Real-time simulation method, system and equipment for dynamic friction coefficient of runway based on multi-source data

Abstract

The invention belongs to the field of analog simulation, and particularly relates to a runway dynamic friction coefficient real-time simulation method, system and equipment based on multi-source data. The problem that taxiing simulation is insufficient in authenticity is solved. The method comprises the following steps: acquiring runway surface temperature data, accumulated water depth and / or accumulated snow depth; judging a runway type based on the runway texture features, matching a basic friction coefficient according to the runway type, and dynamically correcting the basic friction coefficient in combination with the accumulated water depth and / or accumulated snow depth and runway surface temperature data to obtain a corrected friction coefficient; acquiring a tire vertical load, an initial sideslip angle and tire pressure data of the aircraft, calculating a tire lateral force, and calculating a final friction coefficient in combination with the updated sideslip angle, the corrected friction coefficient and the sideslip attenuation coefficient; and inputting the data into a ground taxiing model of the flight simulator to realize real-time simulation. Through the attenuation effect of the dynamic process on the friction coefficient, the simulation result better fits the real runway state.

Description

Background Art

[0002] In the field of flight simulation training, the accurate calculation of runway friction coefficient is one of the core technologies to improve the authenticity of taxiing simulation. Currently, the modeling methods for runway friction coefficient μ mainly rely on the following two types of technical solutions: 1. Fixed friction coefficient model This method is based on a preset constant friction coefficient (e.g., μ = 0.6), and different μ values are switched through the runway surface type (such as dry, wet, frozen, snow-covered, or wet rubber residue). Although this method is simple to implement and has high computational efficiency, it is essentially a static model that can only distinguish runway states through limited empirical values and cannot reflect the real-time impact of actual dynamic environmental parameters (such as water depth, snow thickness, oil pollution) on the friction coefficient. In addition, this model completely ignores the non-linear coupling relationship between tire dynamic parameters (such as sideslip angle, tire pressure change) and friction force, resulting in a significant deviation between the simulation results and the actual physical laws.

[0003] 2. Modified model based on empirical formula Such models correct the basic friction coefficient by introducing linear empirical formulas. Although they partially consider the adjustment requirements of environmental factors, the correction coefficients still highly depend on artificial experience and lack clear physical basis. They do not involve the quantitative input of dynamic environmental parameters such as water accumulation and snow accumulation, and do not integrate the real-time perception ability of multi-source sensor data (such as cameras, millimeter-wave radars) for the runway surface state. In addition, the existing models do not establish the coupling relationship between the tire sideslip angle and the vertical load, resulting in a significant increase in the calculation error of the friction coefficient in complex scenarios (such as the superposition of crosswind and water accumulation).

[0004] The above methods have the following key problems in practical applications: (1) Lack of environmental dynamics: Unable to quantify the dynamic attenuation effect of parameters such as water depth and snow thickness on the friction coefficient, resulting in the disconnection between the simulation results and the real runway conditions; (2) Insufficient tire dynamics coupling: Ignoring the non-linear correction requirement of the tire sideslip angle on the friction force and not considering the physical impact of tire pressure change on the tire contact area; (3) Strong empirical dependence: The correction coefficients lack theoretical support based on physical laws or experimental data, and the model generalization ability is limited; (4) Weak real-time linkage ability: Not forming a closed loop with multi-sensor data (such as cameras, radars, thermometers), it is difficult to achieve millisecond-level response in complex coupling scenarios.

[0005] These problems directly reduce the authenticity of the taxiing simulation of flight simulators and the training effect. For example, in the case of the coexistence of crosswind and runway waterlogging, the existing models cannot dynamically integrate environmental parameters and tire sideslip angle correction, resulting in obvious differences between the taxiing resistance feedback by the simulator and the real flight experience, seriously restricting the reliability of pilot training.

[0006] Based on this, the present invention proposes a real-time simulation method, system and device for the dynamic friction coefficient of an aircraft runway based on multi-source data. Summary of the Invention

[0007] In order to solve the above problems in the prior art, that is, the existing runway friction coefficient calculation method cannot dynamically quantify environmental parameters such as waterlogging depth and snow depth, ignores the non-linear influence of tire sideslip angle and tire pressure change, relies on manual experience correction and lacks real-time linkage of multi-sensor data, resulting in insufficient authenticity of taxiing simulation, the present invention provides a real-time simulation method, system and device for the dynamic friction coefficient of an aircraft runway based on multi-source data.

[0008] The first aspect of the present invention proposes a real-time simulation method for the dynamic friction coefficient of an aircraft runway based on multi-source data, and the method includes: Obtain runway image data and runway surface temperature data, extract the waterlogging area and / or snow area and runway texture features of the runway based on the runway image data, and obtain the waterlogging depth and / or snow depth; Judge the runway type based on the runway texture features, match the basic friction coefficient according to the runway type, and dynamically correct the basic friction coefficient in combination with the waterlogging depth and / or snow depth and runway surface temperature data to obtain a corrected friction coefficient; Obtain the tire vertical load, initial sideslip angle and tire pressure data of the aircraft for calculating the theoretical tire lateral force, update the sideslip angle in real time in combination with the deviation between the theoretical tire lateral force and the actual tire lateral force and the iterative algorithm, and calculate the final friction coefficient in combination with the updated sideslip angle, corrected friction coefficient and sideslip attenuation coefficient; Input the final friction coefficient into the ground taxiing model of the flight simulator to achieve real-time simulation.

[0009] Further, the method for dynamically correcting the basic friction coefficient to obtain a corrected friction coefficient is: Establish a linear attenuation factor based on the product of the waterlogging depth and a preset waterlogging attenuation coefficient; Establish an exponential attenuation factor based on the product of the snow depth and a preset snow attenuation coefficient; Establish a temperature compensation factor according to the product of the absolute value of the difference between the runway surface temperature and the reference temperature and the temperature compensation coefficient; The basic friction coefficient is successively compounded and adjusted by three attenuation factors to obtain a corrected friction coefficient.

[0010] Further, for the final friction coefficient, its calculation method is as follows: Based on the corrected friction coefficient, a dynamic attenuation parameter is constructed by combining the product of the absolute value of the real-time updated sideslip angle and the preset sideslip attenuation coefficient. The friction performance of the corrected friction coefficient is reduced under the slip state through the dynamic attenuation parameter to obtain the final friction coefficient representing the actual taxiing condition.

[0011] Further, for the real-time update of the sideslip angle, the method is as follows: Based on the initial sideslip angle, tire vertical load, and tire pressure data, combined with the pre-stored cornering stiffness and main wheel load upper limit parameters, the theoretical lateral force is output through the lateral force calculation model; The theoretical lateral force is compared with the actual tire lateral force to determine whether there is a deviation. If there is no deviation, the sideslip angle is not updated. If there is a deviation, the sideslip angle is iteratively updated based on the iterative algorithm.

[0012] Further, the image data and the runway surface temperature data are respectively obtained by a camera and an infrared thermometer.

[0013] Further, the tire vertical load and the tire pressure data are obtained by a tire pressure sensor.

[0014] Further, for obtaining the water depth and / or snow depth, the method is as follows: Based on the image data, the water accumulation area and / or snow accumulation area are detected, and the millimeter wave radar is used to detect the water depth and / or snow depth.

[0015] In the second aspect of the present invention, a real-time simulation system for the dynamic friction coefficient of an aircraft runway based on multi-source data is proposed. Based on a real-time simulation method for the dynamic friction coefficient of an aircraft runway based on multi-source data, the system includes: A data acquisition module configured to acquire runway image data and runway surface temperature data, extract the water accumulation area and / or snow accumulation area and runway texture features of the runway based on the runway image data, and obtain the water depth and / or snow depth; A corrected friction coefficient calculation module configured to determine the runway type based on the runway texture features, match the basic friction coefficient according to the runway type, and dynamically correct the basic friction coefficient in combination with the water depth and / or snow depth and the runway surface temperature data to obtain the corrected friction coefficient; A final friction coefficient calculation module, configured to obtain the vertical load of the aircraft's tires, the initial sideslip angle, and tire pressure data, for calculating the theoretical tire lateral force, and combining the deviation between the theoretical tire lateral force and the actual tire lateral force and an iterative algorithm to update the sideslip angle in real time, and combining the updated sideslip angle, the corrected friction coefficient, and the sideslip attenuation coefficient to calculate the final friction coefficient; A real-time simulation module, configured to input the final friction coefficient into the ground roll model of a flight simulator to achieve real-time simulation.

[0016] In a third aspect of the present invention, an electronic device is proposed, including: At least one processor; and A memory communicatively connected to at least one of the processors; wherein, The memory stores instructions executable by the processor, and the instructions are used to be executed by the processor to implement the above-mentioned real-time simulation method for the dynamic friction coefficient of an aircraft runway based on multi-source data.

[0017] In a fourth aspect of the present invention, a computer-readable storage medium is proposed, and the computer-readable storage medium stores computer instructions, and the computer instructions are used to be executed by the computer to implement the above-mentioned real-time simulation method for the dynamic friction coefficient of an aircraft runway based on multi-source data.

[0018] Advantages of the present invention: Enhanced dynamic environment parameter quantification ability: Based on image data, the water accumulation depth, snow accumulation thickness, and runway texture features are extracted in real time, and combined with temperature sensor data to construct an environmental attenuation factor to dynamically correct the basic friction coefficient. Compared with the static model, it can accurately reflect the attenuation effect of dynamic processes such as water accumulation penetration and snow compaction on the friction coefficient, making the simulation results more consistent with the real runway state (such as the hydrodynamic pressure effect differences of different depths of water accumulation).

[0019] Improved accuracy of tire dynamics coupling modeling: By introducing the vertical load of the tire, the real-time updated sideslip angle, and tire pressure data, a non-linear relationship model between the lateral force and the sideslip angle is established. By iteratively calculating the sideslip angle correction amount, the problems of ignoring the change in the tire contact area (caused by tire pressure) and the coupling effect between the sideslip angle and friction force in the existing methods are solved, significantly improving the calculation accuracy of the friction coefficient under complex working conditions such as crosswinds and sharp turns.

[0020] Optimized generalization of data-driven physical models: Automatically match the basic friction coefficient based on the runway texture features, and perform physical corrections in combination with multi-source environmental parameters (such as the influence of ice surface temperature on the thickness of the meltwater film), replacing manual adjustment of empirical coefficients. The model parameters have clear physical meanings and support generalization applications for different runway types (asphalt, concrete) and extreme conditions (oil pollution, freeze-thaw cycles).

[0021] Multi-sensor real-time closed-loop response: Integrate data streams from multiple sensors such as cameras (texture recognition), millimeter-wave radars (measurement of water accumulation layer thickness), and thermometers (monitoring of ice surface conditions) to construct a closed-loop with millisecond-level updates. By providing real-time feedback on runway state changes (such as a sudden surge in water accumulation depth caused by heavy rainfall), ensure that the dynamic response delay of the friction coefficient in complex coupled scenarios (crosswind + water accumulation + tire pressure fluctuations) is less than 10 ms, meeting the high real-time requirements of flight simulators.

[0022] Significantly improved training reliability: After the final friction coefficient is input into the ground roll model, it can accurately simulate physical phenomena such as sudden changes in rolling resistance (e.g., when the tires enter a snow-covered area) and critical instability of sideslip. In the scenario of superimposed crosswind and water accumulation, the error between the simulation results and real flight test data is less than 5%, effectively improving the training effect of pilots against complex runway conditions and their emergency response capabilities. Description of the Drawings

[0023] Other features, objects, and advantages of the present application will become more apparent by reading the detailed description of the non-limiting embodiments with reference to the following drawings: Figure 1 It is a schematic flow chart of a method for real-time simulation of the dynamic friction coefficient of an aircraft runway based on multi-source data of the present invention. Detailed Embodiments

[0024] The following further elaborates on the present application in conjunction with the drawings and embodiments. It can be understood that the specific embodiments described herein are only for explaining the relevant invention and not for limiting the invention. Additionally, it should be noted that for ease of description, only parts related to the relevant invention are shown in the drawings.

[0025] It should be noted that, without conflict, the embodiments in the present application and the features in the embodiments can be combined with each other. The following will detail the present application with reference to the drawings and embodiments.

[0026] The present invention provides a method for real-time simulation of the dynamic friction coefficient of an aircraft runway based on multi-source data, which includes: Step S10: Obtain runway image data and runway surface temperature data, extract the water accumulation area and / or snow accumulation area and runway texture features of the runway based on the runway image data, and obtain the water accumulation depth and / or snow accumulation depth; Step S20: Determine the runway type based on the runway texture features, match the basic friction coefficient according to the runway type, and dynamically correct the basic friction coefficient in combination with the water accumulation depth and / or snow accumulation depth and runway surface temperature data to obtain the corrected friction coefficient; Step S30: Obtain the vertical load of the aircraft's tires, the initial sideslip angle, and the tire pressure data, which are used to calculate the theoretical tire lateral force. Combine the deviation between the theoretical tire lateral force and the actual tire lateral force and the iterative algorithm to update the sideslip angle in real time. Combine the updated sideslip angle, the corrected friction coefficient, and the sideslip attenuation coefficient to calculate the final friction coefficient. Step S40: Input the final friction coefficient into the ground roll model of the flight simulator to achieve real-time simulation.

[0027] To more clearly illustrate a real-time simulation method for the dynamic friction coefficient of an aircraft runway based on multi-source data of the present invention, the following will be combined with Figure 1 Expand and detail each step in the embodiment of the present invention.

[0028] A real-time simulation method for the dynamic friction coefficient of an aircraft runway based on multi-source data in the first embodiment of the present invention includes steps S10 - S40, and each step is described in detail as follows: Step S10: Obtain runway image data and runway surface temperature data. Based on the runway image data, extract the water accumulation area and / or snow accumulation area and runway texture features of the runway, and obtain the water accumulation depth and / or snow accumulation depth. In this embodiment, the image data and the runway surface temperature data are respectively obtained according to a camera and an infrared thermometer.

[0029] Extract the water accumulation area and / or snow accumulation area and runway texture features of the runway, which is implemented by a convolutional-attention hybrid network with multi-scale feature fusion. This hybrid network includes a multi-scale feature encoder, a cross-domain adaptation module, and a texture decoder. Multi-scale feature encoder: Input layer: Receive the runway image with the fusion of RGB and near-infrared dual channels, with a size of H×W×4. Multi-branch convolutional module: First branch: 3×3 convolutional kernel, depthwise separable convolution, with the number of output channels C. Second branch: 5×5 dilated convolution (dilation rate 2), grouped convolution (number of groups 4), with the number of output channels C. Third branch: 1×1 convolution → 3×3 max pooling → 3×3 transposed convolution, with the number of output channels C. Feature fusion layer: Concatenate the outputs of each branch along the channel dimension, and compress them to 3C channels through 1×1 convolution. Cross-domain adaptation module: Channel attention weighting unit: Calculate the global average pooling → fully connected layer → Sigmoid activation for each channel of the feature map to generate a channel weight vector. Adversarial discriminator: Consists of a 3-layer fully connected network. The input is the feature vector, and the output is the classification probability of the output domain. Adversarial training is achieved through the gradient reversal layer. Texture decoder: Feature pyramid structure: Bilinearly upsample the feature maps at each stage of the encoder to a unified resolution; Cross-scale attention gate: Calculate the cosine similarity between the low-level features and the high-level features to generate a spatial attention mask; Output layer: 3×3 convolution → Softmax activation to generate a pixel-level texture classification map.

[0030] The domain adaptation module performs the following steps: Multi-scale feature extraction: Use the multi-scale feature encoder to perform multi-branch convolution processing on the input source domain runway image and target domain real-time image respectively to generate a source domain feature map Fs and a target domain feature map Ft; Source domain: Load the annotated runway images in the ASFT dataset, including various lighting conditions such as sunny and cloudy days; Target domain: Runway images in rainy and foggy weather collected in real time, without manual annotation; Through the channel splicing and compression operations of the feature fusion layer, output a fusion feature vector with a unified dimension; Cross-domain adversarial training: Input the fusion feature vector into the adversarial discriminator to calculate the domain classification probabilities of the source domain and the target domain; Construct a domain difference loss function L domain , whose expression is the weighted sum of the source domain classification error and the target domain classification error, where the source domain samples are labeled 0 and the target domain samples are labeled 1; Among them, L domain =CE(D(Fs),0)+CE(D(Ft),1), where CE is the cross-entropy loss.

[0031] Implement inverse gradient propagation through the gradient reversal layer, maximize the discriminator error when updating the encoder, and minimize the classification error when updating the discriminator; Attention-guided feature enhancement: Through the channel attention weighting unit, weight the channel importance of the source domain feature map Fs to generate an attention weight matrix Wc; Wc=σ(FC(GAP(Fs))); FC is the fully connected layer, GAP is the global average pooling, and σ is the Sigmoid function.

[0032] Wc is used to perform channel-level weighting on the source domain feature Fs, strengthen the channels sensitive to runway texture, and suppress the rain and fog noise channels.

[0033] In the feature pyramid structure, use the cross-scale attention gate to calculate the similarity matrix S between feature maps at different levels, and the similarity matrix is used to fuse multi-scale texture features; Multiply the weighted feature map with the similarity matrix S to output an enhanced texture feature vector; Joint optimization and inference: Input the enhanced feature vector into the texture decoder to generate a pixel-level texture segmentation result through bilinear upsampling and cross-scale attention gates; Combine the segmentation loss Lseg, the domain difference loss Ldomain, and the feature consistency loss Lconsistency to jointly optimize the parameters of the multi-scale feature encoder and the texture decoder in an end-to-end manner; In the inference stage, freeze the parameters of the adversarial discriminator and only retain the multi-scale feature encoder and the texture decoder for real-time texture feature extraction.

[0034] Obtain the water accumulation depth and / or snow accumulation depth, and the method is as follows: Detect the water accumulation area and / or snow accumulation area based on the image data, and use a millimeter-wave radar to detect the water accumulation depth and / or snow accumulation depth. Among them, the snow includes dry snow and wet snow, and the water accumulation includes liquid water and ice layers.

[0035] The present invention also installs a polarization imaging sensor on the runway, and combines the millimeter-wave radar to obtain the water accumulation depth and / or snow accumulation depth, and the method is as follows: Step S11, scan the runway surface through the millimeter-wave radar in the water accumulation area and / or snow accumulation area to obtain the medium reflection signal, and at the same time collect the polarization reflection image of the runway area through the polarization imaging sensor, and perform spatio-temporal registration on the detection areas of the two sensors to form fusion detection data covering the same target area; Step S12, perform polarization angle distribution analysis on the polarization reflection image, extract the polarization angle change rate characteristics corresponding to different phase substances, and distinguish the substance phases according to the polarization characteristic differences of liquid water, solid ice, and snow-water mixture; Step S13, input the dielectric characteristics of the medium detected by the millimeter-wave radar and the polarization angle change rate characteristics into the classification model, and identify the runway surface state through the multi-physical quantity fusion decision logic, and the surface state at least includes dry snow, wet snow, ice layer, and liquid water; Step S14, when it is detected that the ice layer covers, perform non-linear compensation on the snow accumulation depth measurement value of the millimeter-wave radar based on the ice layer thickness to generate an equivalent depth value that incorporates the influence of the ice layer; Step S15, arrange a calibration device at a preset position on the runway, periodically compare the fusion detection result with the measured data of the contact sensor, and dynamically adjust the parameters of the depth compensation model according to the deviation.

[0036] Among them, in the step S13, the method of identifying the runway surface state through the multi-physical quantity fusion decision logic is as follows: Step S131: Perform noise suppression and feature extraction on the millimeter-wave radar echo signal to obtain scattering parameters characterizing the dielectric properties of the runway surface; Step S132: Decompose the polarization reflection image into intensity components, polarization degree components, and polarization angle components to generate a multi-channel polarization feature map; Step S133: Perform spatial correlation matching between the radar scattering parameters and the polarization feature map to construct a multi-dimensional fusion feature vector containing dielectric properties, polarization properties, and texture features; Step S134: Use the multi-class evidence theory to perform probability synthesis on the multi-dimensional fusion feature vector, and determine the runway surface state according to the following rules: Liquid water state: Determine when the confidence of liquid water exceeds the first threshold and the probability of ice layer is lower than the critical value; Ice layer state: Determine when the confidence of solid ice exceeds the second threshold and the dielectric properties conform to the ice layer characteristics; Dry snow / wet snow state: Jointly determine through the polarization angle change rate and the dielectric property gradient. Dry snow is characterized by a low dielectric constant and high polarization angle stability, while wet snow is characterized by dielectric constant fluctuations and dynamic polarization angle changes.

[0037] Among them, the generation methods of the first threshold (liquid water confidence) and the second threshold (solid ice confidence) include the following steps: Step S1341: The first threshold is determined through a joint calibration experiment of liquid water dielectric properties and polarization angle stability, and is dynamically adjusted by environmental temperature and humidity; Step S1342: The second threshold is generated based on the mapping relationship between ice layer reflection characteristics and solid dielectric constant, and the light intensity is introduced as a compensation factor; Step S1343: When sensor performance degradation or environmental mutation is detected, trigger the threshold online self-learning mechanism, and update the threshold generation model through the federated learning framework.

[0038] In this embodiment, when an ice-water mixture is detected, the following control steps are executed: Dynamic friction coefficient constraint: Generate a dynamic attenuation factor according to the correlation between ice layer thickness and surface temperature, impose a non-linear constraint on the basic friction coefficient, and limit the maximum available friction coefficient when the attenuation factor exceeds the safety threshold; Three-dimensional visualization warning: In the virtual runway scene of the flight simulator, dynamically render the ice-water mixing area, and characterize the spatial distribution of ice layer thickness through color gradient and transparency change; Multi-modal interaction feedback: When the virtual aircraft slides into the ice-water mixing area, synchronously trigger the force feedback module of the flight control device to generate a tactile warning signal corresponding to the real-time change of the friction coefficient.

[0039] The camera in this embodiment: It is deployed on both sides or at a high place of the runway to ensure full coverage of the runway section. The resolution is not less than 1080P, and it supports the visible light and near-infrared bands, and is used to collect runway surface images all-weather.

[0040] Infrared thermometer: It is installed at the same position as the camera or integrated into the camera module, and collects runway surface temperature data in real time. The temperature measurement range covers -40°C to +80°C, and the accuracy is ±0.5°C.

[0041] Millimeter-wave radar: It uses a 24GHz or 77GHz band radar, which is installed on the runway edge or on a mobile detection device. The beam angle ≤ 5°, and it points vertically to the runway surface, and is used to detect the depth of water accumulation / snow accumulation areas.

[0042] In this embodiment, the water accumulation / snow accumulation areas are identified through image segmentation algorithms (such as U-Net, MaskR-CNN).

[0043] Threshold segmentation is performed based on reflection characteristics (highlight areas) and color space (low saturation and high brightness in HSV). Combining white area detection and texture analysis (surface granularity), and auxiliary judgment of whether the snow is frozen is carried out through infrared temperature data (the snow accumulation mode is triggered when the temperature ≤ 0°C).

[0044] Extract the texture features (such as gray-level co-occurrence matrix, edge density) of the normal runway area, compare with the abnormal area, and exclude interferences such as reflection and shadow.

[0045] Judge the environmental state through the infrared thermometer data: When the temperature > 0°C and there are large areas of high-reflection areas, it is determined as water accumulation; When the temperature ≤ 0°C and the image segmentation is a white area, it is determined as snow accumulation; When the temperature is close to 0°C and there is snow accumulation, a freezing risk warning is triggered.

[0046] Map the coordinates of the water accumulation / snow accumulation areas identified in the image to the radar detection range, and drive the radar beam to scan the target area directionally.

[0047] In this embodiment, according to the time difference of the radar wave at the air-water interface (strong reflection) and the water-runway interface (secondary reflection) , the depth is calculated ; where is the speed of light, is the dielectric constant of water (≈80).

[0048] Based on the attenuation characteristics of the radar wave by the snow layer, through joint modeling of the reflection intensity and time delay, and combining temperature data to calibrate the density parameter of the snow (such as the lower the temperature, the higher the snow density and the more significant the attenuation).

[0049] If the radar detects that the depth does not match the area and texture features of the region in the image (for example, a large area of shallow ponding is misjudged as deep ponding), start rescan or fuse temperature data for dynamic correction.

[0050] Step S20: Determine the runway type based on the runway texture features, match the basic friction coefficient according to the runway type, and dynamically correct the basic friction coefficient by combining the ponding depth and / or snow depth and runway surface temperature data to obtain a corrected friction coefficient. Among them, the runway types include asphalt runways, concrete runways, and ice / snow runways.

[0051] Dynamically correct the basic friction coefficient , and obtain a corrected friction coefficient , and the method is as follows: Establish a linear attenuation factor based on the product of the ponding depth and a preset ponding attenuation coefficient; Establish an exponential attenuation factor based on the product of the snow depth and a preset snow attenuation coefficient; Establish a temperature compensation factor according to the product of the absolute value of the difference between the runway surface temperature and the reference temperature and the temperature compensation coefficient; Compound and adjust the basic friction coefficient through the triple attenuation factors in sequence to obtain a corrected friction coefficient.

[0052] ; Among them, is the ponding depth, is the snow depth, is the runway surface temperature data, is the preset reference temperature, where for asphalt / concrete it is set to 15 °C, and for ice / snow runways it is set to -5 °C, is the ponding attenuation coefficient, by default 0.05 - 0.15 / mm, for every 1 mm increase in ponding, the friction coefficient decreases by 5%, is the snow attenuation coefficient, by default 0.03 - 0.07 / mm, and the exponential attenuation model simulates the influence of the fluffiness of the snow layer, is the temperature compensation coefficient, by default 0.005 - 0.015 / °C, and when the temperature deviates from the reference value, the friction coefficient linearly attenuates.

[0053] If there is both ponding and snow (such as during the snowmelt period), calculate according to the superposition of and ; when is between -2 °C and 2 °C, trigger the detection of the snow - water mixed state and automatically increase the weights of and .

[0054] In this embodiment, the runway images captured by the camera are grayscaled, denoised, and enhanced to extract the texture features of the runway surface: Gray-level co-occurrence matrix (GLCM): Calculate parameters such as contrast, energy, and homogeneity to distinguish asphalt (rough texture, high contrast) from concrete (uniform texture, low contrast); Edge detection: Use the Canny operator to extract the edge density. The ice and snow runway shows sparse edge features due to its smooth surface.

[0055] Adopt a support vector machine (SVM) or a convolutional neural network (CNN) to train a classifier based on the labeled asphalt / concrete / ice and snow runway image dataset, and output the runway type label in real time.

[0056] In this embodiment, the preferred basic friction coefficient for the asphalt runway is [0.7, 0.8], the preferred basic friction coefficient for the concrete runway is [0.6, 0.7], and the preferred basic friction coefficient for the ice and snow runway is [0.1 - 0.2].

[0057] Call the corresponding value according to the classification result. If ice and snow coverage is detected (temperature ≤ 0°C and snow depth > 0), force the runway type to be marked as "ice and snow runway" and update .

[0058] Step S30: Obtain the vertical load of the aircraft's tires, the initial sideslip angle, and the tire pressure data to calculate the theoretical tire lateral force. Combine the deviation between the theoretical tire lateral force and the actual tire lateral force and the iterative algorithm to update the sideslip angle in real time. Combine the updated sideslip angle, the corrected friction coefficient, and the sideslip attenuation coefficient to calculate the final friction coefficient; Update the sideslip angle in real time. The method is as follows: Based on the initial sideslip angle, the vertical load of the tires, and the tire pressure data, combine the pre-stored cornering stiffness and the upper limit parameter of the main wheel load, and output the theoretical lateral force through the lateral force calculation model; Compare the theoretical lateral force with the actual tire lateral force to determine whether there is a deviation. If there is no deviation, do not update the sideslip angle. If there is a deviation, trigger iterative update of the sideslip angle based on the iterative algorithm.

[0059] Among them, the theoretical tire lateral force F y , and its calculation method is: F y =C α ×α0×(1 - F z / F zmax ); Among them, C α is the cornering stiffness, C α =1.2×10^5 N / rad, Fzmax is the upper limit of the main wheel load, and the initial sideslip angle α0 = 2° = 0.0349 rad.

[0060] Among them, the vertical load and tire pressure data of the tire are obtained according to the tire pressure sensor. The initial sideslip angle is provided based on the inertial navigation system.

[0061] Among them, the Newton-Raphson iteration algorithm is used to process the deviation value. The sideslip angle correction amount is positively correlated with the deviation value and inversely correlated with the side slip stiffness and the proportional coefficient of the vertical load relative to the load upper limit.

[0062] Final friction coefficient , and its calculation method is: On the basis of the corrected friction coefficient, a dynamic attenuation parameter is constructed by combining the product of the absolute value of the real-time updated sideslip angle and the preset sideslip attenuation coefficient. The friction performance under the slip state of the corrected friction coefficient is reduced through the dynamic attenuation parameter to obtain the final friction coefficient representing the actual taxiing condition.

[0063] ; Among them, is the sideslip attenuation coefficient, = 0.003~0.008 / deg, is the sideslip angle.

[0064] In this embodiment, the tire pressure sensor measures the vertical load F z = 12000 N; The millimeter-wave radar detects the water accumulation depth = 3 mm; The camera classifies the runway as a concrete runway and selects ; Based on the above data, it is calculated that F y = C α ×α×(1 - F z / F zmax ) = 1.2×10 5 ×0.0349×(1 - 12000 / 15000) = 1675.2 N.

[0065] Assume that the actual sensor measures the current lateral force to be 1800 N, which deviates from the theoretical value of 1675.2 N, and the deviation value is = 1800 - 1675.2 = 124.8 N.

[0066] This embodiment updates the sideslip angle based on the Newton-Raphson iteration algorithm , specifically: .

[0067] 。

[0068] 。

[0069] Step S40: Input the final friction coefficient into the ground roll model of the flight simulator to achieve real-time simulation.

[0070] In this embodiment, the dynamic friction model is encapsulated using C++ for the GRM module to call, as shown in the following example: / / Dynamic friction coefficient calculation (C++ example) double compute_mu_env(double mu0, double h_water, double h_snow,double T) { const double alpha = 0.1; / / Water accumulation attenuation coefficient (1 / mm) const double beta = 0.05; / / Snow accumulation attenuation coefficient (1 / mm) const double gamma = 0.01; / / Temperature compensation coefficient (1 / °C) const double T_ref = 25.0; / / Reference temperature (°C) double mu_wet = mu0 × (1 - alpha×h_water); double mu_snow = mu_wet×exp(-beta×h_snow); double mu_temp = mu_snow×(1 - gamma×fabs(T - T_ref)); return mu_temp; } / / Tire sideslip correction double compute_mu_final(double mu_env, double alpha_deg) { const double k = 0.005; / / Sideslip attenuation coefficient (1 / deg) return mu_env×(1 - k×fabs(alpha_deg)); } The tire coupling model can be tested using Matlab / Simulink or C++. Different runways and different water accumulation depths can be selected. Based on the error between the theoretical value and the actual value, the optimal k value can be selected.

[0071] Although the steps are described in the above order in the above embodiments, those skilled in the art can understand that for the purpose of achieving the effects of this embodiment, different steps do not have to be executed in such an order. They can be executed simultaneously (in parallel) or in a reversed order, and these simple changes are within the protection scope of the present invention.

[0072] A real-time simulation system for the dynamic friction coefficient of an aircraft runway based on multi-source data according to the second embodiment of the present invention is based on a real-time simulation method for the dynamic friction coefficient of an aircraft runway based on multi-source data. The system includes: A data acquisition module configured to acquire runway image data and runway surface temperature data, extract the water accumulation area and / or snow accumulation area and runway texture features of the runway based on the runway image data, and obtain the water accumulation depth and / or snow accumulation depth. A corrected friction coefficient calculation module configured to determine the runway type based on the runway texture features, match the basic friction coefficient according to the runway type, and dynamically correct the basic friction coefficient in combination with the water accumulation depth and / or snow accumulation depth and runway surface temperature data to obtain the corrected friction coefficient. A final friction coefficient calculation module configured to acquire the tire vertical load, initial sideslip angle, and tire pressure data of the aircraft to calculate the theoretical tire lateral force, update the sideslip angle in real time in combination with the deviation between the theoretical tire lateral force and the actual tire lateral force and an iterative algorithm, and calculate the final friction coefficient in combination with the updated sideslip angle, corrected friction coefficient, and sideslip decay coefficient. A real-time simulation module configured to input the final friction coefficient into the ground roll model of a flight simulator to achieve real-time simulation.

[0073] Those skilled in the art of the relevant technical field can clearly understand that for the convenience and conciseness of description, the specific working process and related explanations of the system described above can refer to the corresponding process in the foregoing method embodiment, and will not be elaborated here.

[0074] It should be noted that the real-time simulation system for the dynamic friction coefficient of an aircraft runway based on multi-source data provided in the above embodiments is only illustrated by dividing the above functional modules. In practical applications, the above functions can be assigned to different functional modules according to needs, that is, the modules or steps in the embodiments of the present invention can be further decomposed or combined. For example, the modules in the above embodiments can be combined into one module, or 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 present invention are only for distinguishing each module or step, and are not regarded as an improper limitation of the present invention.

[0075] An electronic device according to a third embodiment of the present invention includes: At least one processor; and A memory communicatively connected to at least one of the processors; wherein, The memory stores instructions executable by the processor, and the instructions are used to be executed by the processor to implement the above real-time simulation method for the dynamic friction coefficient of an aircraft runway based on multi-source data.

[0076] A computer-readable storage medium according to a fourth embodiment of the present invention stores computer instructions, and the computer instructions are used to be executed by the computer to implement the above real-time simulation method for the dynamic friction coefficient of an aircraft runway based on multi-source data.

[0077] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes and related descriptions of the above-described storage device and processing device can refer to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0078] Those skilled in the art should be able to realize that the modules and method steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, computer software, or a combination of the two. The programs corresponding to the software modules and method steps can be placed in a random access memory (RAM), memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, register, hard disk, removable disk, CD-ROM, or any other form of storage medium well-known in the art. To clearly illustrate the interchangeability of electronic hardware and software, the components and steps of each example have been generally described according to their functions in the above description. Whether these functions are executed in the form of 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 to exceed the scope of the present invention.

[0079] The terms "first", "second", etc. are used to distinguish similar objects and are not used to describe or indicate a particular order or sequence.

[0080] The term "comprising" or any other similar term is intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus / device that comprises a series of elements includes not only those elements but also other elements not expressly listed, or elements inherent to those processes, methods, articles, or apparatus / devices.

[0081] So far, the technical solution of the present invention has been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is easily understood by those skilled in the art that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the protection scope of the present invention.

Claims

1. A real-time simulation method for the dynamic friction coefficient of an aircraft runway based on multi-source data, characterized in that, The method includes: Obtaining runway image data and runway surface temperature data, extracting the water accumulation area and / or snow accumulation area and runway texture features of the runway based on the runway image data, and obtaining the water accumulation depth and / or snow accumulation depth; Judging the runway type based on the runway texture features, matching the basic friction coefficient according to the runway type, and dynamically correcting the basic friction coefficient in combination with the water accumulation depth and / or snow accumulation depth and runway surface temperature data to obtain a corrected friction coefficient; Obtaining the vertical load of the aircraft's tires, the initial sideslip angle, and the tire pressure data for calculating the theoretical tire lateral force, and combining the deviation between the theoretical tire lateral force and the actual tire lateral force and an iterative algorithm to update the sideslip angle in real time. Combining the updated sideslip angle, the corrected friction coefficient, and the sideslip attenuation coefficient, calculate the final friction coefficient; Input the final friction coefficient into the ground roll model of the flight simulator to achieve real-time simulation.

2. The real-time simulation method for the dynamic friction coefficient of an aircraft runway based on multi-source data according to claim 1, characterized in that, Dynamically correct the basic friction coefficient to obtain a corrected friction coefficient. The method is as follows: Establish a linear attenuation factor based on the product of the water accumulation depth and a preset water accumulation attenuation coefficient; Establish an exponential attenuation factor based on the product of the snow accumulation depth and a preset snow accumulation attenuation coefficient; Establish a temperature compensation factor according to the product of the absolute value of the difference between the runway surface temperature and the reference temperature and the temperature compensation coefficient; Compound-adjust the basic friction coefficient through three attenuation factors in sequence to obtain a corrected friction coefficient.

3. A real-time simulation method for the dynamic friction coefficient of an aircraft runway based on multi-source data according to claim 1, characterized in that The final friction coefficient, the calculation method is as follows: Based on the corrected friction coefficient, combine the product of the absolute value of the sideslip angle updated in real time and a preset sideslip attenuation coefficient to construct a dynamic attenuation parameter, and reduce the friction performance under the slip state of the corrected friction coefficient through the dynamic attenuation parameter to obtain the final friction coefficient characterizing the actual roll condition.

4. A real-time simulation method for the dynamic friction coefficient of an aircraft runway based on multi-source data according to claim 1, characterized in that, Update the sideslip angle in real time. The method is as follows: Based on the initial sideslip angle, the vertical load of the tires, and the tire pressure data, combine the pre-stored cornering stiffness and the main wheel load upper limit parameters, and output the theoretical lateral force through a lateral force calculation model; Compare the theoretical lateral force with the actual tire lateral force to judge whether there is a deviation. If there is no deviation, the sideslip angle is not updated. If there is a deviation, trigger an iterative update of the sideslip angle based on an iterative algorithm.

5. A real-time simulation method for the dynamic friction coefficient of an aircraft runway based on multi-source data according to claim 1, characterized in that The image data and the runway surface temperature data are obtained according to a camera and an infrared thermometer respectively.

6. The real-time simulation method for the dynamic friction coefficient of an aircraft runway based on multi-source data according to claim 1, characterized in that The vertical load of the tires and the tire pressure data are obtained according to a tire pressure sensor.

7. A real-time simulation method for the dynamic friction coefficient of an aircraft runway based on multi-source data according to claim 5, characterized in that Obtain the water accumulation depth and / or snow accumulation depth. The method is as follows: Detect the water accumulation area and / or snow accumulation area based on the image data, and use a millimeter-wave radar to detect the water accumulation depth and / or snow accumulation depth.

8. A real-time simulation system for the dynamic friction coefficient of an aircraft runway based on multi-source data, based on the real-time simulation method for the dynamic friction coefficient of an aircraft runway based on multi-source data according to any one of claims 1-7, characterized in that, The system includes: A data acquisition module configured to obtain runway image data and runway surface temperature data, extract the water accumulation area and / or snow accumulation area and runway texture features of the runway based on the runway image data, and obtain the water accumulation depth and / or snow accumulation depth; A corrected friction coefficient calculation module configured to judge the runway type based on the runway texture features, match the basic friction coefficient according to the runway type, and dynamically correct the basic friction coefficient in combination with the water accumulation depth and / or snow accumulation depth and runway surface temperature data to obtain a corrected friction coefficient; The final friction coefficient calculation module is configured to obtain the vertical load of the aircraft's tire, the initial sideslip angle, and the tire pressure data, and is used to calculate the theoretical tire lateral force. Combining the deviation between the theoretical tire lateral force and the actual tire lateral force and the iterative algorithm, it updates the sideslip angle in real time. Combining the updated sideslip angle, the corrected friction coefficient, and the sideslip attenuation coefficient, it calculates the final friction coefficient; The real-time simulation module is configured to input the final friction coefficient into the ground roll model of the flight simulator to achieve real-time simulation.

9. An electronic device, characterized in that, Comprising: At least one processor; And A memory communicatively connected to 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 a real-time simulation method for the dynamic friction coefficient of an aircraft runway based on multi-source data according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions, and the computer instructions are used to be executed by the computer to implement a real-time simulation method for the dynamic friction coefficient of an aircraft runway based on multi-source data according to any one of claims 1-7.

Images