Vehicle braking distance prediction method, device, equipment and storage medium

Abstract

This application discloses a vehicle braking distance prediction method, apparatus, device, and storage medium, relating to the field of vehicle driving safety technology. The method includes: acquiring a road surface texture dataset and a friction coefficient dataset; inputting the road surface texture dataset into a three-dimensional road surface reconstruction model; performing data separation, data denoising, and data reconstruction; inputting the friction coefficient dataset and the reconstructed road surface texture data into an anti-skid performance prediction model to obtain a predicted result for the road surface friction coefficient; based on the reconstructed road surface texture data and the predicted road surface friction coefficient, establishing a three-dimensional tire hydroplaning model and performing fluid-structure interaction simulation to obtain the tire-road adhesion characteristic curve of the wet road surface; inputting the tire-road adhesion characteristic curve into a vehicle braking simulation model; and calculating the vehicle braking distance on the wet road surface through dynamic simulation. This solves the problem that traditional braking distance formulas ignore actual tire-road adhesion characteristic variations, achieving high-precision prediction of braking distance in complex rainy scenarios.

Description

Technical Field

[0001] This application relates to the field of vehicle driving safety technology, and in particular to vehicle braking distance prediction methods, devices, equipment and storage media. Background Technology

[0002] Road surface skid resistance refers to the ability of the tire to generate friction between the tire and the road surface. It is a key indicator in the fields of road engineering and traffic safety, and directly affects vehicle driving safety and road service level.

[0003] Currently, the main challenges in traditional road friction performance testing lie in the varying reliability of testing equipment and the lack of standardized evaluation criteria. Traditional methods rely on manual operation, resulting in discontinuous test results, especially in rainy conditions where rainwater reduces road surface roughness. When water accumulates to a certain thickness on the road surface, a water film forms, preventing complete contact between the tire and the road surface. This makes it impossible to accurately predict the vehicle's braking distance, severely impacting vehicle stability and driving safety.

[0004] Therefore, improving the accuracy of vehicle braking distance prediction under slippery road conditions is a problem that urgently needs to be solved. Summary of the Invention

[0005] The main objective of this application is to provide a method, apparatus, device, and storage medium for predicting vehicle braking distance, aiming to solve the technical problem of improving the accuracy of vehicle braking distance prediction values ​​under slippery road conditions.

[0006] To achieve the above objectives, this application proposes a vehicle braking distance prediction method, the method comprising:

[0007] Obtain the road surface texture dataset and friction coefficient dataset;

[0008] The road surface texture dataset is input into the three-dimensional road surface reconstruction model, and data separation, data denoising and data reconstruction are performed to obtain the reconstructed road surface texture data.

[0009] The friction coefficient dataset and the reconstructed road surface texture data are input into the anti-skid performance prediction model to obtain the prediction results of the road surface friction coefficient;

[0010] Based on the reconstructed road surface texture data and the predicted road surface friction coefficient, a three-dimensional tire hydroplaning model is established and fluid-structure interaction simulation is performed to obtain the tire-road adhesion characteristic curve of wet and slippery road surface.

[0011] The tire adhesion characteristic curve is input into the vehicle braking simulation model, and the vehicle braking distance on wet and slippery roads is calculated through dynamic simulation.

[0012] In one embodiment, the step of inputting the road surface texture dataset into a three-dimensional road surface reconstruction model, performing data separation, data denoising, and data reconstruction to obtain reconstructed road surface texture data includes:

[0013] The road surface texture dataset is input into the three-dimensional road surface reconstruction model, and outlier processing is performed to obtain the initial processed data.

[0014] The initial processed data is decomposed into initial macroscopic data and initial microscopic data using a frequency domain filter;

[0015] The initial macroscopic data and the initial microscopic data are respectively subjected to noise reduction processing to obtain macroscopic texture data and microscopic texture data;

[0016] The macroscopic texture data and the microscopic texture data are integrated to obtain the reconstructed road surface texture data.

[0017] In one embodiment, the step of integrating the macroscopic texture data and the microscopic texture data to obtain reconstructed road surface texture data includes:

[0018] The macroscopic texture data and the microscopic texture data are integrated by inverse frequency domain transformation to obtain integrated texture data;

[0019] The integrated texture data is skewed by linear regression to obtain corrected texture data.

[0020] The corrected texture data is mapped to a reference plane for elevation correction to obtain reconstructed road surface texture data.

[0021] In one embodiment, before the step of inputting the friction coefficient dataset and the reconstructed road surface texture data into the anti-skid performance prediction model to obtain the prediction result of the road surface friction coefficient, the following steps are included:

[0022] The friction coefficient dataset and the reconstructed road surface texture data are matched to obtain a matching dataset, and the matching dataset is divided into a training set and a test set.

[0023] An initial anti-slip performance prediction model is established based on the recurrent neural network unit structure. The initial anti-slip performance prediction model includes a reset gate and an update gate.

[0024] The training set and test set are input into the initial anti-skid performance prediction model for model training to obtain the anti-skid performance prediction model.

[0025] In one embodiment, the step of establishing a three-dimensional tire hydroplaning model and performing fluid-structure interaction simulation based on the reconstructed road texture data and the predicted road friction coefficient to obtain the tire-road adhesion characteristic curve of the wet road surface includes:

[0026] Obtain two-dimensional tire cross-sectional data, perform rotational simulation processing on the two-dimensional tire cross-sectional data, and obtain a three-dimensional tire model;

[0027] The Eulerian domain model is partitioned according to a preset material matching method to obtain a water film model;

[0028] Based on the predicted road surface friction coefficient, the three-dimensional tire model, the three-dimensional road surface reconstruction model, and the water film model are brought into contact through node control to obtain a three-dimensional tire hydroplaning model.

[0029] Fluid-structure interaction simulation was performed on the three-dimensional tire hydroplaning model to obtain the tire-road adhesion characteristic curve of wet road surface.

[0030] In one embodiment, the step of performing fluid-structure interaction simulation on the three-dimensional tire hydroplaning model to obtain the tire road adhesion characteristic curve of the wet road surface includes:

[0031] In the three-dimensional tire hydroplaning model, the motion state of the water film fluid domain is simulated using the Euler method;

[0032] The mechanical behavior of tires and road surface solid domains was simulated using the Lagrange method;

[0033] The translation of the three-dimensional road surface reconstruction model and the water film model along the x-axis is defined as the first constraint condition;

[0034] The rotation of the 3D tire model around the y-axis is defined as the second constraint condition.

[0035] Based on the motion state of the water film fluid domain, the mechanical behavior of the tire and road surface solid domain, the first constraint condition, and the second constraint condition, the slip ratio is calculated according to the translational velocity and the tire angular velocity, and the tire-road adhesion characteristic curves under different working conditions are obtained by adjusting the slip ratio.

[0036] In one embodiment, the step of inputting the tire adhesion characteristic curve into a vehicle braking simulation model and calculating the vehicle braking distance on a wet road surface through dynamic simulation includes:

[0037] The tire adhesion characteristic curve is input into the vehicle braking simulation model, which includes a preset tire model.

[0038] Set driver control parameters and road geometry parameters;

[0039] Based on the driver control parameters and road geometry parameters, the vehicle braking simulation model is used to perform braking dynamics simulation and output the vehicle braking distance on wet and slippery roads.

[0040] Furthermore, to achieve the above objectives, this application also proposes a vehicle braking distance prediction device, which includes:

[0041] The data acquisition module is used to acquire road surface texture datasets and friction coefficient datasets;

[0042] The data reconstruction module is used to input the road surface texture dataset into the three-dimensional road surface reconstruction model, perform data separation, data noise reduction and data reconstruction to obtain the reconstructed road surface texture data.

[0043] The performance prediction module is used to input the friction coefficient dataset and the reconstructed road surface texture data into the anti-skid performance prediction model to obtain the prediction result of the road surface friction coefficient.

[0044] The coupled simulation module is used to establish a three-dimensional tire hydroplaning model and perform fluid-structure interaction simulation based on the reconstructed road texture data and the predicted road friction coefficient to obtain the tire-road adhesion characteristic curve of wet road surface.

[0045] The distance prediction module is used to input the tire adhesion characteristic curve into the vehicle braking simulation model and calculate the vehicle braking distance on wet and slippery roads through dynamic simulation.

[0046] In addition, to achieve the above objectives, this application also proposes a vehicle braking distance prediction device, the device comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the vehicle braking distance prediction method as described above.

[0047] In addition, to achieve the above objectives, this application also proposes a storage medium, which is a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the steps of the vehicle braking distance prediction method described above.

[0048] One or more technical solutions proposed in this application have at least the following technical effects:

[0049] By acquiring road surface texture and friction coefficient datasets, and simultaneously collecting road surface texture and friction coefficient data through high-precision 3D laser scanning and dynamic friction testing, the problems of data discreteness and asynchronousness in traditional methods are solved, providing high-precision input for subsequent models.

[0050] By inputting the road surface texture dataset into the 3D road surface reconstruction model, data separation, data denoising, and data reconstruction were performed. This effectively removed noise while preserving key texture features, solving the problem of 3D road surface representation distortion caused by noise interference in the original data, and providing a high-fidelity 3D road surface reconstruction model for subsequent simulations.

[0051] By inputting the friction coefficient dataset and the reconstructed pavement texture data into the anti-skid performance prediction model, the predicted results of the pavement friction coefficient are obtained. By adjusting the time step to adapt to a limited number of samples and capturing the temporal correlation between texture and friction coefficient, the problem that traditional empirical models cannot dynamically reflect the decline of pavement anti-skid performance is solved, and continuous and accurate prediction of friction coefficient is achieved.

[0052] Based on the reconstructed road texture data and the predicted road friction coefficient, a three-dimensional tire hydroplaning model was established and fluid-structure interaction simulation was performed to obtain the tire-road adhesion characteristic curve of wet road surface. This solved the problem that the traditional model ignores the dynamic behavior of water film and the dynamic changes of tire-road friction characteristics, and accurately simulated the nonlinear changes of tire adhesion in rainy weather.

[0053] The tire adhesion characteristic curve is input into the vehicle braking simulation model, and the vehicle braking distance on wet and slippery roads is calculated through dynamic simulation. By inputting the adhesion characteristic curve into the CARSIM vehicle model for braking dynamics simulation, and by coupling driver control, vehicle dynamics, and road parameters, the problem of traditional braking distance formulas ignoring actual changes in tire adhesion characteristics is solved, and high-precision prediction of braking distance in complex rainy scenarios is achieved. Attached Figure Description

[0054] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0055] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0056] Figure 1 This is a flowchart illustrating the first embodiment of the vehicle braking distance prediction method of this application;

[0057] Figure 2 This is a flowchart illustrating the second embodiment of the vehicle braking distance prediction method of this application;

[0058] Figure 3 This is a schematic diagram of the separated macroscopic and microscopic textures in an embodiment of this application;

[0059] Figure 4 This is a schematic diagram of the three-dimensional texture information data reconstruction process in an embodiment of this application;

[0060] Figure 5 This is a schematic diagram of the anti-skid prediction model structure in an embodiment of this application;

[0061] Figure 6 This is a schematic diagram showing the distribution of prediction results from the anti-skid prediction model in an embodiment of this application;

[0062] Figure 7 This is a flowchart illustrating the third embodiment of the vehicle braking distance prediction method of this application;

[0063] Figure 8 This is a schematic diagram of the ABAQUS-CARSIM co-simulation braking model according to an embodiment of this application;

[0064] Figure 9 This is a schematic diagram of the module structure of the vehicle braking distance prediction device according to an embodiment of this application;

[0065] Figure 10 This is a schematic diagram of the equipment structure of the hardware operating environment involved in the vehicle braking distance prediction method in the embodiments of this application.

[0066] The purpose, features, and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0067] It should be understood that the specific embodiments described herein are merely illustrative of the technical solutions of this application and are not intended to limit this application.

[0068] To better understand the technical solution of this application, the vehicle braking distance prediction method, vehicle braking distance prediction device, vehicle braking distance prediction equipment and storage medium provided in this application will be described in detail below with reference to the accompanying drawings and through specific embodiments and application scenarios.

[0069] It should be noted that the vehicle braking distance prediction method can be applied to the terminal, and can be executed by the hardware or software in the terminal.

[0070] The terminal includes, but is not limited to, portable communication devices such as mobile phones or tablets with touch-sensitive surfaces (e.g., touchscreen displays and / or touchpads). It should also be understood that, in some embodiments, the terminal may not be a portable communication device, but rather a desktop computer with touch-sensitive surfaces (e.g., touchscreen displays and / or touchpads).

[0071] The following embodiments describe a terminal including a display and a touch-sensitive surface. However, it should be understood that the terminal may include one or more other physical user interface devices such as a physical keyboard, mouse, and joystick.

[0072] Based on this, embodiments of this application provide a vehicle braking distance prediction method, referring to... Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of the vehicle braking distance prediction method of this application.

[0073] In this embodiment, the vehicle braking distance prediction method includes steps S10 to S50:

[0074] Step S10: Obtain the road surface texture dataset and friction coefficient dataset.

[0075] For example, a standard rutted pavement specimen can be identified, and a circular accelerated-load abrasion device can be used to wear the specimen. The asphalt pavement texture of the specimen is scanned using an LS-40 high-precision 3D laser texturing device to obtain a pavement texture dataset (which can be a 3D asphalt pavement texture dataset). The frictional properties of the specimen are measured using a dynamic friction tester to obtain a friction coefficient dataset.

[0076] Step S20: Input the road surface texture dataset into the 3D road surface reconstruction model, perform data separation, data noise reduction and data reconstruction to obtain the reconstructed road surface texture data.

[0077] It should be noted that the 3D road surface reconstruction model is used to process road surface texture data, including data separation, data denoising, and data reconstruction. It can separate the road surface texture data into macro and micro textures, denoise the separated data, and then integrate them to obtain the reconstructed road surface texture data.

[0078] Step S30: Input the friction coefficient dataset and the reconstructed road surface texture data into the anti-skid performance prediction model to obtain the prediction results of the road surface friction coefficient.

[0079] For example, the anti-skid performance prediction model can be a time-series prediction model based on a gated recurrent unit (GRU) network, used to capture the nonlinear relationship between the friction coefficient and texture decay. The predicted road friction coefficient is the output value of the anti-skid performance prediction model, representing the predicted road friction coefficient.

[0080] Step S40: Based on the reconstructed road surface texture data and the predicted road surface friction coefficient, a three-dimensional tire hydroplaning model is established and fluid-structure interaction simulation is performed to obtain the tire-road adhesion characteristic curve of the wet road surface.

[0081] It should be noted that the 3D tire hydroplaning model can be a coupled model built in ABAQUS software. The predicted road friction coefficient can be imported into the tire-road contact algorithm to simulate tire behavior on flooded roads. Fluid-structure interaction simulation can quantify the interference of water film on tire-road contact by solving the fluid-solid interaction. The tire-road adhesion characteristic curve can be understood as a function curve output from the simulation, representing the change in longitudinal adhesion force of the tire under different slip ratios.

[0082] Step S50: Input the tire adhesion characteristic curve into the vehicle braking simulation model, and calculate the vehicle braking distance on the wet and slippery road surface through dynamic simulation.

[0083] For example, a vehicle braking simulation model can be understood as a multi-module coupled system built in CARSIM software. The vehicle braking simulation model may include tire parameters, driver behavior parameters, road environment parameters, etc. Braking behavior is predicted by solving the vehicle's motion equations (such as acceleration and force balance), and the output is the vehicle braking distance on a wet road surface. The vehicle braking distance on a wet road surface represents the distance the vehicle travels from the start of braking to complete stop on a wet road surface.

[0084] In this embodiment, road surface texture and friction coefficient data are simultaneously acquired through high-precision 3D laser scanning and dynamic friction testing, solving the problems of data discreteness and asynchrony in traditional methods and providing high-precision input for subsequent models. The road surface texture dataset is input into the 3D road reconstruction model for data separation, noise reduction, and reconstruction, effectively removing noise while preserving key texture features. This solves the problem of 3D road surface distortion caused by noise interference in the original data, providing a high-fidelity 3D road surface reconstruction model for subsequent simulations. The friction coefficient dataset and reconstructed road surface texture data are input into the anti-skid performance prediction model to obtain the predicted road friction coefficient. By adjusting the time step to adapt to a limited sample and capturing the temporal correlation between texture and friction coefficient, the problem of traditional empirical models being unable to dynamically reflect the decay of road anti-skid performance is solved, achieving continuous and accurate prediction of the friction coefficient. Based on the reconstructed road surface texture data and the predicted road friction coefficient, a 3D tire hydroplaning model is established and fluid-structure interaction simulation is performed to obtain the tire-road adhesion characteristic curve of the wet road surface. This solves the problem of traditional models ignoring the dynamic behavior of water film and the dynamic changes in tire-road friction characteristics, accurately simulating the nonlinear changes in tire adhesion in rainy weather. By inputting the adhesion characteristic curve into the CARSIM vehicle model for braking dynamics simulation, and by coupling driver control, vehicle dynamics and road parameters, the problem of neglecting the actual tire-road adhesion characteristic changes in the traditional braking distance formula is solved, and high-precision prediction of braking distance in complex rainy scenarios is achieved.

[0085] Reference Figure 2 , Figure 2This is a flowchart illustrating the second embodiment of the vehicle braking distance prediction method of this application, based on the above. Figure 1 The first embodiment shown illustrates a second embodiment of the vehicle braking distance prediction method of this application.

[0086] In the second embodiment, step S20 includes:

[0087] Step S201: Input the road surface texture dataset into the three-dimensional road surface reconstruction model, perform outlier processing, and obtain the initial processed data.

[0088] It should be noted that outlier processing can be performed using Median Absolute Deviation (MAD) to identify and remove outlier data points. The initial processed data represents the optimized dataset after outlier processing, which eliminates sampling errors and provides clean input for subsequent decomposition.

[0089] For example, outlier values ​​in the road surface texture dataset are removed by outlier processing, as shown in Equations (1) and (2).

[0090] (1)

[0091] (2)

[0092] In formulas (1) and (2), x To fit the data vector of the window size, n To determine the parameter, it is generally set to 3.

[0093] Step S202: The initial processed data is decomposed into initial macroscopic data and initial microscopic data by using a frequency domain filter.

[0094] For example, a Fast Fourier Transform can be used to decompose the initial processed data into initial macroscopic data and initial microscopic data (such as...) using a Butterworth filter. Figure 3 As shown in the figure, it is specifically represented by formula (3).

[0095]

[0096] In formula (3), For transfer functions, For signal frequency, Here, n is the cutoff frequency, and n is the filter order.

[0097] Step S203: Denoise reduction is performed on the initial macro data and initial micro data respectively to obtain macro texture data and micro texture data.

[0098] For example, the discrete wavelet thresholding method can be used to denoise the initial macroscopic data to obtain macroscopic texture data; and threshold filtering can be used to perform threshold filtering on the initial microscopic data to obtain microscopic texture data.

[0099] Step S204: Integrate the macroscopic texture data and the microscopic texture data to obtain the reconstructed road surface texture data.

[0100] For example, the data reconstruction process is as follows: Figure 4 As shown, macroscopic and microscopic texture data can be merged using the frequency domain transformation method to obtain reconstructed road surface texture data.

[0101] In this embodiment, the MAD method is used for outlier processing, which solves the distortion problem caused by sampling noise in the original data and improves the reliability of the initial dataset. Frequency domain filter decomposition (FFT and Butterworth filter) solves the problem of blurred anti-skid performance evaluation caused by the inability of traditional methods to separate macro and micro texture features, achieving accurate isolation of texture features. Wavelet denoising and threshold filtering are used for noise reduction, solving the problems of micro-noise amplification and macro-feature distortion, thus enhancing the fidelity of texture data. Inverse Fourier transform integrates macro- and micro-texture data, enabling the construction of a high-precision 3D road surface model and providing a realistic foundation for subsequent simulations.

[0102] In one implementation, step S204 includes: integrating macroscopic and microscopic texture data through inverse frequency domain transformation to obtain integrated texture data; correcting the tilt of the integrated texture data through linear regression to obtain corrected texture data; and mapping the corrected texture data to a reference surface for elevation correction to obtain reconstructed road surface texture data.

[0103] It should be noted that an inverse Fourier transform can be used to merge macroscopic and microscopic texture data in the frequency domain to obtain integrated texture data. Linear regression, based on a least-squares mathematical model, can be used to quantify and correct angle errors during data acquisition. The corrected texture data is the tilt-corrected dataset, eliminating systematic biases caused by equipment tilt. The reference plane can be a pre-defined ideal horizontal reference plane, serving as the benchmark for elevation correction. The reconstructed road surface texture data is the final high-fidelity 3D texture dataset, integrating macroscopic and microscopic features and eliminating noise and tilt errors, and can be directly used for fluid-structure interaction simulations.

[0104] For example, the process of using the least squares method to correct the tilt of the integrated texture data can be expressed as formula (4) and formula (5).

[0105] (4)

[0106] (5)

[0107] In formulas (4) and (5), Indicates the measured deflection angle. For bias terms, This indicates that the acquisition points in the scan line are i The corresponding elevation value, N This represents the total number of acquisition points for a single scan line.

[0108] In this embodiment, inverse Fourier transform technology is employed to solve the phase misalignment problem caused by frequency domain separation of macro and micro textures, achieving lossless fusion of texture data. Tilt correction is performed using linear regression to address the systematic bias introduced by the tilt of the measurement equipment. Elevation correction is achieved by mapping the corrected texture data to a reference plane, resolving the inconsistency between the spatial reference of the reconstructed model and the simulation environment, and providing a high-precision input foundation for subsequent coupled simulations.

[0109] In one implementation, before step S30, the following steps are included: matching the friction coefficient dataset and the reconstructed road surface texture data to obtain a matching dataset, and dividing the matching dataset into a training set and a test set; establishing an initial anti-skid performance prediction model based on the recurrent neural network unit structure, the initial anti-skid performance prediction model including a reset gate and an update gate; and inputting the training set and the test set into the initial anti-skid performance prediction model for model training to obtain the anti-skid performance prediction model.

[0110] For example, a schematic diagram of the anti-skid performance prediction model is shown below. Figure 5 The friction coefficient dataset and reconstructed road texture data are matched, aligning the friction coefficient data with texture data at the same location or time point. The matched dataset is then divided into training and testing sets in an 8:2 ratio. A gated recurrent unit network model for friction performance prediction is constructed, specifically including the following steps:

[0111] (1) Time step selection. It should be noted that, considering the limited data sample, in order to both appropriately expand the data sample and retain some sequence information, the time step can be adjusted from 11 in the original data to 3.

[0112] (2) Reset gate operation. The implicit state of the previous time step is output using the reset gate. Perform a reset to obtain the reset data. , which are represented by formulas (6) and (7).

[0113] (6)

[0114] (7)

[0115] In formulas (6) and (7), To reset the door, It is the sigmoid activation function. These are built-in parameters of the model. As the information carrier of the previous time step, The input features are for the current time step. This is the data after the reset.

[0116] (3) tanh activation function operation. Input at the current moment The data is concatenated and then scaled to the range [-1, 1] using a tanh activation function to obtain the final result. , which is represented by formula (8).

[0117] (8)

[0118] In formula (8), As a carrier containing current feature information, for Activation function These are built-in parameters of the model.

[0119] (4) Update gate operation. The update gate control can be used. The decision on how much historical and current information to use, and how to update the current implicit state, is expressed as formula (9).

[0120]

[0121] In formula (9), To update the door, These are built-in parameters of the model.

[0122] (5) Information carriers that obtain current feature information The final output target value It is obtained by the information carrier through the sigmoid activation function, and is expressed as formula (10) and formula (11).

[0123] (10)

[0124] (11)

[0125] In formulas (10) and (11), As a carrier of current information after remembering past information, The target value to be output. These are built-in parameters of the model.

[0126] For example, the GRU parameters can be optimized using the backpropagation algorithm, and the mean squared error loss function can be used to narrow the gap between the predicted friction coefficient and the measured value. The anti-skid performance prediction model is a converged GRU network that can output the predicted road friction coefficient. Figure 6 .

[0127] In this implementation, spatiotemporal synchronized data matching strictly aligns the friction coefficient dataset with the reconstructed texture data, resolving the sampling position offset problem in traditional methods. By proportionally dividing the training and test sets, the model's generalization ability is verified while ensuring sufficient training, overcoming the overfitting bottleneck of static empirical models. An initial prediction model is constructed based on a GRU-gated recurrent unit structure. The reset gate filters noise interference during texture decay, and the update gate dynamically fuses historical states with current features, overcoming the challenge of quantifying the temporal decay law of the friction coefficient. Finally, backpropagation training optimization enables the model to accurately capture the nonlinear mapping between microscopic texture wear and macroscopic friction decay, providing reliable input for subsequent coupled simulations.

[0128] Reference Figure 7 , Figure 7 This is a flowchart illustrating the third embodiment of the vehicle braking distance prediction method of this application, based on the above. Figure 2 The third embodiment shown presents a third embodiment of the vehicle braking distance prediction method of this application.

[0129] In the third embodiment, step S40 includes:

[0130] Step S401: Obtain the two-dimensional cross-sectional data of the tire, perform rotation simulation processing on the two-dimensional cross-sectional data of the tire, and obtain a three-dimensional tire model.

[0131] It should be noted that the two-dimensional cross-sectional data of a tire can be its two-dimensional geometric contour information, including the cross-sectional shape data of the tread, sidewall, etc. The three-dimensional tire model can be a solid tire model with a three-dimensional geometric structure obtained through rotational simulation. For example, a three-dimensional tire model can be generated by rotating a two-dimensional cross-section 360 degrees around its central axis in ABAQUS.

[0132] Step S402: Divide the Eulerian domain model into partitions according to the preset material matching method to obtain the water film model.

[0133] It should be noted that the water film model can be an Eulerian mesh region created in ABAQUS to simulate fluids (water films). For example, an Eulerian domain model is created in ABAQUS, and partitions are set. Through ABAQUS material assignment, the Eulerian domain is divided according to fluid properties, such as assigning hydrodynamic parameters (density, viscosity) to the water flow partition and gas parameters to the air partition.

[0134] Step S403: Based on the predicted road surface friction coefficient, the three-dimensional tire model, the three-dimensional road surface reconstruction model, and the water film model are brought into contact through node control to obtain a three-dimensional tire hydroplaning model.

[0135] It should be noted that node control can be understood as a technique in finite element method software for setting contact pairs, achieving interaction between models by constraining the degrees of freedom of nodes. The 3D tire hydroplaning model is a coupled model of a 3D tire model, a 3D road reconstruction model, and a water film model. For example, node control is used to bring the 3D tire model, the 3D road reconstruction model, and the water film model into contact. The vertical contact between the 3D tire model and the 3D road reconstruction model is selected as hard contact, and the tangential contact is selected based on the predicted road friction coefficient.

[0136] Step S404: Perform fluid-structure interaction simulation on the three-dimensional tire hydroplaning model to obtain the tire-road adhesion characteristic curve of the wet road surface.

[0137] For example, the Eulerian-Lagrange method can be used to simulate the fluid-structure interaction behavior between tires, water flow, and road surface, and the degrees of freedom of each sub-model can be constrained and controlled through boundary conditions. The tire-road adhesion characteristic curve of wet road surface, i.e. the tire-road adhesion characteristic curve under rainy conditions, is used to simulate the driving conditions of tires on wet road surfaces.

[0138] In this embodiment, a three-dimensional tire model is generated by parameterizing a two-dimensional cross-section using rotational simulation technology, solving the problems of low efficiency and large geometric distortion in traditional manual modeling and achieving high-precision digital representation of the tire. An Eulerian domain water film model is constructed using a preset material partitioning strategy (water flow or air), overcoming the shortcomings of traditional methods that ignore the dynamic characteristics of the water film and accurately quantifying the interference of water accumulation rheological behavior on adhesion. A contact algorithm based on the friction coefficient predicted by GRU is defined, and a three-phase model is coupled through node control, overcoming the dual limitations of empirically assigning friction coefficients and the lack of water film effects in traditional simulations, thus constructing a physically realistic tire, water film, and road surface interaction system. The use of Eulerian-Lagrange fluid-structure interaction co-simulation solves the problem that a single-field theory cannot describe the coupling effect between fluid and solid.

[0139] In one embodiment, based on the third embodiment described above, step S403 includes: simulating the motion state of the water film fluid domain in the three-dimensional tire hydroplaning model using the Euler method; simulating the mechanical behavior of the tire and road surface solid domains using the Lagrange method; defining the translation of the three-dimensional road surface reconstruction model and the water film model along the x-axis as a first constraint condition; defining the rotation of the three-dimensional tire model around the y-axis as a second constraint condition; and calculating the slip ratio based on the motion state of the water film fluid domain, the mechanical behavior of the tire and road surface solid domains, the first constraint condition, and the second constraint condition, according to the translational velocity and the tire angular velocity, and obtaining tire-road adhesion characteristic curves under different working conditions by adjusting the slip ratio.

[0140] For example, the motion state of the water film fluid domain can be the dynamic response of the water film under the action of the tire, including the water flow velocity field, pressure field, and splash behavior. The real physical process of surface water on the road surface is simulated by partitioning the Eulerian domain (the water flow parameters are assigned density and viscosity). The Lagrangian method can simulate the mechanical behavior of the tire and road surface solid domain by defining the tire and road surface as Lagrangian domains (deformable meshes). The stress, strain, and displacement are solved by the finite element method to quantify the interaction between tire deformation and road surface texture. For example, the slip ratio calculation formula is expressed as formula (12).

[0141]

[0142] In formula (12), s For slip ratio, V Let x be the initial velocity applied along the x-direction to the road surface model and the water film model. ω The angular velocity applied to the tire model, R This refers to the tire radius. Adjustment is made... V and ω By adjusting the ratio of the components, the longitudinal adhesion of the tires under different slip ratios is obtained, which is the tire-road adhesion characteristic curve under rainy conditions, thus simulating the tire's driving conditions on rainy roads.

[0143] In this embodiment, the Eulerian method is used to simulate the motion state of the water film fluid domain in the three-dimensional tire hydroplaning model, solving the problem of neglecting the dynamic behavior of the water film in traditional single-field simulations and achieving accurate quantification of water flow pressure and velocity distribution. The Lagrangian method is used to simulate the mechanical behavior of the tire and road surface solid domains, overcoming the deficiency of missing solid deformation and fluid interaction in traditional techniques and accurately capturing stress-strain response. By defining the first constraint (translation along the x-axis) and the second constraint (rotation around the y-axis), the physical distortion problem caused by redundancy in simulation degrees of freedom is solved, ensuring that the model conforms to real driving constraints. Combining slip ratio calculation and continuous adjustment, the tire-road adhesion characteristic curve is obtained, overcoming the limitation of relying on experience for predicting adhesion performance under rainy conditions.

[0144] In one embodiment, based on the above embodiments and implementation methods, step S50 includes: inputting the tire adhesion characteristic curve into the vehicle braking simulation model, the vehicle braking simulation model including a preset tire model; setting driver control parameters and road geometry parameters; and performing braking dynamics simulation of braking conditions through the vehicle braking simulation model based on the driver control parameters and road geometry parameters, and outputting the vehicle braking distance on the wet and slippery road surface.

[0145] It should be noted that the preset tire model can be the tire mechanics model built into CARSIM. The vehicle braking simulation model can also include a driver module and a road module. The driver module allows input of the vehicle's initial speed, master cylinder pressure, and path following mode, while the road module allows input of the road alignment and lateral and longitudinal slopes. By solving for vehicle force balance (tire forces, air resistance), torque balance (braking torque), and motion equations, the braking distance on wet and slippery surfaces is predicted. A schematic diagram of the ABAQUS / CARSIM co-simulation braking model can be found here. Figure 8 .

[0146] In this embodiment, by embedding the tire adhesion characteristic curve generated by ABAQUS into a preset tire model, the problem of distorted adhesion prediction in rainy weather caused by the reliance on empirical friction coefficients in traditional braking models is solved. In the driver and road parameter configuration stage, the limitation of static models failing to reflect dynamic driving behavior is overcome by parametrically inputting initial speed, braking pressure, road alignment, and gradient. Finally, in the dynamic simulation stage, based on CARSIM multi-module coupled solution, an end-to-end mapping from adhesion characteristics to braking distance is achieved, improving the accuracy of braking distance prediction in rainy weather compared to traditional empirical formulas, and providing reliable data support for active safety systems.

[0147] It should be noted that the above examples are only for understanding this application and do not constitute a limitation on the vehicle braking distance prediction method of this application. Any simple modifications based on this technical concept are within the protection scope of this application.

[0148] This application also provides a vehicle braking distance prediction device, please refer to... Figure 9 The vehicle braking distance prediction device includes:

[0149] Data acquisition module 10 is used to acquire road surface texture dataset and friction coefficient dataset;

[0150] The data reconstruction module 20 is used to input the road surface texture dataset into the three-dimensional road surface reconstruction model, perform data separation, data noise reduction and data reconstruction to obtain reconstructed road surface texture data;

[0151] The performance prediction module 30 is used to input the friction coefficient dataset and the reconstructed road surface texture data into the anti-skid performance prediction model to obtain the prediction result of the road surface friction coefficient.

[0152] The coupled simulation module 40 is used to establish a three-dimensional tire hydroplaning model and perform fluid-structure interaction simulation based on the reconstructed road texture data and the predicted road friction coefficient to obtain the tire-road adhesion characteristic curve of the wet road surface.

[0153] The distance prediction module 50 is used to input the tire adhesion characteristic curve into the vehicle braking simulation model and calculate the vehicle braking distance on wet and slippery roads through dynamic simulation.

[0154] The vehicle braking distance prediction device provided in this application, employing the vehicle braking distance prediction method in the above embodiments, can solve the technical problem of how to improve the accuracy of vehicle braking distance prediction values ​​under slippery road conditions. Compared with the prior art, the beneficial effects of the vehicle braking distance prediction device provided in this application are the same as those of the vehicle braking distance prediction method provided in the above embodiments, and other technical features in the vehicle braking distance prediction device are the same as those disclosed in the methods of the above embodiments, and will not be repeated here.

[0155] This application provides a vehicle braking distance prediction device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the vehicle braking distance prediction method in the above embodiment 1.

[0156] The following is for reference. Figure 10 The diagram illustrates a structural schematic suitable for implementing a vehicle braking distance prediction device according to embodiments of this application. The vehicle braking distance prediction device in embodiments of this application may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital radio receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Description), PMPs (Portable Media Players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. Figure 10 The vehicle braking distance prediction device shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of this application.

[0157] like Figure 10As shown, the vehicle braking distance prediction device may include a processing unit 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1002 or a program loaded from a storage device 1003 into a random access memory (RAM) 1004. The RAM 1004 also stores various programs and data required for the operation of the vehicle braking distance prediction device. The processing unit 1001, ROM 1002, and RAM 1004 are interconnected via a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to the I / O interface 1006: input devices 1007 including, for example, a touchscreen, touchpad, keyboard, mouse, image sensor, microphone, accelerometer, gyroscope, etc.; output devices 1008 including, for example, a liquid crystal display (LCD), speaker, vibrator, etc.; storage devices 1003 including, for example, magnetic tape, hard disk, etc.; and communication devices 1009. The communication device 1009 allows the vehicle braking distance prediction device to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 10 Vehicle braking distance prediction devices with various systems are shown; however, it should be understood that implementation or possession of all the systems shown is not required. More or fewer systems may be implemented alternatively.

[0158] The vehicle braking distance prediction device provided in this application, employing the vehicle braking distance prediction method in the above embodiments, can solve the technical problem of how to improve the accuracy of vehicle braking distance prediction values ​​under slippery road conditions. Compared with the prior art, the beneficial effects of the vehicle braking distance prediction device provided in this application are the same as those of the vehicle braking distance prediction method provided in the above embodiments, and other technical features in this vehicle braking distance prediction device are the same as those disclosed in the previous embodiment method, and will not be repeated here.

[0159] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0160] This application provides a computer-readable storage medium having computer-readable program instructions (i.e., a computer program) stored thereon, the computer-readable program instructions being used to execute the vehicle braking distance prediction method in the above embodiments.

[0161] The computer-readable storage medium provided in this application may be, for example, a USB flash drive, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, system, or device. The program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination thereof.

[0162] Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a Local Area Network (LAN) or a Wide Area Network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0163] The readable storage medium provided in this application is a computer-readable storage medium that stores computer-readable program instructions (i.e., a computer program) for executing the above-described vehicle braking distance prediction method. This solves the technical problem of improving the accuracy of vehicle braking distance prediction values ​​under slippery road conditions. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in this application are the same as those of the vehicle braking distance prediction method provided in the above embodiments, and will not be repeated here.

[0164] The above description is only a part of the embodiments of this application and does not limit the patent scope of this application. All equivalent structural transformations made under the technical concept of this application and using the contents of the specification and drawings of this application, or direct / indirect applications in other related technical fields, are included in the patent protection scope of this application.

Claims

1. A vehicle braking distance prediction method characterized by, The method includes: Obtain the road surface texture dataset and friction coefficient dataset; The road surface texture dataset is input into a 3D road surface reconstruction model, and data separation, data denoising, and data reconstruction are performed to obtain reconstructed road surface texture data. The friction coefficient dataset and the reconstructed road surface texture data are input into the anti-skid performance prediction model to obtain the prediction result of the road surface friction coefficient. The anti-skid performance prediction model is a time-series prediction model based on a gated recurrent unit network, which is used to capture the nonlinear relationship between the friction coefficient and the texture decay. Based on the reconstructed road texture data and the predicted road friction coefficient, a three-dimensional tire hydroplaning model is established and fluid-structure interaction simulation is performed to obtain the tire-road adhesion characteristic curve of wet road surface. The tire adhesion characteristic curve is input into the vehicle braking simulation model, and the vehicle braking distance on the wet and slippery road surface is calculated through dynamic simulation. The step of establishing a three-dimensional tire hydroplaning model and performing fluid-structure interaction simulation based on the reconstructed road texture data and the predicted road friction coefficient to obtain the tire-road adhesion characteristic curve of the wet road surface includes: Obtain two-dimensional tire cross-sectional data, perform rotational simulation processing on the two-dimensional tire cross-sectional data, and obtain a three-dimensional tire model; The Eulerian domain model is partitioned according to a preset material matching method to obtain a water film model; Based on the predicted road surface friction coefficient, the three-dimensional tire model, the three-dimensional road surface reconstruction model, and the water film model are brought into contact through node control to obtain a three-dimensional tire hydroplaning model. Fluid-structure interaction simulation was performed on the three-dimensional tire hydroplaning model to obtain the tire-road adhesion characteristic curve of wet road surface.

2. The method as described in claim 1, characterized in that, The steps of inputting the road surface texture dataset into a 3D road surface reconstruction model, performing data separation, data denoising, and data reconstruction to obtain reconstructed road surface texture data include: The road surface texture dataset is input into the three-dimensional road surface reconstruction model, and outlier processing is performed to obtain the initial processed data. The initial processed data is decomposed into initial macroscopic data and initial microscopic data using a frequency domain filter; The initial macroscopic data and the initial microscopic data are respectively subjected to noise reduction processing to obtain macroscopic texture data and microscopic texture data; The macroscopic texture data and the microscopic texture data are integrated to obtain the reconstructed road surface texture data.

3. The method as described in claim 2, characterized in that, The step of integrating the macroscopic texture data and the microscopic texture data to obtain the reconstructed road surface texture data includes: The macroscopic texture data and the microscopic texture data are integrated by inverse frequency domain transformation to obtain integrated texture data; The integrated texture data is skewed by linear regression to obtain corrected texture data. The corrected texture data is mapped to a reference plane for elevation correction to obtain reconstructed road surface texture data.

4. The method as described in claim 1, characterized in that, Before the step of inputting the friction coefficient dataset and the reconstructed road texture data into the anti-skid performance prediction model to obtain the prediction result of the road friction coefficient, the following steps are included: The friction coefficient dataset and the reconstructed road surface texture data are matched to obtain a matching dataset, and the matching dataset is divided into a training set and a test set. An initial anti-slip performance prediction model is established based on the recurrent neural network unit structure. The initial anti-slip performance prediction model includes a reset gate and an update gate. The training set and test set are input into the initial anti-skid performance prediction model for model training to obtain the anti-skid performance prediction model.

5. The method as described in claim 1, characterized in that, The step of performing fluid-structure interaction simulation on the three-dimensional tire hydroplaning model to obtain the tire adhesion characteristic curve of the wet road surface includes: In the three-dimensional tire hydroplaning model, the motion state of the water film fluid domain is simulated using the Euler method; The mechanical behavior of tires and road surface solid domains was simulated using the Lagrange method; The translation of the three-dimensional road surface reconstruction model and the water film model along the x-axis is defined as the first constraint condition; The rotation of the 3D tire model around the y-axis is defined as the second constraint condition. Based on the motion state of the water film fluid domain, the mechanical behavior of the tire and road surface solid domain, the first constraint condition, and the second constraint condition, the slip ratio is calculated according to the translational velocity and the tire angular velocity, and the tire-road adhesion characteristic curves under different working conditions are obtained by adjusting the slip ratio.

6. The method according to any one of claims 1 to 5, characterized in that, The step of inputting the tire adhesion characteristic curve into the vehicle braking simulation model and calculating the vehicle braking distance on a wet road surface through dynamic simulation includes: The tire adhesion characteristic curve is input into the vehicle braking simulation model, which includes a preset tire model. Set driver control parameters and road geometry parameters; Based on the driver control parameters and road geometry parameters, the vehicle braking simulation model is used to perform braking dynamics simulation and output the vehicle braking distance on wet and slippery roads.

7. A vehicle braking distance prediction device, characterized in that, The vehicle braking distance prediction device includes: The data acquisition module is used to acquire road surface texture datasets and friction coefficient datasets; The data reconstruction module is used to input the road surface texture dataset into the three-dimensional road surface reconstruction model, perform data separation, data noise reduction and data reconstruction to obtain the reconstructed road surface texture data. The performance prediction module is used to input the friction coefficient dataset and the reconstructed road surface texture data into the anti-skid performance prediction model to obtain the prediction result of the road surface friction coefficient. The anti-skid performance prediction model is a time-series prediction model based on a gated recurrent unit network, which is used to capture the nonlinear relationship between the friction coefficient and texture decay. The coupled simulation module is used to establish a three-dimensional tire hydroplaning model and perform fluid-structure interaction simulation based on the reconstructed road texture data and the predicted road friction coefficient to obtain the tire-road adhesion characteristic curve of the wet road surface. The distance prediction module is used to input the tire adhesion characteristic curve into the vehicle braking simulation model and calculate the vehicle braking distance on wet and slippery roads through dynamic simulation. The coupled simulation module is also used to acquire two-dimensional tire cross-sectional data, perform rotational simulation processing on the two-dimensional tire cross-sectional data to obtain a three-dimensional tire model; partition the Eulerian domain model according to a preset material matching method to obtain a water film model; based on the predicted road friction coefficient, bring the three-dimensional tire model, the three-dimensional road reconstruction model, and the water film model into contact through node control to obtain a three-dimensional tire hydroplaning model; and perform fluid-structure interaction simulation on the three-dimensional tire hydroplaning model to obtain the tire-road adhesion characteristic curve of the wet road surface.

8. A vehicle braking distance prediction device, characterized in that, The device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the vehicle braking distance prediction method as described in any one of claims 1 to 6.

9. A storage medium, characterized in that, The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, it implements the steps of the vehicle braking distance prediction method as described in any one of claims 1 to 6.

Images