Flight simulation stress assessment method and system based on fusion of MR and AI, and medium

Abstract

The invention provides a flight simulation stress assessment method and system based on MR fused AI, and a medium, and relates to the technical field of stress assessment, and the method comprises the steps: constructing a virtual flight environment, fusing a flight simulator with the virtual flight environment, carrying out flight control simulation in the virtual flight environment, and constructing a flight training platform; constructing a stress evaluation model; acquiring real-time physiological data, control data of a flight simulator and simulated flight state data when a pilot executes simulated flight in different flight scenes and different flight stages on a complete time axis in the flight training platform; the real-time physiological data, the control data and the simulated flight state data are extracted to be used for representing mutation characteristics of the cardiovascular stress level, the stress state of the pilot is recognized through a stress evaluation model, and the stress state of the pilot, the flight scene and the flight stage of the stress state and the corresponding stress degree are determined. According to the method, pre-flight stress assessment can be effectively carried out.

Description

Technical Field

[0001] The present invention relates to the technical field of stress assessment, and in particular to a flight simulation stress assessment method, system, and medium based on MR fusion and AI. Background Art

[0002] With the rapid advancement of modern technology and the rapid development of aviation equipment, flight missions have become increasingly complex and diverse. These missions not only require pilots to possess superb flying skills, but also place unprecedented demands on their physical and mental fitness. In modern warfare, flight missions are often accompanied by high-intensity combat environments and extremely high psychological pressure. Pilots are required to make accurate judgments and execute a series of high-precision operations in a very short period of time, which greatly tests their mental endurance and physical coordination.

[0003] In the high-intensity, high-stress flight environment, pilots face not only external challenges but also tests of their own physiological functions. Cardiovascular stress response is a key factor affecting flight safety. When pilots are nervous, anxious, or fearful, their cardiovascular system responds with a series of stress reactions, such as increased heart rate and blood pressure. While these physiological changes can help pilots cope with emergencies to a certain extent, excessive stress responses can cause discomfort such as palpitations, shortness of breath, and dizziness, potentially jeopardizing flight safety.

[0004] Currently, the presence of stress is typically determined by monitoring pilots' physiological data during flight missions. However, individual pilots experience varying levels of stress during different flight phases, necessitating a pre-flight assessment of stressful behaviors to prevent them from occurring during the actual mission. However, existing technologies lack pre-flight stress assessment methods. Summary of the Invention

[0005] The present invention provides a flight simulation stress assessment method based on MR fusion AI. This method comprehensively assesses the pilot's stress level in different flight scenarios and flight phases by simulating physiological response data generated during multiple flight phases, flight simulator control data, and simulated flight status data after flight simulator operation, thereby achieving pre-flight stress assessment. The present invention specifically includes the following steps:

[0006] Build a virtual flight environment, integrate the flight simulator with the virtual flight environment, perform flight control simulation in the virtual flight environment, including gravity, lift and thrust simulation, and build a flight training platform; Adjust the virtual flight environment, flight control simulation data and flight parameters to create different flight scenarios; the virtual flight environment includes multiple virtual environments for complex weather, emergency situations and night flights; Construct a stress assessment model based on neural network; Acquire real-time physiological data and flight simulator control data from pilots performing simulated flights in different flight scenarios and phases on a complete timeline in the flight training platform, and associate them through graph neural networks to form a multidimensional feature matrix. The stress assessment model is used to extract mutation features in the multidimensional feature matrix to characterize the cardiovascular stress level, identify the pilot's stress state, determine the pilot's stress state, the flight scenario and flight stage in which the stress state occurs, and the corresponding stress degree.

[0007] Preferably, the real-time physiological data is collected in real time by a wearable device to obtain real-time physiological data based on a complete time axis; the real-time physiological data includes heart rate, electrocardiogram, blood oxygen saturation, blood pressure and respiratory rate.

[0008] Preferably, the acquisition of the control data of the flight simulator includes the following steps: The sensors installed on the flight simulator collect the pilot's control data in the flight simulation, including climbs, dives, left fuselage movements, right fuselage movements, left turns, right turns, horizontal left turns, and horizontal right turns along the complete timeline. The panel operations of restarting the engine, turning on and off the power, parachuting, and mid-air braking are collected in the flight simulator cockpit. The sensors collect the joystick pressure, control surface response accuracy, heading correction frequency, and panel operation frequency at this time as control data for evaluating cardiovascular stress.

[0009] Preferably, the identification of the pilot's stress state comprises the following steps; Filter, denoise and interpolate the real-time physiological data, and perform data standardization and normalization to obtain pre-processed physiological data; Align physiological data with control data in time series; dynamically associate physiological data with control data through graph neural network modeling to construct a multi-dimensional feature matrix; Construct a convolutional neural network to identify specific features such as mutation features and fluctuation data features in the multidimensional feature matrix as cardiovascular stress response characteristics, determine the pilot's stress state and the time point when cardiovascular stress response occurs; Based on the time point when cardiovascular stress response occurs, determine the flight scenario and flight stage where the stress state occurs and the corresponding stress level.

[0010] Preferably, the construction of the virtual flight environment comprises the following steps: Using Unreal Engine, satellite maps are converted into three-dimensional terrain to construct the flight environment; By combining satellite photos with terrain information to form a terrain map, the terrain map is imported into the flight environment in the form of an elevation map. Using the terrain control, multiple undulating plots are formed in the flight environment based on the terrain map. Multiple plots are stitched together according to longitude and latitude coordinates to form a ground map in the flight environment; The rendering of the ground map is optimized, and a sky box is added to the scene to simulate the sky, parallel light is added to simulate sunlight, and atmospheric fog, exponential fog and particle cloud are set to complete the overall construction of the flight environment.

[0011] Preferably, the rendering optimization of the ground map comprises the following steps: Use pre-load mode for all plots, loading the map at the lowest quality when entering the scene; Based on the pilot's position, the terrain quality is adjusted using a view tracking algorithm: The pilot's visual direction is determined by the view tracking algorithm. For plots in the opposite direction of the view, the lowest display mode is used, reducing the texture quality and the number of model faces. For the plots in the perspective cache area, that is, within a 30° angle outside the 180° directly in front, medium map quality is used; The land in front is rendered at the highest level. The map quality is adjusted according to the display range. The display range is spherical with the user as the center. The highest quality is within 1000 meters, the medium quality is within 5000 meters, and the lowest quality is within 10000 meters. The multi-resolution levels of each plot are managed using a quadtree.

[0012] Preferably, the construction of the flight training platform includes the following steps: MR equipment is used to provide pilots with an interactive virtual flight environment, and the flight simulator and the virtual flight environment are integrated and displayed.

[0013] Preferably, performing flight control simulation in the virtual flight environment includes the following steps: In a virtual flight environment, gravity is simulated through stance nodes. These stance nodes are placed in the scene and have properties of gravity and air resistance. When the controlled object enters these stances, it will be affected accordingly. Simulate lift by calculating the aircraft's lift index based on its pitch and bank angles. When the aircraft's pitch angle is positive, the aircraft will climb upward, while when it is negative, the aircraft will dive downward. When the aircraft's bank angle changes, the lift index affects the aircraft's flight state in a direction 90° perpendicular to the fuselage. Thrust is simulated through the flight simulator's cockpit.

[0014] The present invention also proposes a flight simulation stress assessment system based on MR fusion AI, the system comprising: processor; a memory having stored thereon a computer program executable on the processor; Among them, when the computer program is executed by the processor, the steps of the flight simulation stress assessment method based on MR fusion AI are implemented.

[0015] The present invention also proposes a computer-readable storage medium, on which a data processing program is stored. When the data processing program is executed by a processor, the steps of the flight simulation stress assessment method based on MR fusion AI are implemented.

[0016] Beneficial effects of the present invention: The present invention proposes a flight simulation stress assessment method based on MR fusion AI. The method constructs a virtual flight environment and combines it with a flight simulator to build a virtual and real flight training platform, which can provide a relatively realistic flight training environment for different flight scenarios and flight phases. The method comprehensively assesses the pilot's stress level in different flight scenarios and flight phases by simulating the physiological response data generated during multiple flight phases, the control data of the flight simulator, and the simulated flight status data after the flight simulator is operated, thereby realizing pre-flight stress assessment. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 Flowchart of a flight simulation stress assessment method based on MR fusion AI according to an embodiment of the present invention; Figure 2 This is a system architecture execution logic diagram of the flight simulation stress assessment method based on MR fusion AI in an embodiment of the present invention. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0019] Example 1 This embodiment provides a flight simulation stress assessment method based on MR fusion AI, the flow chart is as follows: Figure 1 As shown, the system architecture execution logic diagram is as follows Figure 2 The specific steps include:

[0020] S1: Build a virtual flight environment, integrate the flight simulator with the virtual flight environment, perform flight control simulation in the virtual flight environment, including gravity, lift and thrust simulation, and build a flight training platform.

[0021] Building a virtual flight environment includes the following steps: S1.1: Use Unreal Engine to convert satellite maps into 3D terrain to construct the flight environment.

[0022] S1.2: Satellite photos are combined with terrain information to form a terrain map, which is then imported into the flight environment as an elevation map. Using terrain controls, multiple undulating plots of land are created in the flight environment based on the terrain map.

[0023] S1.3: Combine multiple plots of land according to their longitude and latitude coordinates to form a ground map of the flight environment.

[0024] S1.4: Optimize the rendering of the ground map, add a skybox to the scene to simulate the sky, add parallel light to simulate sunlight, and set atmospheric fog, exponential fog, and particle cloud to complete the overall construction of the flight environment.

[0025] Furthermore, the rendering optimization of the ground map includes the following steps: all plots are pre-loaded and the map is loaded at the lowest quality when entering the scene; the quality of the plots is adjusted according to the pilot's position through the perspective tracking algorithm: the pilot's visual direction is determined by the perspective tracking algorithm, and for plots in the opposite direction of the perspective, the lowest display mode is adopted to reduce the texture quality and the number of model faces; for plots in the perspective cache area, that is, within a 30° angle range outside 180° in front, medium map quality is adopted; for plots in front, the highest quality is adopted; the map quality is adjusted according to the display range, which is spherical with the user as the center, with the highest quality within 1000 meters, medium quality within 5000 meters, and the lowest quality within 10,000 meters; wherein, the multi-resolution levels of each plot are managed using a quadtree.

[0026] Furthermore, MR equipment is used to provide pilots with a virtual flight environment in which they can interact, and the flight simulator and the virtual flight environment are integrated and displayed.

[0027] Flight control simulation in a virtual flight environment includes the following steps: in the virtual flight environment, gravity is simulated through stand nodes. These stand nodes are placed in the scene and have the properties of gravity and air resistance. When the controlled object enters these stands, it will be affected accordingly; lift is simulated, and the lift index of the aircraft is calculated based on the aircraft's pitch angle and bank angle. When the aircraft's pitch angle is positive, the aircraft will climb upward, and when it is negative, the aircraft will dive downward; when the aircraft's bank angle changes, the effect of the lift index will affect the aircraft's flight state in a direction 90° perpendicular to the fuselage; thrust is simulated through the cockpit of the flight simulator.

[0028] S2: Adjusting the virtual flight environment, flight control simulation data and flight parameters to form different flight scenarios; the virtual flight environment includes multiple virtual environments for complex weather conditions, emergency situations and night flights.

[0029] S3: Construct a stress assessment model based on neural network.

[0030] S4: Obtain the real-time physiological data and flight simulator control data of the pilot when performing simulated flight in different flight scenarios and different flight phases on the complete timeline in the flight training platform, and associate them through the graph neural network to form a multi-dimensional feature matrix.

[0031] Among them, real-time physiological data is collected in real time through wearable devices to obtain real-time physiological data based on a complete timeline. Real-time physiological data includes heart rate, electrocardiogram, blood oxygen saturation, blood pressure and respiratory rate.

[0032] The sensors installed on the flight simulator collect the pilot's control data in the flight simulation, including climbs, dives, left fuselage movements, right fuselage movements, left turns, right turns, horizontal left turns, and horizontal right turns along the complete timeline. The panel operations of restarting the engine, turning on and off the power, parachuting, and mid-air braking are collected in the flight simulator cockpit. The sensors collect the joystick pressure, control surface response accuracy, heading correction frequency, and panel operation frequency at this time as control data for evaluating cardiovascular stress.

[0033] S5: Extract mutation features from the multidimensional feature matrix used to characterize cardiovascular stress levels through the stress assessment model, identify the pilot's stress state, determine the pilot's stress state, the flight scenario and flight phase in which the stress state occurs, and the corresponding stress degree.

[0034] Specifically: Identifying the pilot's stress state includes the following steps; S5.1: Filter, denoise, and interpolate the real-time physiological data, and perform data standardization and normalization to obtain preprocessed physiological data; S5.2: Align the physiological data with the manipulation data in time series; dynamically associate the physiological data with the manipulation data through graph neural network modeling to construct a multidimensional feature matrix; S5.3: Construct a convolutional neural network to identify specific features such as mutation features and fluctuation data features in the multidimensional feature matrix as cardiovascular stress response features, and determine the pilot's stress state and the time point when the cardiovascular stress response occurs; S5.4: Based on the time point when cardiovascular stress response occurs, determine the flight scenario and flight stage in which the stress state occurs and the corresponding stress level.

[0035] The present invention collects physiological data through wearable physiological detection equipment, specifically: Wearable physiological monitoring device hardware is a set of intelligent devices that integrate advanced sensor technology, data processing, and transmission capabilities. They are primarily used to monitor and record various physiological indicators of the human body in real time. These devices are typically lightweight and wearable, making them easy for users to wear during daily activities or in specific testing environments. The following is a detailed description of the wearable physiological monitoring device hardware:

[0036] (1) Sensor technology Multi-sensor fusion: The device has multiple built-in sensors, such as electrocardiogram (ECG), blood oxygen saturation (SpO2), electrical skin conduction (EDA), respiration (RESP), skin temperature (SKT), and a nine-axis attitude sensor, which can synchronously collect multi-channel physiological signals.

[0037] High-precision measurement: The sensor is precisely calibrated to ensure the accuracy and reliability of measurement data, such as precise monitoring of heart rate and real-time measurement of blood oxygen saturation.

[0038] (2) Lightweight and wearable design Portability: The device is easy to carry and wear, and can be adapted to the flight environment without affecting the flight crew's operations.

[0039] Comfort: Made of soft, skin-friendly material, it ensures users’ comfort for long-term wearing and reduces skin irritation and discomfort.

[0040] (3) Data collection and transmission Real-time data acquisition: Built-in high-performance processor, capable of collecting physiological data collected by sensors in real time, extracting key parameters and performing preliminary classification.

[0041] Wireless transmission: Supports wireless communication technologies such as Bluetooth and Wi-Fi to achieve instant data transmission to smartphones, tablets or cloud servers, allowing users to view and analyze data at any time.

[0042] Wearable physiological monitoring devices rely on a variety of advanced sensing technologies and algorithms to measure cardiovascular data, including: (1) Photoplethysmography (PPG) for heart rate measurement: An optical sensor emits green or red light that penetrates the skin and is absorbed by the blood. As the heart beats, blood flow and volume change, causing the amount of light absorbed to also change. The sensor monitors these changes in light, indirectly reflecting the heartbeat waveform and calculating the heart rate.

[0043] (2) Oscillometric method of blood pressure measurement: Parameters such as static pressure and repeatability of blood pressure indications are tested using standard pressure gauges, blood pressure simulators and other equipment to improve accuracy.

[0044] (3) Photoelectric sensing technology for measuring blood oxygen saturation: The device emits red and infrared light and measures how these lights are absorbed by the blood after passing through the skin tissue. Since oxyhemoglobin and deoxyhemoglobin have different light absorption capacities, an estimated blood oxygen saturation value can be calculated by measuring the ratio of red light to infrared light in the reflected light.

[0045] (4) Electrocardiogram (ECG) measurement: Single-lead or multi-lead ECG: ECG monitoring devices typically use single-lead or multi-lead ECG technology to collect ECG signals through ECG electrodes and transmit these signals to the device for processing and analysis. These devices can monitor the heart's electrical activity in real time and provide key information such as heart rate, heart rhythm, and ST segment changes.

[0046] (5) Respiratory rate measurement: Respiratory rate is detected by measuring the rise and fall of the chest and abdomen or changes in airflow. These methods can provide relatively accurate respiratory rate data and help assess the health of the respiratory system.

[0047] The flight simulator of the present invention can simulate various flight scenarios including but not limited to flight instrument operation, emergency landing procedures, and flight under complex weather conditions, so as to comprehensively test the pilot's cardiovascular stress response in different situations.

[0048] The flight simulator perfectly replicates in-flight maneuvers such as climbing, diving, left and right fuselage positions, turning left and right, horizontal left and right turns, and cockpit controls such as restarting the engine, turning the power on and off, parachuting, and airbrakes are all replicated in the cockpit. Mixed reality technology achieves a superposition of real and virtual, enabling the virtual environment to simultaneously simulate the pilot's in-flight operations in the real world, allowing the pilot to control the aircraft directly from the physical cockpit. This further enhances the realism of the controls and allows the pilot to address stressors through corresponding cockpit functions (restarting the engine, resetting the power, airbrakes, etc.).

[0049] This invention uses MR technology to provide pilots with an interactive virtual flight environment, integrating the flight simulator with the virtual flight environment. Specifically, it utilizes mixed reality technology to construct a highly realistic flight simulation environment, including varying climates, altitudes, speeds, and other conditions, allowing pilots to experience the flight process immersively. Through the MR technology, pilots can operate within the virtual environment and interact with it in real time, simulating the stresses of flight. The real and virtual environments are displayed simultaneously, integrating the real cockpit environment with the virtual flight environment through technology, making it possible to control the flight directly from the actual simulated cockpit.

[0050] The main technical features used are: MR technology can not only overlay virtual information on the real environment, but also create a completely virtual immersive environment. In flight simulation, MR technology uses head-mounted displays (HMDs), projection mapping, gesture recognition, and other equipment to enable pilots to simultaneously see the real cockpit layout and virtual flight scenes, such as clouds, terrain, and other aircraft, providing an unprecedented training experience.

[0051] High-precision modeling is fundamental to creating a realistic flight environment. This includes meticulous modeling of aircraft models, terrain, weather conditions, airport facilities, and more. Leveraging 3D scanning, computer graphics, and physics engine technology, we can create highly realistic aircraft interiors and exteriors, complex and dynamic weather systems, and precise surface details. This high-precision modeling not only enhances the realism of simulations but also ensures the accuracy and reliability of the results.

[0052] Flight dynamics simulation is key to simulating aircraft behavior during flight. Complex mathematical models and algorithms are used to simulate aircraft performance under various conditions, such as takeoff, climb, cruise, descent, landing, and maneuvering. These models consider multiple factors, including aerodynamics, mass distribution, engine performance, and control systems, ensuring that simulated flight behavior conforms to the laws of actual physics.

[0053] Sensors and interaction technologies serve as the bridge between pilots and the simulation environment. Sensors capture pilots' operational commands (such as joystick, throttle, and switches) and physiological responses (such as heart rate and eye movements), while interaction technologies feed this information back to the simulation system in real time, generating corresponding visual, auditory, and tactile feedback. This two-way interaction enhances the immersiveness and interactivity of the simulation, allowing pilots to interact more naturally with the simulated environment.

[0054] Data processing and analysis are crucial for optimizing simulation performance and evaluating training outcomes. By collecting and analyzing large amounts of data from simulations (such as flight trajectories, operational records, and performance indicators), pilot skill levels can be assessed, potential issues identified, and improvement recommendations made. Furthermore, this data can be used to optimize simulation models, improving accuracy and efficiency.

[0055] Environmental simulation and generation technology creates diverse flight scenarios. This includes simulations of different times (day and night, seasons), weather conditions (sunny, rainy, foggy, etc.), terrains (oceanic, mountainous, urban), and special environments (e.g., electromagnetic interference, war zones). Dynamic generation and real-time rendering ensure a high degree of realism and diversity in the simulated environments, meeting diverse training and testing requirements.

[0056] This embodiment also provides a computer-readable storage medium, which stores a computer program that can be used to execute the above Figure 1 Provided is a flight simulation stress assessment method based on MR fusion AI.

[0057] Those skilled in the art will appreciate that all or part of the processes in the above-described method embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the above-described method embodiments. Any reference to memory, storage, database, or other media used in the embodiments provided herein may include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory, or optical storage. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0058] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A flight simulation stress assessment method based on MR fusion AI, characterized in that: The following steps are involved: Build a virtual flight environment, integrate the flight simulator with the virtual flight environment, perform flight control simulation in the virtual flight environment, including gravity, lift and thrust simulation, and build a flight training platform; Adjust the virtual flight environment, flight control simulation data and flight parameters to create different flight scenarios; the virtual flight environment includes multiple virtual environments for complex weather, emergency situations and night flights; Construct a stress assessment model based on neural network; Acquire real-time physiological data and flight simulator control data from pilots performing simulated flights in different flight scenarios and phases on a complete timeline in the flight training platform, and associate them through graph neural networks to form a multidimensional feature matrix. The stress assessment model is used to extract mutation features in the multidimensional feature matrix to characterize the cardiovascular stress level, identify the pilot's stress state, determine the pilot's stress state, the flight scenario and flight stage in which the stress state occurs, and the corresponding stress degree.

2. The flight simulation stress assessment method based on MR fusion AI according to claim 1, characterized in that: The real-time physiological data is collected in real time by wearable devices to obtain real-time physiological data based on a complete timeline; the real-time physiological data includes heart rate, electrocardiogram, blood oxygen saturation, blood pressure and respiratory rate.

3. The flight simulation stress assessment method based on MR fusion AI according to claim 2, characterized in that: The acquisition of the control data of the flight simulator includes the following steps: The sensors installed on the flight simulator collect the pilot's control data in the flight simulation, including climbs, dives, left fuselage movements, right fuselage movements, left turns, right turns, horizontal left turns, and horizontal right turns along the complete timeline. The panel operations of restarting the engine, turning on and off the power, parachuting, and mid-air braking are collected in the flight simulator cockpit. The sensors collect the joystick pressure, control surface response accuracy, heading correction frequency, and panel operation frequency at this time as control data for evaluating cardiovascular stress.

4. The flight simulation stress assessment method based on MR fusion AI according to claim 3 is characterized in that: said identifying the pilot's stress state, The following steps are included; Filter, denoise and interpolate the real-time physiological data, and perform data standardization and normalization to obtain pre-processed physiological data; Temporally align physiological data with manipulation data; Through graph neural network modeling, physiological data and control data are dynamically associated to construct a multi-dimensional feature matrix; Construct a convolutional neural network to identify specific features such as mutation features and fluctuation data features in the multidimensional feature matrix as cardiovascular stress response characteristics, determine the pilot's stress state and the time point when cardiovascular stress response occurs; Based on the time point when cardiovascular stress response occurs, determine the flight scenario and flight stage where the stress state occurs and the corresponding stress level.

5. The flight simulation stress assessment method based on MR fusion AI according to claim 1, characterized in that: The construction of the virtual flight environment includes the following steps: Using Unreal Engine, satellite maps are converted into three-dimensional terrain to construct the flight environment; By combining satellite photos with terrain information to form a terrain map, the terrain map is imported into the flight environment in the form of an elevation map. Using the terrain control, multiple undulating plots are formed in the flight environment based on the terrain map. Multiple plots are stitched together according to longitude and latitude coordinates to form a ground map in the flight environment; The rendering of the ground map is optimized, and a sky box is added to the scene to simulate the sky, parallel light is added to simulate sunlight, and atmospheric fog, exponential fog and particle cloud are set to complete the overall construction of the flight environment.

6. The flight simulation stress assessment method based on MR fusion AI according to claim 5, characterized in that: The rendering optimization of the ground map includes the following steps: Use pre-load mode for all plots, loading the map at the lowest quality when entering the scene; Based on the pilot's position, the terrain quality is adjusted using a view tracking algorithm: The pilot's visual direction is determined by the view tracking algorithm. For plots in the opposite direction of the view, the lowest display mode is used, reducing the texture quality and the number of model faces. For the plots in the perspective cache area, that is, within a 30° angle outside the 180° directly in front, medium map quality is used; The land in front is rendered at the highest level. The map quality is adjusted according to the display range. The display range is spherical with the user as the center. The highest quality is within 1000 meters, the medium quality is within 5000 meters, and the lowest quality is within 10000 meters. The multi-resolution levels of each plot are managed using a quadtree.

7. The flight simulation stress assessment method based on MR fusion AI according to claim 6, characterized in that: The construction of the flight training platform includes the following steps: MR equipment is used to provide pilots with an interactive virtual flight environment, and the flight simulator and the virtual flight environment are integrated and displayed.

8. The flight simulation stress assessment method based on MR fusion AI according to claim 6, characterized in that: The flight control simulation in the virtual flight environment includes the following steps: In a virtual flight environment, gravity is simulated through stance nodes. These stance nodes are placed in the scene and have properties of gravity and air resistance. When the controlled object enters these stances, it will be affected accordingly. Simulate lift by calculating the aircraft's lift index based on its pitch and bank angles. When the aircraft's pitch angle is positive, the aircraft will climb upward, while when it is negative, the aircraft will dive downward. When the aircraft's bank angle changes, the lift index affects the aircraft's flight state in a direction 90° perpendicular to the fuselage. Thrust is simulated through the flight simulator's cockpit.

9. A flight simulation stress assessment system based on MR fusion AI, characterized by: The system comprises: processor; a memory having stored thereon a computer program executable on the processor; Wherein, when the computer program is executed by the processor, the steps of the flight simulation stress assessment method based on MR fusion AI as described in any one of claims 1 to 8 are implemented.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a data processing program, which, when executed by a processor, implements the steps of the flight simulation stress assessment method based on MR fusion AI as described in any one of claims 1 to 8.

Images