A vehicle driving optimization method based on a road surface friction coefficient intelligent prediction model

Abstract

The application discloses a vehicle driving optimization method based on a road surface friction coefficient intelligent prediction model and belongs to the technical field of urban traffic road engineering and vehicle control. Real-time state parameters of a front road surface are collected in real time by a vehicle vision system, including a road surface type and a road surface condition; real-time prediction friction coefficient values between the vehicle and the road surface are output by using a friction coefficient intelligent prediction model according to the real-time state parameters of the front road surface, vehicle self running state parameters and environmental state parameters; and the real-time prediction friction coefficient values are transmitted to a vehicle control system to adjust power distribution of the vehicle. The application can predict the road surface friction coefficient values in real time, reasonably adjust the power distribution of the vehicle in combination with the current state of the vehicle, reasonably use the vehicle power source, and further improve the passability, driving economy and running stability of the vehicle, prolong the driving mileage of the vehicle and ensure that the switching of the vehicle driving mode has a basis.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of urban traffic road engineering and vehicle control technology, and particularly relates to a vehicle driving optimization method based on a road surface friction coefficient intelligent prediction model. BACKGROUND

[0002] With the increase of traffic flow and vehicle speed on high-grade highways and urban roads in China, the requirement for road surface skid resistance is increasingly improved, and the influence of road surface skid resistance on traffic safety is increasingly valued by highway construction and maintenance management departments. At present, the following methods are mainly used for testing and evaluating the road surface skid resistance in China: sanding method, pendulum friction tester method, DF tester method, etc. However, these methods use manual detection, have low detection efficiency, and the construction depth is only an indirect index, which cannot directly reflect the actual friction performance. In addition, although the existing laser scanning construction depth method and transverse force coefficient tester method have high automation degree and detection efficiency, the test results are easily affected by external environments such as accumulated water, accumulated snow, and mud, and need to be tested under closed traffic or low traffic, which has poor real-time performance. In addition, the setting distance method and locked wheel trailer method are commonly used for testing the friction resistance between the vehicle tire and the road surface after the vehicle is braked and the wheels are locked, which has high automation level and efficiency, but can only be performed in an annual inspection manner, has a long idle period, and the working condition scene is insufficient to reflect the real-time performance of the actual driving environment.

[0003] Due to the low detection efficiency of the traditional manual detection method, the need for testing in a specific environment, the lag of the detection results, the insufficient application real-time performance, the limited working condition scene, and the influence of the external environment, the method cannot meet the high traffic flow and high vehicle speed requirements of modern traffic.

[0004] CN119000520A discloses a longitudinal road surface friction coefficient measurement system, which is mainly used for measuring and evaluating the skid resistance of highways. The system drives the measurement wheel to rotate along the driving direction by using the vehicle platform, and applies additional resistance to the measurement wheel by using the braking system, so that the measurement wheel produces a certain sliding when rolling on the road surface. The system obtains the vertical load, the horizontal resistance of the forward movement, the sliding speed, and the friction force between the tire and the road surface of the measurement wheel, calculates the original road surface friction coefficient. In addition, the system also obtains the vibration acceleration information of the measurement wheel in the vertical road surface direction by using the vertical load loading system and the imu module, corrects the original road surface friction coefficient, and thus improves the measurement accuracy. The system mainly improves the existing measurement method by designing the vertical load loading, vibration acceleration acquisition, friction coefficient correction, and automatic adjustment of water spraying amount, but still has the difficulties of system complexity, high implementation difficulty, difficult data processing, and adaptability to be optimized, and is not further applied to the prediction and change of the vehicle running state in the real-time driving process of the vehicle.

[0005] CN118977714A discloses a driving mode switching method, which intelligently determines and switches the target driving mode (economy mode, sports mode, off-road mode or wet road mode) of the vehicle by comprehensively combining multiple sensor data and environmental detection results, but does not consider the road surface predicted friction coefficient in the driving mode switching condition and reasonably distribute the optimized vehicle power accordingly.

[0006] CN115081927A discloses a road surface friction coefficient evaluation and prediction method, which provides real-time road surface friction coefficient warning for vehicles and provides medium and long-term solutions, but does not predict the vehicle road surface friction coefficient in the future extreme time and does not reasonably apply the predicted value to achieve the effect of strengthening the vehicle operation performance. SUMMARY

[0007] The purpose of the present application is to solve the problem that the friction coefficient between the tire and the ground during the vehicle driving process cannot be measured in real time and fed back to the vehicle to adjust the vehicle operation state parameters. A vehicle driving optimization method based on a road surface friction coefficient intelligent prediction model is proposed. The road surface friction coefficient value is predicted in real time, and the vehicle power distribution is reasonably adjusted in combination with the current state of the vehicle.

[0008] To achieve the above purpose, the technical scheme adopted by the present application is as follows:

[0009] A vehicle driving optimization method based on a road surface friction coefficient intelligent prediction model, comprising:

[0010] The real-time state parameters of the front road surface are collected in real time by the vehicle vision system, including the road surface type and the road surface condition; the distance range of the specified distance segment ΔY of the front road section collected by the vehicle relative to the vehicle is:

[0011]

[0012] Wherein, π1(t) is the driving speed in the geodetic coordinate system collected at the current time, TTC min , TTC max represent the upper and lower limits of the collision time, TTC min , TTC max represent the minimum and maximum values of the distance range;

[0013] According to the real-time state parameters of the front road surface, the vehicle's own operation state parameters and the environmental state parameters, the friction coefficient intelligent prediction model is used to output the real-time predicted friction coefficient value between the vehicle and the road surface;

[0014] The real-time predicted friction coefficient value is transmitted to the vehicle control system to adjust the power distribution of the vehicle.

[0015] The present application has the following beneficial effects:

[0016] The present application is based on vehicle dynamics model, tire model and other vehicle engineering technology, combined with computer vision, deep learning and neural network, and through the design of vehicle vision and sensing system, real-time collection of front road surface type (material and gradation, texture characteristics, etc.) and condition (wear degree, medium between tire and road surface, etc.), and transmission of the collected data to big data processing and intelligent prediction module through vehicle network. In the intelligent prediction model of road surface friction coefficient designed in the big data processing and intelligent prediction module, the input of the model is the front road surface type and condition collected by the vehicle vision and sensing system, the vehicle itself running state parameters (driving speed, driving acceleration, side slip angle, slip rate, tire type and wear degree, tire inflation pressure and load, etc.) and environmental state parameters (humidity, temperature, wind resistance, etc.), and the output is the real-time predicted friction coefficient value between the vehicle and the road surface. The value is transmitted to the vehicle control system, and then the vehicle control system adjusts the vehicle power distribution reasonably through the predicted value, so as to reasonably use the vehicle power source, and to improve the passability, driving economy and running stability of the vehicle, prolong the driving mileage of the vehicle and ensure the switching of the vehicle driving mode.

[0017] DRAWINGS

[0018] Figure 1 The overall architecture diagram of the intelligent prediction, calculation and execution system of road surface friction coefficient shown in the embodiment of the present application

[0019] Figure 2 The running architecture diagram of the vehicle vision system shown in the embodiment of the present application

[0020] Figure 3 The overall architecture diagram of the big data processing and intelligent prediction module shown in the embodiment of the present application

[0021] Figure 4 The flow chart of vehicle power distribution optimization shown in the embodiment of the present application DETAILED DESCRIPTION

[0022] The present application will be further described below in conjunction with specific embodiments. It should be understood that these embodiments are only used to illustrate the present application and not to limit the scope of the present application. In addition, it should be understood that after reading the content taught by the present application, those skilled in the art can make various modifications or modifications to the present application, and these equivalent forms also fall within the scope defined by the appended claims of the present application.

[0023] The application is based on vehicle dynamics model, tire model and other vehicle engineering technology, and combines computer vision, deep learning and neural network, and proposes a vehicle driving optimization method based on intelligent prediction model of road surface friction coefficient. The application mainly relates to four modules: vehicle vision and sensor acquisition system module, big data processing and intelligent prediction module, vehicle control center system module, instruction execution VCU module, and the overall architecture is as shown in the accompanying Figure 1 .

[0024] Firstly, the vehicle obtains the front road type (material and gradation, texture characteristics, etc.) and condition (wear degree, medium between tire and road surface, etc.) parameters of the specified position in front of the vehicle through the vehicle vision and sensor acquisition system module.

[0025] The specified position is a preset value road point, and the value of the road point depends on the preset vehicle and road point collision time TTC. Assuming that the time for the vehicle processor to calculate an intelligent prediction value of the road surface friction coefficient and issue an optimized adjustment vehicle dynamic instruction is Δt1, the preset safety prediction constraint interval is Ω=[Δt min ,Δt max ], and the collision time is specifically constrained as follows:

[0026] TTC min ≤Δt1+Δt min ≤TTC max ≤Δt1+Δt max

[0027] Wherein, Δt min is the preset safety prediction constraint reserved minimum time, Δt max is the preset safety prediction constraint reserved maximum time, TTC nib and TTC max represent the upper and lower limits of the collision time.

[0028] Then, the specified distance section ΔY of the value road point of the front road section collected by the vehicle relative to the distance range of the vehicle is:

[0029]

[0030] Wherein, π1(t) is the driving speed in the geodetic coordinate system collected at the current time, which can be converted from the speed v(t) in the vehicle coordinate system. In order to further facilitate control, the coordinates in the vehicle coordinate system and the geodetic coordinate system are converted as follows:

[0031]

[0032] Wherein, x and y respectively represent the vehicle longitudinal coordinate and transverse coordinate in the vehicle coordinate system, The angle between the line connecting the origins of the two coordinate systems and the horizontal axis of the geodetic coordinate system.

[0033] Further, the vehicle motion related parameters can be expressed as:

[0034]

[0035] where w represents the wheel roll angular velocity, r represents the wheel radius. π2(t) represents the current time collected in the geodetic coordinate system under the vehicle acceleration; π3(t) represents the current time collected vehicle side slip angle; π4(t) represents the current time collected vehicle slip rate; π5(t) represents the current time collected vehicle tire type and wear degree, which is according to the vehicle tire type τ type , tire wear rate τ wear and the remaining service life τ life to make fuzzy mathematical evaluation about the actual mileage and time, and calculate the evaluation score belonging to [0, 1] as its return value, F fce (τ type ,τ wear ,τ life ) represents the fuzzy mathematical evaluation process; π6(t) represents the current time collected vehicle tire inflation pressure and load; π7(t) represents the current time collected other indirect influence on the vehicle's own operating state of the parameter variable weighted value, the default value is 0, when the vehicle detects other indirect influence on the vehicle's own operating state of the parameter variable, all influence on the vehicle's own operating state of the parameter variable in [0, 1) fuzzy evaluation after taking the weighted value as its value at this time; τ Pressure and τ Load respectively refer to the vehicle tire pressure and vehicle load anomaly index, vehicle tire pressure anomaly index τ Pressure is the percentage of the real-time pressure of the vehicle tire deviating from the standard pressure, and the specific calculation method is:

[0036]

[0037] where P(t) represents the pressure of the vehicle tire at the current time, P min represents the minimum pressure of the tire under normal circumstances, P max represents the maximum pressure of the tire under normal circumstances. The calculation method of the vehicle load anomaly index τ Load is similar:

[0038]

[0039] where L(t) represents the load of the vehicle at the current time, L max represents the rated maximum load of the vehicle.

[0040] Thus the vehicle's own operating state parameter matrix can be obtained:

[0041] Π(t) = [π1(t), π2(t), π3(t), π4(t), π5(t), π6(t), π7(t)]

[0042] As shown in the vehicle visual system operation architecture diagram, the vehicle visual system mainly includes two main and auxiliary cameras, and the main steps of acquiring the real-time road surface state parameter matrix are as follows: Figure 2

[0043] Step one: acquiring the target road section image obtained by high-speed shooting of the vehicle main camera by the vehicle visual system. The image is the road surface condition image obtained by shooting when the vehicle reaches the preset distance ΔY.

[0044] Step two: inputting the target road section image into the pre-trained convolutional neural network to obtain the pre-identified type and overall recognition probability of the target road surface feature. The attributes of the target road surface feature should include but are not limited to road surface type (material and gradation, texture characteristics, etc.) and condition (wear degree, tire and road surface medium, etc.) parameters. Here, the pre-trained convolutional neural network can be realized based on existing technology.

[0045] Step three: if the overall recognition probability of the target road surface is greater than or equal to the preset probability threshold c, skip step four and execute step five.

[0046] Step four: if the overall recognition probability of the target road surface is lower than the preset road condition probability threshold c, then acquire the road surface image obtained by the vehicle auxiliary camera to make a second determination. The road surface image is a combination of road surface image sequences obtained by continuous shooting of the auxiliary camera within a preset time period.

[0047] If the determination result is the same as the first determination result, mark this frame as empty and send a road surface missing report, and store the road surface image in the database after manual marking for regular updating of the convolutional neural network; return to step one to shoot a new road surface image.

[0048] Otherwise, execute step five using the target road surface obtained by the vehicle auxiliary camera.

[0049] Step five: converting the feature attributes of the target road surface into a real-time road surface state parameter matrix and sending it to the big data processing and intelligent prediction module.

[0050] The real-time road surface state parameter matrix Θ(t) is represented as:

[0051] Θ(t) = [θ1(t), θ2(t), θ3(t), θ4(t), θ5(t)]

[0052] ​Wherein, θ1(t) represents the current time collected road surface material and gradation level parameters, θ2(t) represents the current time collected road surface texture feature parameters, θ3(t) represents the current time collected wear degree level, θ4(t) represents the current time collected between the tire and the road surface medium type parameters, θ5(t) represents the current time collected other indirect influence of real-time state of the road parameter variable weighted value.

[0053] Synchronously, the real-time state parameter matrix E(t) of the environment can be obtained by the environment sensor carried by the vehicle itself, which should be:

[0054] E(t) = [ε1(t), ε2(t), ε3(t), ε4(t)]

[0055] Wherein, ε1(t) represents the current time collected environmental humidity value, ε2(t) represents the current time collected environmental temperature value, ε3(t) represents the current time collected environmental wind resistance value, and ε4(t) represents the current time collected other indirect influence of vehicle driving resistance environmental variable weighted value.

[0056] Then, as shown in Figure 3 The vehicle big data processing and intelligent prediction module receives the vehicle itself running state parameter matrix Π(t), the real-time state parameter matrix Θ(t) of the road surface, and the real-time state parameter matrix E(t) of the environment from the same time stamp, and inputs them into the road surface friction coefficient intelligent prediction model to obtain the feedback of the road surface friction coefficient prediction value μ(t+1).

[0057] The construction method of the road surface friction coefficient intelligent prediction model is as follows:

[0058] Step one: obtain the road surface friction coefficient training sample under different driving conditions to construct the sample database, each sample data contains all state parameters of the vehicle itself running state parameter matrix Π(t), the real-time state parameter matrix Θ(t) of the road surface, and the real-time state parameter matrix E(t) of the environment, a total of 16 components;

[0059] Step two: establish a road surface friction coefficient intelligent prediction model based on BP neural network, and perform network initialization. Wherein, the network initialization includes setting the maximum training times N, the hidden node number m, the threshold value c, the initial weight value w, the initial learning rate v, the learning expected accuracy p and other parameters.

[0060] Step three: training sample input.

[0061] The training sample input is a training sample matrix of n x 16:

[0062]

[0063] Step four: the intelligent prediction BP neural network of the road friction coefficient is established for the driving sample. First, the training sample is evenly distributed to each Map node. Then, the network weight record is read from the Hadoop distributed file system by calling the Map function. Each Map task will obtain and output the corresponding weight. Next, the Reduce function reads the network weight record from the Hadoop distributed file system and receives the weight output by the Map task. The Reduce task determines whether the next iteration is needed by comparing the difference between the updated network weight of the Map node and the weight stored in the Hadoop file system.

[0064] Step five: if the network weight is no longer updated, the intelligent prediction model BP neural network of the road friction coefficient is trained, and the construction is preliminarily completed.

[0065] Step six: a prior model data packet supplement library is set. In the running, if a new road type or other variable parameters are found, they can be further supplemented to the library. At the same time, according to the comparison between the actual running data and the true value of big data, the optimization and correction are continuously carried out.

[0066] As shown in Figure 4 After obtaining the predicted value of the road friction coefficient μ(t+1), the vehicle power distribution can be adjusted through the vehicle control center system module to optimize the vehicle driving. Specifically, the vehicle control center system module obtains the predicted value of the road friction coefficient, the vehicle control center system module determines the target driving mode of the vehicle based on the vehicle speed, the state of the vehicle suspension, the state of the accelerator pedal and the brake pedal, the wheel speed of each wheel, and the road friction force, and outputs the target driving mode to the instruction execution VCU module, and the instruction execution VCU module controls the vehicle driving mode to switch to the target driving mode.

[0067] The specific power distribution strategy and driving mode switching design scheme and algorithm design are as follows:

[0068] The core of the power distribution strategy is to reasonably adjust the driving force, braking force and lateral force of the front and rear axles according to the real-time predicted road friction coefficient value and the current state of the vehicle.

[0069] For the driving force / braking force distribution of the front and rear axles, the total driving force is obtained according to the product of the current acceleration / deceleration and the vehicle weight, and the maximum available driving force of the front and rear axles is calculated according to the front and rear axle load ratio and the friction coefficient prediction value. If the total driving force does not exceed the sum of the maximum available driving force of the front and rear axles, the total driving force is distributed according to the front and rear axle load ratio, otherwise the maximum available driving force of the front and rear axles is taken as the distributed driving force / braking force, and the total driving force is reduced by reducing the engine power, using the braking force or adjusting the brake system pressure.

[0070] For lateral force distribution, the total lateral force is calculated based on the current driving speed, slip angle, vehicle wheelbase, and vehicle weight. The maximum available lateral force of the front and rear axles is calculated based on the front and rear axle load ratio and the predicted value of the friction coefficient. If the total lateral force does not exceed the sum of the maximum available lateral forces of the front and rear axles, the total lateral force is distributed according to the front and rear axle load ratio. Otherwise, the maximum available lateral force of the front and rear axles is used as the distributed lateral force, and the total lateral force is reduced by adjusting the vehicle's steering angle or speed.

[0071] The specific steps are as follows:

[0072] For drive force distribution: input data includes the predicted road friction coefficient μ pred Current vehicle speed v, current acceleration a, vehicle weight m, front and rear axle load ratio k f and k r (Front axle load ratio k) f Ratio of rear axle load k r (Total = 1), the medium parameters between the tire and the road surface: such as water film thickness, mud and sand coverage, etc.

[0073] Calculate the total driving force:

[0074] F drive_total =m·a

[0075] Calculate the maximum available driving force on the front and rear axles:

[0076] F drive_f_max =μ pred ·k f ·m·g

[0077] F drive_r_max =μ pred ·k r ·m·g

[0078] Where g is the acceleration due to gravity (approximately 9.81 m / s²). 2 ).

[0079] Determine the driving force distribution between the front and rear axles: Compare the calculated total driving force F drive_total With the maximum available driving force F of the front and rear axles drive_f_max and F drive_r_max If F drive_total ≤F drive_f_max +F drive_r_max Then the driving force can be distributed proportionally:

[0080] F drive_f =k f ·F drive_total

[0081] F drive_r =k r ·Fdrive_total

[0082] If F drive_total > F drive_f_max +F drive_r_max then measures need to be taken to reduce the total driving force, ensuring safety:

[0083] F drive_f =F drive_f_max

[0084] F drive_r =F drive_r_max

[0085] F drive_reduce =F drive_total -(F drive_f_max +F drive_r_max )

[0086] The reduction of F drive_reduce is achieved by reducing the engine power or using the braking force.

[0087] For the braking force distribution: the input data include the predicted road surface friction coefficient μ pred , the current driving speed v, the current deceleration a b , the vehicle weight m, the load ratio of front and rear axles k f and k r ;

[0088] The total driving force is calculated:

[0089] F brake_total =m·a b

[0090] The maximum available driving force of the front and rear axles is calculated:

[0091] F brake_f_max =μ pred ·k f ·m·g

[0092] F brake_r_max =μ pred ·k r ·m·g

[0093] The driving force distribution of the front and rear axles is determined: the calculated total driving force F brake_total is compared with the maximum available driving force of the front and rear axles F brake_f_max and F brake_r_max , if F brake_total ≤ F brake_f_max +F brake_r_max , the driving force can be distributed proportionally:

[0094] F brake_f =k f ·Fbrake_total

[0095] F nrake_r = k r · F brake_total

[0096] If F brake_total > F brake_f_max +F brake_r_max then measures need to be taken to reduce the total driving force, ensuring safety:

[0097] F brake_f = F brake_f_max

[0098] F brake_r = F brake_r_max

[0099] F brake_reduce = F brake_total -(F brake_f_max +F brake_r_max )

[0100] The reduction of F brake_reduce is achieved by adjusting the pressure of the braking system.

[0101] For lateral force distribution: input data includes the predicted road surface friction coefficient μ pred , the current driving speed v, the current side slip angle δ, the vehicle weight and center position m and h, the wheel base t, the load ratio of front and rear axles k f and k r ;

[0102] Calculate the total lateral force:

[0103]

[0104] Where L is the wheel base of the vehicle.

[0105] Calculate the maximum available lateral force of the front and rear axles:

[0106] F lateral_f_mac = μ pred · k f · m· g

[0107] F lateral_r_max = μ pred · k r · m· g

[0108] Determine the lateral force distribution of the front and rear axles: compare the calculated total lateral force F lateral_total with the maximum available lateral force of the front and rear axles F lateral_f_max and F lateral_r_max , if F lateral_total ≤ F lateral_f_max +Flateral_r_max Then the lateral force can be proportionally distributed:

[0109] F lateral_f = k f · F lateral_total

[0110] F lateral_r = k r · F lateral_total

[0111] If F lateral_total > F lateral_f_max + F lateral_r_max , measures need to be taken to reduce the total lateral force to ensure safety:

[0112] F lateral_f = F lateral_f_max

[0113] F lateral_r = F lateral_r_max

[0114] F lateral_reduce = F laterak_total - (F lateral_f_max + F lateral_r_max )

[0115] The reduction of F lateral_reduce is achieved by adjusting the steering angle or speed of the vehicle.

[0116] In addition, in one specific implementation of the present application, it is also necessary to intelligently determine the most suitable driving mode according to the road surface friction coefficient prediction value and the current state of the vehicle, and to inform the vehicle control system to adjust the corresponding parameters; when intelligently determining the most suitable driving mode, a predefined evaluation rule is used to evaluate the current most suitable driving mode, and if the evaluated driving mode is different from the current mode and the switching conditions meet the requirements, switching is performed; otherwise, no switching is performed. The driving modes include normal mode, wet mode, snow mode, and mud mode; each mode has a predefined parameter adjustment scheme.

[0117] In this embodiment, the driving modes are defined as follows:

[0118] Normal Mode: suitable for dry and flat road surfaces.

[0119] Wet Mode: suitable for wet and slippery road surfaces.

[0120] Snow Mode: suitable for icy and snowy road surfaces.

[0121] Mud Mode: suitable for muddy and soft road surfaces.

[0122] Establish evaluation rules:

[0123] Using decision tree or rule engine, evaluate the most suitable driving mode according to different road friction coefficient values, vehicle status and environmental conditions.

[0124] For example: if μ pred <0.4 and humidity > 80%, switch to wet mode; if μ pred <0.2 and temperature < 0℃, and road type is ice and snow, switch to snow mode; if μ pred <0.3 and sand coverage > 50%, switch to muddy mode.

[0125] When making mode switching decisions, first evaluate in real time, and every certain time (such as 0.1 seconds) get the latest friction coefficient prediction value and other related data from the intelligent prediction module. Use pre-defined evaluation rules to evaluate the most suitable driving mode.

[0126] Switching conditions such as large predicted friction coefficient changes, high vehicle speed, driver requests, etc. determine whether to switch modes.

[0127] For example: if switching from normal mode to wet mode, ensure that the vehicle speed is below 60 km / h; if switching from wet mode to snow mode, ensure that the vehicle speed is below 40 km / h.

[0128] The invention sends mode switching instructions through the vehicle control center system (such as ECU or other control unit).

[0129] The instructions include the new driving mode and its corresponding parameter adjustment requirements.

[0130] In one proposed implementation of the invention, different driving models have pre-defined parameter adjustment schemes for engine management system (EMS), brake control system (BCS), steering control system (SCS), and suspension system (Suspension System) respectively.

[0131] Engine management system (EMS): adjust engine output power to ensure safety and economy.

[0132] Normal mode: normal power output.

[0133] Wet mode: reduce power output to prevent tire slip.

[0134] Snow mode: further reduce power output and use low gear.

[0135] Muddy mode: adjust torque output, use low gear, and increase anti-slip control.

[0136] Brake Control System (BCS): Adjusts brake force distribution to ensure balance between front and rear axles.

[0137] Normal Mode: Standard brake force distribution.

[0138] Wet Mode: Increases rear axle brake force and reduces front axle brake force to prevent front wheel spin.

[0139] Snow Mode: Maintains balance between front and rear axle brake forces and increases Electronic Stability Program (ESP) intervention.

[0140] Mud Mode: Increases front axle brake force and reduces rear axle brake force and increases Anti-lock Braking System (ABS) intervention.

[0141] Steering Control System (SCS): Adjusts steering angle and steering assist to optimize vehicle handling.

[0142] Normal Mode: Standard steering assist.

[0143] Wet Mode: Moderately increases steering assist for improved handling stability.

[0144] Snow Mode: Substantially increases steering assist and reduces steering angle changes.

[0145] Mud Mode: Moderately reduces steering assist and increases limits on steering angle changes.

[0146] Suspension System: Adjusts suspension damping and stiffness to optimize vehicle ride stability.

[0147] Normal Mode: Standard suspension settings.

[0148] Wet Mode: Increases suspension damping and reduces body roll.

[0149] Snow Mode: Further increases suspension damping for improved ride stability.

[0150] Mud Mode: Reduces suspension damping and increases suspension stiffness for improved traction.

[0151] After mode switching, the vehicle state and road conditions continue to be monitored in real-time. If new changes in road conditions or vehicle state are detected, parameters can be dynamically adjusted or mode switching can be triggered again. For example, if dry road conditions are detected while in Wet Mode, the system can re-evaluate and switch back to Normal Mode.

[0152] The above examples are just specific embodiments of the invention. It is clear that the invention is not limited to the above examples and can have many variations. All variations that can be directly derived or inferred from the disclosure of the invention by those skilled in the art should be considered within the scope of the invention.

Claims

1. A vehicle driving optimization method based on an intelligent prediction model of road surface friction coefficient, characterized in that, include: The vehicle's vision system collects real-time road surface parameters, including road type and road condition, in real time. The vehicle collects data at a specified distance along the road ahead. The distance range relative to the vehicle is: in, This represents the vehicle speed in the current geodetic coordinate system. Indicates the upper and lower limits of the collision time. Indicates the minimum and maximum values ​​within the distance range; The upper and lower limits of the collision time satisfy the following conditions: in, , These are the minimum and maximum times of the preset safety prediction constraint interval, respectively. To predict the friction coefficient value in real time and issue optimized vehicle power adjustment commands at the appropriate time. Based on the real-time road surface status parameters, vehicle operating status parameters, and environmental status parameters, the intelligent friction coefficient prediction model outputs the real-time predicted friction coefficient value between the vehicle and the road surface. The real-time predicted friction coefficient value is transmitted to the vehicle control system to adjust the vehicle's power distribution.

2. The vehicle driving optimization method based on the intelligent prediction model of road surface friction coefficient according to claim 1, characterized in that, The process of collecting real-time road surface status parameters through the vehicle's vision system is as follows: (1) Obtain the target road segment image by high-speed shooting of the vehicle's main camera through the vehicle vision system, that is, the road condition image captured when the vehicle reaches the preset distance; (2) Input the target road segment image into a pre-trained convolutional neural network for feature recognition to obtain the pre-recognition type and overall recognition probability of the target road surface features; (3) If the overall recognition probability of the target road surface is greater than or equal to the preset probability threshold , proceed directly to step (4); If the overall recognition probability of the target road surface is lower than a preset road condition probability threshold Then, the road surface image captured by the vehicle's secondary camera is used for secondary judgment. The road surface image is a combination of road surface image sequences continuously captured by the secondary camera within a preset time period. If the judgment result is the same as the first judgment result, then mark this frame as empty and send a road surface missing report. At the same time, manually mark the road surface image and store it in the database for periodic updates of the convolutional neural network; return to step (1) and re-capture a new road surface image; otherwise, use the target road surface captured by the vehicle's secondary camera to execute step (4). (4) Transform the characteristic attributes of the target road surface into a real-time state parameter matrix of the road surface.

3. The vehicle driving optimization method based on an intelligent prediction model of road surface friction coefficient according to claim 1 or 2, characterized in that, The real-time road surface status parameters include at least the road surface material and gradation, texture characteristics, road surface wear degree, and the medium between the tire and the road surface.

4. The vehicle driving optimization method based on the intelligent prediction model of road surface friction coefficient according to claim 1, characterized in that, The vehicle's own operating parameters include vehicle speed, vehicle acceleration, sideslip angle, slip ratio, tire type and wear degree, tire inflation pressure and load, and the calculation formula is as follows: in, This represents the vehicle speed in the current geodetic coordinate system. These represent the vehicle's longitudinal coordinates in the geodetic coordinate system. Horizontal coordinates The first derivative, , These represent the vehicle's longitudinal and lateral coordinates in the vehicle coordinate system, respectively. This represents the vehicle acceleration in the geodetic coordinate system at the current moment. express The first derivative, express The second derivative; This indicates the vehicle sideslip angle collected at the current moment; This represents the vehicle slip ratio collected at the current moment. This represents the angular velocity of the wheel. Indicates the radius of the wheel; This indicates the type and wear level of the vehicle tires collected at the current moment, which depends on the type of vehicle tires. Tire wear rate and remaining service life A fuzzy mathematical evaluation is performed on the actual mileage and time traveled, and the evaluation score belonging to the interval [0,1] is calculated as its return value. This represents the fuzzy mathematical evaluation process; This indicates the vehicle tire inflation pressure and load collected at the current moment. This represents the weighted value of other parameters that indirectly affect the vehicle's operating status at the current moment, with a default value of 0.

5. The vehicle driving optimization method based on the intelligent prediction model of road surface friction coefficient according to claim 1, characterized in that, The environmental parameters mentioned include humidity, temperature, and wind resistance.

6. The vehicle driving optimization method based on the intelligent prediction model of road friction coefficient according to claim 1, characterized in that, The aforementioned intelligent prediction model for friction coefficient employs a BP neural network.

7. The vehicle driving optimization method based on the intelligent prediction model of road surface friction coefficient according to claim 1, characterized in that, The aforementioned method transmits the real-time predicted friction coefficient value to the vehicle control system to adjust the vehicle's power distribution, including the distribution of driving force, braking force, and lateral force between the front and rear axles. For the distribution of driving force / braking force between the front and rear axles, the total driving force is obtained by multiplying the current acceleration / deceleration and the vehicle weight, and the maximum available driving force between the front and rear axles is calculated based on the load ratio of the front and rear axles and the predicted value of the friction coefficient. If the total driving force does not exceed the sum of the maximum available driving forces of the front and rear axles, the total driving force is distributed according to the load ratio of the front and rear axles; otherwise, the maximum available driving force of the front and rear axles is used as the distributed driving force / braking force, and the total driving force is reduced by reducing engine power, using braking force, or adjusting the pressure of the braking system. For lateral force distribution, the total lateral force is calculated based on the current driving speed, slip angle, vehicle wheelbase and vehicle weight, and the maximum available lateral force of the front and rear axles is calculated based on the front and rear axle load ratio and the predicted value of the friction coefficient. If the total lateral force does not exceed the sum of the maximum available lateral forces of the front and rear axles, the total lateral force is distributed according to the load ratio of the front and rear axles; otherwise, the maximum available lateral forces of the front and rear axles are used as the distributed lateral force, and the total lateral force is reduced by adjusting the vehicle's steering angle or speed.

8. The vehicle driving optimization method based on the intelligent prediction model of road friction coefficient according to claim 1, characterized in that, Also includes: Based on the predicted value of the road surface friction coefficient and the current state of the vehicle, the system intelligently determines the most suitable driving mode and notifies the vehicle control system to adjust the corresponding parameters. When the system intelligently determines the most suitable driving mode, it uses predefined evaluation rules to evaluate the most suitable driving mode at present. If the evaluated driving mode is different from the current mode and the switching conditions are met, then the switching is performed. Otherwise, do not switch.

9. The vehicle driving optimization method based on the intelligent prediction model of road surface friction coefficient according to claim 8, characterized in that, The driving modes include normal mode, wet mode, snow mode, and mud mode; each mode has a predefined parameter adjustment scheme.

Images