A vehicle hydroplaning risk identification method based on surface water film distribution prediction

Abstract

The present invention relates to a method for identifying vehicle hydroplaning risk based on surface water film distribution prediction. The method comprises the following steps: using mobile or static laser radar measurement equipment to acquire three-dimensional point cloud data of the road section to be predicted; intercepting, filtering, and performing coordinate conversion on the three-dimensional point cloud of the road section to be predicted, and constructing a three-dimensional elevation matrix of the road surface for the section to be predicted through gridding; importing rainfall intensity data, calculating the peak water film thickness of each road surface grid by solving a two-dimensional shallow water equation, and forming a two-dimensional surface water film thickness distribution; verifying the reliability of the surface water film distribution prediction; collecting typical vehicle parameters for the section to be predicted, and calculating the critical hydroplaning speed for the section to be predicted based on a modified LuGre friction model, thereby identifying the hydroplaning risk of the section to be predicted. Compared with existing technologies, the present invention can achieve rapid perception and identification of hydroplaning risk across a wide range of road surfaces and can simulate different rainfall intensities according to different regional environments, thus having greater applicability.

Description

Technical Field

[0001] The present invention relates to the field of road engineering, and in particular to a vehicle hydroplaning risk identification method based on surface water film distribution prediction. Background Art

[0002] Hydroplaning caused by pavement water film is a major cause of traffic accidents. Excessively thick pavement water film can cause tires to separate from the road surface at high speeds, leading to loss of control. Rapidly and accurately sensing and predicting the pavement water film thickness distribution and further assessing the risk of hydroplaning can provide timely and effective guidance for safe driving and prevent traffic accidents caused by hydroplaning.

[0003] Existing methods for estimating water film thickness and aquaplaning risk rely on laser-based sensing equipment installed on the roadside to detect waterlogging at fixed locations or areas, or by embedding sensors in the road surface to monitor water film thickness. These methods offer high accuracy and real-time performance, but they have limited monitoring or detection ranges. Large-scale water film monitoring requires additional fixed monitoring points, increasing sensing costs.

[0004] LiDAR mapping equipment can quickly capture the three-dimensional morphological characteristics of a road surface. Simulating varying rainfall intensities and constructing a road drainage equation can then determine the distribution of water film across the road surface under various scenarios. Empirical formulas can also be used to analyze the speed of hydroplaning, providing an effective and reliable method for estimating water film thickness and the risk of hydroplaning. However, a mature, comprehensive method or technology for this purpose is currently lacking. Summary of the Invention

[0005] The purpose of the present invention is to overcome the defects of the above-mentioned prior art and provide a vehicle hydroplaning risk identification method based on surface water film distribution prediction.

[0006] The purpose of the present invention can be achieved by the following technical solutions:

[0007] A method for identifying vehicle hydroplaning risk based on surface water film distribution prediction, wherein the surface water film distribution is the distribution of water film thickness on the road surface in a two-dimensional plane, comprises the following steps:

[0008] Step 1: Determine the road section to be predicted;

[0009] Step 2: Using a mobile or static laser radar measurement device to obtain the original three-dimensional point cloud data of the road section to be predicted;

[0010] Step 3: intercepting, filtering, and coordinate converting the original 3D point cloud data of the road section to be predicted to obtain 3D point cloud data of the road surface within the road curb, gridding the 3D point cloud of the road surface, and constructing a 3D elevation matrix of the road surface of the road section to be estimated;

[0011] Step 4: Based on the three-dimensional pavement elevation matrix, a surface water film thickness calculation model is constructed and rainfall intensity data is imported; a two-dimensional shallow water equation is constructed and solved to calculate the peak water film thickness of each pavement grid, thereby forming an estimated surface water film thickness distributed on a two-dimensional plane;

[0012] Step 5: Use a laser-based water film thickness detection device to sample and detect the actual surface water film thickness of the predicted road section to obtain the measured surface water film thickness; calculate the difference between the measured surface water film thickness and the estimated surface water film thickness. If the difference is less than a preset accuracy judgment threshold, the surface water film thickness prediction result is considered feasible and the next step is executed. Otherwise, repeat the previous step and adjust the parameters until the difference is less than the preset accuracy judgment threshold.

[0013] Step 6: Collect typical vehicle parameters of the road section to be predicted, and calculate the critical hydroplaning speed of the road section to be predicted based on the modified LuGre friction model;

[0014] Step 7: Identify the hydroplaning risk of the vehicle on the road section to be predicted based on the critical hydroplaning speed of the road section to be predicted.

[0015] Furthermore, the original three-dimensional point cloud data covers the entire width of the road section to be predicted, and the original three-dimensional point cloud data is stored in a three-dimensional coordinate format, such as in an X, Y, and Z coordinate format.

[0016] Furthermore, the step 3 specifically includes the following steps:

[0017] Step 301: importing original 3D point cloud data, determining a height threshold H of the road surface area, and removing 3D point cloud data above the height threshold H from the original 3D point cloud data to obtain pre-screened 3D point cloud data;

[0018] Step 302: extracting the three-dimensional point cloud data belonging to the road surface from the initially screened three-dimensional point cloud data;

[0019] Step 303: extracting 3D point cloud data of a road cross section from the 3D point cloud data of the road surface plane, segmenting the curb edge, extracting 3D point cloud data of the road surface within the curb, and outputting the data in a 3D coordinate format, such as in X, Y, Z format;

[0020] Step 304: performing coordinate transformation on the road surface three-dimensional point cloud data, converting it from the original coordinate system to the local coordinate system;

[0021] Step 305: setting a grid size d, gridding the road surface three-dimensional point cloud, and performing interpolation calculation on the road surface three-dimensional point cloud data in three-dimensional coordinate format to obtain road surface three-dimensional data in grid form;

[0022] Step 306: Filter the three-dimensional point cloud data of the road surface in the grid format to remove abnormal points on the road surface;

[0023] Step 307: Obtain the filtered road surface three-dimensional point cloud data in grid form, and output it as the road surface three-dimensional elevation matrix of the road section to be predicted.

[0024] Furthermore, in step 301 , a straight-through filtering method is used to remove three-dimensional point cloud data with a height higher than a height threshold H from the original three-dimensional point cloud data.

[0025] Furthermore, in step 302, a random sampling consistency algorithm is used to extract the three-dimensional point cloud data belonging to the road surface plane from the initially screened three-dimensional point cloud data.

[0026] Furthermore, in step 304, the calculation formula for the coordinate transformation is:

[0027]

[0028]

[0029] In the formula, x, y, and z represent the coordinates of the road surface three-dimensional point cloud data, x t 、y t 、z t Indicates the coordinate after translation, x r 、y r 、z r represents the rotated coordinates, n is the total number of road 3D point cloud data, and θ is the angle between the road extension direction and the y-axis.

[0030] Furthermore, in step 4, the two-dimensional shallow water equation is:

[0031] U=(h,uh,vh) T

[0032]

[0033]

[0034]

[0035]

[0036] Where h is the thickness of the water film (m), u is the component of the water velocity along the x direction (m / s), v is the component of the water velocity along the y direction (m / s), and q r is the rainfall intensity (m / s), g is the acceleration of gravity (m / s 2 ), Z is the road surface three-dimensional elevation matrix (s), n c is the Manning coefficient.

[0037] Furthermore, the step 4 specifically includes the following steps:

[0038] S401: Based on the three-dimensional road elevation matrix, a surface water film thickness calculation model is constructed and model parameters are input. The model parameters include the three-dimensional road elevation matrix Z, rainfall intensity q r , Manning coefficient n c ;

[0039] S402: Setting the model solution time T, time step Δt, space step Δx, and space step Δy;

[0040] S403: Initialize the water film thickness h, the component u of the water flow velocity along the x direction, the component v of the water flow velocity along the y direction, and the model time t;

[0041] S404: Use the approximate Riemann solution of the HLL format to solve the boundary flux E of the road surface grid unit Ω. The calculation formula is:

[0042]

[0043] Where L and R represent the left and right sides of the road grid unit Ω, respectively, and S L is the left wave speed, S R is the right wave speed, S L With S R The calculation formulas are:

[0044]

[0045]

[0046] Where n=(n x ,n y ) is the outward normal vector of the road mesh unit Ω, v=(u,v) is the velocity vector, v L is the left velocity vector, v R is the right velocity vector, h L is the thickness of the left water film, h R is the thickness of the right water film, and u * and The calculation formulas are:

[0047]

[0048]

[0049] S405: Using the second-order Runge-Kutta method to perform time discretization on the model, and advancing the model time to t=(n+1)Δt, n>0, the calculation formula is:

[0050]

[0051]

[0052] Where Γ is the boundary of the pavement grid unit Ω, and E = (F, G) is the boundary flux of the pavement grid unit Ω;

[0053] S406: Verify the CFL condition, the calculation formula is:

[0054]

[0055] S407: If the CFL condition is met, execute step S408; if the CFL condition is not met, reduce the time step Δt or increase the spatial step Δx and spatial step Δy, roll back the model time to t = nΔt, and execute step S408;

[0056] S408: Repeat steps S404 to S407 until the model time t≥ the model solution time T, and then output the model calculation value h, that is, the estimated surface water film thickness distributed on the two-dimensional plane.

[0057] Furthermore, in step 5, the parameter is adjusted to the Manning coefficient n c Make adjustments.

[0058] Furthermore, the modified LuGre friction model in step 6 is:

[0059]

[0060] v r =wr-v

[0061]

[0062]

[0063]

[0064] v s =1.46e -690h+0.57 +5.24

[0065] Where μ is the adhesion coefficient between the tire and the road (m), v ris the relative speed of the tire (m / s), w is the angular velocity of the tire (rad / s), v is the speed of the tire (m / s), r is the radius of the tire (m), v s is the Stribeck velocity (m / s), h is the water film thickness (m), γ is the viscosity of water (Pa·s), r0 is the equivalent radius of the tire mark (m), L is the tire mark length (m), p is the tire pressure (Pa), and h min is the minimum water film thickness between the tire and the road surface (m), h0=0.01h is the initial water film thickness in the squeezed water film area (m), ε is the average cross-sectional depth of the asphalt pavement (m), and ρ is the density of water (kg / m 3) , w t is the tire width (m), F n is the vertical force acting on the tire (N).

[0066] Furthermore, the calculation of the critical hydroplaning speed in step 6 specifically includes the following steps:

[0067] Step 601: Determine tire parameters, wheel track geometry parameters, road surface texture parameters, water film thickness parameters, and vehicle motion parameters required to modify the LuGre friction model;

[0068] Step 602: Calculate the adhesion coefficient between the tire and the road surface at different vehicle speeds using the modified LuGre friction model;

[0069] Step 603: Based on the calculation results of the tire-road adhesion coefficient at different vehicle speeds obtained in step 602, the minimum vehicle speed when the adhesion coefficient is 0 is the critical hydroplaning speed.

[0070] Compared with the prior art, the present invention has the following beneficial effects:

[0071] (1) The present invention provides a complete implementation method, which uses lidar mapping equipment to quickly obtain the three-dimensional morphological characteristics of the road surface. On this basis, by simulating different rainfall intensities and constructing road drainage equations, the water film distribution on the road surface under different scenarios can be solved, and the road surface hydroplaning speed can be further analyzed through empirical formulas. It is an effective and reliable method for identifying water film thickness and hydroplaning risk.

[0072] (2) The present invention uses laser radar equipment to obtain the three-dimensional point cloud data of the road section to be predicted, which can realize the rapid and full-width collection of three-dimensional point cloud information of the road surface, and has the advantages of high precision, strong real-time performance and wide monitoring range.

[0073] (3) The present invention provides and deduces through a two-dimensional shallow water equation, which can solve the change process of road surface water film thickness under different rainfall intensities. Compared with the traditional point-type water film thickness sensing method based on roadside and embedded sensors, the method proposed by the present invention can achieve rapid perception and prediction of large-scale road surface aquaplaning risks, and can simulate different rainfall intensities according to different regional environments, with better applicability.

[0074] (4) The present invention identifies the risk of hydroplaning of a vehicle based on the prediction of the thickness of the water film on the road surface. The identification result can provide reliable support for vehicle driving guidance and driving information assistance, and has good practicality.

[0075] (5) By collecting typical vehicle parameters of the road section to be predicted, the present invention proposes a modified LuGre friction model for calculating the critical hydroplaning speed of the road section to be predicted. This model can solve the critical hydroplaning speed under different vehicle driving conditions and different road conditions. It has richer application scenarios and more reliable calculation results. BRIEF DESCRIPTION OF THE DRAWINGS

[0076] Figure 1 Flow chart of the implementation of the present invention;

[0077] Figure 2 It is the process of processing the original 3D point cloud data;

[0078] Figure 3 It is a three-dimensional point cloud map of the road surface;

[0079] Figure 4 The calculation process of the water film thickness on the pavement surface;

[0080] Figure 5 Schematic diagram of the three-dimensional shape of the road surface;

[0081] Figure 6 Schematic diagram of water film distribution on the pavement surface;

[0082] Figure 7 Schematic diagram of the risk distribution of aquaplaning on roads. DETAILED DESCRIPTION

[0083] The present invention is described in detail below with reference to the accompanying drawings and specific embodiments. This embodiment is implemented based on the technical solution of the present invention, and provides a detailed implementation method and specific operation process, but the protection scope of the present invention is not limited to the following embodiments.

[0084] like Figure 1 As shown, the present invention provides a vehicle hydroplaning risk identification method based on the prediction of surface water film distribution. The surface water film distribution is the distribution of water film thickness on the road surface in a two-dimensional plane, comprising the following steps:

[0085] Step 1: Determine the road section to be predicted;

[0086] Step 2: Use a mobile or static LiDAR measurement device to obtain the original three-dimensional point cloud data of the road section to be predicted. The original three-dimensional point cloud data is stored in a three-dimensional coordinate format, such as an X, Y, and Z coordinate format.

[0087] Step 3: The original 3D point cloud data of the road section to be predicted is intercepted, filtered, and coordinate transformed to obtain the 3D point cloud data of the road surface within the road curb. The 3D point cloud of the road surface is gridded to construct the 3D elevation matrix of the road surface of the road section to be estimated.

[0088] Step 4: Using the 3D elevation matrix of the road surface to be predicted as input, a surface water film thickness calculation model is constructed, and rainfall intensity data is imported. By solving the 2D shallow water equation, the peak water film thickness of each road surface grid is calculated, thereby forming an estimated surface water film thickness distributed on a 2D plane.

[0089] Step 5: Use laser water film thickness detection equipment to sample and detect the actual surface water film thickness of the predicted road section to obtain the measured surface water film thickness; calculate the difference between the measured surface water film thickness and the estimated surface water film thickness. If the difference is less than the preset accuracy judgment threshold, it is considered that the surface water film thickness prediction result is feasible and the next step is executed. Otherwise, repeat the previous step and adjust the parameters, that is, adjust the Manning coefficient n in step 3. c Adjustments are made until the difference is less than a preset accuracy determination threshold;

[0090] Step 6: Collect typical vehicle parameters for the road section to be predicted, calculate the critical hydroplaning speed of the road section to be predicted based on the modified LuGre friction model; and identify the hydroplaning risk of vehicles on the road section to be predicted based on the critical hydroplaning speed of the road section to be predicted.

[0091] In this embodiment, for step 2, to ensure the effectiveness of the water film thickness prediction, the density of the collected raw 3D point cloud data must be sufficiently high, the raw 3D point cloud data must cover the entire width of the road section to be predicted, and the point cloud data collection interval must be less than 0.5m. In a specific embodiment, the collection method can be mobile or static collection. Mobile collection uses vehicle-mounted or drone-mounted LiDAR equipment, while static collection uses stand-mounted LiDAR equipment. In the mobile collection method, the LiDAR equipment should be equipped with an elevation correction module to avoid road surface point cloud measurement errors caused by the ups and downs of the vehicle or drone.

[0092] In this embodiment, step 3 analyzes the original point cloud data obtained, converts it into a standard grid format that can be used to calculate the surface water film distribution, and performs filtering to remove abnormal point cloud data. The processing flow is as follows: Figure 2 As shown, the specific steps include:

[0093] Step 301: Import Figure 3 The original 3D point cloud data shown is used to determine the height threshold H of the road area, and the 3D point cloud data with a height higher than the threshold H is removed from the original 3D point cloud data using a straight-through filtering method to obtain the primary screening 3D point cloud data;

[0094] Step 302: Using a random sampling consistency algorithm, setting a distance threshold XX, and extracting the 3D point cloud data belonging to the road surface from the pre-screened 3D point cloud data;

[0095] Step 303: extracting 3D point cloud data of the road surface cross section from the 3D point cloud data of the road surface plane, segmenting the curb edge, extracting the 3D point cloud data of the road surface within the curb, and outputting it in a 3D coordinate format, such as in X, Y, Z format;

[0096] Step 304: performing coordinate transformation on the road surface 3D point cloud data, converting it from the original coordinate system to the local coordinate system;

[0097] Step 305: Setting a grid size d, meshing the road surface 3D point cloud, and performing interpolation calculation on the road surface 3D point cloud data in 3D coordinate format to obtain 3D road surface data in grid form. In specific implementation, the interpolation method can be selected from cubic, spline interpolation, or linear interpolation.

[0098] Step 306: filtering the grid-shaped 3D point cloud data to remove abnormal points on the road surface. In specific implementation, the filtering method may be wavelet filtering, bandpass filtering, median filtering, or mean filtering.

[0099] Step 307: Obtain the filtered road surface 3D point cloud data in grid form as the output of the road surface 3D elevation matrix of the road section to be predicted, such as Figure 4 shown.

[0100] In this embodiment, for step 301, the originally collected point cloud data usually includes trees, lamp poles, buildings and other facilities on the roadside. Therefore, a straight-through filtering method is used to eliminate the point cloud data in higher areas. The height threshold H is determined according to the road surface morphology of the road section to be estimated. In order to retain the point cloud data of the road surface area to the greatest extent, H is usually taken as 2 to 5 meters. In this embodiment, the height threshold H is 2 meters.

[0101] In this embodiment, for step 302, the random sampling consistency algorithm requires setting a distance threshold XX, an initial number of iterations, and a model confidence level. The point cloud data corresponding to the optimal road surface point cloud index is then saved through iterative calculation. In this embodiment, the distance threshold, initial number of iterations, and confidence level are selected as 0.2m, 99.5%, and 15 times, respectively.

[0102] In this embodiment, for step 304, the coordinate transformation may be performed using the following calculation formula:

[0103]

[0104]

[0105] In the formula, x, y, and z represent the coordinates of the road surface three-dimensional point cloud data, x t 、y t 、z t Indicates the coordinate after translation, x r 、y r 、z r represents the rotated coordinates, n is the total number of road 3D point cloud data, and θ is the angle between the road extension direction and the y-axis.

[0106] In step 306, this embodiment employs a filtering method based on wavelet decomposition. First, the gridded 3D road surface data is decomposed twice in both the x and y directions using the db4 wavelet basis. This yields nine wavelet decomposition components, including one residual component and eight 2D components in different spatial frequency bands. Observation reveals that each of these eight 2D components contains some anomalies caused by foreign objects on the road surface and measurement errors. Therefore, only one residual component is retained in this step, which serves as the output of the 3D road surface elevation matrix for the road section to be estimated in step 307.

[0107] In step 4, the two-dimensional shallow water equation is:

[0108] U=(h,uh,vh) T

[0109]

[0110]

[0111]

[0112]

[0113] Where h is the thickness of the water film (m), u is the component of the water velocity along the x direction (m / s), v is the component of the water velocity along the y direction (m / s), and q r is the rainfall intensity (m / s), g is the acceleration of gravity (m / s 2 ), Z is the road surface three-dimensional elevation matrix (s), n c In this embodiment, U, F, Q and G are intermediate parameters for solving the water film thickness h.

[0114] For step 4, by deriving the two-dimensional shallow water equation and using the numerical simulation solution method, the peak water film thickness of each road surface grid under any rainfall environment can be calculated. The processing flow is as follows: Figure 5 As shown, the specific steps include:

[0115] S401: Based on the pavement 3D elevation matrix, a surface water film thickness calculation model is constructed and model parameters are input. The model parameters include the pavement 3D elevation matrix Z, rainfall intensity q r , Manning coefficient n c In this embodiment, the rainfall intensity q r To measure the rainfall intensity, set the Manning coefficient n of the road section c is 0.013;

[0116] S402: Set the model solution time T, time step Δt, spatial step Δx, and spatial step Δy. In this embodiment, T is set to 120s, Δt is set to 0.025s, Δx is set to 0.1m, and Δy is set to 0.1m.

[0117] S403: Initialize the water film thickness h, the component u of the water flow velocity along the x-direction, the component v of the water flow velocity along the y-direction, and the model time t. In this embodiment, the model parameters are initialized as t=h=u=v=0.

[0118] S404: Use the approximate Riemann solution of the HLL format to solve the boundary flux E of the road surface grid unit Ω. The calculation formula is:

[0119]

[0120] Where L and R represent the left and right sides of the road grid unit Ω, respectively, and S L is the left wave speed, S R is the right wave speed, S L With S R The calculation formulas are:

[0121]

[0122]

[0123] Where n=(n x ,n y ) is the outward normal vector of the road grid unit Ω, v=(u,v) is the velocity vector, v L is the left velocity vector, v R is the right velocity vector, h L is the thickness of the left water film, h R is the thickness of the right water film, and u * and The calculation formulas are:

[0124]

[0125]

[0126] S405: Use the second-order Runge-Kutta method to discretize the model in time and advance the model time to t = (n + 1)Δt, n>0. The calculation formula is:

[0127]

[0128]

[0129] Where Γ is the boundary of the pavement grid unit Ω, and E = (F, G) is the boundary flux of the pavement grid unit Ω;

[0130] S406: Verify the CFL condition, the calculation formula is:

[0131]

[0132] S407: If the CFL (Courant-Friedrichs-Lewy) condition is met, step S408 is executed. If the CFL condition is not met, the time step Δt is reduced or the spatial step Δx and spatial step Δy are increased, the model time is rolled back to t = nΔt, and step S408 is executed.

[0133] S408: Repeat steps S404 to S407 until the model time t≥120s, then output the model calculation value h, that is, the estimated surface water film thickness distributed on the two-dimensional plane, such as Figure 6 shown.

[0134] For step 401 , the measured rainfall intensity data may be measured by a tipping bucket rain gauge, a siphon rain gauge, or a weighing rain gauge, and the unit of the rainfall intensity data is mm / min.

[0135] The modified LuGre friction model in step 6 is:

[0136]

[0137] v r =wr-v

[0138]

[0139]

[0140]

[0141] v s =1.46e -690h+0.57+5.24

[0142] Where μ is the adhesion coefficient between the tire and the road (m), v r is the relative speed of the tire (m / s), w is the angular velocity of the tire (rad / s), v is the speed of the tire (m / s), r is the radius of the tire (m), v s is the Stribeck velocity (m / s), h is the water film thickness (m), γ is the viscosity of water (Pa·s), r0 is the equivalent radius of the tire mark (m), L is the tire mark length (m), p is the tire pressure (Pa), and h min is the minimum water film thickness between the tire and the road surface (m), h0=0.01h is the initial water film thickness in the squeezed water film area (m), ε is the average cross-sectional depth of the asphalt pavement (m), and ρ is the density of water (kg / m 3) , w t is the tire width (m), F n is the vertical force acting on the tire (N).

[0143] The squeezed water film area refers to the area where the tire contacts the water film; for rectangular wheel prints, the equivalent radius of the wheel print is b is the wheel track width (m).

[0144] In this embodiment, the calculation of the critical hydroplaning speed in step 6 specifically includes the following steps:

[0145] Step 601: Determine tire parameters, wheel print geometry parameters, road surface texture parameters, water film thickness parameters, and vehicle motion parameters required for modifying the LuGre friction model based on actual conditions;

[0146] Step 602: Calculate the adhesion coefficient between the tire and the road surface within the vehicle speed range of 10 km / h to 150 km / h using the modified LuGre friction model at a calculation interval of 0.1 km / h.

[0147] Step 603: Based on the calculation results of the tire-road adhesion coefficient at different vehicle speeds obtained in step 602, the minimum vehicle speed when the adhesion coefficient is 0 is the critical hydroplaning speed.

[0148] For steps 5 and 6, this embodiment is directed to Figure 6 The water film distribution on the road surface output in the experiment was sampled and verified, and the critical hydroplaning speed was calculated using the modified LuGre friction model to evaluate the hydroplaning risk of the road section in this embodiment. Figure 7 The results show that when the vehicle speed reaches 100 km / h or above, hydroplaning is very likely to occur in some areas. This evaluation result can provide reliable support for vehicle driving guidance and driving information assistance.

[0149] The above describes in detail the preferred embodiments of the present invention. It should be understood that those skilled in the art can make numerous modifications and variations based on the concepts of the present invention without inventive effort. Therefore, any technical solutions that can be derived by those skilled in the art through logical analysis, reasoning, or limited experimentation based on the concepts of the present invention and the prior art should be within the scope of protection defined by the claims.

Claims

1. A method for identifying vehicle hydroplaning risk based on surface water film distribution prediction, wherein the surface water film distribution is the distribution of water film thickness on the road surface in a two-dimensional plane, characterized in that: The following steps are involved: Step 1: Determine the road section to be predicted; Step 2: Using a laser radar measurement device to obtain the original three-dimensional point cloud data of the road section to be predicted; Step 3: Analyze and process the original three-dimensional point cloud data to construct a three-dimensional elevation matrix of the road surface of the road section to be predicted; Step 4: Based on the three-dimensional road elevation matrix, a surface water film thickness calculation model is constructed; a two-dimensional shallow water equation is constructed, and the surface water film thickness calculation model is solved by solving the two-dimensional shallow water equation to obtain an estimated surface water film thickness distributed on a two-dimensional plane; Step 5: Use water film thickness detection equipment to detect the actual surface water film thickness of the predicted road section to obtain the measured surface water film thickness; Calculate the difference between the measured water film thickness and the estimated water film thickness. If the difference is less than a preset accuracy threshold, proceed to the next step. Otherwise, repeat the previous step and adjust the parameters until the difference is less than the preset accuracy threshold. Step 6: Collect typical vehicle parameters of the road section to be predicted, and calculate the critical hydroplaning speed of the road section to be predicted based on the modified LuGre friction model; Step 7: Identify the hydroplaning risk of the vehicle on the road section to be predicted based on the critical hydroplaning speed of the road section to be predicted.

2. The method for identifying vehicle hydroplaning risk based on surface water film distribution prediction according to claim 1, characterized in that: The original three-dimensional point cloud data covers the entire width of the road section to be predicted, and the original three-dimensional point cloud data is stored in a three-dimensional coordinate format.

3. The method for identifying vehicle hydroplaning risk based on surface water film distribution prediction according to claim 1, characterized in that: Step 3: Analyze the original three-dimensional point cloud of the road section to be predicted, wherein the analysis includes interception, filtering and coordinate conversion processing to obtain three-dimensional point cloud data of the road surface within the road curb, grid the three-dimensional point cloud data of the road surface, and construct a three-dimensional elevation matrix of the road section to be predicted; The specific steps include: Step 301: importing original 3D point cloud data, determining a height threshold H of the road surface area, and removing 3D point cloud data above the height threshold H from the original 3D point cloud data to obtain pre-screened 3D point cloud data; Step 302: extracting the three-dimensional point cloud data belonging to the road surface from the initially screened three-dimensional point cloud data; Step 303: extracting 3D point cloud data of a road cross section from the 3D point cloud data of the road surface plane, segmenting the curb edge, extracting 3D point cloud data of the road surface within the curb, and outputting it in a 3D coordinate format; Step 304: performing coordinate transformation on the road surface three-dimensional point cloud data, converting it from the original coordinate system to the local coordinate system; Step 305: setting a grid size d, gridding the road surface three-dimensional point cloud, and performing interpolation calculation on the road surface three-dimensional point cloud data in three-dimensional coordinate format to obtain road surface three-dimensional data in grid form; Step 306: Filter the grid-formatted 3D point cloud data to remove abnormal points on the road surface. Step 307: Obtain the filtered road surface three-dimensional point cloud data in grid form, and output it as the road surface three-dimensional elevation matrix of the road section to be predicted.

4. The method for identifying vehicle hydroplaning risk based on surface water film distribution prediction according to claim 3, characterized in that: In step 301 , a straight-through filtering method is used to remove three-dimensional point cloud data with a height higher than a height threshold H from the original three-dimensional point cloud data.

5. The method for identifying vehicle hydroplaning risk based on surface water film distribution prediction according to claim 3, characterized in that: In step 302, a random sampling consistency algorithm is used to extract the three-dimensional point cloud data belonging to the road surface plane from the initially screened three-dimensional point cloud data.

6. The method for identifying vehicle hydroplaning risk based on surface water film distribution prediction according to claim 1, characterized in that: In step 4, the two-dimensional shallow water equation is: U=(h,uh,vh) T Where h is the thickness of the water film, u is the component of the water velocity along the x direction, v is the component of the water velocity along the y direction, and q r is the rainfall intensity, g is the acceleration of gravity, Z is the three-dimensional elevation matrix of the road surface, n c is the Manning coefficient.

7. The method for identifying vehicle hydroplaning risk based on surface water film distribution prediction according to claim 6, characterized in that: The step 4 comprises the following steps: Based on the three-dimensional elevation matrix of the road surface, a surface water film thickness calculation model is constructed, and the model parameters are input. The model parameters include the three-dimensional elevation matrix Z of the road surface, the rainfall intensity q r , Manning coefficient n c ; Set the model solution time T, time step Δt, space step Δx and space step Δy; Initialize the water film thickness h, the component of the water velocity along the x direction u, the component of the water velocity along the y direction v and the model time t; Verify the CFL condition, the calculation formula is: If the CFL condition is met, proceed to the next step. If the CFL condition is not met, reduce the time step Δt or increase the spatial step Δx and spatial step Δy, set the model time to t = nΔt, and proceed to the next step. Verify whether the model time t is greater than or equal to the model solution time T. If so, output the model calculation value h, which is the estimated surface water film thickness distributed on the two-dimensional plane.

8. The method for identifying vehicle hydroplaning risk based on surface water film distribution prediction according to claim 6, characterized in that: In step 5, the parameters are adjusted to the Manning coefficient n c Make adjustments.

9. The method for identifying vehicle hydroplaning risk based on surface water film distribution prediction according to claim 1, characterized in that: The modified LuGre friction model in step 6 is: v r =wr-v v s =1.46e -690h+0.57 +5.24 Where μ is the adhesion coefficient between the tire and the road, v r is the relative speed of the tire, w is the angular velocity of the tire, v is the speed of the tire, r is the radius of the tire, v s is the Stribeck velocity, h is the water film thickness, γ is the viscosity of water, r0 is the equivalent radius of the tire mark, L is the tire mark length, p is the tire pressure, h min is the minimum water film thickness between the tire and the road surface, h0=0.01h is the initial water film thickness in the squeezed water film area, ε is the average cross-sectional depth of the asphalt pavement, ρ is the density of water, and w t is the tire width, F n is the vertical force acting on the tire.

10. The vehicle hydroplaning risk identification method based on surface water film distribution prediction according to claim 9, characterized in that: The calculation of the critical hydroplaning speed in step 6 includes the following steps: Step 601: Determine tire parameters, wheel track geometry parameters, road surface texture parameters, water film thickness parameters, and vehicle motion parameters required to modify the LuGre friction model; Step 602: Calculate the adhesion coefficient between the tire and the road surface at different vehicle speeds using the modified LuGre friction model; Step 603: Based on the calculation results of the tire-road adhesion coefficient at different vehicle speeds obtained in step 602, the minimum vehicle speed when the adhesion coefficient is 0 is the critical hydroplaning speed.

Images