CN115511010A - Method and system for classifying pilot driving style based on aircraft landing process - Google Patents
Method and system for classifying pilot driving style based on aircraft landing process Download PDFInfo
- Publication number
- CN115511010A CN115511010A CN202211451872.0A CN202211451872A CN115511010A CN 115511010 A CN115511010 A CN 115511010A CN 202211451872 A CN202211451872 A CN 202211451872A CN 115511010 A CN115511010 A CN 115511010A
- Authority
- CN
- China
- Prior art keywords
- pilot
- landing
- altitude
- clustering
- flight data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Landscapes
- Traffic Control Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to a method and a system for classifying pilot driving style based on an aircraft landing process, wherein the method comprises the following steps: step S1: dividing the aircraft landing process into different altitude intervals according to the ground clearance of the aircraft, and clustering the flight data subsequences in each altitude interval to obtain landing semantic codes; step S2: respectively calculating the conditional probability of any two landing semantic codes of each pilot in all adjacent altitude intervals, and constructing a pilot driving style vector based on the conditional probability; and step S3: and clustering the pilot style vectors to obtain the pilot style category. The method provided by the invention can analyze the driving operation style of the pilot and classify the pilot so as to help civil aviation operation to improve the safety level and guide an airline company to improve the pilot culture scheme.
Description
Technical Field
The invention relates to the field of civil aviation flight safety, in particular to a method and a system for classifying pilot driving style based on an aircraft landing process.
Background
The guarantee of civil aviation flight safety is the premise of carrying out all specific civil aviation activities. Of all flight phases, the accident incidence and the accident mortality rate are significantly higher in the landing phase than in the other phases, being one of the most dangerous flight phases. The flight accident is caused by the combined action of the pilot, the environment and the airplane. Therefore, how to analyze the driving style of the pilot to help civil aviation operation to improve the safety level and instruct an airline company to improve the pilot culture scheme becomes an urgent problem to be solved.
Disclosure of Invention
In order to solve the technical problem, the invention provides a method and a system for classifying pilot driving style based on an aircraft landing process.
The technical solution of the invention is as follows: a method for classifying pilot driving style based on an aircraft landing process comprises the following steps:
step S1: dividing the aircraft landing process into different altitude intervals according to the ground clearance of the aircraft, and clustering the flight data subsequences in the altitude intervals to obtain landing semantic codes;
step S2: respectively calculating the conditional probability of any two landing semantic codes of each pilot in all adjacent altitude intervals, and constructing a pilot driving style vector based on the conditional probabilities;
and step S3: and clustering the pilot style vectors to obtain the pilot style category.
Compared with the prior art, the invention has the following advantages:
the invention discloses a method for classifying pilot driving style based on an aircraft landing process, which can calculate the preference degree of a pilot for controlling the state transition of an aircraft in different aircraft states in the aircraft landing process, is used for analyzing the driving operation style of the pilot and classifying the pilot, and helps civil aviation to improve the safety level and guide an airline company to improve the pilot culture scheme.
Drawings
FIG. 1 is a flowchart of a method for classifying the driving style of a pilot based on the landing process of an aircraft according to an embodiment of the present invention;
fig. 2 is a block diagram of a pilot driving style classification system based on an aircraft landing process according to an embodiment of the present invention.
Detailed Description
The invention provides a method for classifying pilot driving style based on an airplane landing process, which can analyze the driving operation style of a pilot and classify the pilot so as to help civil aviation operation to improve the safety level and guide an airline company to improve the pilot culture scheme.
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings.
Example one
As shown in fig. 1, a method for classifying a pilot driving style based on an aircraft landing process according to an embodiment of the present invention includes the following steps:
step S1: dividing the aircraft landing process into different altitude intervals according to the ground clearance of the aircraft, and clustering the flight data subsequences in each altitude interval to obtain landing semantic codes;
step S2: respectively calculating the conditional probability of any two landing semantic codes of each pilot in all adjacent altitude intervals, and constructing a pilot driving style vector based on the conditional probability;
and step S3: and clustering the pilot style vectors to obtain the pilot style category.
In one embodiment, the step S1: according to the ground clearance of the airplane, dividing the landing process of the airplane into different height intervals, clustering flight data subsequences in each height interval to obtain landing semantic codes, and specifically comprising the following steps:
step S11: acquiring a flight data set M, wherein the flight data set M comprises P pilots, and each pilot has flight data of F flights;
wherein the content of the first and second substances,indicating the ith flightFlight data for the member's jth flight; order toWherein, in the step (A),flight data, T, representing the 1 st second of the ith pilot's jth flight during landing ij Indicating the duration of the jth flight of the ith pilot in the landing stage; wherein the content of the first and second substances,,a value representing the nth flight data for the ith pilot for the jth flight during the 1 st second of landing;
step S12: respectively calculating mean value mean of single flight data of all flights in M n And standard deviation ofstd n :
Using mean n Andstd n for flight dataStandardized processing is carried out to obtain standardized flight data(ii) a Wherein the content of the first and second substances,;
partitioning M into two subdata sets M 1 And M 2 Wherein, in the process,M 1 and M 2 All contain all pilots in M, and each pilot contains the same amount of flight data;
step S13: dividing the landing process of the airplane into K altitude intervals according to the ground clearance of the airplane: { (H) i-1 ,H i ) I =1,2,3 \8230K }, wherein H i Is a division value of the height interval; flight data that standardizes primary flights
Is shown asWherein, in the step (A),flight data representing the flight in the first altitude interval is represented asWherein, in the step (A),a radio altitude representing the u-th second recorded in the flight data;
step S14: using a Dynamic Time Warping (DTW) -based k-means + + time sequence clustering method to M 1 Clustering the flight data of all flights in different height intervals, and adjusting the number of each clustering center and initializing each clustering center initial value to make the clustering outline coefficient as small as possible to obtain the landing semantic code number of the airplane in the kth height intervalτ k Then the set of all landing semantic codes for the kth altitude interval is represented asThus, flight data for the j-th flight of the i pilots can be normalizedIs shown asWherein, in the process,。
in one embodiment, the above S2: respectively calculating the conditional probability of any two landing semantic codes of each pilot in all adjacent altitude intervals, and constructing a pilot driving style vector based on the conditional probability, wherein the method specifically comprises the following steps:
step S21: calculating M 1 The conditional probability of occurrence of any two landing semantic codes of each pilot in all adjacent altitude intervals;
wherein, the first and the second end of the pipe are connected with each other,expressing that the ith pilot semantically codes the ith code for the ith altitude interval in the kth altitude intervalUnder the condition (1), the landing semantic code of the (k + 1) th altitude interval is the r-th code of the altitude intervalThe probability of (d);card{ } is the operation of calculating the number of collection elements;
step S22: each height interval hasN i The landing semantic code is used for constructing the ith pilotDriving style vector of dimensionWherein, in the step (A),the probability matrix representing all landing semantic code transitions of the ith pilot from the kth altitude interval to the (k + 1) th altitude interval is represented as。
In one embodiment, the step S3: clustering the pilot style vectors to obtain the pilot style categories, which specifically comprises the following steps:
step S31: using KMeans clustering method to determine driving style vectorV 1 ,V 2 ,V 3 …V P ]Clustering to obtain the class of pilot's driving styleS 1 ,S 2 ,S 3 …S P ]Wherein, in the process,S i a driving style category for the ith pilot;
step S32: to M is aligned with 2 Repeating the steps S2 to S3 to obtain all the pilot style categories(ii) a Order toAnd then, the L is minimized by adjusting the initial clustering centers and the number of the clustering centers of the Kmeans clustering method in the S3, and finally all the pilot style categories are obtained.
The flight data of different batches of pilots with longer flight time can be classified by the classification method provided by the embodiment of the invention, so that the difference of the classification results is minimum. In addition, the clustered categories can be labeled by flight instructors or experienced pilots, and finally, the label of each category is obtained.
The invention discloses a method for classifying pilot driving style based on an airplane landing process, which can calculate the preference degree of a pilot for controlling airplane state transition in different airplane states in the airplane landing process, is used for analyzing the driving operation style of the pilot and classifying the pilot, and helps civil aviation operation to improve the safety level and guides an airline company to improve the pilot culture scheme.
Example two
As shown in fig. 2, an embodiment of the present invention provides a system for classifying a pilot's driving style based on an aircraft landing process, which includes the following modules:
the obtaining landing semantic code module 41 is configured to divide the aircraft landing process into different altitude intervals according to the altitude of the aircraft from the ground, and cluster the flight data subsequences in each altitude interval to obtain a landing semantic code;
a pilot driving style vector construction module 42, configured to calculate conditional probabilities of any two landing semantic codes of each pilot in all adjacent altitude sections, respectively, and construct a pilot driving style vector based on the conditional probabilities;
and the pilot style category obtaining module 43 is configured to cluster the pilot style vectors to obtain the pilot style categories.
The above examples are provided for the purpose of describing the present invention only and are not intended to limit the scope of the present invention. The scope of the invention is defined by the appended claims. Various equivalent substitutions and modifications can be made without departing from the spirit and principles of the invention, and are intended to be within the scope of the invention.
Claims (5)
1. A method for classifying pilot driving style based on an aircraft landing process is characterized by comprising the following steps:
step S1: dividing the aircraft landing process into different altitude intervals according to the ground clearance of the aircraft, and clustering the flight data subsequences in the altitude intervals to obtain landing semantic codes;
step S2: respectively calculating the conditional probability of any two landing semantic codes of each pilot in all adjacent altitude intervals, and constructing a pilot driving style vector based on the conditional probabilities;
and step S3: and clustering the pilot style vectors to obtain the pilot style category.
2. The method for classifying pilot' S driving style based on aircraft landing procedure according to claim 1, wherein the step S1: dividing the aircraft landing process into different altitude intervals according to the ground clearance of the aircraft, clustering the flight data subsequences in the altitude intervals to obtain landing semantic codes, and specifically comprising the following steps:
step S11: acquiring a flight data set M, wherein the flight data set M comprises P pilots, and each pilot has flight data of F flights;
wherein, the first and the second end of the pipe are connected with each other,flight data representing the ith pilot's jth flight; order toWherein, in the step (A),flight data, T, representing the 1 st second of the ith pilot's jth flight during landing ij Indicating the duration of the jth flight of the ith pilot in the landing stage; wherein the content of the first and second substances,,a value representing the nth flight data for the ith pilot for the jth flight during the 1 st second of landing;
step S12: respectively calculating mean value mean of single flight data of all flights in M n And standard ofDifference betweenstd n :
Using mean n Andstd n for flight dataStandardized processing is carried out to obtain standardized flight data(ii) a Wherein, the first and the second end of the pipe are connected with each other,;
partitioning M into two subdata sets M 1 And M 2 Wherein M is 1 And M 2 All contain all pilots in M, and each pilot contains the same amount of flight data;
step S13: dividing the landing process of the airplane into K altitude intervals according to the ground clearance of the airplane: { (H) i-1 ,H i ) I =1,2,3 \ 8230k }, wherein H i Is a division value of the height interval; normalizing flight data for a flightIs shown asWherein, in the step (A),flight data representing the flight in the first altitude interval, denoted asWherein, in the process,represents the radio altitude of the u second recorded in the flight data;
step S14: using a k-means + + time sequence clustering method based on dynamic time warping to M 1 Clustering the flight data of all flights in different height intervals, and adjusting the number of each clustering center and initializing each clustering center initial value to make the clustering outline coefficient as small as possible to obtain the landing semantic code number of the airplane in the kth height intervalτ k Then the set of all landing semantic codes for the kth altitude interval is represented asThus, flight data for the j-th flight of the i pilots can be normalizedIs shown asWherein, in the step (A),。
3. the method for classifying pilot' S driving style based on aircraft landing procedure according to claim 2, wherein the step S2: respectively calculating the conditional probability of any two landing semantic codes of each pilot in all adjacent altitude intervals, and constructing a pilot driving style vector based on the conditional probabilities, wherein the method specifically comprises the following steps:
step S21: calculating M 1 In all adjacent altitude areas for each pilotThe conditional probability of any two landing semantic codes in between;
wherein the content of the first and second substances,expressing that the ith pilot semantically codes the ith code for the ith altitude interval in the kth altitude intervalUnder the condition (1), the landing semantic code of the (k + 1) th altitude interval is the r-th code of the altitude intervalThe probability of (d);card{ } is an operation to count the number of collection elements;
step S22: each height interval hasN i Landing semantic code is adopted to construct the model of the ith pilotDriving style vector of dimensionWherein, in the process,the probability matrix representing all landing semantic code transitions of the ith pilot from the kth altitude interval to the (k + 1) th altitude interval is represented as。
4. The method for classifying pilot' S driving style based on aircraft landing procedure according to claim 3, wherein the step S3: clustering the pilot style vectors to obtain pilot style categories, specifically comprising:
step S31: the driving style vector [ 2 ] using KMeans clustering methodV 1 ,V 2 ,V 3 …V P ]Clustering to obtain the class of pilot's driving styleS 1 ,S 2 ,S 3 …S P ]Wherein, in the step (A),S i a driving style category for the ith pilot;
step S32: to M 2 Repeating the steps S2 to S3 to obtain all the pilot style categories(ii) a Order toAnd then, the L is minimized by adjusting the initial clustering centers and the number of the clustering centers of the Kmeans clustering method in the S3, and finally all the pilot style categories are obtained.
5. A pilot driving style classification system based on an airplane landing process is characterized by comprising the following modules:
the system comprises an obtaining landing semantic coding module, a classifying module and a classifying module, wherein the obtaining landing semantic coding module is used for dividing the landing process of the airplane into different altitude intervals according to the ground clearance of the airplane, and clustering flight data subsequences in the altitude intervals to obtain landing semantic codes;
the pilot driving style vector building module is used for respectively calculating the conditional probability of any two landing semantic codes of each pilot in all adjacent altitude intervals and building a pilot driving style vector based on the conditional probability;
and the pilot style category acquisition module is used for clustering the pilot style vectors to obtain the pilot style categories.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211451872.0A CN115511010B (en) | 2022-11-21 | 2022-11-21 | Method and system for classifying pilot driving style based on aircraft landing process |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211451872.0A CN115511010B (en) | 2022-11-21 | 2022-11-21 | Method and system for classifying pilot driving style based on aircraft landing process |
Publications (2)
Publication Number | Publication Date |
---|---|
CN115511010A true CN115511010A (en) | 2022-12-23 |
CN115511010B CN115511010B (en) | 2023-04-07 |
Family
ID=84513867
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202211451872.0A Active CN115511010B (en) | 2022-11-21 | 2022-11-21 | Method and system for classifying pilot driving style based on aircraft landing process |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115511010B (en) |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
RU2436164C1 (en) * | 2010-10-01 | 2011-12-10 | Федеральное государственное образовательное учреждение высшего профессионального образования "Военный авиационный инженерный университет" (г. Воронеж) Министерства обороны Российской Федерации | Method of assessing quality of piloting aeroplane by pilot during landing phase based on data from standard on-board recording device |
CN106548294A (en) * | 2016-11-11 | 2017-03-29 | 中国民航大学 | A kind of landing maneuver Performance Evaluation Methods and device based on flying quality |
CN107909106A (en) * | 2017-11-14 | 2018-04-13 | 北京航空航天大学 | A kind of detection method of aircraft flight environment |
CN111340388A (en) * | 2020-03-13 | 2020-06-26 | 中国民航大学 | Pilot operation quality evaluation method based on flight QAR data |
CN114004292A (en) * | 2021-10-29 | 2022-02-01 | 重庆大学 | Pilot flat-floating ejector rod behavior analysis method based on flight parameter data unsupervised clustering |
-
2022
- 2022-11-21 CN CN202211451872.0A patent/CN115511010B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
RU2436164C1 (en) * | 2010-10-01 | 2011-12-10 | Федеральное государственное образовательное учреждение высшего профессионального образования "Военный авиационный инженерный университет" (г. Воронеж) Министерства обороны Российской Федерации | Method of assessing quality of piloting aeroplane by pilot during landing phase based on data from standard on-board recording device |
CN106548294A (en) * | 2016-11-11 | 2017-03-29 | 中国民航大学 | A kind of landing maneuver Performance Evaluation Methods and device based on flying quality |
CN107909106A (en) * | 2017-11-14 | 2018-04-13 | 北京航空航天大学 | A kind of detection method of aircraft flight environment |
CN111340388A (en) * | 2020-03-13 | 2020-06-26 | 中国民航大学 | Pilot operation quality evaluation method based on flight QAR data |
CN114004292A (en) * | 2021-10-29 | 2022-02-01 | 重庆大学 | Pilot flat-floating ejector rod behavior analysis method based on flight parameter data unsupervised clustering |
Also Published As
Publication number | Publication date |
---|---|
CN115511010B (en) | 2023-04-07 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107086935B (en) | People flow distribution prediction method based on WIFI AP | |
CN111008669A (en) | Deep learning-based heavy landing prediction method | |
CN111881213B (en) | System for storing, processing and using flight big data | |
CN111915930B (en) | Airborne empty pipe enhancement co-processing system and method | |
US9475573B2 (en) | Methods for matching flight data | |
CN106910376A (en) | A kind of air traffic operation regulation instructs monitoring method and system | |
CN111985685B (en) | Flight plan optimization method and device based on big data analysis and electronic equipment | |
CN112966555A (en) | Remote sensing image airplane identification method based on deep learning and component prior | |
CN113761226A (en) | Ontology construction method of multi-modal airport data | |
CN110264787A (en) | A kind of flight Route reform time reliability evaluation method and system | |
CN115470796A (en) | Text instruction generation method and equipment for air traffic control simulation training | |
CN110659773A (en) | Flight delay prediction method based on deep learning | |
Chou et al. | A machine learning application for predicting and alerting missed approaches for airport management | |
CN113127503A (en) | Automatic information extraction method and system for aerospace information | |
CN117995027A (en) | Flight plan drop-preparation recommendation method, system, equipment and medium | |
CN115511010B (en) | Method and system for classifying pilot driving style based on aircraft landing process | |
Zhang et al. | Research on spectrum sensing system based on composite neural network | |
Chakrabarti et al. | Clustering Aircraft Trajectories According to Air Traffic Controllers' Decisions | |
CN116579552A (en) | Self-adaptive decision method for airport runway configuration | |
CN112948161B (en) | Deep learning-based aviation message error correction and correction method and system | |
CN112348225B (en) | Flight path prediction method, electronic device, and non-transitory computer-readable storage medium | |
Chu et al. | Hierarchical Method for Mining a Prevailing Flight Pattern in Airport Terminal Airspace | |
Khattak et al. | Missed Approach, a Safety‐Critical Go‐Around Procedure in Aviation: Prediction Based on Machine Learning‐Ensemble Imbalance Learning | |
CN116993137B (en) | Method and device for determining stand, electronic equipment and medium | |
Clarke et al. | Natural Language Understanding and Extraction of Flight Constraints Recorded in Letters of Agreement |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |