Flight visual simulation system based on sensor technology and QAR data

Abstract

With the development of wireless sensor technology, more and more fields are applied to wireless sensor technology, and the design and implementation of the flight scene system are in it. The purpose of this study is to design and implement a flight visual simulation system using sensor technology and QAR data. The method of this study is to first establish the mathematical model of the aircraft and then calculate the aerodynamic force and torque of each part of the aircraft respectively to obtain the nonlinear dynamic model of the aircraft, and finally realize the whole simulation model on the MATLAB platform. The results show that the main motion variables corresponding to the conjugate complex root (0.0530.287i) are the velocity and pitch angle of the aircraft, and they all change slowly and long period. The period is 21.9 seconds, and the frequency doubling time (LN2 / 0.053 + 0.287i) is 11.0 seconds, which meets the requirements of flight quality specification. It is concluded that the flight visual simulation system in this study can well describe the flight trajectory and observe the external scene. The simulation system uses computer graphics and image technology to realize real-time and accurate simulation and reproduction of the flight status and trajectory of the aircraft, which can provide an intuitive and convenient simulation analysis method for the exploration of new aerospace technology and engineering design.

Keywords

Sensor technology QAR data flight visual system simulation design

1 Introduction

The data source of the flight quality monitoring system (FOQA) is relatively simple and does not need professional training personnel. The data source is QAR data which records flight parameters in real-time. Today, most domestic aircraft are equipped with QAR recording devices to facilitate the acquisition of data sources. However, compared with the LOSA system, one of the disadvantages of the FOQA system is that it can only reflect the accidents or events in-flight data, but cannot explain the reasons. The method of using QAR data to explain the cause of events (defined as limiting events and irregular operations) is a topic that needs to be solved as soon as possible, and it is also a topic with great possibility to be studied.

The simulation system plays an important role in aerospace engineering because it has the incomparable characteristics of its corresponding physical system. Specifically, it is about operability, observability, and economy and prediction possibility. Some data show that the input-output ratio of simulation and engineering may reach a high level of 1:100. In order to carry out cosmic activities faster and better, the research and development of simulation technology should play a more important role. In the design of the space launch mission, theoretical analysis is the basis, and simulation is the main method to determine the important position of 3D simulation technology.

In the design method of the flight visual simulation system, Hosman r introduces an advanced design method suitable for the specification of the flight simulator prompt system. This process is based on the analysis of the pilot model, which includes visual and vestibular feedback, as well as aircraft dynamics [1]. After replacing the model with a simulated aircraft, he used analysis tools to adjust the parameters of the scour filter to restore the pilot’s control behavior. This procedure allows you to specify a motion prompt algorithm. Then, according to the flight documents representing the operational flight envelope, the required motion system space is determined. Based on the practical constraints and the stability criteria of the platform under singular conditions, he established the basic geometry of motion. In this process, the characteristics of the aircraft, the task to be simulated and the task itself are considered when defining the motion prompt system of the simulator. His method is not stable [2]. El Gabri A R uses 3D visualization technology to model the real scene and observe the response of different tracking algorithms. By changing the simulation conditions, he can predict the behavior of the actual system. Visual simulation first needs to develop a model to represent the characteristics of the selected physical system or process. The model represents the system itself, but in order to avoid complexity and traceability, it represents the system itself in a simpler way, while the simulation represents the operation of the system over time. The simulation includes the missile and the target object created by Multigen Creator. The six degrees of freedom mathematical model of the missile’s dynamic parameters in the flight trajectory is realized by visual c++, and all of them are integrated into the infrared simulation environment implemented by Vega. Finally, the tracking algorithm that needs to be tracked is evaluated. His method model is too complex to be used in calculation and construction [3]. In order to improve the safety and efficiency of flight control implementation, Xian B developed low-cost real-time hardware in the loop simulation test-bed for quadrotor UAV. In order to realize the HILS test-bed, he took a micro quadrotor as the main body, equipped with a micro AHRS (attitude and heading reference system) unit and self-built DSP (digital signal processor) board. HILS is realized with xPC Target. The target computer is a compact PC / 104 computers, and the upper computer is a notebook computer. The flight data of the UAV [4], such as the flight path of the quad UAV, the flight path of the helicopter and the flight data of the helicopter are visualized by using the PC, the flight data of the UAV and the flight data of the helicopter are displayed by using the flight data of the Google and the helicopter. The test-bed can be used to simulate various flight control algorithms without losing safety and reliability. In order to verify the effectiveness of the proposed test-bed, he verified it on the HILS test-bed. Under the simulation of flight trajectory, his method has a large error and is not accurate enough [5].

This paper first introduces the 3D Flight Scene System and explains its working principle. This study also describes the large-scale scene model generation and collision detection in-flight scene. The three main detection technologies are: rectangular bounding box or spherical bounding box; polygon combination; object trajectory tracking. After that, this study describes the control mode and algorithm of the flight visual system. Through the analysis of the three-dimensional schematic diagram, the motion trajectory and each position point are confirmed, and the model building and system simulation are better carried out. The experimental process includes QAR data preprocessing, motion modeling and simulation data analysis of the flight scene. Combined with the experimental simulation platform, the system stability analysis, coupling analysis, demand analysis and simulation example analysis are carried out to verify the practicability of the system.

2 QAR data and flight visual simulation system

2.1 3D flight visual system

The 3D vision system is the terminal graphic display subsystem of the UAV flight control simulation environment [6]. It consists of three parts: model construction, scene driving and driving data acquisition. Collect 3D graphics objects in the computer, including hierarchy, LOD, material, color, lighting, texture setting and processing [7]. In this study, a multign creator was used to construct the model. The so-called scene driven mode means that the simulation program generates real-time scene based on the established scene model and the state parameters of various simulation objects in the scene under execution, and outputs the simulation results in the form of images. In the process of simulation, the state parameters of the simulation object can be controlled by user input, and also depends on the law of motion change of the simulation object itself [8, 9].

This paper uses vegprime as the scene drove software platform, uses VC++ 6.0 integrated environment, and designs the system with an object-oriented software development method. Operation data collection is to collect all kinds of real-time UAV flight control data as the operation data of the UAV model of a 3D vision system through data transmission technology from the UAV flight control simulation environment. RS232 serial port communication is used to achieve the driver data acquisition [10].

2.2 Working principle of the visual system

The operation principle of the system is that the user interacts with the system through the software user interface to realize data configuration, real-time or theoretical driven switch, 3D resource scheduling, scene management and other operations. The system receives real-time aerospace survey data through LAN multicast and uses it as input to realize the real-time 3D simulation of the whole process of space launch [11].

Through the construction of the resource server, we can realize the network structure-function of large-scale image and elevation data, and provide data services such as image, elevation and model to multiple terminal software. In operation, the so earth engine library is used for real-time scheduling to ensure the stability and efficiency of the system.

2.3 Large scale scene model generation

For large-scale scene models, the vision system can’t read all-terrain into memory at once [12]. In fact, due to the directivity of the line of sight, the limitation of vision and the interaction between objects, what the eye can see is often only a part of the scene. Therefore, the scene model is divided into smaller model units. Only a part of the scene model surrounding the viewpoint is read into the memory and is described in real-time through the dynamic update function. The whole database for flight is saved in the hard disk, and these model blocks are updated in real-time according to the view mode. In order to ensure the quality of the graphics, some real-time modifications of the scene are needed during the actual operation, and the replacement of the solid model and texture is performed according to the distance between the scene and the viewpoint, so the image display becomes smoother and more realistic [13–14].

2.4 Collision detection of flight scene

Collision detection is an indispensable element in the construction of a 3D vision system. Users can operate the scene objects of the 3D vision system in a more natural way. If no collision is detected, when an object encounters another object, it will “pass through the wall” without being affected by the collision, but it does not exist. Therefore, when building a 3D vision system, it is necessary to accurately judge whether the objects in the scene are in real-time conflict [15]. In a 3D vision system, the interaction between dynamic objects and static objects or between dynamic objects and dynamic objects is basically collision detection [16].

The general method of collision detection in a 3D vision system is to use a bounding box. Each line segment of the bounding box is parallel to the coordinate axis. The bounding box is the smallest box around a dummy object. The biggest advantage of using a bounding box to detect a collision is that it can realize fast collision detection, but in many practical applications, it is not enough to rely on a bounding box to realize natural interaction. If you need to prove that two objects do not intersect, using a bounding box is very effective. However, if the bounding boxes of two objects intersect, the bounding box is only part of the object boundary, so there is no need for the two objects to cross. Therefore, collision detection based on the bounding box is very rough and inaccurate.

Precise conflict detection can be used to perform accurate conflict detection in a 3D vision system. In other words, a bounding box is used to represent the boundary of a virtual object, and multiple polygons are also used to represent the bounding box, which surrounds the virtual object. Obviously, the more subdivided the polygon, the more accurate the boundary representation. Therefore, the boundaries of 3D virtual objects are represented by a series of polygons [17]. When determining whether each object intersects, determines whether the two polygon sets intersect. Only after the bounding boxes of two objects are confirmed to cross, it is necessary to confirm whether the corresponding two polygon combinations intersect [18, 19].

Therefore, in order to achieve various precision conflict detection in a 3D vision system, the following techniques or their combination can be used: rectangular or spherical bounding box, polygon combination, tracking object trajectory [20].

2.5 Control mode of the visual system

In order to meet the needs of users, the system has set up three visual control modes (random tracking mode, fixed tracking mode and side forward tracking mode). These three modes can use the “B” key to freely switch the perspective.

The height control of the three modes is obtained from the simulated trajectory data, and the displacement control is obtained by the step displacement. In the first two modes, the program changes the position of the viewpoint randomly by determining the distance between the viewpoint and the target point [21]. The distance of the side feed-forward model is stable. The change of the line of sight vector corresponds to the change of the radius of the circle centered on the Y-axis. The program uses dist to represent the rotation radius of the viewpoint centered on the Y-axis. The program sets the viewpoint position by judging the dist value [22, 23].

In the random tracking mode, the viewpoint is always aligned with the target, the position of the viewpoint remains unchanged, the target can fly autonomously, and the distance between the target and the viewpoint is increasing. When the distance is more than 300, the system will automatically reset the viewpoint position randomly. The viewpoint may be located in different directions and heights of the target, but the height must be greater than zero. The next pattern is based on the viewpoint of random generation, and the position of viewpoint is also different. There are numerous random patterns [24]. A scene walkthrough can be performed in this mode.

In the fixed tracking mode, the viewpoint is always aligned with the target, and the position of the viewpoint will not change. When dist > 1000, the system randomly selects and generates 12 kinds of specified location information to assign new positions to viewpoints. Please select these 12 kinds of location information based on the vision system’s more effective location. Screenshots of several partitions selected in fixed tracking mode [25].

Side to front tracking mode: the viewpoint is always set on the side and front of the target, and the distance between the viewpoint and the target is the same, so the target can fly autonomously, and the viewpoint follows the whole action [26].

2.6 Flight scene roaming control algorithm

The scene roaming function is mainly realized by setting the gloookat() function to control the position of viewpoint in the scene. In the random model of the system, the angle of view changes randomly, and the position of the viewpoint is generated randomly [27]. Therefore, when moving the target, you can set the change of the viewpoint in each direction through the program and operate with the keyboard. The purpose of viewpoint moving is to realize the roaming control of the scene. In the fixed mode and infeed mode, the viewpoint is set to a fixed position relative to the target through the program, and the scene cannot be roamed [28, 29].

The motion of the viewpoint in the scene can be divided into translation motion and rotation motion. Both movements are based on the target and move around the target’s local coordinate system. The translation movement is relatively simple. After determining the moving direction, the relative displacement of the original component is directly increased or decreased. The rotation motion is relatively complex, so the corresponding coordinate system transformation calculation should be considered. First, the local coordinate system is converted into the local coordinate system [30].

The rotation algorithm adopted in this system considers the coordinate system of the viewpoint centered on the target, that is, the rotation algorithm based on the object coordinate system (local coordinate system). The line of sight distance r between the viewpoint and the target does not change, and the viewpoint rotates around the Y-axis. However, in this system, the rotation around the x-axis and z-axis is not considered. Suppose that point O is the target position, point a is the initial position of the viewpoint, point B is the position after the clockwise rotation angle θ with y-axis as the center, the position vector of point an in three directions is V_3A=(V_ax, V_ay, V_az), and the position vector value of point B in three directions is V_3B=(V_bx, V_by, V_bz). Suppose l represents the rotation radius of the viewpoint centered on the y-axis [31, 32]. As shown in Fig. 1 is a 3D schematic diagram of clockwise rotation of viewpoint.

Fig. 1

3D diagram of clockwise rotation of viewpoint.

Since the viewpoint is rotated around the Y-axis, the vector of the Y direction of the viewpoint will not change. Therefore, in the research of rotation algorithm, x-y-z3-d coordinate system can be reduced to x-z2-d coordinate system. In the two-dimensional coordinate system, other settings remain unchanged. The angle through which point a passes clockwise along the positive direction of the x-axis is α, the two-dimensional vector value of point a is V_2A=(V_ax, V_az), and the two-dimensional vector value of point B is V_2B=(V_bx, V_bz). Figure 2 is a two-dimensional diagram of the clockwise rotation of viewpoint [33].

Fig. 2

Two-dimensional diagram of clockwise rotation of viewpoint.

We know the coordinate value of point a, so we need to obtain the coordinate value of point B. Use the basic formula of the triangle to calculate the value of each component of point a and the value of each component of point B: $\begin{matrix} V_{ax} = L cos α \\ V_{az} = L sin α \end{matrix}$ (1) $\begin{matrix} V_{bx} = L cos (α + θ) \\ V_{bz} = L sin (α + θ) \end{matrix}$ (2) $\begin{matrix} V_{bx} = L cos α cos θ - L sin α sin θ \\ V_{bz} = L sin α cos θ + L cos α sin θ \end{matrix}$ (3)

Substituting formula (1) into (2) (3) to obtain (4) $\begin{matrix} V_{bx} = V_{ax} cos θ - V_{az} sin θ \\ V_{bz} = V_{ax} cos θ + V_{az} sin θ \end{matrix}$ (4)

When the viewpoint rotates the angle θ anticlockwise with the y-axis as the center, it is only necessary to add a negative sign before expression (4) θ, as shown in the following formula: $\begin{matrix} V_{bx} = V_{ax} cos (- θ) - V_{az} sin (- θ) \\ V_{bz} = V_{ax} cos (- θ) + V_{az} sin (- θ) \end{matrix}$ (5)

All the above calculations obtain the vector values of the new viewpoint in the object coordinate system. In order to obtain the coordinate position of the new viewpoint in the world coordinate system, it is necessary to add the vector values of the target in each direction relative to the world coordinate system to realize the absolute coordinate transformation of the relative coordinates [34, 35]. Suppose that the target coordinate vector in the world coordinate system is V_o=(V_OX, V_OY, V_OZ), and the coordinate vector of the new viewpoint in the world coordinate system is V_B=(V_BX, V_BY, V_BZ).

The viewpoint rotates clockwise around the y-axis $V_{BX} = V_{OX} + V_{ax} cos θ - V_{az} sin θ$ (6) $V_{BY} = V_{OY}$ (7) $V_{BZ} = V_{OZ} + V_{ax} cos θ + V_{az} sin θ$ (8)

The calculation formula of the viewpoint with Y-axis as the center and counterclockwise rotation is as follows: $V_{BX} = V_{OX} + V_{ax} cos (- θ) - V_{az} sin (- θ)$ (9) $V_{BZ} = V_{OZ} + V_{ax} cos (- θ) + V_{az} sin (- θ)$ (10)

3 Design of flight visual system simulation experiment model

3.1 QAR data preprocessing

The main functions completed in the process are as follows.

Delete unnecessary information in QAR header, explanation of data field, unit, etc. Enter the ID of the blank data. For example, blank data for “DATECLKYR” is represented by “null”. Duplicate field names are marked separately. For example, the original data has two “Mach” fields, and during database conversion, the name of the second field is changed to “Mach 1” to distinguish. Please remove invalid characters from the field. For example, if the raw data is “WFL”, then “. “Is an invalid character in the data field definition? These characters will be replaced in the “WFL.” field. Add the data index “data index” as the original index “frame” [36]. Supplement SF: add preprocessed data to the flight phase flag “phase” and input the subsequent two modules (“manual QAR analysis” and “QAR data monitoring”) for the corresponding processing. In addition, the next time you call the data, you just need to open the access database file directly, without calling the CSV file.

3.2 Motion modeling of flight scene

Action modeling is to describe various action characteristics of UAV. In the vision system, the characteristics of the UAV model include appearance, color, structure, texture, etc., as well as position change, rotation, collision, zoom, etc. It is not enough to build a static 3D geometry of the virtual vision system. The content of motion modeling includes 6 DOF motion of UAV, LADA surfing, collision detection, ion system, etc.

(1) Particle system (special effects)

The modeling methods of the visual system can be divided into two categories: Geometric Modeling and motion modeling. Action modeling is used to describe the actions and actions of objects. The particle system is a process model [37]. In other words, the modeling method of each body element of the model is generated by using various calculation processes. Many process models are based on physical actions and combine geometric modeling with motion modeling.

The main advantages of process modeling are as follows. Using accurate physical models can improve the authenticity of objects. The model contains geometry and actions, and geometry reflects actions. If there is an effective physical model, the object’s action modeling becomes very simple. As long as the current geometric model is realized.

(2) Particle system operation process:

Particle source generates new particles.

Any number of new particles are generated, and the initial attributes are controlled by a random process. All particles have a lifespan, and if you can’t delete some particles, you can give them an unlimited lifespan.

Update particle attributes, update existing particle attributes.

Delete the “dead” particles and confirm their lifespan. If 0, the particles are removed from the system.

Draws particles and displays all existing particles in the particle system. Generally speaking, the geometric characteristics of particles are very simple and can be represented by pixels or small polygons.

The advantages of the particle system: the organic combination of simple body unit and complex object action. With particle systems, you can easily generate tail frames, and it’s easy to implement. The particle system can be defined by providing some parameters in the random process of particle properties. The particle is simple, the display is simple, and the display efficiency is high.

3.3 Simulation data analysis

Measurement data frame includes frame title, frame length, feature, parameter data, GPS information, check word and so on. According to the simulation data protocol, after mastering the data frame format and the data type of each parameter, the simulation data is analyzed and the required parameters are extracted.

In order to facilitate the data operation and conversion between different types of data, in order to realize the conversion from binary text data to an integer, floating-point and multiple precision data, several data structures need to be defined by the program. The most important link is to check the words. In other words, the last byte of data stored in the data frame is compared with the last check word of the frame. If these are the same, then the sending framework is correct and trusted. In different cases, the sending process is not correct. Code and data are not trusted, so it is necessary to judge the check word after data extraction [38].

The simulation data used in the system is the data file created for the simulation of objects and the demonstration of specific frame format aaa.dat in the database. No matter how the data frame format changes, the idea and process of data analysis will not change.

4 Analysis of flight visual simulation system

4.1 Stability analysis of flight visual simulation system

The natural stability of aircraft means that the aircraft can automatically maintain its original flight state when there is no control system. The characteristic roots of the linearized mathematical model are obtained when the aircraft is at a height of 50 m and hovering. As shown in Table 1, the characteristic roots of hovering aircraft are shown.

Table 1
Characteristic roots of hovering aircraft

Portrait Lateral and lateral Longitudinal and transverse coupling

0.053±0.288i 0 0.0653±0.238i

–0.303 –0.223 –0.318

–0.714 –0.0345±0.478 –1.41

–3.76 0

–0.12

–0.153±0.476

–2.96

Portrait	Lateral and lateral	Longitudinal and transverse coupling
0.053±0.288i	0	0.0653±0.238i
–0.303	–0.223	–0.318
–0.714	–0.0345±0.478	–1.41
	–3.76	0
		–0.12
		–0.153±0.476
		–2.96

It can be seen from Table 1 that the aircraft is unstable in hover state, and there is a conjugate complex root (0.0530.287i) of the positive real part, which is common for aircraft with a single rotor and tail rotor. Through the analysis of the characteristic root, it is found that the main motion variables corresponding to the conjugate complex root (0.0530.287i) are the aircraft speed and pitch angle, which show a slow long period change, the period is 21.9 seconds, and the frequency doubling time (LN2/ 0.053 + 0.287i) is 11.0 seconds, which can meet the requirements of ads-33c flight quality specification.

4.2 Coupling analysis of flight visual simulation system

By comparing the longitudinal and transverse separated characteristic roots and the longitudinal and transverse coupling characteristic roots in the table, it can be found that the two are quite different, which indicates that the longitudinal and transverse coupling of the aircraft is very prominent. This strong coupling is one of the main characteristics of aircraft which is different from fixed-wing aircraft. The main reason for the longitudinal and transverse coupling of aircraft is the coupling between rotor aerodynamic forces, which produces the cross derivatives of longitudinal and transverse coupling.

Aircraft coupling can be divided into two forms: state coupling and control coupling. State coupling means that the change of one state variable will cause the change of other state variables, which is caused by the system matrix A. Manipulation coupling means that a certain control operation (such as total distance) not only changes the corresponding state variables but also affects other state variables, which is caused by control matrix B. The state coupling and control coupling can be analyzed by observing the zero input response and zero state response of the aircraft in hover. As shown in Fig. 3, the vertical velocity initial interference is shown, and Fig. 4 is the initial roll rate interference.

Fig. 3

Vertical velocity initial interference.

Fig. 4

Initial roll rate interference.

In Figs. 3 and 4, p q r w represents roll rate, pitch rate, yaw rate and vertical velocity, respectively. From the curves of zero input response and zero state response, it can be seen that there is a serious coupling phenomenon between helicopter channels. In the aspect of state coupling, there is a serious coupling between roll angle velocity and pitch angular velocity; pitch angular velocity, roll angle velocity and vertical velocity all affect yaw angular velocity, among which vertical velocity has the most serious influence on yaw rate coupling.

In the aspect of handling coupling, longitudinal periodic pitch mainly affects pitch angular velocity, but causes serious coupling between roll angle velocity and yaw angular velocity; lateral periodic pitch mainly affects roll angle velocity, but also causes serious coupling of pitch angular velocity; total pitch mainly affects vertical velocity, but also causes pitch angular velocity, roll angle velocity and pitch angle velocity The results show that the coupling of yaw rate is serious; the control of tail pitch mainly results in the change of yaw rate, but there is a certain coupling in pitch angular velocity and roll angular velocity. To sum up, the coupling between the longitudinal and transverse channels is serious, and the coupling between the total distance channel and the heading channel is serious. The heading channel has certain coupling with the longitudinal channel and the transverse channel respectively.

4.3 Requirement analysis of flight visual system

In order to meet the requirements of the real environment, such as the visualization of the background and the simulation environment [39], the system must have the following requirements in order to meet the requirements of the simulation, such as the visualization of the background and the simulation environment.

It is required to realize the real-time driving of model motion by telemetry data, including the animation of aircraft flying with bomb and weapon flying after bomb-dropping;

When the model moves, it is required to switch any view angle and roam the scene, so that the operator can observe the three-dimensional view of each state point of the weapon according to the requirements of the test task;

It is required to realize real-time two-dimensional display of telemetry data, including state information, GPS information, the main parameter information of sensors, etc [40];

Because of the continuous development of new weapon systems, the simulation software must be compatible with many weapon types, that is, the simulation software should be compatible with the third-party modeling software to improve the modeling efficiency and system expansion compatibility;

It is required that the simulation system should be universal. The simulation system can be transplanted to any computer and can provide new algorithms and simulation support of similar characteristics.

4.4 Simulation example and analysis of flight visual system

When the vertical speed command, roll angle command, pitch angle command and yaw angle speed command are the unit step input at the same time, the unit step response curves under 0 knots, 10 knots, 20 knots and 40 knots mathematical models are obtained respectively, as shown in Fig. 5.

Fig. 5

The closed-loop unit step response curve of heading channel.

It can be seen from Fig. 5 that the four curves in each diagram represent the closed-loop unit step response curve when the controlled object is four linear models. It can be seen from the figure that the four channels have good decoupling. Among them, the heading channel has weak signal tracking ability and has a large static error, and the other three channels have good signal tracking ability [41]. Therefore, the designed control system can be analyzed and designed as the controller sacrifices part of the signal tracking performance to ensure the robustness of the whole system [42]. From the simulation results, it can be concluded that the controller designed by the QFT method can guarantee the robustness of the whole control system even when the mathematical model of the plant is quite different.

Table 2 shows the flight simulation log.

Table 2

Simulation flight log

Time(s)	Longitude	Latitude	Altitude(ft)	Roll	Pitch	Heading
1.225000	37.613544	–122.35724	1.51168	–1.88907	6.08993	298.00170
2.243333	37.613546	–122.35724	11.40219	–2.32785	6.12608	298.36588
3.223333	37.613548	–122.35724	23.74400	–2.46108	5.95368	298.34848
4.433333	37.613552	–122.35723	36.71503	–2.52228	5.76615	298.30381
5.246623	37.613550	–122.35723	34.75403	–2.43048	5.60503	298.24663
6.133000	37.613551	–122.35723	33.71653	–2.44018	5.73563	298.14330
7.243333	37.613550	–122.35723	32.73503	–2.47725	5.76545	298.12887
8.260000	37.613550	–122.357230	32.46253	–2.82253	5.71421	297.70636

Figure 6 is a 3D visual simulation example of the walker.

Fig. 6

Example of aircraft 3D scene simulation.

(Figure from http://fxmnq.cn.baimao.com/supply/)

It can be seen from Fig. 6 a 3D visual simulation example of a certain type of aircraft flight test. The three-dimensional visual simulation system of aircraft flight test has a vivid and fluent picture, display frequency of 30 frames per second, good sound effect, and operators can switch between different views through keyboard and mouse.

5 Conclusion

The 3D visual simulation system of the flight test in this study basically meets the requirements of 3D visual simulation of a digital simulation system for aircraft test and evaluation. Friendly human-computer interface, high frame rate, good fidelity. Through this system, the aircraft designer can directly observe the flight process, flight posture and the target movement form of the aircraft, so as to analyze whether the orbit design is appropriate. It is necessary to further enhance the immersion of the system to refine the scene model and realize the natural transformation of texture. In the future research and development, we can import sky box technology and baking texture technology and other visual reality generation technology.

Visual design of flight program, dynamic and intuitive demonstration, and Simulation of terrain and ground objects when the aircraft is designed to fly according to the program. Using sensor technology and QAR data processing, obstacles, standard instrument entry procedures, standard instrument start-up procedures and entry procedures, as well as restricted areas, restricted areas and other navigation elements that need attention. At the same time, it is convenient for the user to change the model individually, which can dynamically simulate the moving track of the aircraft target, overlook the airport from multiple angles, and dynamically roam according to the flight sequence.

Especially in the initial design stage of the aircraft weapon system, if the actual firing test is not allowed according to the objective conditions, the designer will use the 3D visual simulation system of aircraft flight test to simulate the target shooting test and provide a standard value of an excellent application system in the subsequent space mission.

References

Yang

, Wang

, Jiang

, Song

and Meng

, Visual Perception Enabled Industry Intelligence: State of the Art, Challenges and Prospects, in IEEE Transactions on Industrial Informatics.

Hosman

, Advani

and Haeck

, Integrated design of flight simulator motion cueing systems, Aeronautical Journal 109(1091) (2016), 1–12.

El-Gabri

A.R.

, Ouda

A.N.

, Ahmed

A.S.

, et al., Command Guidance Missile Tracking Algorithms Evaluation Based on Visual Simulation, International Journal of Engineering and Technical Research 7(8) (2018), 110–116.

, The security of the Internet of drones, Computer Communications 148 (2019), 208–214.

Xian

, Zhao

, Zhang

, et al., A low-cost hardware-in-the-loop-simulation testbed of quadrotor UAV and implementation of nonlinear control schemes, Robotica 35(3) (2017), 588–612.

Song

, Rawat

, Jeschke

and Brecher

, Cyber-Physical Systems: Foundations, Principles and Applications, ISBN: 978-0-12-803801-7. Boston, MA: Academic Press (2016), 1–514.

Wang

, Cheng

and Zhou

, A multiple hierarchical structure strategy to quantized control of Markovian switching systems, Applied Mathematics and Computation 373 (2020), ID:125037.

Zhou

, Liu

and Li

, Distributed and real-time integrated simulation system on avionics, International Journal of Aeronautical & Spaceences 18(3) (2017), 574–578.

Sanchez-Lopez

J.L.

, Pestana

, Paloma

D.L.P.

, et al., A Reliable Open-Source System Architecture for the Fast Designing and Prototyping of Autonomous Multi-UAV Systems:, Simulation and Experimentation. Journal of Intelligent & Robotic Systems 84(1–4) (2016), 1–19.

10.

Subscribers Only, Multi-function Visual Simulation System of Aircraft Based on Decoupling and Controller Design, Boletin Tecnico/Technical Bulletin 55(2) (2017), 356–365.

11.

Wang

, Kim

, et al., A visual simulation of ocean floating wind power system, Computer Animation & Virtual Worlds 30(2) (2019), 1–15.

12.

Wan

and Goudas

, Faster R-CNN for Multi-class Fruit Detection using a Robotic Vision System, Computer Networks 107036, (2019).

13.

Iii

J.T.J.A.

, Iii

L.J.P.

, Bailey

R.E.

, et al., Evaluation of head-worn and head-up displays for use in enhanced flight vision system operations, Optical Engineering 58(5) (2019), 051802.1–051802.13.

14.

Magree

and Johnson

E.N.

, Factored Extended Kalman Filter for Monocular Vision-Aided Inertial Navigation, Journal of Aerospace Information Systems 13(12) (2016), 475–490.

15.

, Li

and Li

, Virtual reality geographical interactive scene semantics research for immersive geography learning, Neurocomputing 254 (2017), 71–78.

16.

Bartosh

V.S.

, Belago

I.V.

, D’yakov

M.S.

, et al., Simulation of visual instrumental observations of the Earth from the International Space Station, Optoelectronics Instrumentation & Data Processing 52(3) (2016), 252–258.

17.

Yan

and Lv

, The Influence of Immersive Virtual Reality Systems on Online Social Application, Applied Sciences 10(15) (2020), 5058.

18.

Ishikawa

, Visual Simulation of Magnetohydrodynamics, Journal of the Visualization Society of Japan 36(140) (2016), 19–23.

19.

Benedetto

, Nigro

, Bartolomei

, et al., 1062 Does 3D technology affect the visual system? European Urology Supplements 15(3) (2016), e1062–e1062.

20.

Muellhaeuser

and Leissling

, Development and In-Flight Evaluation of a Haptic Torque Protection, Journal of the American Helicopter Society 64(1) (2019), 012003.1–012003.9.

21.

Gao

, Xuan

H.Z.

, Zhang

, Wan

and Choo

K.K.R.

, Adaptive fusion and category-level dictionary learning model for multi-viewhuman action recognition, IEEE Internet of Things Journal (2019).

22.

Kramer

L.J.

, Etherington

T.J.

, Severance

, et al., Assessing Dual-Sensor Enhanced Flight Vision Systems to Enable EquivalentVisual Operations, Journal of Aerospace Computing Information & Communication 14(10) (2017), 533–550.

23.

Zhao

, Lin

, Peng

, et al., Vision-aided Estimation of Attitude, Velocity, and Inertial Measurement Bias for UAV Stabilization, Journal of Intelligent & Robotic Systems 81(3-4) (2016), 531–549.

24.

Shi

, Tang

, Zhong

, Yin

, Huang...

and Wang

, Nonfragile asynchronous control for uncertain chaotic Lurie network systems with Bernoulli stochastic process, International Journal of Robust and Nonlinear Control 28(5) (2018), 1693–1714.

25.

Lood

, H?Ggblom-Kronl?F

and Dellenborg

, Embraced by the past, hopeful for the future: the meaning of health to ageing persons who have migrated from the Western Balkan region to Sweden, Agng & Society 36(03) (2016), 649–665.

26.

Edvardsson

, Sj?Gren

, Lood

, et al., A person-centered and thriving-promoting intervention in nursing homes - study protocol for the U-Age nursing home multi-center, non-equivalent controlled group before-after trial, Bmc Geriatrics 17(1) (2017), 22.

27.

Yuan

Y.H.

, Tsao

S.H.

, Chyou

J.T.

and Tsai

S.B.

, An empirical study on the effects of electronic word-of-mouth and internet risk avoidance on purchase intention - from the perspective of big data, Soft Computing 24 (2020), 5713–5728.

28.

, Sun

, Yan

, et al., Estimation of exhaust pollution emissions for civil aircraft engine based on QAR data, Chinese Journal of Environmental Engineering 11(6) (2017), 3607–3616.

29.

Masahiro

and Yoshiteru

, Sensors, Vol. 16, Pages: Introduction to the Special Issue on “State-of-the-Art Sensor Technology in Japan ”, Sensors 16(9) (2016), 11045–11048.

30.

Wagner

, Vogt

, Jamal

F.I.

, et al., Application of microwave sensor technology in cardiovascular disease for plaque detection, Current Directions in Biomedical Engineering 2(1) (2016), 273–277.

31.

Liu

, Yao

, Zeng

, et al., An RFID and sensor technology-based warehouse center: assessment of new model on a superstore in China, Assembly Automation 39(1) (2019), 86–100.

32.

Hiligsmann

, Advanced sensor technology is pivotal to next-gen automotive design, ECN Electronic Component News 61(7) (2017), 14–16.

33.

Albahri

O.S.

, Albahri

A.S.

, Mohammed

K.I.

, et al., Systematic Review of Real-time Remote Health Monitoring System in Triage and Priority-Based Sensor Technology: Taxonomy, Open Challenges, Motivation and Recommendations, Journal of Medical Systems 42(5) (2018), 1–27.

34.

Shimizu

, Matsukuma

and Gao

, Optical Angle Sensor Technology Based on the Optical Frequency Comb Laser, Appliedences 10(11) (2020), 4047.

35.

de Carvalho

, Martins

V.H.

, Fabris

L.V.M.

, et al., Promoting optical fiber sensor technology with the educational experimental setup, Physics Education 54(4) (2019), 045005.

36.

Wan

, Zhao

, Wang

, Gu

, Abbasi

Q.H.

and Choo

K.K.R.

, Multi-dimensional data indexing and range query processing via the Voronoi diagram for the internet of things,382–391, Future Generation Computer Systems 91 (2019).

37.

Francis Saviour Devaraj

, Elhoseny

, Dhanasekaran

, Laxmi Lydia

and Shankar

, Hybridization of firefly and Improved Multi-Objective Particle Swarm Optimization algorithm for energy-efficient load balancing in Cloud Computing environments, Journal of Parallel and Distributed Computing, (2020), In Press.

38.

Liu

, Yang

, Sun

, Wu

, Lin...

and Chou

, Enhanced embedding capacity for the SMSD-based data-hiding method, Signal Processing: Image Communication 78 (2019), 216–222.

39.

Tang

and Elhoseny

, Computer network security evaluation simulation model based on neural network, Journal of Intelligent and Fuzzy Systems (2019).

40.

Zhou

, Wang

, Liu

and Tian

, An Information-theoretic View of WLAN Localization Error Bound in GPS-denied Environment, IEEE Transactions on Vehicular Technology 68(4) (2019), 4089–4093.

41.

, Yang

H.J.

, Sun

, Cioffi

J.M.

and Yang

, Adaptive Overhearing in Two-Way Multi-Antenna Relay Channels, IEEE Signal Processing Letters 23(1) (2016), 117–120.

42.

Shi

, Tang

, Liu

and Zhong

, Secondary delay-partition approach on robust performance analysis for uncertain time-varying Lurie nonlinear control system, Optimal Control Applications and Methods 38(6) (2017), 1208–1226.

Flight visual simulation system based on sensor technology and QAR data

Abstract

Keywords

1 Introduction

2 QAR data and flight visual simulation system

2.1 3D flight visual system

2.2 Working principle of the visual system

2.3 Large scale scene model generation

2.4 Collision detection of flight scene

2.5 Control mode of the visual system

2.6 Flight scene roaming control algorithm

3.1 QAR data preprocessing

3.2 Motion modeling of flight scene

3.3 Simulation data analysis

4 Analysis of flight visual simulation system

4.1 Stability analysis of flight visual simulation system

Table 1 Characteristic roots of hovering aircraft Portrait Lateral and lateral Longitudinal and transverse coupling 0.053±0.288i 0 0.0653±0.238i –0.303 –0.223 –0.318 –0.714 –0.0345±0.478 –1.41 –3.76 0 –0.12 –0.153±0.476 –2.96

4.4 Simulation example and analysis of flight visual system

References

Table 1
Characteristic roots of hovering aircraft

Portrait Lateral and lateral Longitudinal and transverse coupling

0.053±0.288i 0 0.0653±0.238i

–0.303 –0.223 –0.318

–0.714 –0.0345±0.478 –1.41

–3.76 0

–0.12

–0.153±0.476

–2.96