Method for evaluating competency in course control training process of pilot

By collecting and integrating multiple data of pilots, building standard feature models and evaluating their route maneuverability, the problem of low accuracy of existing evaluation methods is solved, comprehensive and accurate assessment of pilot capabilities is achieved, and flight safety and professionalism are improved. level.

CN119005820AInactive Publication Date: 2024-11-22CIVIL AVIATION FLIGHT UNIV OF CHINA

Patent Information

Application Number
CN202411496122.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-10-25
Publication Date
2024-11-22
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The existing pilot route maneuver competency assessment methods have poor accuracy and cannot fully reflect the pilot's comprehensive capabilities, which affects the effectiveness of the assessment.

Method used

By collecting pilot's flight data, human posture and status information, extracting features related to route manipulation capabilities, building standard feature models, conducting real-time evaluation, combining deep learning and machine learning methods for multi-source data fusion, and evaluating pilot's route manipulation capabilities .

Benefits of technology

It has achieved a comprehensive, accurate and real-time assessment of pilot route maneuverability, improved flight safety and professionalism, provided pilots with a direction for self-improvement, and helped airlines develop personalized training plans.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119005820A_ABST
    Figure CN119005820A_ABST
Patent Text Reader

Abstract

The invention relates to the field of evaluation methods, and discloses a competency evaluation method in a pilot route control training process, which comprises the following steps of: 1, firstly collecting relevant data of a pilot, namely, firstly collecting flight data of the pilot in the route control process; 2, based on the collected flight data, feature extraction is carried out, and features related to the flight route control ability are extracted; 3, constructing a standard feature model by using the extracted features related to the air route control capability; 4, evaluating the flight line control ability of the pilot through the obtained standard feature model, and obtaining real-time evaluation information of the pilot; and 5, evaluating the human body posture to obtain human body posture evaluation information. According to the method, comprehensive, accurate and real-time evaluation can be realized, an airline company can be helped to better understand the ability level of a pilot, and the flight safety level is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the field of evaluation methods, and in particular to a competency evaluation method in a pilot route operation training process. Background Art

[0002] Pilots' flight control competence is a comprehensive evaluation indicator, covering technical operation ability, decision-making ability, emergency handling ability, teamwork and communication ability, etc. In order to improve pilots' flight control competence, comprehensive training and evaluation are needed to ensure that they have the ability to cope with various challenges during flight. In the process of evaluating the pilot's line control competency, the evaluation method will be used to comprehensively evaluate the pilot's line control competency.

[0003] The existing assessment methods have poor assessment accuracy and cannot fully reflect the pilots' route control competence, which has a certain impact on the use of the assessment methods. Therefore, a competence assessment method for pilot route control training is proposed. Summary of the invention

[0004] The technical problem to be solved by the present invention is: how to solve the problem that the existing evaluation method has poor evaluation accuracy and cannot fully reflect the pilot's route control competence, which has brought certain impacts on the use of the evaluation method, and provides a competence evaluation method in the pilot route control training process.

[0005] The present invention solves the above technical problems through the following technical solutions, and the present invention comprises the following steps: Step 1: First collect the pilot's relevant data, that is, first collect the pilot's flight data during the route control process; Step 2: Based on the collected flight data, feature extraction is performed to extract features related to route control capabilities; Step 3: Use the extracted features related to route control capabilities to build a standard feature model; Step 4: Evaluate the pilot's route control ability through the acquired standard feature model and obtain the pilot's real-time evaluation information; Step 5: Evaluate the human body posture and obtain the human body posture evaluation information; Step 6: During the capability assessment process, the pilot's human body related information during the flight is also processed to obtain human body status assessment information.

[0006] The sources of flight data collected by pilots during route control include: The data and voice information during the flight that are automatically recorded by the data recording system and cockpit voice recorder equipped by the aircraft, namely, automatically recorded data; Flight logs filled out by pilots, which record events and feelings during the flight, and flight instructors and evaluators record information about the pilot's performance, i.e., flight log data; Survey and interview data are collected through relevant surveys and interviews.

[0007] Furthermore, after the automatic recording data, the flight log data and the survey interview data are collected, data preprocessing and data integration are performed on the automatic recording data, the flight log data and the survey interview data; Data preprocessing includes data cleaning, processing missing values, correcting erroneous data, deleting invalid data and standardizing data formats; Data integration includes determining data sources, data mapping, data conversion, data merging and data verification.

[0008] Furthermore, the specific process of constructing the standard feature model is as follows: The extracted features related to route control ability are classified and integrated according to different dimensions of route control ability, including flying skills, decision-making ability, teamwork ability and psychological quality; Standardize the extracted features related to the route control capability and convert them into standardized features; Then, according to the evaluation objectives and expert opinions, a corresponding weight is assigned to each standardized feature; Based on the above process, a standard feature model including standardized features and corresponding weights is constructed through a model building method; Finally, the model is verified by using historical data or simulation data to verify the constructed standard feature model. After verification, the standard feature model is exported for evaluation.

[0009] Furthermore, when there are problems with model verification, the modeling method is changed and re-modeling is performed, and the final standard feature model is extracted and re-verified. If there are no problems with the verification this time, the final standard feature model is exported for evaluation. If problems still exist in this verification, the data required for modeling will be collected again.

[0010] Furthermore, the process of obtaining the pilot's real-time evaluation information is as follows: Extracting automatic recording data, flight log data and survey interview data, i.e. real-time pilot data; Perform data cleaning on the extracted real-time pilot data, including data cleaning and removal of duplicate, erroneous or invalid data; At the same time, the real-time pilot data is preprocessed to obtain the preprocessed real-time pilot data; Then, the pre-processed real-time pilot data is subjected to fusion modeling to obtain a real-time flight data model; Then compare and analyze the real-time flight data model with the already constructed standard feature model to obtain the pilot's real-time evaluation information; During the comparison and analysis process between the real-time flight data model and the constructed standard feature model, when the similarity between the real-time flight data model and the constructed standard feature model is greater than the preset value v1, the pilot's real-time evaluation information is evaluated as excellent; When the feature model similarity between the real-time flight data model and the constructed standard feature model is between the preset values ​​v1 and v2, the pilot's real-time evaluation information is considered normal; When the feature model similarity between the real-time flight data model and the constructed standard feature model is less than a preset value v2, the pilot's real-time evaluation information is poorly evaluated.

[0011] Furthermore, the specific process of fusion modeling of the pre-processed real-time pilot data is as follows: First, feature extraction is performed. Feature extraction methods include: Extract corresponding features according to actual needs and obtain pilot data features; Use deep learning models to automatically learn data and obtain pilot data features; Then, multi-source data fusion is performed. The specific process of multi-source data fusion is as follows: Feature-level fusion, after feature extraction, combines features from different data sources; Model-level fusion: training multiple models, each using a different data source, and fusing the outputs of multiple models; Finally, use machine learning or deep learning methods to train the model: Model selection: select the corresponding machine learning or deep learning model according to the task and data characteristics; The training process uses the preprocessed and feature extracted dataset to train the model.

[0012] Furthermore, the human body posture assessment information includes normal human body posture and abnormal human body posture; The specific process of obtaining human posture assessment information is as follows: During the simulated flight or actual flight, an image acquisition device is set up to collect the real-time image information of the pilot, process the real-time image information of the pilot, and obtain the attitude evaluation parameters; When the posture evaluation parameters are abnormal, human posture abnormality is generated; When there is no abnormality in the posture evaluation parameters, the generated human posture is normal.

[0013] Furthermore, the process of acquiring the posture evaluation parameters and determining whether the posture evaluation parameters are abnormal is as follows: Extract the real-time image information of the pilot, extract the facial image of the pilot when he / she enters the flight preparation posture from the real-time image information of the pilot, extract the nose tip point and two outer eye corner points of the pilot at this time, and mark them as point a, point b1 and point b2; Connect the two outer points b1 and b2 to obtain the line segment L1, and then make a perpendicular line segment L2 perpendicular to the line segment L1 with point a as the endpoint; Then, L2 is extended to both ends, one end is extended to the top of the cockpit, and the other end is extended to the bottom of the cockpit to obtain the evaluation line P; After starting the simulated flight or the actual flight, continue to collect facial images, and collect the distance between the vertical line segment L2 in the collected facial image and the evaluation line P every preset time period to obtain multiple evaluation distances; After a single flight is completed, the average of multiple evaluation distances is calculated, and the attitude evaluation parameters are obtained; When the posture evaluation parameter is greater than the preset value, it means that there is an abnormality.

[0014] Furthermore, the human body status assessment information includes normal human body assessment and abnormal human body assessment, and the specific processing process of the human body status assessment information is as follows: collecting pilot human body information during the flight, the pilot human body information includes human blood pressure information, heart rate information, respiratory rate information and voice information during the flight; During the simulated flight or the actual flight, the human body's blood pressure information, heart rate information and respiratory rate information are collected every preset time until the simulated flight or the actual flight ends; The amount of collected human blood pressure information, heart rate information and respiratory rate information is marked as Qi; Then, the blood pressure information of Qi people is converted into a line graph according to the time collection sequence to obtain a blood pressure assessment graph; Convert the Qi heart rate information into a line graph according to the time collection sequence to obtain the heart rate evaluation graph; The Qi respiratory rate information is converted into a line graph according to the time collection sequence to obtain a respiratory rate evaluation graph; Then the voice information is processed to obtain the pilot's speech speed information; When any one of the blood pressure assessment graph, heart rate assessment graph, respiratory rate assessment graph and pilot speech rate information is abnormal, a human body assessment abnormality is generated; When there is no abnormality in the blood pressure assessment graph, heart rate assessment graph, respiratory rate assessment graph and pilot speech rate information, the human body assessment is normal; The process of determining the abnormality of the blood pressure evaluation graph is as follows: analyzing the fluctuation trend of the blood pressure evaluation graph, when the fluctuation trend is greater than a preset amplitude, it indicates that there is an abnormality, and setting a standard blood pressure line in the blood pressure evaluation graph, when the duration of exceeding the standard blood pressure line is greater than a preset duration, it indicates that there is an abnormality; The process of determining the abnormality of the heart rate evaluation graph is as follows: analyzing the fluctuation trend of the heart rate evaluation graph, when the fluctuation trend is greater than a preset amplitude, it indicates that there is an abnormality, and setting a standard heart rate line in the heart rate evaluation graph, when the duration of exceeding the standard heart rate line is greater than a preset value, it indicates that there is an abnormality; The abnormality determination process of the respiratory rate evaluation graph is as follows: the fluctuation trend of the respiratory rate evaluation is analyzed. When the fluctuation trend is greater than a preset amplitude, it indicates that there is an abnormality. A respiratory rate standard line is set in the respiratory rate evaluation graph. When the duration of exceeding the respiratory rate standard line is greater than a preset duration, it indicates that there is an abnormality. The abnormality judgment process of the pilot speech speed information is as follows: when the pilot speech speed information exceeds a preset range, it indicates that there is an abnormality.

[0015] Compared with the prior art, the present invention has the following advantages: the competence assessment method in the pilot route control training process can ensure flight safety. By assessing the pilot's route control competence, it can ensure that the pilot has the skills and knowledge required to perform specific route tasks, thereby greatly improving the safety of flight. The assessment process includes multiple aspects such as theoretical knowledge testing, simulated flight tasks and performance in actual flights. It can comprehensively assess the various capabilities of the pilot, ensure that the pilot can make correct responses and decisions in emergency situations, and improve the pilot's professional level. The assessment results provide an objective evaluation of the pilot's ability level and provide the pilot with a direction for self-improvement and improvement. Airlines or training institutions can formulate personalized training plans for pilots based on the assessment results to help them improve their skills and knowledge in route control, thereby achieving a more comprehensive assessment of the pilot's route control competence. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a flow chart of the present invention. DETAILED DESCRIPTION

[0017] The following is a detailed description of an embodiment of the present invention. This embodiment is implemented on the premise of the technical solution of the present invention, and a detailed implementation method and a specific operation process are given, but the protection scope of the present invention is not limited to the following embodiment.

[0018] like Figure 1 As shown, this embodiment provides a technical solution: a method for evaluating the competency of pilots in the process of flight line operation training, comprising the following steps: Step 1: First collect the pilot's relevant data, that is, first collect the pilot's flight data during the route control process; Step 2: Based on the collected flight data, feature extraction is performed to extract features related to route control capabilities; Step 3: Use the extracted features related to route control capabilities to build a standard feature model; Step 4: Evaluate the pilot's route control ability through the acquired standard feature model and obtain the pilot's real-time evaluation information; Step 5: Evaluate the human body posture and obtain the human body posture evaluation information; Step 6: During the capability assessment process, the pilot's human body related information during the flight is also processed to obtain human body status assessment information; The above process can provide a comprehensive, accurate and real-time assessment, which can help airlines better understand the pilots' ability level and improve flight safety.

[0019] The data sources during the data collection process are as follows: Data recorded by the automatic recording system: data and voice information during the flight are automatically recorded by the flight data recording system and cockpit voice recorder equipped on the aircraft; Manually recorded data: flight logs filled out by pilots, which record events and feelings during the flight, and flight instructors and evaluators record information about pilots’ performance; Data obtained from questionnaire survey: pilot survey and interview data collected through questionnaire survey and interview; After collecting the data recorded by the automatic recording system, the data recorded manually and the data obtained by the questionnaire survey, it is necessary to perform data preprocessing and data integration on the data recorded by the automatic recording system, the data recorded manually and the data obtained by the questionnaire survey; Data preprocessing includes data cleaning, processing missing values, correcting erroneous data, deleting invalid data and standardizing data formats; Data integration includes determining data sources, data mapping, data conversion, data merging and data verification.

[0020] The specific process of feature model construction is as follows: Feature classification and integration: classify the extracted features and integrate them according to different dimensions of route control ability, including flight skills, decision-making ability, teamwork ability and psychological quality; Among them, flying technology includes flight time, flying skills, and flight performance characteristics.

[0021] decision-making abilities, including emergency response and decision-making characteristics; Teamwork skills, including the ability to collaborate and communicate with other crew members, participation in team decision-making, and contribution characteristics; Psychological qualities, including reactions to stress tests and emotional stability characteristics; Feature standardization, standardize the collected features. Common standardization methods include minimum-maximum normalization, Qi-score normalization, etc.

[0022] Feature weight allocation, assigning corresponding weights to each feature based on the evaluation objectives and expert opinions; Feature model construction, based on the above process, a standard feature model including standardized features and corresponding weights is constructed through a model construction method; Model verification and optimization: Use historical data or simulation data to verify the constructed standard feature model. After verification, export the standard feature model for evaluation to check its accuracy and effectiveness. If problems or deviations are found in the model, the model needs to be optimized and adjusted to improve its evaluation accuracy and reliability.

[0023] Based on the above process, a standard feature model including standardized features and corresponding weights is constructed through a model building method; When there are problems with model verification, the modeling method is changed and the modeling is re-performed, and the final standard feature model is extracted and re-verified. If there are no problems with the verification this time, the final standard feature model is exported for evaluation.

[0024] The constructed feature model is applied to the actual route control capability assessment, and the model is updated and improved based on the assessment results and feedback. With the continuous accumulation of data and changes in assessment requirements, the feature model also needs to be continuously updated and optimized.

[0025] The process of obtaining real-time assessment information by the pilot is as follows: Data collection: multi-source data collection, collecting data related to pilot route operations from different data sources, which is real-time pilot data. Real-time pilot data includes flight records, simulation training data, behavioral event interview records, and questionnaire survey results; Data cleaning and preprocessing: cleaning the collected real-time pilot data and removing duplicate, erroneous or invalid data; At the same time, pre-process the real-time pilot data, such as data standardization and normalization, to facilitate subsequent data analysis; Then, the pre-processed real-time pilot data is subjected to fusion modeling to obtain a real-time flight data model; Then compare and analyze the real-time flight data model with the already constructed standard feature model to obtain the pilot's real-time evaluation information; During the comparison and analysis process between the real-time flight data model and the constructed standard feature model, when the similarity between the real-time flight data model and the constructed standard feature model is greater than the preset value v1, the pilot's real-time evaluation information is evaluated as excellent; When the feature model similarity between the real-time flight data model and the constructed standard feature model is between the preset values ​​v1 and v2, the pilot's real-time evaluation information is considered normal; When the feature model similarity between the real-time flight data model and the constructed standard feature model is less than the preset value v2, the pilot's real-time evaluation information is poorly evaluated; v1>v2; The evaluation results will be fed back to pilots and relevant managers to help them understand the pilots' route control capabilities and provide a basis for subsequent selection, training and improvement. At the same time, the evaluation results can also be used to formulate targeted training and improvement plans to help pilots improve their route control capabilities.

[0026] The specific process of fusion modeling of pre-processed real-time pilot data is as follows: Feature extraction: extract corresponding features according to actual needs to obtain pilot data features, that is, use deep learning models such as convolutional neural network CNN, recurrent neural network RNN, Transformer, etc. to automatically learn data and obtain pilot data features; The specific process of multi-source data fusion is as follows: Feature-level fusion, after feature extraction, combines features from different sources; Model-level fusion: training multiple models, each using a different data source or feature set, and fusing the outputs of multiple models; The specific process of machine learning or deep learning model training is as follows: Model selection: select the corresponding machine learning or deep learning model according to the task and data characteristics; The training process uses the preprocessed and feature extracted dataset to train the model.

[0027] Through the above steps, a comprehensive, accurate and reliable assessment of the pilot's flight control competence can be conducted based on multi-source data fusion. This assessment method can make full use of information from different data sources to improve the accuracy and effectiveness of the assessment.

[0028] Human body posture assessment information includes normal human body posture and abnormal human body posture; The specific process of obtaining human posture assessment information is as follows: During the simulated flight or actual flight, an image acquisition device is set up to collect the real-time image information of the pilot, process the real-time image information of the pilot, and obtain the attitude evaluation parameters; When the posture evaluation parameters are abnormal, human posture abnormality is generated; When there is no abnormality in the posture evaluation parameters, the generated human posture is normal.

[0029] The acquisition of the posture evaluation parameters and the determination process of posture evaluation parameter abnormalities are as follows: Extract the real-time image information of the pilot, extract the facial image of the pilot when he / she enters the flight preparation posture from the real-time image information of the pilot, extract the nose tip point and two outer eye corner points of the pilot at this time, and mark them as point a, point b1 and point b2; Connect the two outer points b1 and b2 to obtain the line segment L1, and then make a perpendicular line segment L2 perpendicular to the line segment L1 with point a as the endpoint; Then, L2 is extended to both ends, one end is extended to the top of the cockpit, and the other end is extended to the bottom of the cockpit to obtain the evaluation line P; After starting the simulated flight or the actual flight, continue to collect facial images, and collect the distance between the vertical line segment L2 in the collected facial image and the evaluation line P every preset time period to obtain multiple evaluation distances; After a single flight is completed, the average of multiple evaluation distances is calculated, and the attitude evaluation parameters are obtained; When the posture evaluation parameter is greater than the preset value, it means that there is an abnormality; Through the above process, the facial shaking amplitude of pilots in simulated or actual flight was analyzed to evaluate the stability of vestibular function. The vestibular organ, namely the vestibular receptor, is an important part of the vestibular nervous system, responsible for receiving and processing information related to body movement and head position changes. Complex movements during flight (such as takeoff, landing, turning, etc.) will have a greater stimulation on the vestibular system.

[0030] Evaluation significance: By evaluating the pilot's facial shaking amplitude, the stability of his vestibular function can be indirectly reflected. A smaller shaking amplitude indicates that the pilot can maintain better body stability under vestibular stimulation, which helps to reduce flight illusions and dizziness and improve flight safety.

[0031] To improve the ability to resist vertigo, flight trainees usually need to receive systematic anti-vertigo training to improve their adaptability in complex flight environments. The assessment of facial shaking amplitude can be used as an intuitive indicator of the effectiveness of anti-vertigo training. Small shaking amplitude indicates significant training effects, and pilots have strong anti-vertigo abilities and can stay awake and stable during high-intensity flight missions.

[0032] It can also evaluate the stability of flight operations. During flight, the pilot needs to accurately control the aircraft's attitude and heading to ensure flight safety. The amplitude of facial shaking may indirectly affect the pilot's operating accuracy and stability. By evaluating the amplitude of facial shaking, the unstable factors in the pilot's flight operations can be discovered in time, and corresponding measures can be taken to correct and train, thereby improving the safety and stability of flight operations.

[0033] Flight adaptability is predicted. During the pilot selection process, the assessment of facial shaking amplitude can be used as an important reference indicator for predicting flight adaptability. For pilots with larger facial shaking amplitudes, targeted training and intervention can be carried out in advance to reduce their risks in future flight missions.

[0034] The assessment of the pilot's facial shaking amplitude during the flight assessment is of great significance in assessing the stability of vestibular function, anti-vertigo ability, flight operation stability, guiding personalized training, and predicting flight adaptability. This assessment helps improve the pilot's flying skills and safety, and provides a strong guarantee for the development of the aviation industry.

[0035] The human body status assessment information includes normal human body assessment and abnormal human body assessment. The specific processing process of the human body status assessment information is as follows: collecting pilot human body information during the flight, the pilot human body information includes human blood pressure information, heart rate information, respiratory rate information and voice information during the flight; During the simulated flight or the actual flight, the human body's blood pressure information, heart rate information and respiratory rate information are collected every preset time until the simulated flight or the actual flight ends; The amount of collected human blood pressure information, heart rate information and respiratory rate information is marked as Qi; Then, the blood pressure information of Qi people is converted into a line graph according to the time collection sequence to obtain a blood pressure assessment graph; Convert the Qi heart rate information into a line graph according to the time collection sequence to obtain the heart rate evaluation graph; The Qi respiratory rate information is converted into a line graph according to the time collection sequence to obtain a respiratory rate evaluation graph; Then the voice information is processed to obtain the pilot's speech speed information; When any one of the blood pressure assessment graph, heart rate assessment graph, respiratory rate assessment graph and pilot speech rate information is abnormal, a human body assessment abnormality is generated; When there is no abnormality in the blood pressure assessment graph, heart rate assessment graph, respiratory rate assessment graph and pilot speech rate information, the human body assessment is normal; The process of determining the abnormality of the blood pressure evaluation graph is as follows: analyzing the fluctuation trend of the blood pressure evaluation graph, when the fluctuation trend is greater than a preset amplitude, it indicates that there is an abnormality, and setting a standard blood pressure line in the blood pressure evaluation graph, when the duration of exceeding the standard blood pressure line is greater than a preset duration, it indicates that there is an abnormality; The process of determining the abnormality of the heart rate evaluation graph is as follows: analyzing the fluctuation trend of the heart rate evaluation graph, when the fluctuation trend is greater than a preset amplitude, it indicates that there is an abnormality, and setting a standard heart rate line in the heart rate evaluation graph, when the duration of exceeding the standard heart rate line is greater than a preset value, it indicates that there is an abnormality; The abnormality determination process of the respiratory rate evaluation graph is as follows: the fluctuation trend of the respiratory rate evaluation is analyzed. When the fluctuation trend is greater than a preset amplitude, it indicates that there is an abnormality. A respiratory rate standard line is set in the respiratory rate evaluation graph. When the duration of exceeding the respiratory rate standard line is greater than a preset duration, it indicates that there is an abnormality. The abnormality judgment process of the pilot speech rate information is as follows: when the pilot speech rate information exceeds the preset range, it indicates that there is an abnormality; Through the above process, while the overall method realizes the assessment of pilots, it also provides some auxiliary reference assessment content, thereby achieving a more comprehensive assessment of pilot route control competence.

[0036] To ensure flight safety, by assessing the pilot's route control competency, it can be ensured that the pilot has the skills and knowledge required to perform specific route tasks, thereby greatly improving flight safety.

[0037] The assessment process includes theoretical knowledge tests, simulated flight missions and actual flight performance, which can comprehensively evaluate the pilot's various capabilities and ensure that he can make correct responses and decisions in emergency situations.

[0038] Improve the professional level of pilots. The evaluation results provide an objective evaluation of the pilots' ability level and provide pilots with direction for self-improvement and improvement.

[0039] Airlines or training institutions can develop personalized training plans for pilots based on the assessment results to help them improve their skills and knowledge in route operations.

[0040] To promote the all-round development of pilots, the assessment of route control competency not only focuses on the pilot's flying skills, but also includes knowledge application, emergency knowledge, meteorological knowledge and other aspects, which helps promote the all-round development of pilots.

[0041] Through competency assessment, pilots can develop the required knowledge and skills more comprehensively, improving their ability to perform flight missions and flight safety.

[0042] Providing decision support for airlines, the evaluation results provide airlines with a reference for pilots' ability levels, helping airlines make more reasonable flight personnel deployment and flight mission arrangements.

[0043] At the same time, the evaluation results can also serve as an important basis for airlines to recruit, select and train pilots, thereby improving the overall operational efficiency and service quality of the airlines.

[0044] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In the description of the present invention, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0045] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0046] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present invention. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present invention.

Claims

1. Competence assessment method for pilots in line operation training, It is characterized in that The following steps are involved: Step 1: First collect the pilot's relevant data, that is, first collect the pilot's flight data during the route control process; Step 2: Based on the collected flight data, feature extraction is performed to extract features related to route control capabilities; Step 3: Use the extracted features related to route control capabilities to build a standard feature model; Step 4: Evaluate the pilot's route control ability through the acquired standard feature model and obtain the pilot's real-time evaluation information; Step 5: Evaluate the human body posture and obtain the human body posture evaluation information; Step 6: During the capability assessment process, the pilot's human body related information during the flight is also processed to obtain human body status assessment information.

2. The method for evaluating the competency of pilots in the course of flight line operation training according to claim 1, Features: The sources of flight data collected by pilots during route control include: The data and voice information during the flight that are automatically recorded by the data recording system and cockpit voice recorder equipped by the aircraft, namely, automatically recorded data; Flight logs filled out by pilots, which record events and feelings during the flight, and flight instructors and evaluators record information about the pilot's performance, i.e., flight log data; Survey and interview data are collected through relevant surveys and interviews.

3. The method for evaluating the competency of pilots in the course of flight line operation training according to claim 2, Features: After collecting the automatic recording data, flight log data and survey interview data, the automatic recording data, flight log data and survey interview data are preprocessed and integrated; Data preprocessing includes data cleaning, processing missing values, correcting erroneous data, deleting invalid data and standardizing data formats; Data integration includes determining data sources, data mapping, data conversion, data merging and data verification.

4. The method for evaluating the competency of pilots in the course of flight line operation training according to claim 1, Features: The specific process of constructing the standard feature model is as follows: The extracted features related to route control ability are classified and integrated according to different dimensions of route control ability, including flying skills, decision-making ability, teamwork ability and psychological quality; Standardize the extracted features related to the route control capability and convert them into standardized features; Then, according to the evaluation objectives and expert opinions, a corresponding weight is assigned to each standardized feature; Based on the above process, a standard feature model including standardized features and corresponding weights is constructed through a model building method; Finally, the model is verified by using historical data or simulation data to verify the constructed standard feature model. After verification, the standard feature model is exported for evaluation.

5. The method for evaluating the competency of pilots in the course of flight line operation training according to claim 1, Features: When there are problems with model verification, the modeling method is changed and re-modeling is performed, and the final standard feature model is extracted and re-verified. If there are no problems with the verification this time, the final standard feature model is exported for evaluation. If problems still exist in this verification, the data required for modeling will be collected again.

6. The method for evaluating the competency of pilots in the course of flight line operation training according to claim 1, Features: The process of obtaining the pilot's real-time evaluation information is as follows: Extracting automatic recording data, flight log data and survey interview data, i.e. real-time pilot data; Perform data cleaning on the extracted real-time pilot data, including data cleaning and removal of duplicate, erroneous or invalid data; At the same time, the real-time pilot data is preprocessed to obtain the preprocessed real-time pilot data; Then, the pre-processed real-time pilot data is subjected to fusion modeling to obtain a real-time flight data model; Then compare and analyze the real-time flight data model with the already constructed standard feature model to obtain the pilot's real-time evaluation information; During the comparison and analysis process between the real-time flight data model and the constructed standard feature model, when the similarity between the real-time flight data model and the constructed standard feature model is greater than the preset value v1, the pilot's real-time evaluation information is evaluated as excellent; When the feature model similarity between the real-time flight data model and the constructed standard feature model is between the preset values ​​v1 and v2, the pilot's real-time evaluation information is considered normal; When the feature model similarity between the real-time flight data model and the constructed standard feature model is less than a preset value v2, the pilot's real-time evaluation information is poorly evaluated.

7. The method for evaluating the competency of pilots in the course of flight line operation training according to claim 5, Features: The specific process of fusion modeling of pre-processed real-time pilot data is as follows: First, feature extraction is performed. Feature extraction methods include: Extract corresponding features according to actual needs and obtain pilot data features; Use deep learning models to automatically learn data and obtain pilot data features; Then, multi-source data fusion is performed. The specific process of multi-source data fusion is as follows: Feature-level fusion, after feature extraction, combines features from different data sources; Model-level fusion: training multiple models, each using a different data source, and fusing the outputs of multiple models; Finally, use machine learning or deep learning methods to train the model: Model selection: select the corresponding machine learning or deep learning model according to the task and data characteristics; The training process uses the preprocessed and feature extracted dataset to train the model.

8. The method for evaluating the competency of pilots in the course of flight line operation training according to claim 1, Features: Human body posture assessment information includes normal human body posture and abnormal human body posture; The specific process of obtaining human posture assessment information is as follows: During the simulated flight or actual flight, an image acquisition device is set up to collect the real-time image information of the pilot, process the real-time image information of the pilot, and obtain the attitude evaluation parameters; When the posture evaluation parameters are abnormal, human posture abnormality is generated; When there is no abnormality in the posture evaluation parameters, the generated human posture is normal.

9. The method for evaluating the competency of pilots in the course of flight line operation training according to claim 8, Features: The acquisition of the posture evaluation parameters and the determination process of posture evaluation parameter abnormalities are as follows: Extract the real-time image information of the pilot, extract the facial image of the pilot when he / she enters the flight preparation posture from the real-time image information of the pilot, extract the nose tip point and two outer eye corner points of the pilot at this time, and mark them as point a, point b1 and point b2; Connect the two outer points b1 and b2 to obtain the line segment L1, and then make a perpendicular line segment L2 perpendicular to the line segment L1 with point a as the endpoint; Then, L2 is extended to both ends, one end is extended to the top of the cockpit, and the other end is extended to the bottom of the cockpit to obtain the evaluation line P; After starting the simulated flight or the actual flight, continue to collect facial images, and collect the distance between the vertical line segment L2 in the collected facial image and the evaluation line P every preset time period to obtain multiple evaluation distances; After a single flight is completed, the average of multiple evaluation distances is calculated, and the attitude evaluation parameters are obtained; When the posture evaluation parameter is greater than the preset value, it means that there is an abnormality.

10. The method for evaluating the competency of pilots in the course of flight line operation training according to claim 1, Features: The human body status assessment information includes normal human body assessment and abnormal human body assessment. The specific processing process of the human body status assessment information is as follows: collecting pilot human body information during the flight, the pilot human body information includes human blood pressure information, heart rate information, respiratory rate information and voice information during the flight; During the simulated flight or the actual flight, the human body's blood pressure information, heart rate information and respiratory rate information are collected every preset time until the simulated flight or the actual flight ends; The amount of collected human blood pressure information, heart rate information and respiratory rate information is marked as Qi; Then, the blood pressure information of Qi people is converted into a line graph according to the time collection sequence to obtain a blood pressure assessment graph; Convert the Qi heart rate information into a line graph according to the time collection sequence to obtain the heart rate evaluation graph; The Qi respiratory rate information is converted into a line graph according to the time collection sequence to obtain a respiratory rate evaluation graph; Then the voice information is processed to obtain the pilot's speech speed information; When any one of the blood pressure assessment graph, heart rate assessment graph, respiratory rate assessment graph and pilot speech rate information is abnormal, a human body assessment abnormality is generated; When there is no abnormality in the blood pressure assessment graph, heart rate assessment graph, respiratory rate assessment graph and pilot speech rate information, the human body assessment is normal; The process of determining the abnormality of the blood pressure evaluation graph is as follows: analyzing the fluctuation trend of the blood pressure evaluation graph, when the fluctuation trend is greater than a preset amplitude, it indicates that there is an abnormality, and setting a standard blood pressure line in the blood pressure evaluation graph, when the duration of exceeding the standard blood pressure line is greater than a preset duration, it indicates that there is an abnormality; The process of determining the abnormality of the heart rate evaluation graph is as follows: analyzing the fluctuation trend of the heart rate evaluation graph, when the fluctuation trend is greater than a preset amplitude, it indicates that there is an abnormality, and setting a standard heart rate line in the heart rate evaluation graph, when the duration of exceeding the standard heart rate line is greater than a preset value, it indicates that there is an abnormality; The abnormality determination process of the respiratory rate evaluation graph is as follows: the fluctuation trend of the respiratory rate evaluation is analyzed. When the fluctuation trend is greater than a preset amplitude, it indicates that there is an abnormality. A respiratory rate standard line is set in the respiratory rate evaluation graph. When the duration of exceeding the respiratory rate standard line is greater than a preset duration, it indicates that there is an abnormality. The abnormality judgment process of the pilot speech speed information is as follows: when the pilot speech speed information exceeds a preset range, it indicates that there is an abnormality.

Citation Information

Patent Citations

  • Cockpit man-machine interaction system based on mental fatigue monitoring

    CN114742090A

  • Method for introducing physiological information into flight training evaluation system

    CN115227214A

  • Training presentation and evaluation system based on flight simulator

    CN115938189A

  • Pilot in-cabin motion digital twinning system

    CN116188673A

  • Method for analyzing ability of flight personnel based on flight parameter data

    CN117875766A

Cited By

  • Flight personnel space orientation training evaluation method

    CN119692869A

  • Artificial intelligence-based flight control quality evaluation method and system

    CN120373936A