CN112684884A - Flight training action automatic identification and evaluation system - Google Patents

Flight training action automatic identification and evaluation system Download PDF

Info

Publication number
CN112684884A
CN112684884A CN202011532200.3A CN202011532200A CN112684884A CN 112684884 A CN112684884 A CN 112684884A CN 202011532200 A CN202011532200 A CN 202011532200A CN 112684884 A CN112684884 A CN 112684884A
Authority
CN
China
Prior art keywords
gesture
sequence
flight
preset
actions
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.)
Pending
Application number
CN202011532200.3A
Other languages
Chinese (zh)
Inventor
郭乐江
李�浩
彭晓明
何松
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Air Force Early Warning Academy
Original Assignee
Air Force Early Warning Academy
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Air Force Early Warning Academy filed Critical Air Force Early Warning Academy
Priority to CN202011532200.3A priority Critical patent/CN112684884A/en
Publication of CN112684884A publication Critical patent/CN112684884A/en
Pending legal-status Critical Current

Links

Landscapes

  • Measurement Of Velocity Or Position Using Acoustic Or Ultrasonic Waves (AREA)

Abstract

The invention discloses an automatic recognition and evaluation system for flight training actions, which comprises a gesture ultrasonic acquisition device, a data processing system and a gesture database, wherein each gesture action model is preset in the database; assigning values to each gesture action in sequence according to the correct operation sequence of the aircraft; the gesture ultrasonic acquisition device is used for adopting gesture actions of a pilot; the data processing compares the acquired gesture with gesture model parameters in a database, judges the gesture action type and assigns a value to the gesture action type; and arranging the recognized gesture actions according to the time sequence. The invention identifies the gestures and actions of the pilots, records the sequence of the gestures and the actions, and finally scores and evaluates the whole flight actions according to the operation sequence and the missing action ratio, so that the evaluation process is completely objective, and a large amount of human resources required by manual evaluation are reduced.

Description

Flight training action automatic identification and evaluation system
Technical Field
The invention relates to the technical field of flight teaching, in particular to an automatic identification and evaluation system for flight training actions.
Background
The flight performance is the most intuitive and effective method for judging the level of daily training (including simulated flight training) of a pilot, and currently, the judgment of the flight performance is mainly carried out by a marker by taking a capability flight training outline as reference and subjectively evaluating, so that the existing flight performance evaluation mode has the defects of strong subjectivity, insufficient system, slow and comprehensive scoring and the like. In addition, the scoring personnel need to have detailed knowledge of the powerful flight training outline to evaluate the flight performance, and the workload of the scoring personnel is large.
Disclosure of Invention
The technical problem to be solved by the invention is to solve the existing defects and provide an automatic evaluation system for flight simulation training or real training.
In order to solve the technical problems, the invention adopts the technical scheme that: the utility model provides a flight training action automatic identification and evaluation system, includes gesture ultrasonic acquisition device and data processing system, its characterized in that:
the gesture recognition system also comprises a gesture database, wherein each gesture action model is preset in the database;
assigning values to the gesture actions in sequence according to the correct operation sequence of the aircraft, wherein the values are 1, 2 and 3 … n in sequence to obtain a sequence Q;
the gesture ultrasonic acquisition device is used for adopting gesture actions of a pilot;
the data processing compares the acquired gesture with gesture model parameters in a database, judges the gesture action type and assigns a value to the gesture action type;
arranging the recognized gesture actions according to the time sequence to obtain a sequence xi
According to the sequence xiScreening the array Q to screen out the array xiThe intersection with the sequence Q and sorting by size to obtain a sequence Si
If the degree of mixing is ρ, then:
Figure BDA0002852367930000011
wherein m is the sequence SiL is a series xiThe item number difference value with the number sequence Q, wherein a and b are set parameters;
and evaluating and scoring the action according to the magnitude of the rho value.
Further, a gesture probability vector of the gesture features is obtained by utilizing a pre-trained gesture classification model, and the gesture probability vector is composed of probabilities that the gesture features belong to various preset gestures; determining a context probability vector related to the gesture feature based on the acquired context information and a predetermined context feature matrix, wherein the context feature matrix is composed of the probability of the occurrence of each preset gesture under each context factor, and the context probability vector is composed of the probability of the occurrence of each preset gesture under the current context; and determining the probability that the gesture features belong to various preset gestures in the current situation based on the gesture probability vector and the situation probability vector, and identifying the gesture corresponding to the maximum probability as the gesture corresponding to the acquired ultrasonic signals.
Furthermore, the position of the aircraft is obtained in real time according to the GPS, the flight track of the aircraft is obtained according to the position, and the correlation r between the aircraft and the preset track is judged according to the flight track.
Furthermore, a coordinate system is established by taking longitude and latitude as an x axis and a y axis respectively, a flight track and a preset track are displayed in the coordinate system,
Figure BDA0002852367930000021
wherein X is a latitude value corresponding to the flight trajectory, and Y is a latitude value corresponding to the preset trajectory;
and scoring and evaluating the flight video according to the r value.
The technical scheme shows that the invention has the following advantages: the invention identifies the gestures and actions of the pilots, records the sequence of the gestures and the actions, and finally scores and evaluates the whole flight actions according to the operation sequence and the missing action ratio, so that the evaluation process is completely objective, and a large amount of human resources required by manual evaluation are reduced.
Detailed Description
The invention discloses an automatic identification and evaluation system for flight training actions, which comprises a gesture ultrasonic acquisition device and a data processing system.
Firstly, a gesture database is required to be established, each gesture action model is preset in the database, and the collected models are identified through a trained gesture classification model.
The specific identification method is as follows: acquiring a gesture probability vector of gesture features by using a pre-trained gesture classification model, wherein the gesture probability vector is composed of probabilities that the gesture features belong to various preset gestures; determining a context probability vector related to the gesture feature based on the acquired context information and a predetermined context feature matrix, wherein the context feature matrix is composed of the probability of the occurrence of each preset gesture under each context factor, and the context probability vector is composed of the probability of the occurrence of each preset gesture under the current context; and determining the probability that the gesture features belong to various preset gestures in the current situation based on the gesture probability vector and the situation probability vector, and identifying the gesture corresponding to the maximum probability as the gesture corresponding to the acquired ultrasonic signals.
The evaluation process of the flight action is as follows: and according to the correct operation sequence of the aircraft, assigning values to the gesture actions in sequence, wherein the values are 1, 2 and 3 … n in sequence, and obtaining a sequence Q. If the operation flow has one hundred steps, n is 100.
The data processing compares the acquired gesture with gesture model parameters in a database, judges the gesture action type and assigns a value to the gesture action type; arranging the recognized gesture actions according to the time sequence to obtain a sequence xi
According to the sequence xiScreening the array Q to screen out the array xiThe intersection with the sequence Q and sorting by size to obtain a sequence Si
If the degree of mixing is ρ, then:
Figure BDA0002852367930000031
wherein m is the sequence SiL is a series xiThe item number difference value with the number sequence Q, wherein a and b are set parameters; the operator can set the parameters according to experience, and the b parameter mainly reflects the characteristics of missing actions.
And evaluating and scoring the action according to the magnitude of the rho value. According to the formula, automatic scoring of the operation flow in flight training can be completed, and human resources in the manual judging process are reduced.
In addition, the invention also calculates the deviation of the flight path, thereby integrally evaluating the flight training effect. The method comprises the following specific steps of acquiring the position of the aircraft in real time according to a GPS positioning system, acquiring the flight track of the aircraft according to the position, and judging the correlation r between the aircraft and the preset track according to the flight track. Respectively using longitude and latitude as x-axis and y-axis to establish coordinate system, displaying flight track and preset track in the coordinate system,
Figure BDA0002852367930000032
wherein X is a latitude value corresponding to the flight trajectory, and Y is a latitude value corresponding to the preset trajectory;
and scoring and evaluating the flight video according to the r value.

Claims (4)

1. The utility model provides a flight training action automatic identification and evaluation system, includes gesture ultrasonic acquisition device and data processing system, its characterized in that:
the gesture recognition system also comprises a gesture database, wherein each gesture action model is preset in the database;
assigning values to the gesture actions in sequence according to the correct operation sequence of the aircraft, wherein the values are 1, 2 and 3 … n in sequence to obtain a sequence Q;
the gesture ultrasonic acquisition device is used for adopting gesture actions of a pilot;
the data processing compares the acquired gesture with gesture model parameters in a database, judges the gesture action type and assigns a value to the gesture action type;
in time order, for identificationThe gesture movements are arranged to obtain a sequence xi
According to the sequence xiScreening the array Q to screen out the array xiThe intersection with the sequence Q and sorting by size to obtain a sequence Si
If the degree of mixing is ρ, then:
Figure FDA0002852367920000011
wherein m is the sequence SiL is a series xiThe item number difference value with the number sequence Q, wherein a and b are set parameters; and evaluating and scoring the action according to the magnitude of the rho value.
2. The automatic flight training action recognition and assessment system according to claim 1, wherein: acquiring a gesture probability vector of gesture features by using a pre-trained gesture classification model, wherein the gesture probability vector is composed of probabilities that the gesture features belong to various preset gestures; determining a context probability vector related to the gesture feature based on the acquired context information and a predetermined context feature matrix, wherein the context feature matrix is composed of the probability of the occurrence of each preset gesture under each context factor, and the context probability vector is composed of the probability of the occurrence of each preset gesture under the current context; and determining the probability that the gesture features belong to various preset gestures in the current situation based on the gesture probability vector and the situation probability vector, and identifying the gesture corresponding to the maximum probability as the gesture corresponding to the acquired ultrasonic signals.
3. The automatic flight training action recognition and assessment system according to claim 1, wherein: and acquiring the position of the aircraft in real time according to the GPS, acquiring the flight track of the aircraft according to the position, and judging the correlation r between the aircraft and the preset track according to the flight track.
4. The automatic flight training action recognition and assessment system according to claim 3, wherein: respectively using longitude and latitude as x-axis and y-axis to establish coordinate system, displaying flight track and preset track in the coordinate system,
Figure FDA0002852367920000021
wherein X is a latitude value corresponding to the flight trajectory, and Y is a latitude value corresponding to the preset trajectory;
and scoring and evaluating the flight video according to the r value.
CN202011532200.3A 2020-12-22 2020-12-22 Flight training action automatic identification and evaluation system Pending CN112684884A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011532200.3A CN112684884A (en) 2020-12-22 2020-12-22 Flight training action automatic identification and evaluation system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011532200.3A CN112684884A (en) 2020-12-22 2020-12-22 Flight training action automatic identification and evaluation system

Publications (1)

Publication Number Publication Date
CN112684884A true CN112684884A (en) 2021-04-20

Family

ID=75450837

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011532200.3A Pending CN112684884A (en) 2020-12-22 2020-12-22 Flight training action automatic identification and evaluation system

Country Status (1)

Country Link
CN (1) CN112684884A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114529177A (en) * 2022-02-10 2022-05-24 北方天途航空技术发展(北京)有限公司 Training result intelligent judgment method and system for training machine

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114529177A (en) * 2022-02-10 2022-05-24 北方天途航空技术发展(北京)有限公司 Training result intelligent judgment method and system for training machine

Similar Documents

Publication Publication Date Title
CN111160474B (en) Image recognition method based on deep course learning
CN114419736B (en) Experiment scoring method, system, equipment and readable storage medium
CN102053563A (en) Flight training data acquisition and quality evaluation system of analog machine
CN108830332A (en) A kind of vision vehicle checking method and system
CN109711377B (en) Method for positioning and counting examinees in single-frame image monitored by standardized examination room
CN103324937A (en) Method and device for labeling targets
CN105224921A (en) A kind of facial image preferentially system and disposal route
CN114662208B (en) Construction visualization system and method based on Bim technology
CN115810163B (en) Teaching evaluation method and system based on AI classroom behavior recognition
CN115826763B (en) Special combat simulation system and method based on virtual reality
CN110210695A (en) A kind of tower control simulated training appraisal procedure based on support vector machines
CN111523445A (en) Examination behavior detection method based on improved Openpos model and facial micro-expression
CN106529470A (en) Gesture recognition method based on multistage depth convolution neural network
CN117726991B (en) High-altitude hanging basket safety belt detection method and terminal
CN112684884A (en) Flight training action automatic identification and evaluation system
CN116994077A (en) Regression prediction method for flight attitude under action of complex wind field
CN116665841A (en) Directional shooting athlete reaction training device and real-time evaluation system
CN116128681A (en) Intelligent scoring system for oxygen preparation experiment of middle school
CN115829234A (en) Automatic supervision system based on classroom detection and working method thereof
CN115861977A (en) Evaluation method for simulated driving posture and simulated driving device
CN108764029A (en) Model intelligence inspection method
CN111223341A (en) Teaching training intelligent evaluation inspection method and system based on virtualization
CN113409635A (en) Interactive teaching method and system based on virtual reality scene
CN112184040A (en) Platform for evaluating software engineering capability based on behavior and learning data
CN103838793A (en) Image semantic retrieval method based on weighted image pyramid structure and fuzzy support vector machine

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
WD01 Invention patent application deemed withdrawn after publication

Application publication date: 20210420

WD01 Invention patent application deemed withdrawn after publication