Road adhesion coefficient estimation method, system, equipment and medium

Abstract

The invention provides a road adhesion coefficient estimation method, system and device and a medium, and the method comprises the steps: S1, obtaining a road image and vehicle driving state parameters, and recognizing a road type through an image recognition model; s2, constructing a corresponding vehicle tire dynamic model according to the road type, and importing the vehicle driving state parameters into the vehicle tire dynamic model to obtain first estimation parameters; s3, according to the vehicle driving state and the road type, a Burckhardt model is used for estimation, and a second estimation parameter is obtained; and S4, repeating the steps S1 to S3 for multiple times, and then calculating the average value of all the obtained first estimation parameters and second estimation parameters to obtain an estimated road adhesion coefficient. The problem that in the prior art, a method for calculating the road adhesion coefficient by directly collecting factor data influencing the road adhesion coefficient through a sensor is not high in accuracy is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of electric vehicles, and particularly to a method, system, device and medium for estimating road adhesion coefficient. Background Art

[0002] With the rapid development of autonomous driving technology, in order to control the vehicle to drive smoothly under different working conditions, it is necessary to calculate the adhesion coefficient between the vehicle and the ground. The adhesion coefficient between the vehicle and the road surface has a great influence on the longitudinal and lateral control of the vehicle. Introducing the road adhesion coefficient information into the active collision avoidance strategy of the vehicle can improve the adaptability of the AEB system to different road conditions, thereby avoiding or reducing the occurrence of collision accidents. Therefore, it is very necessary to accurately, effectively and quickly estimate the road adhesion coefficient of the vehicle driving.

[0003] In the prior art, data of factors affecting the road adhesion coefficient, such as road gloss, roughness, etc., are directly collected through sensors, and then the results are deduced. However, for the same type of road surface, the road adhesion coefficient is not constant, and the sensors are expensive, increasing the manufacturing cost of the whole vehicle. In addition, this method is easily affected by the quality of training data, resulting in certain limitations in accurately identifying the road adhesion coefficient. Summary of the Invention

[0004] In view of the deficiencies in the prior art, the present invention provides a method for estimating road adhesion coefficient, which solves the problem of low accuracy of the method for directly collecting data of factors affecting road adhesion coefficient through sensors in the prior art to deduce road adhesion coefficient.

[0005] According to an embodiment of the present invention, a method for estimating road adhesion coefficient includes:

[0006] S1: Obtain a road surface image and vehicle driving state parameters, and use an image recognition model to identify the road type;

[0007] S2: Construct a corresponding vehicle tire dynamic model according to the road type, and import the vehicle driving state parameters into the vehicle tire dynamic model to obtain a first estimation parameter;

[0008] S3: Estimate using the Burckhardt model according to the vehicle driving state and road type to obtain a second estimation parameter;

[0009] S4: Repeat steps S1 - S3 multiple times, and then calculate the mean values of all the first estimation parameters and the second estimation parameters obtained to obtain the estimated road adhesion coefficient.

[0010] Preferably, it further includes: S5: According to the road type, calculate the theoretical road surface adhesion coefficient, and then determine whether to use the estimated road surface adhesion coefficient or the theoretical road surface adhesion coefficient as the current road surface adhesion coefficient based on the difference between the estimated road surface adhesion coefficient and the theoretical road surface adhesion coefficient.

[0011] Preferably, the vehicle tire power model includes an ordinary road power model and an extreme road power model;

[0012] When the road category is an icy road surface or a snow-covered road surface, construct an extreme road power model; when the road category is other road surfaces, construct an ordinary road power model.

[0013] Preferably, the calculation method of the first estimated parameter includes:

[0014] Calculate the slip rate of the vehicle tire according to the vehicle driving state parameters;

[0015] Based on the vehicle driving state parameters and the slip rate, use the vehicle tire power model to calculate the longitudinal force of the vehicle tire;

[0016] Take the vehicle vertical load as the lateral force, and calculate the first estimated parameter according to the lateral force and the longitudinal force.

[0017] Preferably, the calculation formula of the ordinary road power model is:

[0018] F z =αsin(βarctan(γ(1 - σ)·s + σarctan(γ·s)))

[0019] The calculation formula of the extreme road power model is:

[0020]

[0021] Among them, α, β, λ, σ are all fitting parameters, s is the slip rate, C x is the longitudinal stiffness, and f(λ) is the correction function.

[0022] Preferably, obtain the corresponding vehicle speed decay coefficient and road surface static adhesion coefficient according to the road surface type, and then calculate the theoretical road surface adhesion coefficient at different vehicle speeds according to the vehicle speed decay coefficient and the road surface static adhesion coefficient.

[0023] Preferably, the calculation formula of the theoretical road surface adhesion coefficient is as follows:

[0024] μ(v)=μ0·e -kv

[0025] Among them, μ0 is the road surface static adhesion coefficient, k is the vehicle speed decay coefficient, and v is the vehicle speed.

[0026] On the other hand, according to an embodiment of the present invention, an estimation system for road surface adhesion coefficient is further provided. The system uses the above-mentioned method for estimating road surface adhesion coefficient, and includes:

[0027] An acquisition module for acquiring vehicle driving state parameters and road surface images;

[0028] An identification module for identifying the road type according to the road surface image;

[0029] A calculation module for constructing a vehicle tire dynamics model and a Burckhardt model, calculating a first estimation parameter and a second estimation parameter respectively according to the vehicle driving state, then calculating the mean value of all the first estimation parameters and the second estimation parameters to obtain an estimated road surface adhesion coefficient, and calculating a theoretical road surface adhesion coefficient;

[0030] A decision module, and the comparison module is used to determine whether to use the estimated road surface adhesion coefficient or the theoretical road surface adhesion coefficient as the current road surface adhesion coefficient according to the difference between the estimated road surface adhesion coefficient and the theoretical road surface adhesion coefficient.

[0031] On the other hand, according to an embodiment of the present invention, a computer is further provided, including at least one processor and a memory. The memory stores a computer program, and the computer program is configured to be executed by the processor to implement the above-mentioned method for estimating road surface adhesion coefficient.

[0032] On the other hand, according to an embodiment of the present invention, a storage medium is further provided. The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. The computer program can be executed by one or more processors to implement the above-mentioned method for estimating road surface adhesion coefficient.

[0033] Compared with the prior art, the present invention has the following beneficial effects:

[0034] The present invention accurately identifies the road type through an image recognition algorithm, provides environmental information for the subsequent vehicle tire dynamics model and Burckhardt model, facilitates the rapid and accurate setting of corresponding environmental parameters for the calculation model, improves the subsequent estimation accuracy, and then estimates the road surface adhesion coefficient of the vehicle driving road by using two methods of the vehicle tire dynamics model and the Burckhardt model respectively according to the driving state of the vehicle, and further reduces the error by calculating the mean value of the road surface estimation coefficients estimated by the two methods, and further improves the estimation accuracy of the road surface adhesion coefficient. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 It is a flowchart for estimating the road surface adhesion coefficient according to an embodiment of the present invention. Detailed implementation manners

[0036] The technical solutions in the present invention will be further described below with reference to the accompanying drawings and embodiments.

[0037] As Figure 1 shown, an estimation method for road surface adhesion coefficient according to an embodiment of the present invention includes:

[0038] S1: Obtain a road surface image and vehicle driving state parameters, and use an image recognition model to identify the road type;

[0039] The vehicle driving state parameters that can be obtained by using various in-vehicle sensors include: vehicle driving speed, longitudinal stiffness of the tire, vehicle vertical load, and wheel angular velocity

[0040] After that, use an in-vehicle camera to obtain a road surface image of the vehicle driving road, and identify the road type through an image recognition model, including: dry asphalt road surface, wet asphalt road surface, dry cement road surface, wet cement road surface, dry cobblestone road surface, wet cobblestone road surface, snow-covered road surface, and ice-covered road surface.

[0041] S2: Construct a corresponding vehicle tire dynamic model according to the road type, and import the vehicle driving state parameters into the vehicle tire dynamic model to obtain a first estimation parameter;

[0042] The vehicle tire dynamic model includes a normal road dynamic model and an extreme road dynamic model. The normal road dynamic model is calculated using the magic formula. Since the road surface adhesion coefficient of snow-covered roads and ice-covered roads is very low, using the normal road dynamic model will result in a large error. Therefore, if the road category is an ice road surface or a snow road surface, an extreme road dynamic model is constructed.

[0043] First, calculate the slip ratio s of the vehicle tire:

[0044]

[0045] where v is the vehicle driving speed, w is the wheel angular velocity, and R is the wheel radius.

[0046] After that, for snow-covered roads and ice-covered roads, construct an extreme road dynamic model to calculate the longitudinal force F received by the wheel z :

[0047]

[0048] where C x is the longitudinal stiffness, and f(λ) is a correction function.

[0049] For other road surfaces, construct a normal road dynamic model to calculate the longitudinal force F received by the wheel z :

[0050] F z = α sin(β arctan(γ(1 - σ)·s + σ arctan(γ·s)))

[0051] Where α, β, λ, and σ are all fitting parameters and can be set as needed.

[0052] Calculate the first estimated parameter according to the following formula:

[0053]

[0054] Where F x is the vertical force on the tire, and the vehicle vertical load can be used as the vertical force.

[0055] S3: According to the vehicle driving state and road type, use the Burckhardt model for estimation to obtain the second estimated parameter;

[0056] The calculation formula of the Burckhardt model is:

[0057]

[0058] If the influence of speed change is ignored, the above formula can be simplified to:

[0059]

[0060] The model parameters c1, c2, and c3 are adjusted according to the road surface conditions (such as dry asphalt, wet asphalt, snow, ice, etc.) to reflect the friction characteristics of different road surfaces. These parameters are obtained by fitting experimental data and affect the performance of the model under different road surface conditions. The parameter values for typical road surfaces are shown in Table 1:

[0061] Table 1 Model parameter values for each road type

[0062]

[0063]

[0064] According to the road type, select the corresponding parameter values of c1, c2, and c3 and substitute them into the Burckhardt model for calculation to obtain the second estimated parameter.

[0065] S4: Calculating the first estimated parameter and the second estimated parameter only once will inevitably lead to calculation errors due to some mechanical errors that cannot be corrected. Therefore, steps S1 - S3 need to be repeated multiple times. The number of repetitions is set according to requirements, and then the mean values of all the first estimated parameters and the second estimated parameters calculated are obtained to get the estimated road surface adhesion coefficient μ(s).

[0066] S5: Calculate the theoretical road surface adhesion coefficient according to the road type, and then determine whether to use the estimated road surface adhesion coefficient or the theoretical road surface adhesion coefficient based on the difference between the estimated road surface adhesion coefficient and the theoretical road surface adhesion coefficient.

[0067] Obtain the corresponding vehicle speed attenuation coefficient and road surface static adhesion coefficient according to the road surface type, and then calculate the theoretical road surface adhesion coefficient at different vehicle speeds as the current road surface adhesion coefficient based on the vehicle speed attenuation coefficient and the road surface static adhesion coefficient.

[0068] The vehicle speed attenuation coefficient is the degree of reduction in the vehicle speed under the influence of various resistances when the vehicle is traveling smoothly, the engine does not apply power to the vehicle, and the vehicle has an initial speed. The vehicle speed attenuation coefficient is also different under different vehicle speeds and road surface environments. See Table 2 (taking dry asphalt road surface as an example):

[0069] Table 2: Relationship between vehicle speed and vehicle speed attenuation coefficient on dry asphalt road surface

[0070] Vehicle speed Vehicle speed attenuation coefficient 50 0.0018 70 0.002 90 0.0026 110 0.0035 130 0.005

[0071] The static adhesion coefficients of various road surfaces are shown in Table 3:

[0072] Table 3: Static adhesion coefficients of various road surfaces

[0073] Road surface type Static adhesion coefficient Dry asphalt pavement 0.9 Wet asphalt pavement 0.75 Dry cement pavement 0.85 Wet cement pavement 0.75 Snow-covered road surface 0.3 Icy road surface 0.2 Dry cobblestones 0.7 Wet cobblestones 0.379

[0074] Calculate the theoretical road surface adhesion coefficient at the current vehicle driving speed according to Table 2 and Table 3:

[0075] μ(v) = μ0·e -kv

[0076] where μ0 is the road surface static adhesion coefficient, k is the vehicle speed attenuation coefficient, and v is the vehicle speed.

[0077] Considering the small fluctuations in the adhesion coefficient of the same road surface in actual driving conditions, in order to avoid frequent switching of the control strategy of the automatic emergency braking system, when the absolute value of the difference between the road surface adhesion coefficients identified by the two algorithms is less than a certain threshold (taking 0.15), it is considered that there is no sudden change in the road surface ahead. At this time, the theoretical road surface adhesion coefficient μ(v) is used as the road surface adhesion coefficient of the current road surface. When the absolute value of the difference between the two is greater than or equal to the threshold, it is considered that there is a sudden change in the road surface ahead. At this time, the estimated road surface adhesion coefficient μ(s) is used as the road surface adhesion coefficient of the current road surface.

[0078] On the other hand, the embodiment of the present invention also provides an estimation system for road surface adhesion coefficient. This system uses the above-mentioned estimation method for road surface adhesion coefficient, including:

[0079] An acquisition module, where the acquisition module is used to acquire vehicle driving state parameters and road surface images;

[0080] An identification module, which is configured to identify the road type according to the road surface image;

[0081] A calculation module, which is configured to construct a vehicle tire dynamic model and a Burckhardt model, calculate a first estimated parameter and a second estimated parameter respectively according to the vehicle driving state, then calculate the mean value of all the first estimated parameters and the second estimated parameters to obtain an estimated road surface adhesion coefficient, and calculate a theoretical road surface adhesion coefficient;

[0082] A decision module, and the comparison module is configured to determine whether to use the estimated road surface adhesion coefficient or the theoretical road surface adhesion coefficient as the current road surface adhesion coefficient according to the difference between the estimated road surface adhesion coefficient and the theoretical road surface adhesion coefficient.

[0083] On the other hand, an embodiment of the present invention further provides a computer, including at least one processor and a memory, the memory stores a computer program, and the computer program is configured to be executed by the processor to implement the above-mentioned method for estimating a road surface adhesion coefficient.

[0084] On the other hand, an embodiment of the present invention further provides a storage medium, the storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium, and the computer program can be executed by one or more processors to implement the above-mentioned method for estimating a road surface adhesion coefficient.

[0085] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered within the scope of the claims of the present invention.

Claims

1. A method for estimating the road surface adhesion coefficient, characterized in that: Including: S1: Obtain the road surface image and vehicle driving state parameters, and use an image recognition model to identify the road type; S2: Construct a corresponding vehicle tire dynamic model according to the road type, and import the vehicle driving state parameters into the vehicle tire dynamic model to obtain the first estimated parameter; S3: Estimate using the Burckhardt model according to the vehicle driving state and road type to obtain the second estimated parameter; S4: Repeat steps S1 - S3 multiple times, and then calculate the mean of all the first estimated parameters and the second estimated parameters obtained to get the estimated road surface adhesion coefficient.

2. The estimation method of a road surface adhesion coefficient according to claim 1, characterized in that: Also including: S5: Calculate the theoretical road surface adhesion coefficient according to the road type, and then determine whether to use the estimated road surface adhesion coefficient or the theoretical road surface adhesion coefficient as the current road surface adhesion coefficient according to the difference between the estimated road surface adhesion coefficient and the theoretical road surface adhesion coefficient.

3. The method for estimating a road surface adhesion coefficient according to claim 1, wherein: The vehicle tire dynamic model includes an ordinary road dynamic model and an extreme road dynamic model; If the road category is an ice - covered road surface or a snow - covered road surface, construct an extreme road dynamic model, and if the road category is other road surfaces, construct an ordinary road dynamic model.

4. The method for estimating a road surface adhesion coefficient according to claim 3, wherein: The calculation method of the first estimated parameter includes: Calculate the slip ratio of the vehicle tire according to the vehicle driving state parameters; Based on the vehicle driving state parameters and the slip ratio, use the vehicle tire dynamic model to calculate the longitudinal force of the vehicle tire; Take the vehicle vertical load as the lateral force, and calculate the first estimated parameter according to the lateral force and the longitudinal force.

5. The method for estimating a road surface adhesion coefficient according to claim 4, wherein: The calculation formula of the ordinary road dynamic model is: F z = α sin(β arctan(γ(1 - σ)·s + σ arctan(γ·s))) The calculation formula of the extreme road dynamic model is: Among them, α, β, λ, and σ are all fitting parameters, s is the slip rate, C x is the longitudinal stiffness, and f(λ) is the correction function.

6. The method for estimating a road surface adhesion coefficient according to claim 2, wherein: Obtain the corresponding vehicle speed decay coefficient and road surface static adhesion coefficient according to the road surface type, and then calculate the theoretical road surface adhesion coefficient at different vehicle speeds according to the vehicle speed decay coefficient and the road surface static adhesion coefficient.

7. The method for estimating a road surface adhesion coefficient according to claim 6, wherein: The calculation formula of the theoretical road surface adhesion coefficient is as follows: μ(v) = μ0·e -kv Wherein, μ0 is the road surface static adhesion coefficient, k is the vehicle speed decay coefficient, and v is the vehicle speed.

8. An estimation system for road surface adhesion coefficient, characterized in that: This system uses the method for estimating a road surface adhesion coefficient according to any one of claims 1 - 7, including: An acquisition module, which is used to acquire vehicle driving state parameters and road surface images; An identification module, which is used to identify the road type according to the road surface image; A calculation module, which is used to construct a vehicle tire dynamic model and a Burckhardt model, and calculate the first estimated parameter and the second estimated parameter respectively according to the vehicle driving state, then calculate the mean of all the first estimated parameters and the second estimated parameters to obtain the estimated road surface adhesion coefficient, and calculate the theoretical road surface adhesion coefficient; A decision-making module, wherein the comparison module is configured to determine whether to use the estimated road surface adhesion coefficient or the theoretical road surface adhesion coefficient as the current road surface adhesion coefficient based on the difference between the estimated road surface adhesion coefficient and the theoretical road surface adhesion coefficient.

9. A computer, characterized in that: It includes at least one processor and a memory, and the memory stores a computer program, and the computer program is configured to be executed by the processor to implement an estimation method for a road surface adhesion coefficient according to any one of claims 1-7.

10. 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, and the computer program can be executed by one or more processors to implement an estimation method for a road surface adhesion coefficient according to any one of claims 1-7.

Images