JP7058300B2 - Evaluation support system for aircraft - Google Patents

Description

The present disclosure relates to a system for evaluating an aircraft, such as aircraft maneuverability and pilot workload.

Pilot suitability is evaluated for aircraft maneuverability and flight control system. This evaluation is performed, for example, in the process of designing and manufacturing a new aircraft. This evaluation is performed by giving the pilot a task and performance requirements, which are quantitative flight conditions for the task. Pilots perform maneuvers using flight simulators or real aircraft according to tasks and performance requirements. Here, the "task" is a command to the pilot regarding the operation and operation of the aircraft.

The results of the flight test are evaluated from the viewpoint of whether or not the pilot was able to maneuver as intended in response to the task request. As a result of the evaluation, evaluation values according to evaluation criteria such as Cooper-Harper Rating Scale (Non-Patent Document 1, hereinafter may be abbreviated as C-H Rating) and PIO Rating Scale are specified. In the C-H Rating, a rating value of 10 levels is given to maneuverability according to the pilot workload and the degree of achievement of performance requirements (see Fig. 2). In support of the assessment, pilot comments and flight data will be included in the assessment report along with the assessment.

Pilot comments include a description of the maneuvering situation, the aircraft's reaction to the pilot's maneuver, the pilot's physical and mental load, etc., summarized in a report and fed back to the aircraft designer. .. The physical and mental load of a pilot is referred to as the Pilot Workload. Here, the "physical load" means that the mechanical load for maneuvering is large. In addition, "mental load" means that there are many monitoring points and it is not possible to focus on one place.

Pilot comments are obtained after the flight simulator or aircraft has been piloted. However, at present, the acquisition of pilot comments is carried out by listening to humans and recording by hand. Since the pilot comments include unique phrases and describe a wide range of test conditions, the recording itself requires a certain level of skill. In addition, since the pilot comments at the time of acquisition, that is, the primary comments, are voice information by the pilot, the conversion to text information for posting them in the evaluation report requires considerable effort. In addition, it is sometimes necessary to confirm the nuances of the recorded pilot comments, which requires more time.

Cooper, G.E. and Harper, R.P. Jr .: The Use of Pilot Rating in the Evaluation of Aircraft Handling Qualities, NASA TN D-5153, 1969.

Based on the above, it is an object of the present disclosure to provide an aircraft evaluation support system that can reduce the labor involved in aircraft evaluation and improve the accuracy of evaluation using objective data.

The evaluation support system for aircraft according to the present disclosure includes an evaluation unit that calculates evaluation reference values for evaluation results by pilots who have conducted simulated flight tests or flight tests with actual aircraft based on predetermined evaluation criteria, and flight tests. It is equipped with a biometric data collection unit that collects biometric data of the pilot during the flight. The evaluation unit according to the present disclosure specifies the evaluation reference value based on the biometric data collected by the biometric data collection unit.

According to the present disclosure, it is possible to provide an aircraft evaluation support system that can reduce the labor related to aircraft evaluation and improve the accuracy of evaluation by objective data.

It is a figure which shows the structure of the evaluation system which concerns on embodiment of this disclosure. It is a flowchart which shows the Cooper-Harper Rating Scale. It is a figure which shows an example of the report of the evaluation in a flight test. It is a figure which shows an example of the flight data in a flight test. It is a figure which shows the representative example of various elements (variables) of a flight data. It is a figure which shows an example which converts the voice data of a pilot comment acquired in a flight test into text data. It is a figure which shows another example which converts the voice data of a pilot comment acquired in a flight test into text data. It is an example of biometric data acquired in a flight test, and is a graph showing fluctuations in heartbeat and fluctuations in respiration. It is an example of biometric data acquired in a flight test, and is a graph showing fluctuations in sweating and movement of a viewpoint. It is a flowchart which shows the procedure of converting a voice pilot comment into text data. It is a flowchart which shows the procedure of specifying an evaluation reference value. It is a flowchart which shows the procedure which graphs flight data. It is a flowchart which shows the procedure of making a graph of biometric data. It is a flowchart which shows the procedure which outputs the report of the evaluation result. It is a figure which shows the example of a part of a report. It is a figure which shows the example of the other part of a report. It is a figure which shows the example of the other part of a report.

Hereinafter, embodiments relating to the evaluation of the aircraft by the pilot when conducting a flight test will be described with reference to the attached drawings.
As shown in FIG. 1, the evaluation support system 1 according to the present embodiment includes an evaluation unit 10 and a storage unit 20 that stores data necessary for evaluation and provides the data to the evaluation unit 10. .. Further, the evaluation support system 1 records flight data with a voice recording unit 30 that records a pilot comment by voice and provides it to the evaluation unit 10, a biological data collection unit 40 that collects biometric data of the pilot and provides it to the evaluation unit 10. It includes a simulator / aircraft 50 that is generated and provided to the evaluation unit 10. The simulator / aircraft 50 is intended to generate flight data by a maneuvering simulator or to actually maneuver an airplane to generate flight data. That is, the flight test in the present disclosure includes both a simulated flight test using a simulator and a flight test for operating an actual airplane. The flight data also includes both data obtained by a simulated flight test and data obtained by a flight test in which an actual airplane is operated.

In the evaluation support system 1, the evaluation unit 10 is the following first based on the data stored in advance in the storage unit 20, the pilot comment provided by the voice recording unit 30, and the biometric data provided by the biometric data collection unit 40. Perform the process and the second process. As a result of the processing including the first processing and the second processing, the evaluation support system 1 prepares a report on the evaluation of the aircraft.
First treatment: Quantitative evaluation of the aircraft. The result of the quantitative evaluation is output as an evaluation reference value.
Second process: Converting pilot comments into text data using voice data. In the process of converting to text data, the expressions and wording of multiple pilots with variations are unified.

The evaluation unit 10 is composed of a computer device provided with a liquid crystal display device as an example, and the liquid crystal display unit constitutes the display unit 60 of the evaluation support system 1. Various information including the evaluation reference value described below is displayed on the display unit 60. The display unit 60 may be separated from the evaluation unit 10.

The terms used in this embodiment mean the following. The term "evaluation" simply means the term used for the entire ground and flight test of the aircraft, including evaluation values, reference evaluation values, pilot comments and flight data. "Evaluation value" means a numerical value subjectively specified by the pilot based on an evaluation standard such as C-H Rating. The "evaluation reference value" means a numerical value specified by the evaluation unit 10 by referring to the acquired biometric data or the like. The "evaluation reference value" is referred to to judge the validity of the "evaluation value".

Hereinafter, the contents of the storage unit 20, the voice recording unit 30, the biological data collection unit 40, and the simulator / aircraft 50 will be described in this order, and then the evaluation by the evaluation unit 10, the content of report creation, and the procedure will be described.

[Memory unit 20]
The storage unit 20 holds two types of data, the following evaluation criteria and past test data. This information is stored prior to the evaluation, but is provided to the evaluation unit 10 for the evaluation.

Rating Scale
As the evaluation standard, a known one such as the above-mentioned CH Rating can be used.
As an example, the storage unit 20 stores a program capable of executing the procedure shown in the flow chart of CH Rating shown in FIG. Hereinafter, CH Rating may be referred to as the Evaluation Criteria. The evaluation criteria are used to evaluate the aircraft in the evaluation unit 10. PIO Rating may be used as the evaluation standard.

The outline of the evaluation criteria is as follows.
The evaluation criteria are, as shown in FIG. 2, whether the pilot is controllable (S10), whether the pilot can achieve sufficient performance with an acceptable pilot workload (S20), and without improvement. It has three levels of criteria, whether the pilot is satisfied or not (S30). If performance cannot be achieved in S20 (S20 No), if the pilot is not satisfied in S30 (S30 No), and if the pilot is satisfied in S30 (S30 Yes), each further has three levels of criteria. .. As described above, according to the Evaluation Criteria, 10-level evaluation values are set for the aircraft, and the pilot who conducted the flight test specifies the evaluation values according to the Evaluation Criteria.

Past test data Past test data includes pilot comments, evaluation values, and flight data obtained in past flight tests. In the present embodiment, for example, artificial intelligence (AI) is used to find the correlation between the test data such as pilot comments obtained in the newly performed test and the evaluation value based on the evaluation criteria. AI uses past test data to make learning, reasoning, and judgment. Past test data may include other data.
It is preferable that the pilot comment, the evaluation value, the flight data, and the biometric data, which are the test data obtained in the newly performed test, are immediately stored in the storage unit 20 for use in the subsequent flight test.

An example of correlation learning by AI is as follows.
[Learning the correlation between pilot comments and evaluation criteria]:
AI extracts keywords related to the pilot's problem awareness and workload from the pilot comments and associates them with Rating. The following are examples of correlations between the keywords in comments and the learning of Rating.
For example, if there are no other points to be pointed out in the comments such as "no particular problem" and "good", CH-R is often evaluated as 1 or 2. In addition, CH-R is often evaluated as 3 when there are expressions such as "slightly increased workload but generally good" or "controllable as intended with minimum modification". In addition, keywords and improvements such as "You can control as intended but always need correction", "Although it is within the Desired request, the workload is a little high, please review the elevator gain", etc. CH-R is often rated as 4 if there is a comment that asks for or the workload exceeds the minimum.

[Learning the correlation between flight data and evaluation criteria]:
From flight data, especially the parameters and requirement range that are task requirements, for example, if the parameters satisfy the Desired task requirements with almost no change, CH-R1-2 and Desired requirements are satisfied, but they fluctuate. Correlation is learned, such as CH-R being 4, if the cage pilot tends to make continuous corrections. Further, when continuous correction is required, the current state of affairs is evaluated as CH-R3 and 4 according to the length of time for correction and the type of parameter requiring correction. By using AI and using a huge amount of test data (flight data and evaluation by pilots) including past data, it is more appropriate to find a highly accurate correlation even at the boundary between CH-R3 and 4, for example. Will be able to do.

[Learning the correlation between biometric data and evaluation criteria:
For example, the biometric data involved in the workload that the pilot cannot tolerate (larger than CH-R7) due to the height of the workload overlaps when the line-of-sight movement exceeds the threshold value and when the heartbeat and respiration exceed a certain value. The correlation that it is when is learned. In addition, the degree of pilot compensation such as minimal, moderate, and considerable in the evaluation criteria will be shown by learning the reference values for the data.

3 and 4 show examples of past test data. FIG. 3 shows pilot comments and CH Ratings, and FIG. 4 shows flight data.
In the data α and the data β, Case No. and Flight No. (FLT No.) are identification data of the test. For example, the test data in the data α is distinguished from other test data by the Case No. “1-1-1” and the flight numbers “1002” and “2003”. This data α shows test data by two pilots “A” and “B”, and FLT No. is given to each pilot. These identification data are input to the evaluation unit 10 prior to the test.

Data α and data β contain data related to pilot comments and CH Rating. In the data β, both pilots “A” and “B” have a CH Rating (evaluation value) of 3.
Data α and data β include data related to performance requirements. For example, in data α, the required value of CH Rating and the bank angle during turning are “Desired performance requirements” and “Adequate”. ) Performance requirements ”are specified in two stages.

In data α and data β, the test items are given names that summarize the tasks, and the altitude, speed, center of gravity and morphology indicate flight specifications. Further, the task requirements in FIG. 3 are performance requirements during each task. Tasks are defined by further subdividing test items. For example, in a landing test, the bank angle is maintained within the given angle in the declub operation (the operation of turning the nose that was swung against the crosswind from around the runway edge toward the runway). It is subdivided into each phase, such as task A, and then task B in which the ascending / descending rate at the time of touchdown is set to be equal to or lower than the given ascending / descending rate through flare operation (operation of raising the nose immediately before touchdown). For task A and task B, a specific numerical request is described in the task request of the data β.

FIG. 4 shows flight speed (Vc), column angle (c), throttle lever right (TR), column operating force (fe), elevator steering angle left (eL), and elevator steering angle right from among many flight conditions. Flight data for (eR) is shown. In the graph of FIG. 4, the horizontal axis indicates time and the vertical axis indicates each value, and the storage unit 20 stores data over time under each flight condition. This flight data is, for example, when the test data of the data α in FIG. 3 is acquired. That is, the data of FIG. 3 and the data of FIG. 4 are associated with each other and stored in the storage unit 20. Although the throttle lever left (TL) and elevator steering angle left (eL) are not shown in FIG. 4, the throttle lever left (TL) and elevator steering angle left (eL) are throttle lever right (TR) and elevator steering. This is because it is overlaid with almost the same value as the right corner (eR).
The flight conditions can include the matters shown in FIG. 5 in addition to the matters shown in FIG.

[Voice recording unit 30]
The voice recording unit 30 records the comments of the pilot who conducted the flight test as voice, and converts the voice data into text data. The converted text data of the pilot comment is corrected by the evaluation unit 10.
The voice recording unit 30 includes a microphone that converts voice into an electric signal in order to record voice data. There are no restrictions on the types of microphones used in this disclosure.

Further, the voice recording unit 30 has a voice recognition function for converting data related to voice converted into an electric signal by a microphone into text data. The voice recognition function converts the voice into text while collating the recorded voice data with the language data.

The voice recognition function generally includes a voice recognition dictionary, which analyzes the recorded voice acoustically and collates the result of the acoustic analysis with the voice recognition dictionary. The speech recognition dictionary includes, for example, an acoustic model, a language model, and a pronunciation dictionary. The acoustic model analyzes the frequency components and time changes of the sound to be recognized, and finds the frequency characteristics. A typical acoustic model is based on statistical processing of thousands of people and thousands of hours of speech. The language model is for evaluating whether a character string or a word string is appropriate for Japanese, and is a collection of many Japanese texts and statistical processing. The pronunciation dictionary serves to connect the words of the language model with the acoustic model. The acoustic model is modeled for each "phoneme", which is the smallest unit of voice, and the phoneme acoustic models are connected according to a pronunciation dictionary to form a word acoustic model corresponding to word utterance. By collecting past pilot comments, it is possible to deepen learning and build a word model that is more specialized in aircraft evaluation.

Here, it is decided that the voice recording unit 30 has a voice recognition function, but a voice recognition function may be provided in a portion other than the voice recording unit 30, for example, the evaluation unit 10. In this case, it suffices if the voice recording unit 30 is provided with a microphone. Further, a speaker may be provided so that the voice recorded by the voice recording unit 30 can be viewed. A speaker is a device that converts an electric signal into voice and outputs it.

An example of converting the voice data of the pilot comment into text data will be described with reference to FIG. As an example, this correction can be performed by the artificial intelligence referring to the past test data stored in the storage unit 20.
This example is a pilot comment on rolls during roll-in / out and Fe during and during bank holding. Among them, a conversion example including some characteristic corrections is shown below. In addition, "Fe" means a column operating force.

The upper part of FIG. 6 is basic text data converted into text data as it is without editing the voice data of the pilot comment. The lower part of FIG. 6 is corrected text data obtained by correcting this basic text data with reference to past test data by artificial intelligence. The same applies to FIG. 7, which will be described next.
In the basic text data, the conclusion that "the aircraft response is slow only in the initial motion" is at the end, while in the corrected text data, this conclusion is placed at the beginning. Further, in the voice data, the abstract expression of "the previous case" is converted into the expression specifically specified as "FLT No. 3004" in the converted final text data. This conversion is performed by the flight number included in the past test data of FIG. 3 shown as an example. It can be executed by collating the flight number in this flight test with. Further, in the voice data, what was expressed as "workload" is clearly converted into "pilot workload" in the converted final text data. This conversion refers to historical test data, especially pilot comments, and is converted to a unified representation, for example, on the basis that the word containing "workload" is "pilot workload" with some exceptions. To.

Another example of converting the voice data of the pilot comment into the text data will be described with reference to FIG. 7.
This example is a pilot comment on the pull steering force during go-around and the push steering force associated with the increase in Flap Up and aircraft speed. Among them, some characteristic conversion examples including correction are shown below.

In the example shown in FIG. 7, the inconsistency between the flight data and the pilot comment is corrected based on the flight data, and the height of the workload is corrected from the correlation between the conventional workload and the pilot comment. ..

Specifically, in the basic text data, "heavy at about 90 pounds, 80 pounds," and "heavy at about 80 pounds" are "about 80 pounds, heavy" by referring to the flight data. It is converted into corrected text data as "heavy as 80 lbs" and "heavy (about 70 to 80 lbs)". Further, in the voice data, "about ± 0.5 °" is converted into "about ± 1 °" and corrected text data by referring to the flight data.

Specifically, regarding the latter, in the basic text data, "the pilot workload is high" is converted with the correction, in the corrected text data, "the pilot workload is very high".

As described above, by referring to the flight data in the flight test, the past test data, and particularly the data regarding the stored past workload, it is possible to reduce the variation of the pilot comment due to the pilot's subjectivity.

[Biological data collection unit 40]
The biometric data collection unit 40 measures the biometric data of the pilot conducting the flight test and transmits it to the evaluation unit 10. The evaluation unit 10 uses biometric data as one element for evaluating the flight test. Biometric data is quantitative data that allows for quantification of pilot workloads.

As an example, a wristwatch-type wearable device is applied to the biometric data collection unit 40. The pilot wears this wearable device on his arm and conducts a flight test. The wearable device continuously measures the biometric data of the pilot conducting the flight test. The measured biometric data is sent to the evaluation unit 10.
The wearable device is not limited to the wristwatch type and may be worn on a part of the body. As another example of the wearable device, an eye tracking device can be used. The eye tracking device recognizes the movement of the pilot's viewpoint as a subject. Eye tracking devices become wearable when worn on eyeglasses or on hats.

As the biological data acquired by the biological data collecting unit 40, heart rate, respiration, sweating amount, and viewpoint movement are exemplified as shown in FIGS. 8 and 9. Biodata can be graphed, as shown in FIGS. 8 and 9, and this graph can be included in the report.
In the graph showing the heartbeat of FIG. 8, the horizontal axis shows the elapsed time and the vertical axis shows the strength of the heartbeat. In the data of FIG. 8, it can be determined that the heartbeat suddenly changed after 6 seconds from the start of the test, and the physical load on the pilot increased.
In the data showing respiration in FIG. 8, the horizontal axis indicates the elapsed time and the vertical axis indicates the depth of respiration. In the data of FIG. 8, it can be determined that the depth of breathing and the frequency of breathing, for example, the depth of breathing exceeds −0.15 around 8 seconds after the start of the test, and the physical load on the pilot has increased.

In the graph of sweating in FIG. 9, the horizontal axis shows the elapsed time and the vertical axis shows the amount of sweating. In the data showing the amount of sweating in FIG. 9, for example, when the threshold value exceeds 0.1, it can be determined that one or both of the physical load and the mental load of the pilot have increased.
In the graph of viewpoint movement in FIG. 9, the horizontal axis indicates the elapsed time, and the vertical axis indicates the movement distance of the viewpoint. In the data showing the viewpoint movement in FIG. 9, for example, if there is a viewpoint movement at a distance indicated by 5 in 0.5 seconds, it can be determined that the pilot's attention is reduced and the mental load is increased.

Historically, pilot workloads have traditionally relied on pilot comments, the identification of which has been qualitative. On the other hand, in the evaluation support system 1 according to the present embodiment, the pilot workload is quantitatively adjusted by correcting the pilot comments obtained after the test based on the biological data acquired during the test as described above. Can be identified. For example, the intensity of the heartbeat can be divided into three stages of high, medium, and small, and the size of the pilot workload can be specified as one of large, medium, and small according to this division. Pilot workloads can quantitatively identify pilot workloads based on other physical data such as breathing, sweating, and viewpoint movement, not just heart rate, and combine data on heart rate, breathing, sweating, and viewpoint movement. It can also be specified. In addition, pilot workloads may be quantitatively identified based on biometric data such as heart rate, respiration, sweating and other brain waves excluding viewpoint movements.
In addition, the biometric data can be used to identify the evaluation reference value for determining the validity of the evaluation value so far.

[Simulator / Aircraft 50]
The simulator / aircraft 50 is a flight simulator or a real aircraft to be tested. The simulator is a device for giving the pilot a feeling equivalent to the operation of an actual aircraft and information for the operation, and there is no limitation on the device to be specifically applied. By using a simulator, it is possible to carry out various experiments and evaluation studies on aircraft at low cost while ensuring high safety compared to actually using an aircraft.
If the data processing software and instrument display program used in the simulator are shared with the test aircraft, the processing of the evaluation unit 10 can be shared, and a comprehensive flight simulation and flight experiments using actual aircraft can be combined. It is also possible to carry out various verification tests.

[Evaluation unit 10]
Next, the evaluation unit 10 will be described.
The evaluation unit 10 evaluates the aircraft based on the data stored in the storage unit 20, particularly the past test data and the test data obtained for the current enforcement test. The evaluation procedure in the evaluation unit 10 will be described with reference to FIGS. 10 to 17.
This procedure includes textualization of pilot comments, evaluation reference value output, flight data graphing, biometric data graphing and report output. Hereinafter, the contents of each will be described in this order, but the actual procedure is not limited to this order. At least the output of the report is done at the end of the procedure, but otherwise it can be done in parallel and at the same time.

[Text data conversion of pilot comments (Fig. 10)]
First, the procedure for converting the voice data of the pilot comment into the text data will be described with reference to FIG. This procedure does more than simply convert the voice data to text data, it corrects the converted text data. This is to prevent erroneous information from being included in the voice data due to the mistake of the pilot. This correction is made using flight data operated by the pilot and biometric data of the pilot during the flight operation.

This procedure is started by inputting a command "comment output" to the evaluation unit 10 from the keyboard by the operator performing the evaluation work (FIG. 10 S101).
Upon receiving this command, the evaluation unit 10 converts the already acquired voice data into text data (FIG. 10 S103). The text data converted here is referred to as basic text data. The basic text data is faithfully converted into text data based on the voice recognition technique.

The evaluation unit 10 corrects this basic text data to generate the corrected text data. The correction is performed through the first-stage processing and the second-stage processing as follows using the flight data and the biometric data obtained by operating the simulator. Flight data can also be obtained from an actual aircraft.
As the first stage processing, the part related to the flight condition in the basic text data is compared with the flight data, and it is determined whether or not there is a difference (FIGS. 10 S105 and S107). If there is a difference, the basic text data is modified to match or follow the flight data (FIG. 10 S107 yes, S109). For example, if the pulling operation force in the flight conditions in the basic text data is about 90 pounds, while the pulling operation force in the flight data is 79.2 lbs, it is about 90 pounds in the basic text data. Is rewritten to about 80 lbs and corrected (FIG. 10 S109). If there is no difference, the basic text data is left as it is (FIG. 10 S107 no).
The modified text data and the unmodified basic text data are collectively referred to as the data after the first stage processing.

Next, as a correction in the second stage, the data related to the pilot workload is compared with the biometric data in the data after the processing in the first stage, and it is determined whether or not there is a difference (FIGS. 10 S111, S113). If there is a difference, the data after the first stage processing is modified so as to match the biometric data (FIG. 10 S113 yes, S115). For example, while the workload in the workload in the data after the first stage processing is "high", the heart rate and sweating amount indicating the workload in the biometric data are equivalent to "very high". If the threshold value is exceeded, the expression "high" indicating the workload in the data after the first stage processing is rewritten to "very high" so as to be in line with the biometric data and corrected (FIG. 10 S115). .. If there is no difference, the data after the first stage processing is left as it is (FIG. 10 S113 no).

After performing the two-step correction described above, the pilot's comment is output (FIG. 10 S117). The output includes basic text data, which is a textualized version of pure audio data, in addition to the corrected pilot comments. This allows engineers and pilots to check the content of the correction and make corrections. Eventually, the revised comment will be published in the report. By learning what has been fixed, the system will be able to make more accurate corrections thereafter.

[Evaluation reference value output (Fig. 11)]
Next, the procedure for outputting the evaluation reference value will be described with reference to FIG. This procedure is executed by a command input from an operator who operates the evaluation support system 1 (FIG. 11S201).
When the command is input from the operator, the following three processes are executed in parallel.
The first process compares the flight data with the task request (FIG. 11 S203). This process is performed to confirm how well the flight data meets the criteria for the task request.
The second process compares the biometric data and pilot comments with the workload (FIG. 11 S205). This process is performed to confirm the consistency with the workload representation of the Rating scale.
The third process collates the pilot comment with the keyword (FIG. 11 S207). This process is performed in order to consider whether the evaluation is different from other test cases from the pilot comments, or whether there is a request for improvement of the aircraft.

After performing the above three processes in parallel, the rating scale and the pilot comment are compared (FIG. 11 S209).
As a result of the comparison, if the evaluation value of the pilot is different from the Rating scale (FIG. 11 S211 Y), the evaluation value of the pilot is corrected based on the Rating scale (FIG. 11 S213). The corrected evaluation value is used as an evaluation reference value (FIG. 11 S215). As a result of the comparison, if the evaluation value of the pilot is not different from the Rating scale (FIG. 11 S211 N), the evaluation value of the pilot is used as the evaluation reference value as it is (FIG. 11 S215).
More specifically, the correction of each item is as follows.
For comparison of flight data and task requirements. For example, when the Desired request is ± 100 ft and the Adjust request is ± 200 ft in the holding task at an altitude of 2500 ft, if the altitude deviates up to 2634 ft, CH-R is corrected to 5 or 6 from the reference of FIG. Nevertheless, if the pilot's evaluation value is CH-R of 4, the evaluation reference value is presented as CH-R5 (or 6). Similarly, when the altitude is changed to 2250ft, the evaluation reference value presents a value larger than CH-R7.

For comparison of biometric data and pilot comments with workloads, the following is exemplified. In the comment from the pilot, "The workload is quite high and always requires corrective steering", but if the evaluation value from the pilot is 3, it is understood that it is not the minimum compensation operation of 3. Furthermore, at this time, if the heart rate and sweating amount of the biometric data also confirm that the workload is considerably high, the pilot's evaluation value is corrected to a value larger than CH-R4.

An example of changing the evaluation reference value based on the matching with the keyword of the pilot comment is as follows. The comment states, "The response to steering is slow and it is necessary to review the elevator gain in order to obtain a satisfactory pitch control as intended." Nevertheless, if the CH-R from the pilot is 3, the evaluation reference value is corrected to a value greater than 4 based on the keyword "review". As another example, in a series of test items. Although the comment of FLT No. A12 compared with other specifications of FLT No. A12 is "It is not stable compared to FLT No. A13 and the wobbling continues", the evaluation value is CH-R for FLT No. A12. 3 and FLT No. A13 is CH-R2, CH-2 of FLT No. A13 is used as an evaluation reference value of 3 or 4 based on the keyword "Compare with FLT No. **". It is also necessary to correct it to.
Through the above processing, the evaluation reference value is specified (FIG. 11 S215).
If the three processes are in the same direction, the process is clear, but if they are in opposite directions, weighting between the processes may be possible to adjust the process and request a retest if necessary. The same direction means that the evaluation reference value is worse than the evaluation value of the pilot, for example. The contradictory directions mean, for example, a case where the evaluation is made on the bad side in S203, while the evaluation is made on the good side in S205.

[Graphing flight data (Fig. 12)]
Next, with reference to FIG. 12, a procedure for outputting flight data as a graph will be described. This procedure is executed by a command input from an operator who operates the evaluation support system 1 (FIG. 12 S301).
First, the flight data and the task request are compared (FIG. 12 S303). For example, if there is a task request "hold altitude 2500ft until Gear Down", altitude ± 100ft is Desired request, ± 200ft is Advertise request, the reference line is 2600ft / 2400ft / 2700ft / 2300ft in the altitude data. Can be displayed and highlighted on the display unit 60 at a location outside the required range.
Next, among the flight data, the basic data and the data related to the task request are graphed (FIG. 12 S305). Here, the basic data among the flight data are preset altitude, speed, pitch angle, bank angle, control stick input amount, and the like. In addition, flight data with special notes are extracted from the pilot comments and graphed (Fig. 12 S307). Through the above, the flight data is graphed (FIG. 12 S309).

[Graphing of biometric data (Fig. 13)]
Next, with reference to FIG. 13, a procedure for graphing and outputting biometric data will be described. This procedure is executed by a command input from an operator who operates the evaluation support system 1 (FIG. 13 S401).
First, the biometric data is graphed (FIG. 13 S403). In addition, biometric data with special notes and data with remarkable changes are extracted from the pilot comments and graphed (FIG. 13 S405). Through the above, the biometric data is graphed (FIG. 13 S407).

[Report output (Fig. 14)]
After the pilot comment is converted into text data, the evaluation reference value is output, the flight data is graphed, and the biometric data is graphed, the report is output by the command input from the operator (FIG. 14 S501). In order to output the report, pilot comments converted into text data, evaluations (reference values), graphed flight data, and graphed biometric data are prepared (FIG. 14 S503). The prepared data are input to a predetermined position in the report format (FIG. 14 S505), and when the input is completed, they are displayed on an image display device such as an LCD (Liquid Crystal Display) provided in the evaluation support system 1 for printing. The report is output by printing on a predetermined sheet of paper by the machine (FIG. 14 S507).

An example of the output report is shown separately in FIGS. 15, 16 and 17.
As shown in FIG. 15, the report describes test items, morphology, flight specifications, center of gravity, pilot, flight number, pilot comments, evaluation values, task requirements and evaluation reference values.
This report includes two people, Pilot D and Pilot B, for the same test items shown in FIG.
FIG. 16 is an example of flight data output of FLT No. D74 in FIG. 15, and FIG. 17 is a flight data output and biometric data output of FLT No. B57 in FIG. Both flight data are basic data such as velocity (calibrated airspeed and pitch angle), task requirement altitude (reference line is drawn at ± 100 ft of Desired requirement), and thrust mentioned in the comment. Thrust lever angle and N1 (engine speed) have been added. Further, in FIG. 17, the workload is described in the comment, and since the threshold value is exceeded from the place where the heart rate data is present, the heart rate data is selected and displayed. The evaluation reference value of FLT No. B57 in FIG. 15 has been changed to 5 because the altitude deviates from the Desired requirement and the workload is high.

[effect]
Next, the effect of the evaluation support system 1 according to the present embodiment will be described.
[Reduction of labor related to evaluation]
It took a certain amount of skill and time to record comments, but the efficiency of recording accurate comments is improved by being able to convert the pilot's voice data into text.
In addition, by automating the work of comparing flight data with evaluation criteria and required performance, the burden on engineers is reduced and mistakes are less likely to occur.

[Improvement of evaluation system by objective data]
Quantitative output of the workload from the sensed biometric data supports the height of the workload in the pilot comments. In addition, it is possible to suppress variations among pilots.
Finally, the subjective evaluation result of the pilot is compared with the evaluation reference value calculated by the system, and the correction is made to prevent the evaluation error due to the pilot's assumption and obtain a more accurate evaluation with a minimum of effort. be able to.

[Additional Notes]
The evaluation support system 1 described in the above embodiment is grasped as follows.
[Evaluation support system 1 according to the first aspect]
The evaluation support system 1 according to the first aspect has an evaluation unit 10 that specifies an evaluation reference value for an evaluation result by a pilot who has performed a flight test based on a predetermined evaluation standard, and a flight test is being performed. It is provided with a biometric data collecting unit 40 for collecting biometric data of a pilot. The evaluation unit 10 specifies an evaluation reference value based on the biometric data collected by the biometric data collection unit 40.

According to the evaluation support system 1 according to the first aspect, the evaluation result of the subjective pilot is compared with the evaluation reference value calculated by the system, and the evaluation error due to the pilot's assumption is prevented by making a correction, and the accuracy is improved. Can be highly evaluated with a minimum of effort.

[Evaluation support system 1 according to the second aspect]
The evaluation unit 10 of the evaluation support system 1 according to the second aspect identifies the evaluation reference value by comparing the pilot workload, which is the physical and mental load of the pilot, with the biometric data.
According to the evaluation support system 1 according to the second aspect, more accurate evaluation can be obtained.

[Evaluation support system 1 according to the third aspect]
The evaluation unit 10 of the evaluation support system 1 according to the third aspect compares the flight data in the flight test with the task request, corrects the evaluation result and uses it as an evaluation reference value, or the flight is converted into text data. The evaluation result is corrected based on the keywords included in the comments of the pilot who conducted the test and used as the evaluation reference value.
The evaluation support system 1 according to the third aspect can also obtain a more accurate evaluation.

[Evaluation support system 1 according to the fourth aspect]
The evaluation unit 10 of the evaluation support system 1 according to the fourth aspect compares the workload information regarding the pilot's workload included in the comments of the pilot who conducted the flight test as text data with the biological data, and the workload. Correct the information so that it is in line with the biometric data.
The evaluation support system 1 according to the fourth aspect can also obtain a more accurate evaluation.

[Evaluation support system 1 according to the fifth aspect]
The evaluation unit 10 of the evaluation support system 1 according to the fifth aspect includes a voice recording unit for acquiring comments by the voice of the pilot. The evaluation unit 10 converts the voice comment into text data, and compares the workload information and the biometric data with respect to the converted text data.
The evaluation support system 1 according to the fifth aspect can also obtain a more accurate evaluation while reducing labor.

[Evaluation support system 1 according to the sixth aspect]
The evaluation unit 10 of the evaluation support system 1 according to the sixth aspect compares the flight information regarding the flight conditions included in the comments of the pilot who conducted the flight test as text data with the flight data in the flight test, and the flight information. Is corrected to match or follow the flight data.
The evaluation support system 1 according to the sixth aspect can also obtain a more accurate evaluation.

In addition to the above, it is possible to select the configuration described in the above embodiment or change it to another configuration as appropriate.
In other words, this disclosure can be applied not only to the evaluation of aircraft by pilots, but also to the automation of skill evaluation in crew training. In the skill evaluation of the occupants, the instructor makes a comment on the pilot or the co-pilot's operation. Therefore, the acquisition of the pilot comment in the above embodiment may be changed to the comment from the instructor. Then, flight data, instructor's comments, and evaluation points in the conventional evaluation are stored in a database, and the automatic evaluation device is made to learn. In addition, it can be incorporated into a training simulator to support evaluation and output the results as a report. It is possible to perform a more fair and objective skill evaluation.

In addition, the height of the pilot workload can be quantitatively obtained by associating the acquired biometric data with other test data and storing them. That is, the height of the pilot workload can be quantitatively determined by collating the accumulated past biometric data with the correlation with the pilot comment.

1 Evaluation support system 10 Evaluation unit 20 Storage unit 30 Voice recording unit 40 Biometric data acquisition unit 50 Simulator / Aircraft 60 Display unit

Claims (7)

An evaluation unit that calculates evaluation reference values for evaluation results by pilots who have conducted simulated flight tests or flight tests with actual aircraft based on predetermined evaluation criteria.
A biometric data collection unit that collects biometric data of the pilot during the flight test is provided.
The evaluation unit is an evaluation support system for an aircraft that specifies the evaluation reference value based on the biometric data collected by the biometric data collection unit.
The evaluation unit
The evaluation reference value is specified by comparing the pilot workload, which is the physical and mental load of the pilot, with the biometric data.
The evaluation support system according to claim 1.
The evaluation unit
The flight data in the flight test is compared with the task request, and the evaluation result is corrected to obtain the evaluation reference value, or
The evaluation result is corrected to be the evaluation reference value based on the keyword included in the comment of the pilot who performed the flight test as text data.
The evaluation support system according to claim 1 or 2.
The evaluation unit
The workload information regarding the pilot's workload included in the comment of the pilot who performed the flight test as text data is compared with the biometric data, and the workload information is corrected so as to be in line with the biometric data. ,
The evaluation support system according to any one of claims 1 to 3.
A voice recording unit for acquiring the comment by the voice of the pilot is provided.
The evaluation unit
The comment by voice is converted into the text data, and the workload information and the biometric data are compared with respect to the converted text data.
The evaluation support system according to claim 4.
The evaluation unit
The flight information regarding the flight conditions included in the comment of the pilot who conducted the flight test, which is the text data, is compared with the flight data in the flight test, and the flight information is corrected so as to match or follow the flight data. do,
The evaluation support system according to any one of claims 3 to 5.
A display unit for displaying the evaluation reference value specified by the evaluation unit is provided.
The evaluation support system according to any one of claims 1 to 6.

Images