] Specifically, the category division refers to the process of dividing the real-time psychological state of the pilots in training into three types, the cluster center refers to the mean value of each feature of all data points in the category division result, the high stress sensitivity type refers to the first category in the category division result, which is used to represent the type of pilots whose psychological state is tense when facing flight mission pressure, and the emotionally stable type refers to the type of pilots whose psychological state is normal and relaxed when facing flight mission pressure. The initial selection refers to the process of specifying the cluster center. This embodiment does not limit the initial selection. Relevant technical personnel in this field can freely select according to actual needs, such as random selection. The initial cluster center cj`={cj1, cj2, cj3, ..., cjh} refers to the data set obtained after the initial selection of the cluster center, where cj1 is the value of the first data point in the initial cluster center, cj2 is the value of the second data point in the initial cluster center, cj3 is the value of the third data point in the initial cluster center, cjh is the value of the hth data point in the initial cluster center, and h is the value of the data point in the initial cluster center. The training data feature xi={xi1, xi2, xi3, ..., xin} refers to the data set containing the features of the training data obtained after feature extraction of the training data, wherein xi1 is the value of the first data point in the training data feature, xi2 is the value of the second data point in the training data feature, xi3 is the value of the third data point in the training data feature, xin is the value of the nth data point in the training data feature, n is the order of the data points in the training data feature, and the target distance collection refers to the distance between each data point in the initial cluster center. The distance between the point and each data point in the training data feature, the category division result closest to the data point refers to the category division result to which the initial cluster center with the shortest distance from a single data point in the training data feature belongs. The cluster update refers to the process of taking the mean of all data points in the category division result as the new initial cluster center. For example, if there are gy data points in the j`` category division result, the j`` category division result Xj``={xj``1,xj``2,xj``3,...,xj``nj``}, the updated cluster center , where j``=1, 2, ..., nj``, j`` is the order of data points in the category division result, xj``1 is the value of the first data point of the j``th class in the category division result, xj``2 is the value of the second data point of the j``th class in the category division result, xj``3 is the value of the third data point of the j``th class in the category division result, xj``nj`` is the value of the nj``th data point of the j``th class in the category division result, and nj`` is the order of data points in the j``th class division result. The number of iterations refers to the number of times steps B01 to B06 are repeated. The preset number of iterations Dd0 refers to a preset value for judging whether the number of iterations meets the standard. The compliance of the number of iterations refers to the degree of compliance of the number of iterations judged based on the number of iterations and the preset number of iterations. The compliance of the number of iterations includes the compliance of the number of iterations being unsatisfactory and the compliance of the number of iterations being satisfactory.

[0111] Specifically, the personalized matching training program module obtains the pilot type classification results so as to reasonably match the training program according to the pilot type classification results, thereby improving real-time training efficiency.

[0112] Specifically, when the personalized matching training program module obtains the training program according to the pilot type classification result, if the pilot type classification result is high pressure sensitivity, low-intensity flight is output as the training program;

[0113] When the pilot type classification result is emotionally stable, high-intensity flight is output as the training plan.

[0114] Specifically, the low-intensity flight refers to a training program that matches the high-stress sensitive type. This embodiment does not limit the specific training content of low-intensity flight. Relevant technical personnel in this field can freely choose according to actual needs, such as setting low-intensity flight to daily flight scenarios with simple flight missions and short flight time. The high-intensity flight refers to a training program that matches the emotionally stable type. This embodiment does not limit the specific training content of high-intensity flight. Relevant technical personnel in this field can freely choose according to actual needs, such as setting high-intensity flight to daily flight scenarios with complex flight missions and long flight time.

[0115] Specifically, the personalized matching training program module obtains the training program and reasonably matches the training program according to the individual differences of pilots, thereby improving the training effect.

[0116] Specifically, when the personalized matching training program module adjusts the psychological state and training program according to the comprehensive training index, the comprehensive training index Zx is calculated according to the performance assessment score Jk, the psychological state index Hx and the physiological index evaluation value Sp to obtain the comprehensive training index Zx, and Zx=0.3×Jk+0.4×Hx0+0.3×Sp is set. The comprehensive training index Zx is compared with the preset comprehensive training index Zx0, and 0.61≤Zx0≤0.74 is set. The compliance of the comprehensive training index is judged according to the comparison result, and the preset number of flights Cf0 and the preset number of iterations Dd0 are adjusted according to the judgment result.

[0117] When Zx≥Zx0, the personalized matching training program module determines that the comprehensive training index is up to standard, and adjusts the preset number of flights Cf0 according to the first training adjustment coefficient. Adjust the preset flight times Cf0 and set =1.77-0.23×e -0.38×（Zx-Zx0） , where e is the base of the natural logarithm, and the preset number of flights after training adjustment is Cf01, set Cf01=Cf0× , replacing the preset number of flights Cf0 with the preset number of flights Cf01 after training adjustment, and re-comparing the number of flights Cf with the preset number of flights Cf01 after training adjustment;

[0118] When Zx<Zx0, the personalized matching training program module determines that the comprehensive training index is not up to standard, and adjusts the preset number of iterations Dd0 according to the second training adjustment coefficient. Adjust the preset number of iterations Dd0 for training, set =1.27-0.14×e -0.7×（Zx0-Zx） , where e is the base of the natural logarithm, and the preset number of iterations after training adjustment is obtained, Dd0`, set Dd0`=Dd0× , where Dd0` is a natural positive integer, and the integer value is rounded off to one decimal place. The preset number of iterations Dd0 is replaced by the preset number of iterations after training adjustment Dd0`, and the number of iterations Dd is re-compared with the preset number of iterations after training adjustment Dd0`.

[0119] Specifically, the performance appraisal score refers to a numerical value reflecting the comprehensive performance of the pilot in performing a training flight mission. This embodiment does not limit the specific method for obtaining the performance appraisal score. Relevant technical personnel in this field can freely choose according to actual needs, such as through team scoring. The physiological indicator evaluation value refers to a numerical value reflecting the comprehensive changes in physiological data when the pilot performs a training flight mission. This embodiment does not limit the specific method for obtaining the physiological indicator evaluation value. Relevant technical personnel in this field can freely choose according to actual needs, such as software calculation. The preset comprehensive training index refers to a preset value for judging the compliance of the comprehensive training index. The compliance of the comprehensive training index refers to the degree of compliance of the comprehensive training index judged based on the comprehensive training index and the preset comprehensive training index. The compliance of the comprehensive training index includes the compliance of the comprehensive training index being up to standard and the compliance of the comprehensive training index being not up to standard.

[0120] Specifically, the personalized matching training program module judges whether the comprehensive training index meets the standard. When the comprehensive training index meets the standard, the preset number of flights that affect the psychological state is increased as the comprehensive training index increases, so as to strengthen the training intensity of the pilot and accurately match the training program. When the comprehensive training index does not meet the standard, the preset number of iterations is increased as the comprehensive training index decreases, so as to increase the accuracy of pilot type classification and avoid the impact of inaccurate type classification on the training program, thereby improving training efficiency.

[0121] Specifically, when the personalized matching training program module performs training updates on the training adjustment process according to the number of training times, the number of training times Cx is compared with the preset number of training times Cx0, and 9 times ≤ Cx0 ≤ 12 times is set. The compliance of the number of training times is judged based on the comparison result, and the preset comprehensive training index Zx0 is updated based on the judgment result, wherein:

[0122] When Cx≤Cx0, the personalized matching training program module determines that the number of training times does not meet the standard, and does not perform training updates on the preset comprehensive training index Zx0;

[0123] When Cx>Cx0, the personalized matching training program module determines that the number of training times has reached the standard, and updates the preset comprehensive training index Zx0 according to the training update coefficient £, setting £=1.38-0.21×e -0.7×（Cx-Cx0）, where e is the base of the natural logarithm, and the preset comprehensive training index Zx01 after training update is obtained, and Zx01=Zx0×£ is set. The preset comprehensive training index Zx0 is replaced with the preset comprehensive training index Zx01 after training update, and the comprehensive training index Zx is re-compared with the preset comprehensive training index Zx01 after training update.

[0124] Specifically, the number of training times refers to the number of times a pilot completes a training plan. This embodiment does not limit the specific method for obtaining the number of training times. Relevant technicians in this field can freely choose according to actual needs, such as software records. The preset number of training times refers to a preset value for judging whether the number of training times meets the standard. The compliance status of the number of training times refers to the degree of compliance of the number of training times judged based on the number of training times and the preset number of training times. The compliance status of the number of training times includes the compliance status of the number of training times being unsatisfactory and the compliance status of the number of training times being satisfactory.

[0125] Specifically, the personalized matching training plan module determines whether the number of training times meets the standard. When the number of training times meets the standard, it increases the comprehensive training index to strengthen the training intensity of the pilots, avoids the situation where the number of training times is too many but the pilot training effect is not improved, and thus improves training efficiency.

[0126] Specifically, the team collaboration module matches pilots' roles according to the psychological state index and the comprehensive training index through a role matching method, and the role matching method includes:

[0127] Step C01: Obtain the pilot's characteristic vector P based on the psychological state index and the comprehensive training index, and set P = {P1, P2, P3, ..., Pf};

[0128] Step C02, set the role demand vector G = {G1, G2, G3, ..., Gv};

[0129] Step C03, calculate the role matching degree Mp according to the pilot feature vector P = {P1, P2, P3, ..., Pf} and the role requirement vector G = {G1, G2, G3, ..., Gv}, and obtain the role matching degree Mp, and set , where nu is the total number of features of the role requirement vector, and ny is the total number of features of the pilot feature vector;

[0130] Step C04: compare the role matching degree Mp with the preset role matching degree Mp0, set 0.68≤Mp0≤0.71, judge the matching status of the role matching degree according to the comparison result, and output the role matching result according to the judgment result, wherein:

[0131] When Mp < Mp0, the team collaboration module determines that the matching condition of the role matching degree is mismatched, and outputs the pilot and the role as mismatched as the role matching result.

[0132] When Mp ≥ Mp0, the team collaboration module determines that the matching condition of the role matching degree is matched, and outputs the pilot and the role as matched as the role matching result.

[0133] Specifically, the pilot feature vector P = {P1, P2, P3,..., Pf} refers to a feature set obtained by setting the characteristics of the pilot according to the psychological state index and the comprehensive training index, such as a psychological state of a nervous state and a comprehensive training index of 0.6, wherein P1 is the value of the first feature in the pilot feature vector, P2 is the value of the second feature in the pilot feature vector, P3 is the value of the third feature in the pilot feature vector, Pf is the value of the fth feature in the pilot feature vector, f is the order of the features in the pilot feature vector, the role requirement vector G = {G1, G2, G3,..., Gv} refers to a feature set required for role matching, and the present embodiment does not limit the role requirement vector, and those skilled in the related art can freely select according to actual requirements, such as a psychological state of a relaxed state and a comprehensive training index of 0.7, wherein G1 is the value of the first feature in the role requirement vector, G2 is the value of the second feature in the role requirement vector, G3 is the value of the third feature in the role requirement vector, Gv is the value of the vth feature in the role requirement vector, and v is the order of the features in the role requirement vector, the preset role matching degree refers to a preset value for judging the matching condition of the role matching degree, and the matching condition of the role matching degree refers to the matching degree of the pilot and the role, and the matching condition of the role matching degree includes the matching condition of the role matching degree being mismatched and the matching condition of the role matching degree being matched.

[0134] Specifically, the team collaboration module obtains the role matching result by judging the matching condition of the role matching degree, and preliminarily allocates the role of the pilot in the team according to the individual differences of the pilot, so as to improve the completion degree of the team to the task when the team cooperates, thereby improving the task execution efficiency.

[0135] Specifically, when the team collaboration module adjusts the role matching method according to the team state value, the team state index is obtained according to the team state index obtaining method, and the team state index obtaining method includes:

[0136] Step K01: Calculate the pilot's current pressure value Ps based on heart rate HR, resting heart rate HRrest, maximum heart rate HRmax, galvanic skin response GSR, resting galvanic skin response GSRrest, maximum galvanic skin response GSRmax, and EEG pressure value EEGs to obtain the pilot's current pressure value Ps and set ;

[0137] Step K02: Calculate the pressure average value Pt based on the pressure mean value Pp and the pressure standard deviation σp to obtain the pressure average value Pt, and set Pt=Pp+1.23×σp;

[0138] Step K03, calculate the team status index S according to the pilot's current pressure value Ps and the pressure average value Pt, and obtain the team status index S, set .

[0139] Specifically, the heart rate refers to the number of times the pilot's heart beats per minute when performing a team flight mission, the resting heart rate refers to the pilot's heart rate in a relaxed state, and the maximum heart rate refers to the highest number of beats per minute that the pilot's heart can reach when in a tense state. This embodiment does not limit the specific method of obtaining the heart rate, resting heart rate, and maximum heart rate. Relevant technical personnel in this field can freely choose according to actual needs, such as obtaining the heart rate, resting heart rate, and maximum heart rate through a dynamic electrocardiogram. The skin electrical response refers to a value that reflects the degree of change in the surface conductivity of the pilot's skin when it is stimulated by external stimuli. The resting skin electrical response refers to the skin electrical response of the pilot in a relaxed state, and the maximum skin The electrical response refers to the highest value of the pilot's skin electrical response when the pilot is in a tense state. This embodiment does not limit the specific method of obtaining the skin electrical response, resting skin electrical response, and maximum skin electrical response. Relevant technicians in this field can freely choose according to actual needs, such as obtaining it through a skin electrical response tester. The brain wave pressure value refers to a value reflecting the changes in neuronal electrical activity caused by cerebral cortex pressure on the pilot. This embodiment does not limit the brain wave pressure value. Relevant technicians in this field can freely choose according to actual needs, such as obtaining it through scalp electrodes. The pressure mean refers to the mean of the pilot's brain wave pressure values. For example, setting the brain wave pressure value EEGs={E1, E2, E3,..., E ik}, then the mean pressure , where N is the total number of brainwave pressure values, ik is the order of the brainwave pressure values, E1 is the first value in the brainwave pressure value, E2 is the second value in the brainwave pressure value, E3 is the third value in the brainwave pressure value, and E ik is the ikth value in the brainwave pressure value, and the pressure standard deviation σp refers to the standard deviation of the pilot's brainwave pressure value. .

[0140] Specifically, the team collaboration module obtains the team status index and presents the completion status of the team's tasks in a digital form, so as to reasonably arrange the roles in the team and thus improve the training effect.

[0141] Specifically, when the team collaboration module adjusts the role matching method according to the team status value, the team status index S is compared with the preset team status index S0, and 0.82≤S0≤0.94 is set. The team status is judged according to the comparison result, and the preset role matching degree Mp0 is adjusted according to the judgment result, where:

[0142] When S≥S0, the team collaboration module determines that the team status is up to standard and does not adjust the team to the preset role matching degree Mp0;

[0143] When S<S0, the team collaboration module determines that the team status is not up to standard, and adjusts the preset role matching degree Mp0 according to the team adjustment index eko, setting eko=1.77-0.4×e -0.7 ×（S0-S） , where e is the base of the natural logarithm, and the preset role matching degree Mp0` after team adjustment is obtained. Mp0`=Mp0×eko is set, and the preset role matching degree Mp0 is replaced by the preset role matching degree Mp0` after team adjustment, and the role matching degree Mp is re-compared with the preset role matching degree Mp0` after team adjustment.

[0144] Specifically, the preset team status index refers to a preset value for judging whether the team status meets the standard. The compliance of the team status refers to the degree of compliance of the team cooperation judged based on the team status index and the preset team status index. The compliance of the team status includes the team status meeting the standard and the team status meeting the standard.

[0145] Specifically, the team collaboration module judges whether the team status meets the standards. When the team status does not meet the standards, the preset role matching degree increases as the team status index decreases, so as to reduce the impact of insufficient pilot role matching on the team status, thereby improving training efficiency.

[0146] Specifically, when the team collaboration module performs team updates on the team adjustment process according to changes in team members, when there are no team member changes, the team adjustment process is not updated;

[0147] When there are changes in team members, the team adjustment process is updated according to the team update coefficient. Update the team status index S and set it to 0.71≤ ≤0.77, get the team status index S` after team update, set S`=S× , compare the updated team status index S` with the preset team status index S0, and re-judge whether the team status meets the standards.

[0148] Specifically, the team collaboration module updates the team status index according to changes in team members. When there are changes in team members, the team status index is reduced to avoid the impact of changes in team members on team collaboration, thereby improving the efficiency of team training.

[0149] Thus far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art may make equivalent changes or substitutions to the relevant technical features, and the technical solutions after such changes or substitutions will fall within the scope of protection of the present invention.

Claims

1. An intelligent system for pilot psychological assessment and personalized training, characterized by: The system comprises: Psychological data collection module, used to collect training data; The psychological data processing module is used to extract features from the training data to obtain training data features, and is also used to perform data fusion on the training data features to obtain training data fusion vectors; A psychological assessment module is used to obtain a psychological state index based on the training data fusion vector, and to obtain the psychological state based on the psychological state index, and to perform task adjustment on the psychological state, and to perform experience adjustment on the task adjustment process; A personalized training program matching module is used to classify pilot types according to their psychological state, obtain pilot type classification results, and obtain training programs based on the pilot type classification results. It is also used to adjust the psychological state and training program based on the comprehensive training index, and to update the training adjustment process; A team collaboration module is used to match pilots to roles based on the psychological state index and comprehensive training index through a role matching method, to adjust the role matching method as a team, and to update the team during the team adjustment process; The psychological data processing module collects training data; When the psychological data processing module extracts features from the training data, the training data is input into the data feature extraction model to obtain training data features, wherein the training data features include physiological data features hrv, behavioral data features hry and environmental data features hru; When the psychological data processing module performs data fusion on the feature extraction results, it calculates the training data fusion vector Fto based on the physiological data feature hrv, the behavioral data feature hry, the environmental data feature hru, the pilot communication ability index Gton, the physiological data feature weight w1, the behavioral data feature weight w2, the environmental data feature weight w3 and the pilot communication ability index weight wt to obtain the training data fusion vector Fto, and sets Fto=hrv×w1+hry×w2+hru×w3+Gton×wt.

2. The intelligent system for pilot psychological assessment and personalized training according to claim 1, characterized in that: When the psychological assessment module obtains the psychological state index according to the training data fusion vector, the training data fusion vector is input into the psychological state prediction model to obtain the psychological state index; When the psychological assessment module obtains the psychological state according to the psychological state index, it compares the psychological state index Hx with the first preset psychological state index Hx1 and the second preset psychological state index Hx2, sets Hx1=35 and Hx2=65, judges the state of the psychological state index according to the comparison result, and outputs the psychological state according to the judgment result, wherein: When Hx<Hx1, the psychological assessment module determines that the psychological state index is low and outputs the relaxation state as the psychological state; When Hx1≤Hx<Hx2, the psychological assessment module determines that the state of the psychological state index is moderate and outputs the normal state as the psychological state; When Hx≥Hx2, the psychological assessment module determines that the psychological state index is high and outputs the tension state as the psychological state; When the psychological assessment module adjusts the psychological state according to the complexity of the flight mission, it compares the flight mission complexity Ff with the preset flight mission complexity Ff0, sets 0.53≤Ff0≤0.69, judges the complexity of the flight mission based on the comparison result, and adjusts the first preset psychological state index Hx1 and the second preset psychological state index Hx2 based on the judgment result, wherein: When Ff≤Ff0, the psychological assessment module determines that the complexity of the flight mission is not complex, and does not adjust the first preset psychological state index Hx1 and the second preset psychological state index Hx2; When Ff>Ff0, the psychological assessment module determines that the complexity of the flight mission is complex, and adjusts the first preset psychological state index Hx1 and the second preset psychological state index Hx2 according to the complexity coefficient α, setting α=0.38+0.3×e -0.7 ×（Ff-Ff0） , obtain the adjusted first preset mental state index Hx1` and the adjusted second preset mental state index Hx2`, set Hx1`=Hx1×α, Hx2`=Hx2×α, replace the first preset mental state index Hx1 with the adjusted first preset mental state index Hx1`, replace the second preset mental state index Hx2 with the adjusted second preset mental state index Hx2`, and re-compare the mental state index Hx with the adjusted first preset mental state index Hx1` and the adjusted second preset mental state index Hx2`.

3. The intelligent system for pilot psychological assessment and personalized training according to claim 2, characterized in that: When the psychological assessment module makes empirical adjustments to the task adjustment process based on the number of flights, the number of flights Cf is compared with the preset number of flights Cf0, and 15 times ≤ Cf0 ≤ 23 times is set. The number of flights is judged based on the comparison result, and the preset flight task complexity Ff0 is empirically adjusted based on the judgment result, wherein: When Cf ≥ Cf0, the psychological assessment module determines that the number of flights is sufficient and does not perform empirical adjustments to the preset flight mission complexity Ff0; When Cf<Cf0, the psychological assessment module determines that the number of flights is insufficient, and empirically adjusts the preset flight mission complexity Ff0 according to the empirical adjustment coefficient γ, setting γ=0.95-0.4×e -0.7 ×（Cf / Cf0） , where e is the base of the natural logarithm. The adjusted preset flight mission complexity Ff0` is obtained, and Ff0`=Ff0×γ is set. The preset flight mission complexity Ff0 is replaced with the adjusted preset flight mission complexity Ff0`, and the flight mission complexity Ff is recompared with the adjusted preset flight mission complexity Ff0`.

4. The intelligent system for pilot psychological assessment and personalized training according to claim 3, characterized in that: The personalized matching training program module classifies pilot types according to psychological states using a pilot type classification method, wherein the pilot type classification method includes: Step B01, classifying the pilots according to their psychological states to obtain classification results, wherein the classification results include high stress sensitivity and emotional stability; Step B02: Initially select the cluster centers to obtain the initial cluster centers cj`={cj1,cj2,cj3,...,cjh}; Step B03, calculating the target distance set d = {d1, d2, d3, ..., dn} based on the initial cluster centers cj` = {cj1, cj2, cj3, ..., cjh} and the training data features xi = {xi1, xi2, xi3, ..., xin}; Step B04: assign each data point in the training data feature to the category classification result that is closest to the data point, to obtain an updated category classification result; Step B05, clustering and updating the initial cluster centers to obtain updated cluster centers; Step B06: Replace the initial cluster center with the updated cluster center, and repeat steps B01 to B06 according to the number of iterations Dd, setting Dd=1; Step B07: Compare the number of iterations Dd with the preset number of iterations Dd0, setting 3 ≤ Dd0 ≤ 5. Based on the comparison result, determine whether the number of iterations meets the standard, and output the pilot type classification result based on the determination result, where: When Dd<Dd0, the personalized matching training program module determines that the number of iterations does not meet the standard, and continues to repeat steps B01 to B06 until Dd≥Dd0; When Dd≥Dd0, the personalized matching training program module determines that the number of iterations has met the standard and outputs the updated category division result as the pilot type classification result.

5. The intelligent system for pilot psychological assessment and personalized training according to claim 4, characterized in that: The personalized matching training program module obtains the training program according to the pilot type classification result, and outputs low-intensity flight as the training program when the pilot type classification result is high pressure sensitivity; When the pilot type classification result is emotionally stable, high-intensity flight is output as the training plan; When the personalized matching training program module adjusts the psychological state and training program according to the comprehensive training index, the comprehensive training index Zx is calculated according to the performance assessment score Jk, the psychological state index Hx, and the physiological index evaluation value Sp to obtain the comprehensive training index Zx, and Zx=0.3×Jk+0.4×Hx+0.3×Sp is set. The comprehensive training index Zx is compared with the preset comprehensive training index Zx0, and 0.61≤Zx0≤0.74 is set. The compliance of the comprehensive training index is judged based on the comparison result, and the preset number of flights Cf0 and the preset number of iterations Dd0 are adjusted according to the judgment result. When Zx≥Zx0, the personalized matching training program module determines that the comprehensive training index is up to standard, and adjusts the preset number of flights Cf0 according to the first training adjustment coefficient. Adjust the preset flight times Cf0 and set =1.77-0.23×e -0.38×（Zx-Zx0） , where e is the base of the natural logarithm, and the preset number of flights after training adjustment is Cf01, set Cf01=Cf0× , replacing the preset number of flights Cf0 with the preset number of flights Cf01 after training adjustment, and re-comparing the number of flights Cf with the preset number of flights Cf01 after training adjustment; When Zx<Zx0, the personalized matching training program module determines that the comprehensive training index is not up to standard, and adjusts the preset number of iterations Dd0 according to the second training adjustment coefficient. Adjust the preset number of iterations Dd0 for training, set =1.27-0.14×e -0.7×（Zx0-Zx） , where e is the base of the natural logarithm, and the preset number of iterations after training adjustment is obtained, Dd0`, set Dd0`=Dd0× , replace the preset number of iterations Dd0 with the preset number of iterations Dd0' after training adjustment, and re-compare the number of iterations Dd with the preset number of iterations Dd0' after training adjustment.

6. The intelligent system for pilot psychological assessment and personalized training according to claim 5, characterized in that: When the personalized matching training program module performs training updates on the training adjustment process according to the number of training times, the number of training times Cx is compared with the preset number of training times Cx0, and 9 times ≤ Cx0 ≤ 12 times is set. The compliance of the number of training times is judged based on the comparison result, and the preset comprehensive training index Zx0 is updated based on the judgment result, wherein: When Cx≤Cx0, the personalized matching training program module determines that the number of training times does not meet the standard, and does not perform training updates on the preset comprehensive training index Zx0; When Cx>Cx0, the personalized matching training program module determines that the number of training times has reached the standard, and updates the preset comprehensive training index Zx0 according to the training update coefficient £, setting £=1.38-0.21×e -0.7×（Cx-Cx0） , where e is the base of the natural logarithm, and the preset comprehensive training index Zx01 after training update is obtained, and Zx01=Zx0×£ is set. The preset comprehensive training index Zx0 is replaced with the preset comprehensive training index Zx01 after training update, and the comprehensive training index Zx is re-compared with the preset comprehensive training index Zx01 after training update.

7. The intelligent system for pilot psychological assessment and personalized training according to claim 6, characterized in that: The team collaboration module performs role matching for pilots according to the psychological state index and the comprehensive training index through a role matching method, wherein the role matching method includes: Step C01: Obtain the pilot's characteristic vector P based on the psychological state index and the comprehensive training index, and set P = {P1, P2, P3, ..., Pf}; Step C02, set the role demand vector G = {G1, G2, G3, ..., Gv}; Step C03, according to the pilot feature vector P = {P1, P2, P3, ..., Pf} and the role requirement vector G = {G1, G2, G3, ..., Gv} role matching Mp is calculated to obtain the role matching Mp; Step C04: compare the role matching degree Mp with the preset role matching degree Mp0, set 0.68≤Mp0≤0.71, judge the matching status of the role matching degree according to the comparison result, and output the role matching result according to the judgment result, wherein: When Mp<Mp0, the team collaboration module determines that the role matching degree is mismatched, and outputs the mismatch between the pilot and the role as a role matching result; When Mp≥Mp0, the team collaboration module determines that the role matching degree is a match, and outputs the matching between the pilot and the role as a role matching result.

8. The intelligent system for pilot psychological assessment and personalized training according to claim 7, characterized in that: When the team collaboration module performs team adjustment on the role matching method according to the team status value, the team status index is obtained according to the team status index obtaining method, and the team status index obtaining method includes: Step K01, calculating the pilot's current stress value Ps based on the heart rate HR, resting heart rate HRrest, maximum heart rate HRmax, galvanic skin response GSR, resting galvanic skin response GSRrest, maximum galvanic skin response GSRmax, and EEG pressure value EEGs to obtain the pilot's current stress value Ps; Step K02: Calculate the pressure average value Pt based on the pressure mean value Pp and the pressure standard deviation σp to obtain the pressure average value Pt, and set Pt=Pp+1.23×σp; Step K03: Calculate the team status index S based on the pilot's current pressure value Ps and the average pressure value Pt to obtain the team status index S.

9. The intelligent system for pilot psychological assessment and personalized training according to claim 8, characterized in that: When the team collaboration module adjusts the role matching method according to the team status value, the team status index S is compared with the preset team status index S0, and 0.82≤S0≤0.94 is set. The team status is judged according to the comparison result, and the preset role matching degree Mp0 is adjusted according to the judgment result, wherein: When S≥S0, the team collaboration module determines that the team status is up to standard and does not adjust the team to the preset role matching degree Mp0; When S<S0, the team collaboration module determines that the team status is not up to standard, and adjusts the preset role matching degree Mp0 according to the team adjustment index eko, setting eko=1.77-0.4×e -0.7 ×（S0-S） , where e is the base of the natural logarithm, and the preset role matching degree after team adjustment Mp0` is obtained, Mp0`=Mp0×eko is set, the preset role matching degree Mp0 is replaced with the preset role matching degree after team adjustment Mp0`, and the role matching degree Mp is re-compared with the preset role matching degree after team adjustment Mp0`; The team collaboration module updates the team adjustment process according to the changes in team members. If there are no changes in team members, the team adjustment process will not be updated. When there are changes in team members, the team adjustment process is updated according to the team update coefficient. Update the team status index S and set it to 0.71≤ ≤0.77, get the team status index S` after team update, set S`=S× , compare the updated team status index S` with the preset team status index S0, and re-judge whether the team status meets the standards.

Images