A vehicle-cloud integrated wet and slippery road surface vehicle driving risk early warning method and system

Abstract

The present application relates to a kind of vehicle cloud integrated wet and slippery road vehicle driving risk early warning method and system, comprising the following steps: obtaining the forward video data of vehicle, the type of current driving road is identified according to forward video data by road segmentation and road classification multi-task model, and road adhesion coefficient is obtained by pre-established road adhesion coefficient reference table and uploaded to cloud end;Continuously statistics global road adhesion coefficient, establish road adhesion coefficient fusion model, fuse global road adhesion coefficient by road section matching and introducing time difference factor, generate wet and slippery road map;The early warning area corresponding to any wet and slippery road section on wet and slippery road map is determined by dynamic calculation method;For wet and slippery road section, the phase plane stability boundary of different types of vehicles is calculated, typical driving behavior is simulated, instability risk is evaluated and early warning information is sent out.Compared with prior art, the present application realizes the observation of wet and slippery road, has the advantages such as high early warning precision, strong pertinence etc.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of driving risk early warning, in particular to a vehicle-cloud integrated wet road driving risk early warning method and system. BACKGROUND

[0002] Highway traffic accidents related to wet road surfaces such as water accumulation, snow accumulation, slush, and ice account for a significant proportion of traffic accidents, and wet road surfaces seriously affect driving safety.

[0003] Currently, fixed roadside meteorological monitoring equipment is mainly relied on to measure meteorological values such as short-term rainfall, snowfall, and air temperature, and then driving risk levels are divided according to experience, and finally standardized early warning information is pushed to drivers through variable message boards. However, the existing technology lacks direct observation means for wet road surfaces, cannot continuously observe road conditions, and has limited application scenarios. In addition, the risk level divided according to experience has a coarse granularity, and the targeting of the early warning information is not strong, resulting in low compliance of drivers and poor risk prevention and control effect.

[0004] With the increasing penetration rate of advanced auxiliary driving systems, intelligent connected vehicles have rich, diverse perception and data collection, transmission and reception capabilities, providing the possibility for direct observation of road conditions and accurate early warning of high-risk vehicles. Therefore, it is very important to develop a risk early warning system with strong targeting of early warning information. SUMMARY

[0005] The purpose of the present application is to overcome the lack of observation means for wet road surfaces and the poor targeting of early warning information in the prior art, and to provide a vehicle-cloud integrated wet road driving risk early warning method and system.

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

[0007] A vehicle-cloud integrated wet road driving risk early warning method, comprising the following steps: S1: acquiring forward video data of a vehicle, identifying the type of the current driving road according to the forward video data through a road segmentation and road classification multi-task model, obtaining the road adhesion coefficient through a pre-established road adhesion coefficient reference table, and uploading the road adhesion coefficient to the cloud;

[0008] S2: repeating S1, continuously counting the road adhesion coefficient of the entire domain, establishing a road adhesion coefficient fusion model, fusing the road adhesion coefficient of the entire domain through road segment matching and introducing a time difference factor, and generating a wet road map;

[0009] S3: constructing a warning area dynamic calculation method of a cloud early warning system, and determining the corresponding warning area of any wet road section on the wet road map;

[0010] S4: For specific slippery road sections, the phase plane stability boundary of different types of vehicles is calculated, typical driving behavior is simulated, the instability risk of each type of vehicle is evaluated, the warning object is determined, and warning information is issued according to the vehicle type.

[0011] Further, step S1 comprises the following steps:

[0012] S11: Obtain the front video data of the vehicle, unify the image to a fixed size using a bilinear difference method, then use Z-score standardization for data scaling to obtain a preprocessed image;

[0013] S12: Based on the multi-task learning architecture of the shared bottom feature extraction network, a road surface segmentation and classification multi-task model is constructed to identify the road type corresponding to the preprocessed image;

[0014] S13: Through the pre-established road surface type adhesion coefficient reference table, the road surface adhesion coefficient corresponding to the road type is found and uploaded to the cloud.

[0015] Further, step S2 comprises the following steps:

[0016] S21: Repeat S1 to obtain the global road adhesion coefficient uploaded by multiple vehicles at multiple locations and multiple times;

[0017] S22: Establish a time series neural network to fuse the global road adhesion coefficient by calculating the time difference factor, and output the road adhesion coefficient fusion value of each road section;

[0018] S23: According to the road adhesion coefficient fusion value of each road section, a slippery road map is generated.

[0019] Further, the calculation expression of the time difference factor is:

[0020]

[0021] In the formula, k Δt is the time difference factor, and Δt is the time difference between two different road adhesion coefficients, with the unit of hours.

[0022] Further, the calculation expression of the time series neural network is:

[0023] μ t,f =f x (μ t ,μ t-1,f ,k Δt c t-1,f )+b x ,

[0024] c t,f =f c (μ t,μ t-1,f ,c t-1,f )+b c

[0025] In the formula, μ t ,μ t-1,f ,c t-1,f The inputs are the estimated values ​​of the road surface adhesion coefficient μ for a certain road segment at time t. t Historical road surface adhesion coefficient μ at time t-1 t-1,f Historical network memory value c t-1,f μ t,f ,c t,f The outputs are the fused road surface adhesion coefficient μ at time t. t,f The updated network memory value c t,f f x (·) represents the model parameters for the predicted output, b x To predict the model bias in the output, f c (·) represents the model parameters for updating memory values, b c The model bias used to update memory values.

[0026] Furthermore, step S3 includes the following steps:

[0027] S31: Determine the real-time average vehicle speed and speed limit based on the wet road surface map, and calculate the speed reduction range;

[0028] S32: Calculate the braking distance required after the driver receives the warning information;

[0029] S33: Construct a dynamic calculation method for the warning area of ​​a cloud-based early warning system, which calculates the length of the warning area based on the deceleration rate, braking distance, and traffic density.

[0030] Furthermore, the expression for calculating the deceleration rate is as follows:

[0031]

[0032] In the formula, v t Let v be the average speed of the traffic flow at time t, in kilometers per hour. l Let μ be the fusion road surface adhesion coefficient at time t. t,f The corresponding speed limit, in kilometers per hour, v a This refers to the rate of decrease in speed;

[0033] The formula for calculating braking distance is:

[0034]

[0035] In the formula, d t Braking distance, in meters, vt is the average speed of traffic flow at time t, with the unit of kilometer per hour, v l is the speed limit, with the unit of kilometer per hour, τ1 is the reaction time of the driver after receiving the early warning information, τ2 is the response time of the brake, and a is the comfortable braking deceleration of the driver;

[0036] The calculation expression of the dynamic calculation method of the early warning area is as follows:

[0037] d ROI = d0+ d t × (1+ v a × α)

[0038] In the formula, d ROI is the length of the early warning area, with the unit of meter; d0 is a fixed distance, representing the minimum early warning area; and α is a coefficient set based on the traffic density.

[0039] Further, the step S4 comprises the following steps:

[0040] S41: establishing a multi-type vehicle dynamics parameter library, calculating the center of mass side slip angle-yaw rate phase plane of each type of vehicle in the early warning area under the condition of maintaining the average speed v t at time t and the fusion road adhesion coefficient at time t, and drawing the stable boundary by using the envelope method to obtain the driving stability domain; t

[0041] S42: using numerical simulation software to calculate the center of mass side slip angle and yaw rate of each type of vehicle under the lane changing behavior, and drawing on the phase plane, if it is in the stable domain, it is safe, otherwise if it exceeds the stable domain, there is a risk of instability, and it is determined as a warning object;

[0042] S43: the cloud issues early warning information to the early warning object in the early warning area.

[0043] Further, the envelope method comprises a center of mass side slip angle constraint and a yaw rate constraint, and the calculation expression of the envelope method is as follows:

[0044] β lim = α r,sat + l r r / v t ,

[0045]

[0046] In the formula, β lim is the center of mass side slip angle constraint, r lim is the yaw rate constraint, α r,sat is the rear wheel saturation side slip angle, l r is the distance from the center of mass to the rear axle, and r is the yaw angle, v​t is the average vehicle speed, and t,f is the road adhesion coefficient; g is the acceleration of gravity.

[0047] In a second aspect of the present application, a vehicle-cloud integrated wet road vehicle driving risk early warning system comprises a cloud server and a client, the cloud server comprises a processor, and the processor executes any one of the vehicle-cloud integrated wet road vehicle driving risk early warning methods when running; the client comprises a video acquisition module and an early warning notification module, the video acquisition module records and uploads the forward video data of the vehicle and the road adhesion coefficient to the cloud server, and the early warning notification module receives and reports the early warning information.

[0048] Compared with the prior art, the present application has the following beneficial effects:

[0049] 1) The present application continuously observes the road based on the forward video data of the vehicle, constructs a road adhesion coefficient fusion model, generates a wet road map, overcomes the limitations of traditional observation methods, provides support for vehicle wet road driving risk early warning, determines the early warning area through an early warning area dynamic calculation method, calculates the phase plane stability boundary of different vehicle types on the wet road, evaluates the instability risk of different vehicle types, determines the early warning object, and is beneficial to improving the pertinence of risk early warning information pushing.

[0050] 2) The present application determines the phase plane stability boundary by calculating the centroid side slip angle constraint and the yaw rate constraint, provides a more accurate calculation method for vehicle lateral instability risk evaluation on the wet road, improves the risk prevention and control effect, and avoids the vehicle entering an instability state.

[0051] 3) The present application uses the vehicle in driving to collect the road adhesion coefficient, reflects the current road conditions in real time, realizes global observation through cloud integration of data, reduces the practical difficulty of large-scale installation of roadside sensing devices, accurately delivers early warning information to each type of vehicle, refines the risk evaluation result, and achieves the purpose of balancing safety and traffic capacity. BRIEF DESCRIPTION OF DRAWINGS

[0052] Figure 1 is a structural schematic diagram of the present application;

[0053] Figure 2 is a road segmentation and road classification multi-task model architecture schematic diagram of the shared feature extraction network established in the embodiment;

[0054] Figure 3 is a double moving line test lane size schematic diagram in the embodiment;

[0055] Figure 4 is a stability domain schematic diagram in the embodiment;

[0056] Figure 5A schematic diagram of the architecture of the vehicle-cloud integrated vehicle wet road driving risk early warning system in the embodiment; DETAILED DESCRIPTION

[0057] The present application will be described in detail below with reference to the accompanying drawings and specific embodiments. The present embodiment is implemented on the premise of the technical solution of the present application, and gives a detailed implementation and specific operation process, but the protection scope of the present application is not limited to the following embodiments.

[0058] Embodiment 1

[0059] The present application is a vehicle-cloud integrated wet road driving risk early warning method, as shown in the following steps: Figure 1

[0060] S1: Obtain the front video data of the vehicle, identify the type of the current driving road according to the front video data through the road segmentation and road classification multi-task model, obtain the road adhesion coefficient through the pre-established road adhesion coefficient reference table and upload it to the cloud;

[0061] S11: Obtain the front video data of the vehicle, unify the image to a fixed size by using the bilinear interpolation method, and then use Z-score standardization for data scaling to obtain the preprocessed image;

[0062] In the present embodiment, the front road image is obtained by the vehicle-mounted front camera The size is (720x1080), the image is unified to a fixed size (224x224) by using bilinear interpolation, and then Z-score standardization is used for data scaling to obtain the preprocessed input image I(x) norm ;

[0063] The specific calculation formula of bilinear interpolation is as follows:

[0064] I(x) resize =Bilinear(I(x),target_size)

[0065] In the formula, Bilinear(·) is a bilinear interpolation function, target_size is the target size of the image, and the output.

[0066] In the present embodiment, Z-score standardization is used to adjust the pixel value on the three channels of the image RGB to the standard normal distribution, so as to accelerate the convergence of the model, and the specific calculation formula is as follows:

[0067]

[0068] In the formula, x i is the image I(x)​resize Input on i-channel, The mean value of the luminance value μ and the mean value of the standard deviation of the luminance value σ are set as i ={0.471, 0.448, 0.408}, and i ={0.234, 0.239, 0.242} in this embodiment.

[0069] In this embodiment, the specific calculation formula of the road surface adhesion coefficient μ is as follows:

[0070] μ = Tab(MulNet(I(x) norm ))

[0071] In the formula, Tab is a road surface type adhesion coefficient reference table, and in this embodiment, the interval median value corresponding to the adhesion coefficient is selected as μ.

[0072] S12: Based on the multi-task learning architecture of the shared bottom feature extraction network, a road surface segmentation and road surface classification multi-task model is constructed to identify the road type corresponding to the preprocessed image;

[0073] In this embodiment, a lightweight convolutional neural network MobileNetV2 is used as the shared bottom feature extraction network, and then a road surface segmentation and road surface classification task head is built to construct a road surface segmentation and road surface classification multi-task model MulNet to quickly identify the road type as snow and light compaction snow.

[0074] S13: Through the pre-established road surface type adhesion coefficient reference table, the road surface adhesion coefficient corresponding to the road type is found and uploaded to the cloud.

[0075] The road surface type adhesion coefficient reference table in this embodiment is shown in Table 1. The corresponding road surface adhesion coefficient interval is [0.25, 035] through the road type, and the interval median value is selected as the road surface adhesion coefficient μ = 0.30.

[0076] Table 1 Road surface type-adhesion coefficient reference table

[0077]

[0078] S2: Repeat S1 to continuously calculate the global road surface adhesion coefficient, establish a road surface adhesion coefficient fusion model, match the road section and introduce a time difference factor, fuse the global road surface adhesion coefficient, and generate a slippery road map;

[0079] S21: Repeat S1 to obtain the global road surface adhesion coefficient uploaded by multiple locations, multiple times, and multiple vehicles;

[0080] The cloud matches different road sections according to the vehicle positions, and then collects the road adhesion coefficients uploaded by different vehicles. The road adhesion coefficients in the whole region are obtained by continuously observing a plurality of vehicles in the observation range.

[0081] S22: A time series neural network is established, the time difference factor is calculated, the road adhesion coefficients in the whole region are fused, and the road adhesion coefficient fusion values of each road section are output.

[0082] In this embodiment, a time series neural network based on LSTM (Long Short-Term Memory) as shown in FIG. 2 is established. The time difference factor is proposed to eliminate the possible large error of the road adhesion coefficient of a road section in a long time range. The calculation expression of the time difference factor is as follows: Figure 2

[0083]

[0084] In the formula, k Δt is the time difference factor, and Δt is the time difference between two different road adhesion coefficients, in hours.

[0085] The calculation expression of the time series neural network is as follows:

[0086] μ t,f = f x (μ t , μ t-1,f , k Δt c t-1,f ) + b x ,

[0087] c t,f = f c (μ t , μ t-1,f , c t-1,f ) + b c

[0088] In the formula, μ t , μ t-1,f , c t-1,f are inputs, which are the road adhesion coefficient estimation value μ t of a road section at time t, the historical road adhesion coefficient μ t-1,f at time t-1, and the historical network memory value c t-1,f = θ t-1 , respectively. μ t,f , c t,f are outputs, which are the fused road adhesion coefficient μ t,f at time t and the updated network memory value c t,f , respectively. f x (·) is a model parameter for predicting the output, and b x ​The model bias quantity for predicting the output is f c The model parameter for updating the memory value is b c The model bias quantity for updating the memory value is f

[0089] S23: Generate a wet road map according to the road surface adhesion coefficient fusion value of each road section.

[0090] According to the road surface adhesion coefficient fusion value of different road sections output by the time sequence neural network, a wet road map with low adhesion coefficient is generated at the initial time, and then the map is updated according to the road surface adhesion coefficient fusion value at the next time.

[0091] S3: Construct a warning area dynamic calculation method of the cloud early warning system to determine the warning area corresponding to any wet road section on the wet road map;

[0092] S31: Determine the real-time traffic flow average speed and speed limit value based on the wet road map, and calculate the speed reduction amplitude;

[0093] In this embodiment, based on the low adhesion coefficient wet road map, the road section corresponding to μ t,f = 0.36 is determined, the real-time traffic flow average speed v t = 60 km / h is obtained according to the road section speed measurement, the speed limit value v l = 30 km / h under the current adhesion working condition is obtained by looking up the table, and the speed reduction amplitude v a = 50% is calculated;

[0094] The calculation expression of the speed reduction amplitude is:

[0095]

[0096] In the formula, v t is the traffic flow average speed at time t, with the unit of kilometer per hour, v l is the fusion road surface adhesion coefficient μ t,f at time t, with the unit of kilometer per hour, v a is the speed limit value corresponding to the fusion road surface adhesion coefficient μ t is the speed reduction amplitude;

[0097] S32: Calculate the braking distance required by the vehicle driver after receiving the warning information;

[0098] The calculation expression of the braking distance is:

[0099]

[0100] In the formula, d t is the braking distance, with the unit of meter, v lis the reaction time of the driver after receiving the warning information, τ2=0.2 s is the brake response time, and a=1.67 m / s 2 is the comfortable braking deceleration of the driver;

[0101] In this embodiment, the braking distance d t required by the driver after receiving the warning information is calculated by considering the braking characteristics of the driver ROI =90.7 m;

[0102] S33: A dynamic calculation method of a warning area of a cloud warning system is constructed, and the length of the warning area is calculated according to the deceleration amplitude, the braking distance, and the traffic density.

[0103] The calculation expression of the dynamic calculation method of the warning area is as follows:

[0104] d ROI = d0+d ROI ×(1+v t ×α)

[0105] In the formula, d t is the length of the warning area, in meters; d0 is a fixed interval, representing the minimum warning area; and α is a coefficient set based on the traffic density. In this embodiment, α is the traffic operation safety margin value under different traffic flow types. In the free flow state, the vehicle spacing is large, and the margin value is 1; in the synchronous flow state, the vehicle spacing is small, and the margin value is 1.5; and in the congested flow state, the vehicle spacing is the smallest, and the margin value is set to 2. The calculation result is d t =168.7 m.

[0106] S4: For a specific wet and slippery road section, the phase plane stability boundary of each type of vehicle is calculated, the typical driving behavior is simulated, the instability risk of each type of vehicle is evaluated, the warning object is determined, and the warning information is issued according to the vehicle type.

[0107] S41: A dynamic parameter library of multiple types of vehicles is established, the phase plane of the center side angle-yaw angular velocity of each type of vehicle in the warning area is calculated under the average speed v t,f at t, and the envelope method is used to draw the stability boundary to obtain the driving stability domain;

[0108] In this embodiment, a dynamic parameter library of five types of vehicles, including passenger cars, box trucks, buses, SUVs, and trucks, with a high market share is established, and the phase plane of the center side angle-yaw angular velocity of each type of vehicle in the warning area is calculated under the average speed v lim at t and the adhesion coefficient μ r,sat, calculate the center of mass side slip angle-yaw rate phase plane of each type of vehicle under the condition of maintaining the existing speed, and draw the stability boundary using the envelope method to obtain the driving stability domain;

[0109] The envelope method includes the center of mass side slip angle constraint and the yaw rate constraint, and the calculation expression of the envelope method is:

[0110] β lim = a r,sat + l r r / v t ,

[0111]

[0112] In the formula, β lim is the center of mass side slip angle constraint, r lim is the yaw rate constraint; a r,sat is the rear wheel saturation side slip angle, l r is the distance from the center of mass to the rear axle, r is the yaw angle, v z is the average vehicle speed, μ t,f is the road adhesion coefficient; and g is the gravitational acceleration.

[0113] S42: using numerical simulation software, calculate the center of mass side slip angle and yaw rate of each type of vehicle under the lane changing behavior, and draw on the phase plane, if in the stable domain, it is safe, otherwise if it exceeds the stable domain, there is a risk of instability, and is determined as a warning object;

[0114] As shown in Figure 3 , a double lane shift test track is used to simulate typical driving behaviors, and then Matlab is used to calculate the real-time state of the vehicle center of mass side slip angle and yaw rate of the vehicle when driving on the double lane shift test track, and draw on the phase plane. As shown in Figure 4 part (a), the real-time state of the vehicle is in the stable domain, so the vehicle of this type can safely pass through the wet road at the current speed; otherwise, as shown in Figure 4 part (b), it exceeds the stable domain, so the vehicle of this type has a risk of instability when driving on the wet road, and is a warning object.

[0115] S43: the cloud issues a warning information to the warning object in the warning area.

[0116] In this embodiment, the update period of the warning area and the warning object is 10s. Therefore, after the vehicle in the warning area receives the initial warning information, if the real-time state of the vehicle in the subsequent any period is in the stable domain when performing the double lane shift test, the vehicle of this type does not need to be warned again; otherwise, the warning is continued and iteratively updated, the accurate push of the warning information is realized, the compliance of the driver is improved, and the risk prevention effect is improved.

[0117] Example 2

[0118] Based on Embodiment 1, the present invention also provides a vehicle-cloud integrated vehicle driving risk warning system for slippery roads, including a cloud server and an in-vehicle client. The cloud server includes a processor. When the processor runs, it executes the vehicle driving risk warning method for slippery roads based on Embodiment 1. The client includes a video acquisition module and a warning notification module. The video acquisition module records and uploads forward video data of the vehicle and the road surface adhesion coefficient to the cloud server. The warning notification module receives and notifies the warning information.

[0119] In this embodiment, the architecture of the vehicle-cloud integrated vehicle slippery road driving risk warning system is as follows: Figure 5 As shown, the vehicle estimates the coefficient of friction for slippery road surfaces using video data and uploads it to the cloud. The cloud, leveraging its powerful data aggregation and computing capabilities, generates a slippery road surface map, identifies warning areas and vehicle types, and finally pushes warning information to the entire vehicle fleet, forming a data flow from vehicle to cloud to vehicle fleet. Furthermore, vehicles within the warning area will also upload their estimated coefficient of friction values ​​for slippery road surfaces in the future, thus enabling continuous observation of road conditions and updates to the slippery road surface map.

[0120] The preferred embodiments of the present invention have been described in detail above. It should be understood that those skilled in the art can make numerous modifications and variations based on the concept of the present invention without creative effort. Therefore, all technical solutions that can be obtained by those skilled in the art based on the concept of the present invention through logical analysis, reasoning, or limited experimentation on the basis of existing technology should be within the scope of protection defined by the claims.

Claims

1. A vehicle-cloud integrated wet and slippery road surface vehicle driving risk early warning method, characterized in that, The method comprises the following steps: S1: acquiring forward video data of a vehicle, identifying the type of the current driving road according to the forward video data through a road surface segmentation and classification multi-task model, obtaining a road surface adhesion coefficient through a pre-established road surface adhesion coefficient reference table, and uploading the road surface adhesion coefficient to the cloud; S2: repeating S1 to continuously calculate the road surface adhesion coefficient of the whole region, establishing a road surface adhesion coefficient fusion model, fusing the road surface adhesion coefficient of the whole region through road section matching and introducing a time difference factor, and generating a wet and slippery road surface map; S3: constructing a dynamic calculation method of a warning area of a cloud warning system to determine the warning area corresponding to any wet and slippery road section on the wet and slippery road surface map; S4: for a specific wet and slippery road section, calculating the phase plane stability boundary of different types of vehicles, simulating typical driving behaviors, evaluating the instability risk of each type of vehicle, determining the warning object, and issuing warning information according to the vehicle type; The step S2 comprises the following steps: S21: repeating S1 to obtain the road surface adhesion coefficient of the whole region uploaded by multiple locations, multiple times and multiple vehicles; S22: establishing a time sequence neural network, fusing the road surface adhesion coefficient of the whole region by calculating the time difference factor, and outputting the road surface adhesion coefficient fusion value of each road section; S23: generating a wet and slippery road surface map according to the road surface adhesion coefficient fusion value of each road section; The calculation expression of the time difference factor is: wherein is a time difference factor, is a time difference between two different road adhesion coefficients, in hours; The calculation expression of the time sequence neural network is: In the formula, is the input, respectively, the road surface adhesion coefficient estimation value of a certain section at moment , moment historical road surface adhesion coefficient , historical network memory value , is the output, respectively, moment fusion road surface adhesion coefficient , updated network memory value , is the model parameter of the prediction output, is the model bias of the prediction output, is the model parameter of the updated memory value, is the model bias of the updated memory value. 2.The vehicle-cloud integrated wet road vehicle driving risk early warning method according to claim 1, characterized in that, The step S1 comprises the following steps: S11: acquiring forward video data of a vehicle, uniformly converting the image into a fixed size by using a bilinear difference method, and then performing data scaling by using Z-score standardization to obtain a preprocessed image; S12: constructing a road surface segmentation and classification multi-task model based on a multi-task learning architecture of a shared bottom feature extraction network to identify the road type corresponding to the preprocessed image; S13: searching for the road surface adhesion coefficient corresponding to the road type through a pre-established road surface type adhesion coefficient reference table and uploading the road surface adhesion coefficient to the cloud. 3.The vehicle-cloud integrated wet road vehicle driving risk early warning method according to claim 1, characterized in that, The step S3 comprises the following steps: S31: determining the real-time traffic flow average speed and speed limit value based on the wet and slippery road surface map, and calculating the speed reduction amplitude; S32: calculating the braking distance required after the vehicle driver receives the warning information; S33: constructing a dynamic calculation method of a warning area of a cloud warning system, and calculating the length of the warning area according to the speed reduction amplitude, the braking distance and the traffic flow density. 4.The vehicle-cloud integrated wet road vehicle driving risk early warning method according to claim 3, characterized in that, The calculation expression of the speed reduction amplitude is: In the formula, is the average speed of traffic flow at the moment, in units of kilometers per hour, is the fusion road adhesion coefficient at the moment the corresponding speed limit value, in units of kilometers per hour, is the speed reduction amplitude; The calculation expression of the braking distance is: wherein is the braking distance in meters, is the is the average speed of the traffic flow at the moment in time in km / h, is the speed limit in km / h, is the reaction time of the driver after receiving the warning information, is the brake response time, is the comfortable deceleration of the driver; The calculation expression of the dynamic calculation method of the warning area is: In the formula, is the length of the warning area, in meters; is the fixed interval, representing the minimum warning area; is the coefficient set based on the traffic density. 5.The vehicle-cloud integrated wet road vehicle driving risk early warning method according to claim 1, characterized in that, The step S4 comprises the following steps: S41: Establish a multi-type vehicle dynamics parameter library, according to Instantaneous traffic flow average speed And The fusion road adhesion coefficient at the moment, calculate the centroid side slip angle-yaw rate phase plane of each type of vehicle in the warning area under the condition of maintaining the average speed Draw the stability boundary by using the envelope method to obtain the driving stability domain; S42: using numerical simulation software to calculate the center of mass side slip angle and yaw angular velocity of each type of vehicle under lane changing behavior, and plotting on the phase plane, if in the stable domain, it is safe, otherwise if out of the stable domain, there is an instability risk, and the warning object is determined; S43: the cloud issues a warning information to the warning object in the warning area. 6.The vehicle-cloud integrated wet road vehicle driving risk early warning method according to claim 5, characterized in that, The envelope method comprises a center of mass side slip angle constraint and a yaw angular velocity constraint, and the calculation expression of the envelope method is: wherein is a center of mass side slip angle constraint, is a yaw rate constraint; is a rear wheel saturation side slip angle, is a center of mass to rear axle distance, is a yaw angle, is an average vehicle speed, is a road surface adhesion coefficient; is a gravitational acceleration.

7. A vehicle and cloud integrated wet road surface vehicle driving risk early warning system, characterized in that, The cloud server comprises a processor, and the processor executes the vehicle-cloud integrated wet road vehicle driving risk early warning method according to any one of claims 1-6 when running. The client comprises a video acquisition module and an early warning notification module. The video acquisition module records and uploads the front video data of the vehicle and the road adhesion coefficient to the cloud server. The early warning notification module receives and reports the early warning information.

Images