Full-motion flight simulator instructor seat adaptive control method, system, and apparatus

Abstract

The present application belongs to the technical field of flight simulation training equipment control, and specifically relates to a full-motion flight simulator instructor seat adaptive control method, system and equipment. It aims to solve the problem of passive safety protection, high-frequency vibration interference and lack of dynamic working condition adaptive ability in the prior art, which leads to the risk of instructor posture imbalance. The present application includes: by real-time acquisition of platform motion data, instructor skeletal key points and pressure distribution data, after multi-source fusion processing, the current flight working condition is identified, and the weight proportion of each data source in the deviation calculation is dynamically determined. By calculating the deviation between the weighted current posture of the instructor and the standard posture, a multi-directional adaptive adjustment instruction is generated to drive the seat to adjust the displacement and angle, and based on the residual deviation, a closed-loop adjustment is carried out until the accuracy requirement is met, realizing active safety protection and intelligent adaptive adjustment of the seat.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of flight simulation training equipment control, and particularly relates to a full-motion flight simulator instructor seat adaptive control method, system and device. BACKGROUND

[0002] The full-motion flight simulator is the core equipment of modern pilot training, which can simulate various flight attitudes and working conditions with high fidelity through a six-degree-of-freedom motion platform. The safety, comfort and stability of the instructor seat, which is the key position for the instructor to observe, evaluate and guide the pilot's operation, directly affect the training effect and teaching quality.

[0003] However, the current control system of the full-motion flight simulator instructor seat has obvious deficiencies, mainly in the following aspects:

[0004] Firstly, the safety protection mechanism is missing. The traditional instructor seat only relies on physical safety belts for fixation, and lacks an active safety system based on real-time attitude monitoring. When simulating complex working conditions such as high-frequency vibration and large-angle roll, the instructor may easily adopt non-safe postures such as excessive forward leaning, lateral leaning or even standing up due to focusing on teaching observation, which poses a significant risk of injury due to imbalance.

[0005] Secondly, the adaptability of data acquisition and control is insufficient. Firstly, the data precision and real-time performance are poor: in a high-frequency and high-vibration environment, a single sensor is easily disturbed and has large data noise. The complex algorithm used to compensate for errors introduces unacceptable system delay, causing attitude calculation to drift and failing to meet the demand for real-time and accurate control. Secondly, dynamic adjustment and safety mechanism are missing: the existing seat cannot identify the current flight working condition (such as cruising, jolting, rolling), nor can it dynamically adjust the control strategy (such as prioritizing comfort or safety) according to different working conditions, and lacks a closed-loop control and hierarchical safety linkage mechanism based on attitude data.

[0006] Finally, the system maintenance convenience and traceability are poor. Sensor calibration relies on manual work, and fault diagnosis and positioning takes too long, making it difficult to meet the high availability and high reliability requirements of flight training equipment.

[0007] Therefore, an integrated solution that integrates high-precision data acquisition, intelligent working condition recognition, real-time dynamic matching control and active safety protection is urgently needed to completely solve the above problems and improve the safety and teaching quality of flight simulation training. SUMMARY

[0008] In order to solve the above problems in the prior art, that is, the passive safety protection, high-frequency vibration interference and lack of dynamic working condition adaptive capability in the prior art, resulting in the risk of imbalance of the instructor's posture, the present application provides a full-movement flight simulator instructor seat adaptive control method, system and equipment.

[0009] In a first aspect of the present application, a full-movement flight simulator instructor seat adaptive control method is provided, which comprises:

[0010] Real-time acquisition of motion data of the simulator platform, instructor's skeletal key point coordinates and sitting posture pressure distribution data;

[0011] Temporal and spatial alignment and multi-source data fusion processing are performed on the motion data and the skeletal key point coordinates and sitting posture pressure distribution data to obtain fused device attitude data and instructor attitude data;

[0012] Based on the fused device attitude data and instructor attitude data, the current flight condition is identified, and according to the type of the identified current flight condition, the weight proportion of the device attitude data, the instructor attitude data and the seat attitude data in calculating the attitude deviation is dynamically determined;

[0013] The attitude deviation between the current attitude of the instructor and the standard attitude of the corresponding working condition is calculated, wherein the current attitude of the instructor is obtained by weighted calculation of the fused instructor attitude data based on the weight proportion;

[0014] Based on the attitude deviation and the working condition type, a multi-directional adaptive adjustment instruction of the seat is generated and sent to the seat actuator to drive the seat to displace and adjust the angle;

[0015] The adjusted actual attitude data is collected and compared with the target adjustment value, and if there is a residual deviation, the generation and execution of the adjustment instruction is retriggered until the residual deviation meets the accuracy requirement.

[0016] Further, the fused device attitude data and instructor attitude data are obtained by:

[0017] The platform motion data and the skeletal key point coordinates are time-stamped calibrated and spatially converted to obtain time-space aligned data;

[0018] The device displacement data and the device angle data are extracted from the aligned platform motion data, and the instructor displacement data and the instructor angle data are extracted from the aligned skeletal key point coordinates;

[0019] The displacement fusion weight of the device displacement data and the instructor displacement data, and the angle fusion weight of the device angle data and the instructor angle data are dynamically adjusted according to the real-time acceleration of the platform;

[0020] The device displacement data and the instructor displacement data are fused based on a displacement fusion weight to obtain device posture displacement data, and the device angle data and the instructor angle data are fused based on an angle fusion weight to obtain device posture angle data; and the device posture displacement data and the device posture angle data are taken as the device posture data;

[0021] The key point coordinates of the skeleton are vibration compensated according to the platform vibration acceleration data to obtain compensated instructor posture data, and the compensated instructor posture data is associated and fused with the pressure distribution data to obtain fused instructor posture data.

[0022] Further, the displacement fusion weight of the device displacement data and the instructor displacement data and the angle fusion weight of the device angle data and the instructor angle data are dynamically adjusted according to the real-time acceleration of the platform, and the method is as follows:

[0023] The fusion weight of the device displacement data is set to be inversely proportional to the real-time acceleration of the platform, and the fusion weight of the instructor displacement data is set to be proportional to the real-time acceleration of the platform;

[0024] The fusion weight of the device angle data is set to be inversely proportional to the real-time acceleration of the platform, and the fusion weight of the instructor angle data is set to be proportional to the real-time acceleration of the platform;

[0025] The weight of the seat posture data is a fixed value.

[0026] Further, the compensated instructor posture data is associated and fused with the pressure distribution data, and the method is as follows:

[0027] The forward inclination angle of the instructor's torso is calculated according to the compensated instructor posture data, and the pressure distribution ratio between different areas of the seat cushion is calculated according to the pressure distribution data;

[0028] The correlation between the forward inclination angle and the pressure distribution ratio is established, and when the forward inclination angle exceeds a first angle threshold and the pressure distribution ratio exceeds a first pressure ratio threshold, it is determined that the instructor is in a forward leaning posture;

[0029] Based on the determination result of the correlation, a comprehensive adjustment instruction containing the seat forward movement amount and the backrest angle adjustment amount is generated, and the adjustment instruction is output as the fused instructor posture data.

[0030] Further, the current flight working condition is identified based on the fused device posture data and the instructor posture data, and the weight proportion of the device posture data, the instructor posture data and the seat posture data in calculating the posture deviation is dynamically determined according to the type of the identified current flight working condition, and the method is as follows:

[0031] extracting a multi-dimensional feature vector including a vibration frequency, an acceleration component and a posture change rate from the fused device posture data and the instructor posture data;

[0032] inputting the multi-dimensional feature vector into a pre-trained flight condition classification model for recognition to obtain a type of the current flight condition;

[0033] According to the type of the current flight condition identified, the weight proportions of the device posture data, the instructor posture data and the seat posture data are dynamically allocated when generating the multi-directional adaptive adjustment instruction of the seat.

[0034] Further, the device posture data weight is a first reference value minus a first adjustment amount related to the real-time acceleration value of the platform, the instructor posture data weight is a second reference value plus a second adjustment amount related to the real-time acceleration value of the platform, and the seat posture data weight is a preset fixed value.

[0035] Further, the posture deviation between the current posture of the instructor and the corresponding standard posture of the condition is calculated, and a multi-directional adaptive adjustment instruction of the seat is generated based on the posture deviation and the condition type, and the method is:

[0036] calling a standard sitting posture parameter corresponding to the type of the current flight condition from a pre-stored standard posture database, the standard sitting posture parameter including a standard skeleton key point coordinate and a standard pressure distribution threshold;

[0037] calculating the displacement deviation and the angle deviation between the real-time skeleton key point coordinate in the fused instructor posture data and the standard skeleton key point coordinate;

[0038] Based on the displacement deviation, the angle deviation and the type of the current flight condition, the displacement adjustment amount and the angle adjustment amount of the seat in each axis are calculated in real time by a parameter adaptive model to form the multi-directional adaptive adjustment instruction.

[0039] Further, the motion data of the simulator platform is collected in real time based on a device posture acquisition sensor system, and the skeleton key point coordinates and the sitting posture pressure distribution data of the instructor are collected in real time by an instructor posture acquisition sensor system.

[0040] The second aspect of the present application proposes a full-motion flight simulator instructor seat adaptive control system for realizing a full-motion flight simulator instructor seat adaptive control method, and the system comprises:

[0041] The data acquisition module is configured to collect the motion data of the simulator platform, the skeleton key point coordinates of the instructor and the sitting posture pressure distribution data in real time.

[0042] a data fusion module configured to perform spatio-temporal alignment and multi-source data fusion processing on the motion data, the skeleton key point coordinates and the sitting posture pressure distribution data, to obtain fused device posture data and instructor posture data;

[0043] a weight proportion allocation module configured to identify a current flight working condition based on the fused device posture data and the instructor posture data, and dynamically determine a weight proportion of the device posture data, the instructor posture data and the seat posture data in calculating a posture deviation according to a type of the identified current flight working condition;

[0044] a weighting module configured to calculate a posture deviation between an instructor current posture and a corresponding working condition standard posture, wherein the instructor current posture is obtained by weighting the fused instructor posture data based on the weight proportion;

[0045] an adjustment module configured to generate a multi-directional self-adaptive adjustment instruction of the seat based on the posture deviation and the working condition type, and send the instruction to a seat actuating mechanism to drive the seat to displace and adjust an angle;

[0046] a deviation correction module configured to collect actual posture data after adjustment and compare the actual posture data with a target adjustment value, and if there is a residual deviation, re-trigger generation and execution of the adjustment instruction until the residual deviation meets an accuracy requirement.

[0047] In a third aspect, the present application provides an electronic device, comprising:

[0048] at least one processor; and

[0049] a memory communicatively connected to the at least one processor; wherein

[0050] the memory stores instructions executable by the processor, and the instructions are used to be executed by the processor to implement the full-movement flight simulator instructor seat adaptive control method.

[0051] The present application has the following advantages:

[0052] The present application effectively solves the core problems of lack of safety protection and insufficient dynamic adaptability of the traditional instructor seat by multi-sensor data fusion and working condition adaptive control. By real-time collection and fusion processing of platform motion data and instructor posture data, the accuracy and reliability of the data in a high-frequency vibration environment are significantly improved, laying a solid foundation for precise control.

[0053] Based on the fused data, the flight working condition is intelligently identified and the weight proportion is dynamically adjusted, so that the system can generate optimal adjustment instructions in different working conditions such as cruising, jolting and rolling, and realize the fundamental change from "passive fixation" to "active adaptation".

[0054] By calculating the posture deviation and driving the actuator to adjust in multiple directions, while introducing a closed-loop feedback mechanism, it is ensured that the seat posture can continuously and accurately match the platform movement and the needs of the instructor, completely eliminating the risk of posture imbalance.

[0055] This method not only greatly improves the safety and comfort of the instructor during the simulation training process, but also significantly reduces the dependence of the system on manual maintenance through its highly automated intelligent control process, providing reliable technical support for efficient and high-quality flight simulation training. BRIEF DESCRIPTION OF DRAWINGS

[0056] Other features, objects and advantages of the application will become more apparent from the following detailed description of non-limiting embodiments, made with reference to the attached drawings:

[0057] Figure 1 is a flowchart of a full-motion flight simulator instructor seat adaptive control method of the application;

[0058] Figure 2 is a structural diagram of a computer system of a server for implementing the method, system and device embodiments of the application. DETAILED DESCRIPTION

[0059] The application 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 intended to explain the related application, and not to limit the application. In addition, it should be noted that only parts related to the application are shown in the drawings for ease of description.

[0060] It should be noted that the embodiments and features in the embodiments of the application can be combined with each other without conflict. The application will be described in detail below with reference to the accompanying drawings and embodiments.

[0061] The application provides a first embodiment, which proposes a full-motion flight simulator instructor seat adaptive control method, which comprises:

[0062] Step S10, real-time acquisition of motion data of the simulator platform, bone key point coordinates of the instructor and sitting posture pressure distribution data;

[0063] Step S20, time-space alignment and multi-source data fusion processing of the motion data and the bone key point coordinates and sitting posture pressure distribution data to obtain fused device posture data and instructor posture data;

[0064] Step S30, based on the fused device attitude data and the instructor attitude data, a current flight working condition is identified, and according to the type of the identified current flight working condition, the weight proportions of the device attitude data, the instructor attitude data and the seat attitude data in calculating the attitude deviation are dynamically determined;

[0065] Step S40, an attitude deviation between the current attitude of the instructor and the standard attitude of the corresponding working condition is calculated, wherein the current attitude of the instructor is calculated by weighting the fused instructor attitude data based on the weight proportions;

[0066] Step S50, a multi-directional adaptive adjustment instruction of the seat is generated based on the attitude deviation and the working condition type, and is sent to a seat actuator to drive the seat to displace and adjust the angle;

[0067] Step S60, the adjusted actual attitude data is collected and compared with the target adjustment value, and if there is a residual deviation, the generation and execution of the adjustment instruction are retriggered until the residual deviation meets the accuracy requirement.

[0068] In order to more clearly describe the adaptive control method of the instructor seat of the full-motion flight simulator, the following will be combined with Figure 1 The steps in the embodiment of the present application are described in detail as follows:

[0069] Step S10, real-time collection of motion data of the simulator platform, bone key point coordinates of the instructor and sitting posture pressure distribution data;

[0070] In this embodiment, the motion data of the simulator platform is collected in real time based on a device attitude acquisition sensor system, and the bone key point coordinates of the instructor and the sitting posture pressure distribution data are collected in real time through an instructor attitude acquisition sensor system.

[0071] The training device attitude acquisition sensor system deployed on the six-degree-of-freedom motion platform collects various motion data of the simulator platform in real time, and the system is composed of a high-precision IMU, a three-axis laser displacement sensor, a three-axis magnetic encoder, a high-frequency vibration sensor and a backup acceleration sensor. Among them, the high-precision IMU is installed on a titanium alloy shock-absorbing support at the center of the platform table, which is used to collect three-axis acceleration and angular velocity data of the platform; the laser displacement sensor is fixed on the platform base to obtain absolute displacement data by detecting the displacement change of the platform table reflecting plate; the magnetic encoder is installed at each attitude joint of the platform to provide an absolute angle reference; the high-frequency vibration sensor and the backup acceleration sensor are redundantly arranged on the table to collect high-frequency vibration signals and perform vibration intensity grading. All sensor data is transmitted through an EtherCAT bus and time-synchronized using a TSN protocol.

[0072] Synchronously, the instructor's skeletal key point coordinates and sitting posture pressure distribution data are collected in real time by an instructor posture collection sensor system deployed in the cockpit. The system includes three multi-view infrared dynamic capture cameras arranged on the ceiling and side walls of the cockpit to collect the three-dimensional coordinates of 16 skeletal key points of the instructor at a frame rate of 240 fps; an auxiliary depth camera installed above the instrument panel to supplement the skeletal data when the infrared camera is blocked; a posture feedback sensor installed on the seat connecting shaft to monitor the seat body posture; and a thin film pressure distribution sensor laid on the seat cushion and back surface to collect pressure data from 16 regions at a sampling rate of 200 Hz. After all sensor data is denoised and timestamp calibrated, it is converted to a coordinate system with the seat cushion center as the origin through a unified spatial coordinate mapping matrix to provide standardized input for subsequent data fusion processing.

[0073] In step S20, the motion data, the skeletal key point coordinates and the sitting posture pressure distribution data are subjected to spatio-temporal alignment and multi-source data fusion processing to obtain fused device posture data and instructor posture data.

[0074] In this embodiment, the fused device posture data and instructor posture data are obtained by the following method:

[0075] In step S21, the platform motion data and the skeletal key point coordinates are subjected to timestamp calibration and spatial coordinate conversion to obtain spatio-temporal alignment data.

[0076] In step S22, the device displacement data and the device angle data are extracted from the aligned platform motion data, and the instructor displacement data and the instructor angle data are extracted from the aligned skeletal key point coordinates.

[0077] In step S23, the displacement fusion weight of the device displacement data and the instructor displacement data, and the angle fusion weight of the device angle data and the instructor angle data are dynamically adjusted according to the real-time acceleration of the platform.

[0078] In this embodiment, the displacement fusion weight of the device displacement data and the instructor displacement data, and the angle fusion weight of the device angle data and the instructor angle data are dynamically adjusted according to the real-time acceleration of the platform by the following method:

[0079] In step S231, the fusion weight of the device displacement data is set to be inversely proportional to the first reference value and the real-time acceleration value of the platform, and the fusion weight of the instructor displacement data is set to be directly proportional to the second reference value and the real-time acceleration value of the platform.

[0080] In step S232, the fusion weight of the device angle data is set to be inversely proportional to the third reference value and the real-time acceleration value of the platform, and the fusion weight of the instructor angle data is set to be directly proportional to the fourth reference value and the real-time acceleration value of the platform. The weight of the seat posture data is a fixed value.

[0081] In step S24, the device displacement data and the instructor displacement data are fused based on a displacement fusion weight to obtain device posture displacement data, and the device angle data and the instructor angle data are fused based on an angle fusion weight to obtain device posture angle data; and the device posture displacement data and the device posture angle data are taken as the device posture data.

[0082] In step S25, the skeletal key point coordinates are vibration compensated according to the platform vibration acceleration data to obtain compensated instructor posture data, and the compensated instructor posture data and the pressure distribution data are associated and fused to obtain fused instructor posture data.

[0083] The compensated instructor posture data and the pressure distribution data are associated and fused, specifically as follows:

[0084] In step S251, a forward inclination angle of an instructor torso is calculated according to the compensated instructor posture data, and a pressure distribution ratio between different regions of a seat cushion is calculated according to the pressure distribution data.

[0085] In step S252, an association between the forward inclination angle and the pressure distribution ratio is established, and when the forward inclination angle exceeds a first angle threshold and the pressure distribution ratio exceeds a first pressure ratio threshold, it is determined that the instructor is in a forward leaning posture.

[0086] In step S253, based on the determination result of the association, a comprehensive adjustment instruction including a seat forward movement amount and a backrest angle adjustment amount is generated, and the adjustment instruction is output as the fused instructor posture data.

[0087] Specifically, first, the platform motion data and the skeletal key point coordinates are timestamp calibrated and space coordinate converted to realize time-space alignment, wherein the timestamp calibration is based on a TSN time synchronization signal to ensure a time synchronization error ≤0.5 ms, and the space coordinate conversion converts data collected by different sensors to a unified coordinate system with the center of the seat cushion as the origin through a preset sensor space coordinate mapping matrix, and a space alignment error is also ≤0.5 ms.

[0088] The device displacement data and the device angle data are extracted from the aligned platform motion data, the device displacement data is mainly derived from a three-axis laser displacement sensor and a high-precision IMU, and the device angle data is derived from a three-axis magnetic encoder and an IMU; and the instructor displacement data and the instructor angle data are extracted from the aligned skeletal key point coordinates, and the instructor displacement data and the angle data are calculated from 16 skeletal key point coordinates collected by a multi-view infrared dynamic capture camera and an auxiliary depth camera.

[0089] The displacement fusion weights of equipment displacement data and instructor displacement data, as well as the angle fusion weights of equipment angle data and instructor angle data, are dynamically adjusted based on the platform's real-time acceleration. The fusion weight for equipment displacement data is set to a first baseline value of 0.6, inversely proportional to the platform's real-time acceleration value. The specific calculation formula is W. 设备 =0.6-0.4×a / (2g), the fusion weight of the instructor displacement data is set to the second reference value of 0.3, which is proportional to the real-time acceleration value of the platform. The specific calculation formula is W. 教员 =0.3+0.4×a / (2g), the fusion weight of the equipment angle data is set to the third benchmark value of 0.95, which is inversely proportional to the real-time acceleration value of the platform; the fusion weight of the instructor angle data is set to the fourth benchmark value of 0.05, which is directly proportional to the real-time acceleration value of the platform; and the weight of the seat posture data is a fixed value of 0.1.

[0090] Based on displacement fusion weights, a federated Kalman filter is used to fuse the device displacement data and the instructor displacement data to obtain the device attitude displacement data, in which the laser displacement sensor has a weight of 0.9 and the IMU has a weight of 0.1. Based on angle fusion weights, a quaternion Kalman filter is used to fuse the device angle data and the instructor angle data to obtain the device attitude angle data, in which the magnetic encoder has a weight of 0.95 and the gyroscope has a weight of 0.05. The device attitude displacement data and the device attitude angle data are used as the device attitude data.

[0091] Vibration compensation is performed on the coordinates of key skeletal points based on platform vibration acceleration data to obtain compensated instructor posture data. Vibration compensation is achieved by acquiring platform acceleration 'a' using a high-frequency vibration sensor. plat Combined with the pre-stored transfer coefficient k vib (0.02mm / g) Calculate the theoretical offset Δpos vib =k vib ×a plat From the measured coordinates pos meas The true coordinates pos are obtained by stripping from the middle. real =pos meas -Δpos vib Under special operating conditions (a>2g), LSTM prediction compensation is introduced to further correct vibration disturbances;

[0092] The compensated instructor posture data and pressure distribution data are correlated and fused to obtain fused instructor posture data. Specifically, this includes calculating the instructor's torso forward tilt angle based on the compensated instructor posture data, and simultaneously calculating the pressure distribution ratio P between the front and rear areas of the seat cushion based on the pressure distribution data. front / P rear, the correlation between the forward inclination angle and the pressure distribution ratio is established, when the forward inclination angle exceeds a first angle threshold of 15° and the pressure distribution ratio exceeds a first pressure ratio threshold of 1.5, it is determined that the instructor is in a forward leaning posture, based on the determination result, a comprehensive adjustment instruction including a seat X-axis forward movement of 5-10mm and a backrest backward inclination angle of 2-3° is generated, and the adjustment instruction is output as fused instructor posture data, completing multi-source data fusion and posture determination.

[0093] The present application realizes the cooperation of the instructor posture and the simulator motion working condition through multi-source data fusion and intelligent closed-loop control, effectively overcomes the problems of data precision decline and response delay in high-frequency vibration environment. The hierarchical safety protection mechanism is established based on the human biomechanics model, which can timely identify and intervene in the unsafe posture, and fundamentally eliminates the risk of instructor imbalance. At the same time, the system has good working condition self-adaptive ability, which can dynamically adjust the control strategy according to different flight states, and takes into account the adjustment accuracy and ride comfort. In terms of maintenance, the system greatly reduces the maintenance complexity and time cost through automatic calibration and remote diagnosis function, ensuring the high availability of the training equipment. Overall, the present application enhances the safety and effectiveness of simulation training.

[0094] Step S30, based on the fused device posture data and the instructor posture data, the current flight working condition is identified, and according to the type of the identified current flight working condition, the weight proportion of the device posture data, the instructor posture data and the seat posture data in calculating the posture deviation is dynamically determined;

[0095] In this embodiment, the specific way of step S30 is:

[0096] Step S31, a multi-dimensional feature vector including vibration frequency, acceleration component and attitude change rate is extracted from the fused device posture data and the instructor posture data;

[0097] Step S32, the multi-dimensional feature vector is input into the pre-trained flight working condition classification model for identification, and the type of the current flight working condition is obtained;

[0098] Step S33, according to the type of the identified current flight working condition, when generating the multi-directional adaptive adjustment instruction of the seat, the weight proportion of the device posture data, the instructor posture data and the seat posture data is dynamically allocated.

[0099] Among them, the device posture data weight is a first reference value minus a first adjustment amount related to the platform real-time acceleration value, the instructor posture data weight is a second reference value plus a second adjustment amount related to the platform real-time acceleration value, and the seat posture data weight is a preset fixed value.

[0100] Specifically, first, the vibration frequency of the platform, the X / Y / Z three-axis acceleration components, and the Roll / Pitch / Yaw three-axis angular velocity change rate are extracted from the fused device posture data, the torso pitch angle change rate, the seat cushion pressure distribution change rate, and the skeletal key point displacement change rate are extracted from the fused instructor posture data, which together form an 8-dimensional feature vector; the multi-dimensional feature vector is input into a pre-trained flight condition classification model for recognition, the model uses a BP neural network structure, the input layer has 8 neurons, the hidden layer has 16 neurons, and the output layer has 4 neurons corresponding to the take-off, cruise, turbulence, and roll four flight condition types, the model has an identification accuracy of not less than 98% after being trained with 10,000 samples, and the output is the probability of each condition, and the type with the highest probability and exceeding the 95% threshold is determined as the current flight condition;

[0101] According to the identified current flight condition type, the data weight is dynamically allocated when generating the seat adjustment instruction, wherein the device posture data weight is set to a first reference value minus a first adjustment amount related to the real-time acceleration value of the platform, and the specific calculation formula is Q 设备 =K1-α×(a / 2g), wherein K1 is the device data weight reference value corresponding to the current condition type (such as 0.5 for roll condition, 0.6 for cruise, and 0.45 for turbulence), α is the first adjustment coefficient (0.1-0.4), a is the real-time acceleration value of the platform, and g is the acceleration of gravity; the instructor posture data weight is set to a second reference value plus a second adjustment amount related to the real-time acceleration value of the platform, and the specific calculation formula is Q 教员 =K2+β×(a / 2g), wherein K2 is the instructor data weight reference value corresponding to the current condition type (such as 0.3 for roll condition, 0.15 for cruise, and 0.25 for turbulence), β is the second adjustment coefficient (0.1-0.4); the seat posture data weight is set to a fixed value of 0.1 and does not change with the condition and acceleration; in the above manner, when the platform acceleration increases, the device data weight decreases and the instructor data weight increases, so that the accuracy and adaptability of the posture deviation calculation are dynamically optimized in severe conditions by relying more on the real-time posture data of the instructor and in stable conditions by focusing more on the device posture data.

[0102] Step S40, calculating the posture deviation between the current posture of the instructor and the standard posture corresponding to the condition, wherein the current posture of the instructor is calculated by weighting the fused instructor posture data based on the weight proportion;

[0103] In this embodiment, first, the fused instructor posture data is weighted and calculated based on the dynamically determined weight proportion to obtain a comprehensive representation of the current posture of the instructor, wherein the weighting calculation adopts a linear weighting method, that is, the data of each sensor source is weighted and summed according to its weight, and the specific calculation is based on the weight proportion Q设备 , Q 教员 , and Q 座椅 (50%, 30%, and 20% respectively under rolling conditions);

[0104] The fused data from the device posture sensor (such as a laser displacement sensor, IMU), the instructor posture sensor (such as an infrared camera, pressure distribution sensor), and the seat posture feedback sensor (such as a seat posture sensor) are weighted and fused, and the calculation formula is: current posture data = Q 设备 x device posture data + Q 教员 x instructor posture data + Q 座椅 x seat posture data, thereby obtaining the displacement coordinates (X 当前 , Y 当前 , Z 当前 ) and angle coordinates (θ Roll当前 , θ Pitch当前 , θ Yaw当前 ) of the instructor at the current time in the unified coordinate system;

[0105] Subsequently, the system calls the standard posture parameters corresponding to the current flight condition type (such as the pitching condition) recognized in step S30 from the pre-stored standard posture database, including the standard displacement coordinates (X 标准 , Y 标准 , Z 标准 ) and standard angle coordinates (θ Roll标准 , θ Pitch标准 , θ Yaw标准 ), which is established by collecting standard sitting posture data of multiple experienced instructors under various conditions, and contains auxiliary parameters such as pressure distribution thresholds under various conditions;

[0106] The posture deviation calculation is divided into displacement deviation and angle deviation, and the displacement deviation ΔS is calculated by the three-dimensional Euclidean distance formula:

[0107] ,

[0108] The angle deviation Δθ is calculated by the three-dimensional angle difference sum root formula:

[0109] The calculated displacement deviation ΔS and angle deviation Δθ will be directly input to the subsequent generation of seat multi-directional adaptive adjustment instructions for driving the actuator to perform accurate posture correction.

[0110] Step S50, generate a multi-directional adaptive adjustment instruction for the seat based on the posture deviation and the type of condition, and send it to the seat actuator to drive the seat to adjust the displacement and angle;

[0111] In this embodiment, the method of the multi-directional adaptive adjustment instruction is:

[0112] Step S51, the standard sitting posture parameters corresponding to the type of the current flight working condition are called in the posture database, and the standard sitting posture parameters include standard bone key point coordinates and standard pressure distribution threshold values;

[0113] Step S52, the displacement deviation and the angle deviation between the real-time bone key point coordinates in the fused instructor posture data and the standard bone key point coordinates are calculated;

[0114] Step S53, based on the displacement deviation, the angle deviation and the type of the current flight working condition, the displacement adjustment amount and the angle adjustment amount of the seat in each axial direction are calculated in real time through a parameter adaptive model to form the multi-direction adaptive adjustment instruction.

[0115] In the system initialization stage, a standard posture database needs to be constructed to support posture deviation calculation. In specific implementation, three experienced instructors with a height of 1.7±0.1 meters and a weight of 70±5 kilograms are invited to maintain a standard sitting posture under four typical working conditions of take-off, cruising, turbulence and rolling. The three-dimensional coordinates of 16 bone key points (such as head, shoulders, sternum and hip) are collected by a multi-view infrared camera, and the pressure distribution data of the seat cushion and backrest (such as the cruising working condition with a trunk forward angle ≤15° and a seat cushion front-to-back pressure ratio ≤1.2) are recorded. After filtering and coordinate unification processing, all data are stored in the internal ROM of the FPGA to form a standard posture database (path: / data / standard_posture.db), which provides a reference for subsequent real-time posture deviation calculation.

[0116] The standard sitting posture parameters corresponding to the type of the current flight working condition are called in the standard posture database, which is stored according to the working condition type and includes standard bone key point coordinates and standard pressure distribution threshold values (such as seat cushion front-to-back pressure ratio threshold P front / P rear_std ); the system reads the corresponding standard parameters according to the recognized working condition type (such as turbulence working condition), including standard displacement coordinates (X std , Y std , Z std ) and standard angles (θ Roll_std , θ Pitch_std , θ Yaw_std ), and pressure distribution threshold values, seat cushion front-to-back pressure ratio threshold;

[0117] Then, the displacement deviation and the angle deviation between the real-time bone key point coordinates in the fused instructor posture data and the standard bone key point coordinates are calculated, wherein the displacement deviation ΔS is calculated by the three-dimensional Euclidean distance formula:

[0118] , the angle deviation Δθ is calculated by the three-dimensional angle difference sum root formula:

[0119] ;

[0120] Based on the calculated displacement deviation ΔS, angle deviation Δθ and current flight working condition type, the displacement adjustment amount and angle adjustment amount of the seat in each axis are calculated in real time by a parameter adaptive model. The parameter adaptive model adopts a PID control architecture, and the parameters are dynamically adjusted according to the working condition type and the deviation size: the proportional parameter K p =K {p0} ×(1+0.3×ΔS / 2mm), the integral parameter K i =K {i0} ×(1-0.2×ΔS / 2mm), and the differential parameter K d =K {d0} ×(1+0.1×ΔS / 2mm), wherein K {p0} , K {i0} , and K {d0} are the reference parameters of the current working condition type, such as the reference parameters of the pitching working condition K {p0} =0.6, K {i0} =3.5×10 -3 , and K {d0} =6.0×10 -3 ;

[0121] The final generated multi-direction adaptive adjustment instruction includes the displacement adjustment amount (unit: mm) of the X / Y / Z three axes of the seat and the angle adjustment amount (unit: °) of the Roll / Pitch / Yaw three axes. The instruction is transmitted to the seat MCU main control board through the CAN bus in a specific frame format (frame header 0xCC77+seat state+each axis adjustment amount+timestamp+CRC check+frame trailer 0x77CC) to drive the actuator (such as X / Y / Z axis electric push rod, backrest adjustment motor) to perform accurate adjustment, while superimposing reverse PWM signals to suppress high-frequency vibration, forming a complete adaptive control closed loop.

[0122] In the roll flight working condition (roll ±45°, acceleration 1g–2g), the system determines that the current state is roll by the working condition recognition algorithm, and dynamically adjusts the data weight: the device attitude data accounts for 50%, the instructor attitude data accounts for 30%, and the seat feedback data accounts for 20%. If the instructor roll angle φ is detected to be greater than 10°, a roll angle compensation amount Δθ roll =0.8×φ is generated, while combining feedforward control (LSTM predicts 50ms attitude change) and high-response PID parameters (K p =1.1, K i =7.0×10 -3 , and K d =5.0×10-3 ), driving the seat to synchronously adjust the roll angle, ensuring that the instructor's body is consistent with the platform movement trend. The safety monitoring module detects the roll angle and the standing height in real time. If φ≥20° or h≥150 mm, it immediately triggers the third-level emergency protection (backrest 10° reclined, armrest 50 mm retracted, and cushion 5 mm lifted) and sends a load reduction command to the platform.

[0123] Step S60: Collect the adjusted actual posture data and compare it with the target adjustment value. If there is a residual deviation, re-trigger the generation and execution of the adjustment instruction until the residual deviation meets the accuracy requirement.

[0124] After the seat completes the adjustment action, the actual posture data after adjustment is collected in real time through the seat posture feedback sensor installed at the connection shaft between the seat backrest and the cushion. This sensor collects the actual displacement and angle data of the seat at a sampling rate of 1 kHz, and simultaneously verifies the absolute position of the seat with the aid of the three-axis laser displacement sensor on the platform fixed base. The collected actual posture data is transmitted back to the core controller in the form of a seat state feedback frame. This data frame contains real-time displacement coordinates, angle values, and status identifiers of the seat.

[0125] After the core controller receives the actual posture data, it accurately compares it with the target adjustment value generated in step S40 and calculates the residual deviation. The displacement residual deviation is obtained by calculating the absolute difference between the actual displacement and the target displacement, and the angle residual deviation is obtained by calculating the absolute difference between the actual angle and the target angle. The system's preset accuracy requirement is that the displacement deviation should not exceed 0.1 mm and the angle deviation should not exceed 0.01 degrees.

[0126] If the calculated residual deviation exceeds the above accuracy requirement, the system immediately re-triggers the data processing process starting from step S20: re-performs multi-source data fusion processing, updates the current flight condition recognition result, recalculates the posture deviation based on the latest sensor data and generates a new adjustment instruction, and drives the actuator to perform accurate adjustment again. This cycle continues until the residual deviation of the actual posture data and the target value completely meets the accuracy requirement, forming a complete closed-loop control system and ensuring that the instructor's seat posture continuously matches the device working condition and the instructor's demand.

[0127] During the entire closed-loop control process, the system simultaneously monitors the working state of the actuator in real time, including motor current and temperature parameters. When the current exceeds 5 amperes or the temperature exceeds 85 degrees Celsius, the power is immediately cut off and the fault protection mechanism is triggered to ensure safe operation of the system. All running data and adjustment process records are saved to the local storage and uploaded to the remote operation and maintenance platform through the 4G / 5G network for subsequent analysis and optimization.

[0128] In this embodiment, the multi-directional adaptive adjustment instruction of the seat is generated based on the attitude deviation and the working condition type, and another embodiment further includes:

[0129] A digital twin model representing the dynamic relationship between platform motion, flight working condition and instructor attitude is constructed;

[0130] After identifying the current flight working condition, real-time data is input into the digital twin model to predict the attitude change trend of the instructor in the future period of time;

[0131] The predicted attitude change trend is used as a feedforward amount, and a feedback adjustment amount calculated based on the real-time attitude deviation is fused to generate the multi-directional adaptive adjustment instruction.

[0132] In this embodiment, a digital twin model representing the dynamic relationship between platform motion, flight working condition and instructor attitude is constructed, and the specific method is:

[0133] Based on the multi-body dynamics theory, a physical mechanism model is established to describe the six-degree-of-freedom motion of the simulator platform, the linkage of each joint of the instructor seat, and the motion of the spine and pelvic bone chain related to the sitting posture balance of the instructor's torso. This model constitutes the core dynamics framework of the digital twin model;

[0134] Using the platform motion sequence data and the corresponding instructor skeletal key point coordinate change data collected in the historical operation, a data-driven model is trained with the platform motion parameters as the input and the instructor attitude change as the output. This data-driven model is used to learn and predict the individualized response characteristics and nonlinear disturbances that are not fully described by the physical mechanism model;

[0135] The outputs of the physical mechanism model and the data-driven model are weighted and fused in the model coupling layer, and the weights are adaptively adjusted according to the dynamic intensity of the current flight working condition, thereby forming a hybrid digital twin model;

[0136] Through real-time acquisition of platform motion data and instructor attitude data, the model parameters included in the hybrid digital twin model and the weight coefficients of the weighted fusion are identified and updated online to ensure that the simulation output of the digital twin model and the actual state of the physical system remain synchronized.

[0137] In this embodiment, the predicted attitude change trend is used as a feedforward amount, and a feedback adjustment amount calculated based on the real-time attitude deviation is fused to generate the multi-directional adaptive adjustment instruction, and the method is:

[0138] A feedforward adjustment channel based on the predicted attitude change trend is established, which generates corresponding feedforward compensation instructions according to the predicted displacement and angle change amount;

[0139] In parallel, a feedback adjustment channel based on the real-time attitude deviation is established, which converts the real-time attitude deviation into a feedback correction instruction through a closed-loop control algorithm;

[0140] A dynamic fusion coefficient associated with the current flight condition type is set, and the fusion coefficient is adaptively adjusted according to the change of the real-time motion intensity of the platform;

[0141] The dynamic fusion coefficient is used to weight and fuse the feedforward compensation instruction and the feedback correction instruction to generate a comprehensive adjustment instruction with foresight and stability;

[0142] The comprehensive adjustment instruction is output as the final multi-directional adaptive adjustment instruction to the seat actuator to drive the seat to complete the coordinated adjustment of displacement and angle.

[0143] In specific implementation, first, a digital twin model capable of representing the dynamic relationship between platform motion, flight condition and instructor attitude is constructed. In specific construction, a physical mechanism model is established based on multi-body dynamics theory to describe the six-degree-of-freedom motion of the simulator platform, the linkage of each joint of the instructor seat, and the motion of the spine and pelvic bone chain closely related to sitting posture balance in the instructor's torso. This model constitutes the core dynamics framework of the digital twin model and is used to simulate the basic physical response of the system under ideal conditions.

[0144] At the same time, a large amount of platform motion sequence data such as acceleration, angular velocity time series and corresponding instructor bone key point coordinate change data collected in historical operation are used to train a data-driven model with platform motion parameters as input and instructor attitude change as output, such as long short-term memory network LSTM or Transformer time series model. This model is specifically used to learn and predict individualized habitual response, nonlinear disturbance caused by muscle active force and other complex characteristics that the physical mechanism model cannot accurately describe.

[0145] The output of the physical mechanism model and the output of the data-driven model are weighted and fused in the model coupling layer, where the weighting weight is not fixed but adaptively adjusted according to the dynamic intensity of the currently identified flight condition. The dynamic intensity of the flight condition is taken as an indicator, such as platform combined acceleration or attitude angle change rate. For example, in the smooth cruising condition, the physical mechanism model is more reliable, while in the severe rolling or jolting condition, the data-driven model is given a higher weight, thereby forming a hybrid digital twin model that can reflect the physical nature and capture individual differences.

[0146] More specifically, in the construction process of the hybrid digital twin model, the data-driven model is specifically used to learn and predict complex characteristics that are difficult to accurately describe by the physical mechanism model. For example, the physical mechanism model may simplify the instructor's torso as a uniform mass-spring-damper system, but in reality, different instructors will exhibit unique individualized habitual responses when encountering sudden jolts: some instructors may unconsciously tighten their core muscles to keep their torso rigid, causing the posture to lag behind the platform movement, while some instructors may actively lean forward to observe the instruments due to teaching needs, introducing nonlinear disturbances caused by muscle active force. These dynamics introduced by physiological habits and active behaviors are difficult to describe through pure physical equations. Therefore, the data-driven model learns through analyzing the corresponding relationship in historical data, for example, when the platform appears a specific frequency, such as 2-4Hz, and an amplitude, such as ±0.3g, vertical vibration, a certain instructor's hip joint key point usually appears an additional 3-5mm backward displacement beyond the predicted value of the physical model, and the model can capture and remember this characteristic.

[0147] Subsequently, in the model coupling layer, the output of the physical mechanism model and the output of the data-driven model are weighted and fused. The weighting weight here is not fixed but is adaptively adjusted according to the dynamic intensity of the currently identified flight condition. The dynamic intensity can be represented by quantitative indicators such as platform synthetic acceleration or attitude angle change rate. The system presets an acceleration-based threshold strategy: when the synthetic acceleration is less than 0.3g, corresponding to conditions such as smooth cruising, the system is considered to be in a flat dynamic state, and the physical mechanism model's description of the system's dominant dynamics is relatively reliable, so it is given a higher weight, for example, the physical model weight W phy =0.8, the data-driven model weight W data =0.2, and the physical nature law is more reliable.

[0148] On the contrary, when the synthetic acceleration exceeds 0.7g, corresponding to conditions such as severe rolling, emergency avoidance, or strong jolting, the system is considered to be in a strong nonlinear, strong disturbance interval, and individual differences and active intervention effects are significant, so the data-driven model is given a higher weight, for example, the weight is adjusted to W phy =0.3, W data =0.7, to better capture and predict complex responses that exceed ideal physical assumptions. The weight adjustment process can be continuous, for example, using a sigmoid function of acceleration for smooth transition to avoid command jitter caused by step changes. Through this adaptive weighting fusion strategy, the final hybrid digital twin model not only maintains the explainability and extrapolation ability of the physical mechanism model under normal conditions, but also integrates the precise capture ability of the data-driven model on individual characteristics and nonlinear disturbances under complex dynamics, thereby achieving more comprehensive and accurate dynamic representation and prediction of the instructor's posture.

[0149] To ensure that the model is synchronized with the physical system, the latest platform motion data and instructor attitude data collected in real time are used to identify and update the internal parameters of the model, such as inertia parameters, damping coefficients, and the aforementioned weighted fusion weight coefficients, in the hybrid digital twin model. Specifically, recursive least squares or Kalman filter algorithms can be used to ensure that the simulation output of the digital twin model can continuously track the actual state evolution of the physical system.

[0150] After identifying the current flight condition, the device attitude data and other data collected in real time and after fusion processing are input into the updated digital twin model to predict the attitude change trend of the instructor in a specific time window of the future 100-200 milliseconds, including the displacement and angular offset that the torso may produce.

[0151] Next, the predicted attitude change trend is used as a feedforward quantity and is fused with the feedback adjustment quantity calculated based on the real-time attitude deviation to generate the final multi-directional adaptive adjustment instruction. The specific fusion method is as follows: First, a feedforward adjustment channel based on the predicted attitude change trend is established. This channel generates corresponding feedforward compensation instructions based on the predicted displacement change amount, such as X, Y, and Z axis displacement, and the angle change amount, such as Roll, Pitch, and Yaw angle, with the purpose of pre-compensating for the predicted attitude deviation.

[0152] In parallel, a feedback adjustment channel based on the real-time calculated attitude deviation is established. This channel converts the real-time deviation into a feedback correction instruction through a closed-loop control algorithm, which acts to eliminate the residual error that may still exist after feedforward compensation and the uncertainty of model prediction. For example, a proportional-integral-derivative (PID) controller or its variants;

[0153] A dynamic fusion coefficient associated with the current flight condition type is set, for example, K ff for feedforward instructions, K fb for feedback instructions, and K ff + K fb = 1. The initial value of this coefficient is set according to the type of working condition. For roll working conditions, K ff is set to 0.7 initially, emphasizing forward-looking control; for cruise working conditions, K ff is set to 0.3 initially, focusing on stable tracking, and the real-time motion intensity of the platform and the size of the real-time acceleration are used to adaptively fine-tune this fusion coefficient. For example, the greater the acceleration, the more inclined to increase the front weight K ff for fast response. Then, using this dynamic fusion coefficient, the feedforward compensation instruction and the feedback correction instruction are weighted and fused to calculate the comprehensive adjustment instruction = K ff × feedforward compensation instruction + Kfb The feedback correction instruction is generated, thereby generating a comprehensive adjustment instruction with both forward-looking and system stability guaranteeing. Finally, the comprehensive adjustment instruction is output as a final multi-directional adaptive adjustment instruction to various types of actuators of the seat, such as electric push rods and steering gears in various axes, to drive the seat to accurately complete displacement and angle coordinated adjustment, thereby achieving active, smooth and accurate support and maintenance of the instructor posture.

[0154] Although the above embodiment is described in the above sequence, those skilled in the art can understand that, in order to achieve the effect of the embodiment, the different steps do not have to be executed in such sequence, and can be executed simultaneously (in parallel) or in a reversed sequence, and these simple changes are within the protection scope of the present application.

[0155] The second embodiment of the present application is a full-movement flight simulator instructor seat adaptive control system for realizing a full-movement flight simulator instructor seat adaptive control method, which comprises:

[0156] A data acquisition module configured to acquire motion data of a simulator platform, instructor bone key point coordinates and sitting posture pressure distribution data in real time;

[0157] A data fusion module configured to perform time-space alignment and multi-source data fusion processing on the motion data and the bone key point coordinates and sitting posture pressure distribution data, to obtain fused device posture data and instructor posture data;

[0158] A weight proportion allocation module configured to identify a current flight working condition based on the fused device posture data and instructor posture data, and dynamically determine a weight proportion of the device posture data, the instructor posture data and the seat posture data in calculating a posture deviation according to a type of the identified current flight working condition;

[0159] A weighting module configured to calculate a posture deviation between an instructor current posture and a corresponding working condition standard posture, wherein the instructor current posture is obtained by weighting calculation on the fused instructor posture data based on the weight proportion;

[0160] An adjustment module configured to generate a multi-directional adaptive adjustment instruction of the seat based on the posture deviation and the working condition type, and send the instruction to a seat actuator to drive the seat to perform displacement and angle adjustment;

[0161] A deviation correction module configured to acquire actual posture data after adjustment and compare the data with a target adjustment value, and if there is a residual deviation, re-trigger generation and execution of the adjustment instruction until the residual deviation meets the accuracy requirement.

[0162] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process and related description of the system described above can refer to the corresponding process in the foregoing method embodiment, and will not be repeated here.

[0163] It should be noted that the full-motion flight simulator instructor seat adaptive control system provided in the above embodiment is only exemplified by the division of the above functional modules, and in actual application, the above functions can be completed by different functional modules according to needs, that is, the modules or steps in the embodiment of the application are further decomposed or combined, for example, the modules of the above embodiment can be combined into one module, or can be further split into multiple sub-modules to complete all or part of the functions described above. The names of the modules and steps involved in the embodiments of the application are only for the purpose of distinguishing various modules or steps, and should not be considered as an improper limitation of the application.

[0164] The electronic device of the third embodiment of the application comprises:

[0165] at least one processor; and

[0166] a memory in communication connection with the at least one processor; wherein

[0167] The memory stores instructions executable by the processor, and the instructions are used to be executed by the processor to realize the full-motion flight simulator instructor seat adaptive control method.

[0168] The computer readable storage medium of the fourth embodiment of the application stores computer instructions, and the computer instructions are used to be executed by the computer to realize the full-motion flight simulator instructor seat adaptive control method.

[0169] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process and related description of the storage device and the processing device described above can refer to the corresponding process in the foregoing method embodiment, and will not be repeated here.

[0170] Those skilled in the art should clearly understand that the modules and method steps of each example described in combination with the embodiments disclosed herein can be realized in electronic hardware, computer software or a combination of both. The programs corresponding to the software modules and method steps can be placed in a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art. In order to clearly illustrate the interchangeability of electronic hardware and software, the components and steps of each example have been generally described in the above description. Whether the functions are performed in electronic hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0171] Reference is made below to Figure 2 which shows a structural schematic diagram of a computer system of a server for implementing the embodiments of the method, system and device of the present application. Figure 2 The server shown is merely an example and should not impose any limitation on the functions and use range of the embodiments of the present application.

[0172] As shown in Figure 2 , the computer system includes a central processing unit (CPU) 201 which can perform various appropriate actions and processes according to programs stored in a read-only memory (ROM) 202 or programs loaded from a storage portion 208 into a random access memory (RAM) 203. Various programs and data required for system operation are also stored in the RAM 203. The CPU 201, the ROM 202 and the RAM 203 are connected to each other through a bus 204. An input / output (I / O) interface 205 is also connected to the bus 204.

[0173] The following components are connected to the I / O interface 205: an input section 206 including a keyboard, a mouse, etc.; an output section 207 including a display such as a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 208 including a hard disk, etc.; and a communication section 209 including a network interface card such as a LAN (Local Area Network) card, a modem, etc. The communication section 209 performs communication processing via a network such as the Internet. A drive 210 is also connected to the I / O interface 205 as necessary. A removable medium 211 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc. is attached to the drive 210 as necessary, so that a computer program read out therefrom is installed in the storage section 208 as necessary.

[0174] In particular, according to embodiments of the present disclosure, the processes described above with reference to the flowcharts can be implemented as a computer software program. For example, embodiments of the present disclosure include a computer program product comprising a computer program carried on a computer readable medium, the computer program comprising program code for executing the methods illustrated by the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network by the communication section 209, and / or installed from the detachable medium 211. When the computer program is executed by the central processing unit (CPU) 201, the above-described functions defined in the methods of the present application are executed. It should be noted that the computer readable medium of the present application can be a computer readable signal medium or a computer readable storage medium or any combination of the two. The computer readable storage medium may, for example, be, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus or device, or any suitable combination of the above. More specific examples of the computer readable storage medium can include, but are not limited to, an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, the computer readable storage medium can be any tangible medium that contains or stores a program that can be used by or in connection with an instruction execution system, apparatus or device. In the present application, the computer readable signal medium can include a data signal carried in a baseband or as part of a carrier wave, in which the computer readable program code is carried. Such a propagated data signal can take any of a variety of forms, including but not limited to electro-magnetic, optical, or any suitable combination thereof. The computer readable signal medium can also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate or transport a program for use by or in connection with an instruction execution system, apparatus or device. Program code contained on the computer readable medium can be transmitted by any suitable medium, including but not limited to wireless, wire line, optical fiber cable, RF, etc., or any suitable combination of the above.

[0175] Computer program code for carrying out operations of the present application can be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C++ or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code can execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection can be made to an external computer (for example, through the Internet using an Internet Service Provider).

[0176] The computer program instructions can also be loaded onto a computer or other programmable information processing apparatus to cause a series of operations to be performed on the computer or other programmable information processing apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable information processing apparatus implement the functions / acts specified in the flowchart and / or block diagram block or blocks.

[0177] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the disclosure. For example, singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms "comprises" and / or "comprising," when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof.

[0178] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the disclosure. For example, singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms "comprises" and / or "comprising," when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof.

[0179] The technical scheme of the present application has been described in combination with the preferred embodiments shown in the drawings, but it is easy for those skilled in the art to understand that the protection scope of the present application is obviously not limited to these specific embodiments. Those skilled in the art can make equivalent changes or replacements to the related technical features without departing from the principles of the present application, and the technical schemes after the changes or replacements will all fall within the protection scope of the present application.

Claims

1. A method for adaptive control of a full-movement flight simulator instructor seat, characterized in that, The method comprises: Real-time acquisition of motion data of the simulation machine platform, bone key point coordinates of the instructor, and sitting posture pressure distribution data; Temporal and spatial alignment and multi-source data fusion processing of the motion data, the bone key point coordinates, and the sitting posture pressure distribution data to obtain fused device posture data and instructor posture data: Timestamp calibration and spatial coordinate conversion of the platform motion data and the bone key point coordinates to obtain time-space aligned data; Extraction of device displacement data and device angle data from the aligned platform motion data, and extraction of instructor displacement data and instructor angle data from the aligned bone key point coordinates; Dynamic adjustment of displacement fusion weights of the device displacement data and the instructor displacement data, and angle fusion weights of the device angle data and the instructor angle data according to real-time acceleration of the platform; Fusion of the device displacement data and the instructor displacement data based on the displacement fusion weights to obtain device posture displacement data, fusion of the device angle data and the instructor angle data based on the angle fusion weights to obtain device posture angle data, and taking the device posture displacement data and the device posture angle data as the device posture data; Vibration compensation of the bone key point coordinates according to platform vibration acceleration data to obtain compensated instructor posture data, and associated fusion processing of the compensated instructor posture data and the pressure distribution data to obtain fused instructor posture data; Identification of a current flight working condition based on the fused device posture data and the instructor posture data, dynamic determination of weight proportions of the device posture data, the instructor posture data, and the seat posture data in calculation of attitude deviation according to a type of the identified current flight working condition; Calculation of attitude deviation between a current instructor attitude and a corresponding working condition standard attitude, wherein the current instructor attitude is obtained by weighted calculation of the fused instructor posture data based on the weight proportions; Generation of multi-directional adaptive adjustment instructions of the seat based on the attitude deviation and the working condition type, and sending of the instructions to a seat actuator to drive the seat to adjust in displacement and angle; Collection of actual adjusted attitude data and comparison with a target adjustment value, re-triggering of generation and execution of the adjustment instructions if there is a residual deviation, and stopping until the residual deviation meets accuracy requirements.

2. The adaptive control method for a motion flight simulator instructor seat according to claim 1, wherein Dynamic adjustment of displacement fusion weights of the device displacement data and the instructor displacement data, and angle fusion weights of the device angle data and the instructor angle data according to real-time acceleration of the platform, which comprises: The fusion weight of the device displacement data is set to be inversely proportional to a first reference value and the real-time acceleration of the platform, and the fusion weight of the instructor displacement data is set to be directly proportional to a second reference value and the real-time acceleration of the platform; The fusion weight of the device angle data is set to be inversely proportional to a third reference value and the real-time acceleration of the platform, and the fusion weight of the instructor angle data is set to be directly proportional to a fourth reference value and the real-time acceleration of the platform; The weight of the seat posture data is a fixed value.

3. The adaptive control method for a full-motion flight simulator instructor seat according to claim 1, wherein Associated fusion processing of the compensated instructor posture data and the pressure distribution data, which comprises: According to the compensated instructor posture data, a forward inclination angle of an instructor's torso is calculated, and according to the pressure distribution data, a pressure distribution ratio between different areas of a seat cushion is calculated; A correlation between the forward inclination angle and the pressure distribution ratio is established, and when the forward inclination angle exceeds a first angle threshold and the pressure distribution ratio exceeds a first pressure ratio threshold, it is determined that the instructor is in a forward leaning posture; Based on the determination result of the correlation, a comprehensive adjustment instruction including a seat forward movement amount and a backrest angle adjustment amount is generated, and the adjustment instruction is output as fused instructor posture data.

4. The adaptive control method for a motion flight simulator instructor seat according to claim 1, wherein, Based on the fused device posture data and the instructor posture data, a current flight working condition is identified, and according to the type of the identified current flight working condition, the weight proportions of the device posture data, the instructor posture data and the seat posture data in calculating the posture deviation are dynamically determined, and the method is as follows: From the fused device posture data and the instructor posture data, a multi-dimensional feature vector including a vibration frequency, an acceleration component and an attitude change rate is extracted; The multi-dimensional feature vector is input into a pre-trained flight working condition classification model for identification to obtain the type of the current flight working condition; According to the type of the identified current flight working condition, the weight proportions of the device posture data, the instructor posture data and the seat posture data are dynamically allocated when generating a multi-directional adaptive adjustment instruction of the seat.

5. A method of adaptive control of an instructor seat of a full motion flight simulator according to claim 4, characterized in that, The device posture data weight is a first reference value minus a first adjustment amount related to the real-time acceleration value of the platform, the instructor posture data weight is a second reference value plus a second adjustment amount related to the real-time acceleration value of the platform, and the seat posture data weight is a preset fixed value.

6. The adaptive control method for a motion flight simulator instructor seat according to claim 1, wherein, The method of the multi-directional adaptive adjustment instruction is as follows: A standard sitting posture parameter corresponding to the type of the current flight working condition is called from a pre-stored standard posture database, and the standard sitting posture parameter includes a standard skeleton key point coordinate and a standard pressure distribution threshold; The displacement deviation and the angle deviation between the real-time skeleton key point coordinate in the fused instructor posture data and the standard skeleton key point coordinate are calculated; Based on the displacement deviation, the angle deviation and the type of the current flight working condition, the displacement adjustment amount and the angle adjustment amount of the seat in each axis direction are calculated in real time through a parameter adaptive model to form the multi-directional adaptive adjustment instruction.

7. The adaptive control method for a motion flight simulator instructor seat according to claim 1, wherein, The motion data of the simulator platform is collected in real time based on a device posture acquisition sensor system, and the skeleton key point coordinates and the sitting posture pressure distribution data of the instructor are collected in real time through an instructor posture acquisition sensor system.

8. A self-adaptive control system for a motion flight simulator instructor seat, for implementing the self-adaptive control method for a motion flight simulator instructor seat according to any one of claims 1-7, characterized in that, The system comprises: A data acquisition module configured to collect the motion data of the simulator platform, the skeleton key point coordinates of the instructor and the sitting posture pressure distribution data in real time; A data fusion module configured to perform time-space alignment and multi-source data fusion processing on the motion data, the skeleton key point coordinates and the sitting posture pressure distribution data to obtain fused device posture data and instructor posture data; a weight proportion distribution module configured to identify a current flight working condition based on the fused device attitude data and the instructor attitude data, and dynamically determine a weight proportion of the device attitude data, the instructor attitude data and the seat attitude data in calculating an attitude deviation according to a type of the identified current flight working condition; a weighting module configured to calculate an attitude deviation between an instructor current attitude and a corresponding working condition standard attitude, wherein the instructor current attitude is obtained by weighting the fused instructor attitude data based on the weight proportion; an adjustment module configured to generate a multi-directional self-adaptive adjustment instruction of the seat based on the attitude deviation and the working condition type, and send the instruction to a seat actuator to drive the seat to displace and adjust an angle; a deviation correction module configured to collect actual attitude data after adjustment and compare the actual attitude data with a target adjustment value, and if there is a residual deviation, re-trigger generation and execution of the adjustment instruction until the residual deviation meets a precision requirement.

9. An electronic device, comprising: comprise: at least one processor; and a memory communicatively connected with the at least one processor; wherein the memory stores instructions executable by the processor, and the instructions are used to be executed by the processor to implement the self-adaptive control method of the instructor seat of the full-motion flight simulator according to any one of claims 1-7.

Images