A vehicle road surface state identification method, device and vehicle controller

Abstract

The application discloses a vehicle road surface state recognition method and device and a whole vehicle controller, relates to the field of road surface recognition, and comprises the following steps: acquiring the running state parameters of each wheel of a vehicle, calculating the current road surface adhesion coefficient and the slip rate of each wheel based on the running state parameters, performing asymmetric time sequence updating on the current road surface adhesion coefficient according to the preset slip rate interval in which the slip rate is located, obtaining the output road surface adhesion coefficient of each wheel, recognizing the checkerboard road surface distribution state based on the output road surface adhesion coefficient of each wheel, and inhibiting the output of the left and right side adhesion separation state in response to the recognition of the checkerboard road surface distribution state. According to the asymmetric time sequence updating guided by the slip rate partition, the weak excitation underestimation and the over-slip overestimation are inhibited, the reliable output adhesion coefficient is obtained, the checkerboard road surface distribution state is recognized on the basis, the output of the left and right side adhesion separation state is inhibited, the checkerboard road surface and the stable opposite-opening road surface are effectively distinguished, and the directional error of the whole vehicle control strategy is avoided.

Description

Technical Field

[0001] This invention relates to the field of road surface recognition technology, and more specifically, to a method, device, and vehicle controller for recognizing vehicle road surface conditions. Background Technology

[0002] The coefficient of friction (COP) is a key parameter characterizing the friction between tires and the road surface. It directly determines the boundaries of the vehicle's available longitudinal driving force and lateral steering force, and serves as a crucial input for four-wheel drive torque distribution, anti-lock braking systems (ABS), traction control systems, and vehicle stability control systems. In modern automotive chassis control strategies, accurately sensing road conditions is essential for ensuring driving safety, improving handling stability, and optimizing the driving experience. Existing technical solutions typically rely on vehicle kinematic parameters, such as wheel speed, vehicle speed, acceleration, and yaw rate, combined with tire mechanics models or empirical formulas to construct a COP estimator. These methods dynamically calculate the current road surface adhesion level by real-time monitoring of the relationship between wheel slip ratio and longitudinal force. Furthermore, they analyze the adhesion differences between the left and right wheels to determine complex conditions such as split-wheel surfaces, thus providing decision support for the vehicle controller.

[0003] However, accurate identification of road surface conditions during actual vehicle operation still faces long-standing technical bottlenecks. Particularly under weak excitation conditions, the lack of significant changes in tire longitudinal force and slip ratio can easily lead to an underestimation of road adhesion. Under high slip or overslip conditions, fluctuations in wheel speed, torque impacts, and load estimation errors can easily result in momentarily inflated adhesion coefficients. Furthermore, when faced with complex road surfaces with non-uniform spatial distribution, such as checkerboard patterns or alternating adhesion surfaces, existing technologies often struggle to effectively distinguish them from stable left-right adhesion separation surfaces, easily leading to misjudgments due to the limitations of single characteristic indicators. This instability and uncertainty in the identification results can cause directional errors in vehicle control strategies during torque limiting, torque distribution, or braking intervention, thereby affecting the vehicle's dynamic response performance and safety redundancy. Summary of the Invention

[0004] In view of this, the present invention discloses a vehicle road surface condition recognition method, device and vehicle controller to solve the technical problems of inaccurate estimation of road surface adhesion coefficient and difficulty in effectively distinguishing between checkerboard road surface and stable split road surface during actual vehicle operation, especially under weak excitation or high slip conditions, which leads to control misjudgment.

[0005] A method for vehicle road surface condition recognition includes:

[0006] The operating status parameters of each wheel of the vehicle are obtained, and the current road adhesion coefficient and slip ratio of each wheel are calculated based on the operating status parameters.

[0007] Based on the preset slip ratio range in which the slip ratio is located, an asymmetric time-series update is performed on the current road surface adhesion coefficient to obtain the output road surface adhesion coefficient of each wheel; wherein, the asymmetric time-series update includes: suppressing the decrease of the output road surface adhesion coefficient when the slip ratio is in the low slip ratio range, and suppressing the increase of the output road surface adhesion coefficient when the slip ratio is in the high slip ratio range.

[0008] Based on the output road surface adhesion coefficient of each wheel, the chessboard road surface distribution state is identified;

[0009] In response to the identification of the chessboard surface distribution state, the output of the left and right side attachment separation state is suppressed.

[0010] Optionally, the step of performing asymmetric temporal updates on the current road surface adhesion coefficient based on the preset slip ratio range in which the slip ratio is located, to obtain the output road surface adhesion coefficient for each of the wheels, includes:

[0011] In response to the absolute value of the slip ratio being less than or equal to a first slip ratio threshold, it is determined that the slip ratio is in the low slip ratio range, and the larger value between the current road surface adhesion coefficient and the road surface adhesion coefficient output at the previous moment is taken as the original value of the unfiltered output road surface adhesion coefficient at the current moment.

[0012] In response to the absolute value of the slip ratio being greater than or equal to a second slip ratio threshold, it is determined that the slip ratio is in the high slip ratio range, and the smaller value between the current road surface adhesion coefficient and the road surface adhesion coefficient output at the previous moment is taken as the original value of the output road surface adhesion coefficient.

[0013] In response to the absolute value of the slip ratio being greater than the first slip ratio threshold and less than the second slip ratio threshold, it is determined that the slip ratio is in the intermediate slip ratio range, and the road surface adhesion coefficient output at the previous moment is used as the original value of the output road surface adhesion coefficient.

[0014] The original value of the output road surface adhesion coefficient is filtered to obtain the output road surface adhesion coefficient at the current time.

[0015] Optionally, calculating the current road adhesion coefficient and slip ratio of each wheel based on the operating state parameters includes:

[0016] The current road adhesion coefficient of each wheel is determined based on the wheel-end driving torque, effective tire radius, and wheel-end vertical load of each wheel at the current moment.

[0017] The slip ratio of each wheel is determined based on the reference wheel speed, actual wheel speed, and preset low-speed protection threshold of each wheel.

[0018] Optionally, identifying the chessboard road surface distribution state based on the output road surface adhesion coefficient of each of the wheels includes at least one of the following judgment conditions:

[0019] The difference in road surface adhesion coefficient output by the left and right sides of the front axle has the opposite sign to the difference in road surface adhesion coefficient output by the left and right sides of the rear axle, and the absolute values ​​of both are greater than the preset chessboard threshold.

[0020] The difference between the sum of the output road surface adhesion coefficients of the diagonal wheel combinations is greater than the difference between the sum of the output road surface adhesion coefficients of the left and right wheel combinations.

[0021] The difference in the output road surface adhesion coefficient of the front and rear wheels on the same side both exceed the preset alternation threshold, and the difference in the average output road surface adhesion coefficient between the left and right sides is less than the preset opening threshold.

[0022] Optionally, it also includes:

[0023] In response to the failure to identify the chessboard road surface distribution state, the normalized values ​​of the difference in road surface adhesion coefficient between the left and right sides and the difference in torque between the left and right wheel ends are calculated.

[0024] The normalized values ​​of the difference in road surface adhesion coefficient between the left and right sides and the difference in torque at the left and right wheel ends are normalized and then weighted and fused to obtain the percentage of split road surface.

[0025] The normalized value of the torque difference between the left and right wheel ends is calculated by dividing the difference between the sum of the driving torques at the left wheel end and the sum of the driving torques at the right wheel end by the maximum value among the sum of the driving torques at the left wheel end, the sum of the driving torques at the right wheel end, and the normalized protection threshold.

[0026] Optionally, the calculation of the difference in adhesion coefficients between the left and right road surfaces includes using at least one of the following methods:

[0027] Method 1: In response to the mean output road surface adhesion coefficient of the left side being greater than or equal to the mean output road surface adhesion coefficient of the right side, the smaller value of the output road surface adhesion coefficient of the two left wheels is subtracted from the larger value of the output road surface adhesion coefficient of the two right wheels to obtain the difference in road surface adhesion coefficients between the left and right sides; In response to the mean output road surface adhesion coefficient of the right side being greater than the mean output road surface adhesion coefficient of the left side, the smaller value of the output road surface adhesion coefficient of the two right wheels is subtracted from the larger value of the output road surface adhesion coefficient of the two left wheels to obtain the difference in road surface adhesion coefficients between the left and right sides.

[0028] Method 2: Take half of the difference between the sum of the output road surface adhesion coefficients of the two left wheels and the sum of the output road surface adhesion coefficients of the two right wheels as the original value of the left and right side adhesion difference, and take the absolute value of the original value of the left and right side adhesion difference as the difference of the left and right side road surface adhesion coefficients.

[0029] Optionally, the method further includes post-processing the output road surface adhesion coefficient, the post-processing including at least one of the following:

[0030] The original value of the output road surface adhesion coefficient is subjected to a first-order low-pass filter to obtain the output road surface adhesion coefficient at the current moment.

[0031] The change in the output road surface adhesion coefficient at the current moment relative to the output road surface adhesion coefficient at the previous moment is limited so that the change is between the maximum allowable decrease in a single step and the maximum allowable increase in a single step.

[0032] Optionally, before identifying the chessboard surface distribution state, an enable condition judgment is further performed, the enable condition judgment including:

[0033] If the absolute value of the vehicle's lateral acceleration is greater than a preset lateral acceleration threshold, or the vehicle's required driving torque is less than a preset required torque threshold, or the vehicle speed is lower than a preset low speed threshold, the update of the left and right side attachment separation state is prohibited, and the output result of the previous moment is maintained.

[0034] A vehicle road condition recognition device, comprising:

[0035] The parameter acquisition unit is used to acquire the operating status parameters of each wheel of the vehicle, and calculate the current road surface adhesion coefficient and slip ratio of each wheel based on the operating status parameters.

[0036] An adhesion coefficient update unit is used to perform asymmetric temporal update on the current road surface adhesion coefficient according to the preset slip ratio range in which the slip ratio is located, so as to obtain the output road surface adhesion coefficient of each wheel; wherein, the asymmetric temporal update includes: suppressing the decrease of the output road surface adhesion coefficient when the slip ratio is in the low slip ratio range, and suppressing the increase of the output road surface adhesion coefficient when the slip ratio is in the high slip ratio range.

[0037] A road surface recognition unit is used to identify the chessboard road surface distribution state based on the output road surface adhesion coefficient of each of the wheels.

[0038] The output suppression unit is used to suppress the output of the left and right side attachment separation state in response to the identification of the chessboard road surface distribution state.

[0039] A vehicle controller, the vehicle controller comprising: a memory and a processor;

[0040] The memory is used to store at least one instruction;

[0041] The processor is used to execute at least one instruction to implement the vehicle road condition recognition method described above.

[0042] As can be seen from the above technical solution, the present invention discloses a vehicle road surface condition recognition method, device, and vehicle controller. It acquires the operating state parameters of each wheel of the vehicle and calculates the current road surface adhesion coefficient and slip ratio of each wheel based on these parameters. Furthermore, it performs asymmetric timing updates on the current road surface adhesion coefficient according to the preset slip ratio range. This suppresses the decrease in output road surface adhesion coefficient under weak excitation conditions to prevent underestimation of high-adhesion road surfaces, and suppresses the increase in output road surface adhesion coefficient under high-slip conditions to filter out excessive slippage. This yields the output road surface adhesion coefficient of each wheel, overcoming the problem of inaccurate adhesion coefficient estimation caused by insignificant changes in tire longitudinal force and slip ratio, wheel speed fluctuations, torque impacts, and load estimation errors. This provides a stable and reliable data foundation for subsequent road surface type recognition. Based on this, it identifies the chessboard road surface distribution state based on the output road surface adhesion coefficient of each wheel, and suppresses the output of left and right side adhesion separation states when the chessboard road surface distribution state is identified. This effectively distinguishes the chessboard road surface from stable left and right side adhesion separation roads, solving the problem of misjudgment of chessboard conditions caused by the limitations of a single feature index and avoiding directional errors in the vehicle control strategy. Attached Figure Description

[0043] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the published drawings without creative effort.

[0044] Figure 1 This is a flowchart of a vehicle road surface condition recognition method disclosed in an embodiment of the present invention;

[0045] Figure 2 This is a schematic diagram of the structure of a vehicle road surface condition recognition device disclosed in an embodiment of the present invention;

[0046] Figure 3 This is a schematic diagram of the structure of a vehicle controller disclosed in an embodiment of the present invention. Detailed Implementation

[0047] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0048] See Figure 1This invention discloses a flowchart of a vehicle road condition recognition method. This method is applicable to road condition recognition for new energy vehicles such as pure electric vehicles, hybrid vehicles, and fuel cell vehicles. The method specifically includes:

[0049] Step S101: Obtain the operating status parameters of each wheel of the vehicle, and calculate the current road surface adhesion coefficient and slip ratio of each wheel based on the operating status parameters.

[0050] The operating state parameters of each wheel of a vehicle are a set of physical quantities used to characterize the current motion and force state of the wheel, specifically including the reference wheel speed of each wheel collected at time k. Actual wheel speed Wheel-end drive torque Effective radius of tire and wheel end vertical load lateral acceleration Demand-driven torque And so on, where i represents the wheel index, i∈{LF, RF, LR, RR}, corresponding to the left front wheel, right front wheel, left rear wheel, and right rear wheel, respectively. The vertical load at the wheel end is determined by... The vehicle's mass, wheelbase, track width, center of gravity height, and longitudinal and lateral acceleration are estimated. These parameters can be acquired in real time through the sensor interface of the Vehicle Control Unit (VCU) or the Controller Area Network (CAN) bus. For example, the reference wheel speed can be calculated from the vehicle's reference speed combined with the yaw rate, track width, and steering angle; the actual wheel speed can be measured by wheel speed sensors and converted into linear velocity; the wheel-end drive torque can be obtained from the motor torque output command or the drive shaft torque sensor; and the wheel-end vertical load is estimated based on the vehicle's mass, center of gravity position, and acceleration status.

[0051] The current road adhesion coefficient characterizes the available adhesion limit between the tire and the road surface under current operating conditions. This coefficient is calculated using the wheel-end drive torque, effective tire radius, and wheel-end vertical load, reflecting the degree to which the current tire longitudinal force utilizes the vertical load. For example, the current road adhesion coefficient can be obtained by dividing the wheel-end drive torque by the product of the effective tire radius and the wheel-end vertical load; or it can be derived by combining a tire model and using parameters such as slip ratio and sideslip angle. To ensure the reliability of the calculation results, the operating state parameters are typically low-pass filtered, deburred, and validated before calculation. When the wheel-end vertical load is below a preset threshold, the calculation result from the previous moment can be retained or deemed invalid to avoid numerical anomalies caused by division operations.

[0052] Slip ratio describes the relative motion of a tire to the road surface and is a key indicator for judging the tire's working condition. The slip ratio is calculated based on a reference wheel speed, the actual wheel speed, and a preset low-speed protection threshold. The preset low-speed protection threshold is used to prevent numerical distortion caused by wheel speed signal fluctuations or division by zero under low-speed conditions. For example, the difference between the reference wheel speed and the actual wheel speed is used as the numerator, and the larger value between the reference wheel speed and the preset low-speed protection threshold is used as the denominator to calculate the ratio; alternatively, a sliding window averaging method can be used to smooth the wheel speed before calculating the relative slip. This approach allows for compatibility with slip ratio definitions in different driving and braking directions and effectively suppresses low-speed noise.

[0053] Step S102: Based on the preset slip ratio range in which the slip ratio is located, perform asymmetric time-series update on the current road surface adhesion coefficient to obtain the output road surface adhesion coefficient of each wheel.

[0054] A preset slip ratio range is a defined range based on the magnitude of the slip ratio, used to differentiate the tire's excitation level. Specifically, it includes a low slip ratio range, a high slip ratio range, and an intermediate range between the two. This range is defined by setting a first slip ratio threshold and a second slip ratio threshold, with the first slip ratio threshold being less than the second slip ratio threshold. The preset slip ratio range can be determined, for example, based on calibration results of the linear region's end and the nonlinear region's start of the tire characteristic curve, or based on the statistical distribution of the confidence level of the adhesion coefficient estimation from a large amount of real-vehicle test data. In practical applications, the first slip ratio threshold can be set, for example, to a value between 0.05 and 0.10, and the second slip ratio threshold can be set, for example, to a value between 0.15 and 0.20, to cover the transition range from the linear region to the slip region.

[0055] Asymmetric temporal update refers to the process of dynamically adjusting the output road surface adhesion coefficient by applying different logical combination strategies to the current road surface adhesion coefficient and the output value at the previous moment based on different slip rate intervals. The output road surface adhesion coefficient is the adhesion coefficient value finally used for road surface condition identification after temporal update and filtering. In this embodiment, asymmetric temporal update includes: suppressing the decrease of the output road surface adhesion coefficient when the slip rate is in the low slip rate interval, and suppressing the increase of the output road surface adhesion coefficient when the slip rate is in the high slip rate interval.

[0056] When the slip ratio is in the low slip ratio range, the tire is in a weak excitation state, and the current road adhesion coefficient calculated based on torque is often too low. Direct output will cause high-adhesion roads to be misjudged as low-adhesion roads. Therefore, a strategy to suppress the decline is implemented at this time. This strategy can be, for example, comparing the current road adhesion coefficient with the road adhesion coefficient output at the previous moment and taking the larger value as the updated original value; or using a weighted method, giving higher weight to the output value at the previous moment to avoid the output value being dragged down by the instantaneous low value.

[0057] When the slip ratio is in the high slip ratio range, the tire may enter the overslip or nonlinear region, and wheel speed fluctuations, torque impacts, and load estimation errors can easily lead to an artificially high current road adhesion coefficient. Therefore, a strategy to suppress its increase is implemented at this time. This strategy could be, for example, comparing the current road adhesion coefficient with the road adhesion coefficient output at the previous moment and taking the smaller value as the updated original value; or setting a dynamic upper limit to restrict the increase of the output value with the current calculated value.

[0058] Step S103: Identify the chessboard road surface distribution state based on the output road surface adhesion coefficient of each wheel.

[0059] The checkerboard road surface distribution refers to a special working condition where the road surface adhesion coefficients of the four wheels of a vehicle are distributed alternately front-to-back or diagonally. For example, the left front wheel and right rear wheel are on a high-adhesion road surface, while the right front wheel and left rear wheel are on a low-adhesion road surface. This identification process is achieved by analyzing the specific logical relationship of the spatial distribution of the road surface adhesion coefficients output by the four wheels.

[0060] For example, the difference in road surface adhesion coefficients output by the left and right front axle wheels and the difference in road surface adhesion coefficients output by the left and right rear axle wheels can be calculated. If the signs of the two are opposite and their absolute values ​​are both greater than a preset chessboard threshold, then it is determined to be a chessboard surface. Alternatively, the difference in the sum of the road surface adhesion coefficients output by diagonal wheel combinations (such as left front and right rear, right front and left rear) can be calculated and compared with the difference in the sum of the road surface adhesion coefficients output by left and right wheel combinations. If the former is significantly greater than the latter, then it is determined to be a chessboard surface. The preset chessboard threshold can be determined, for example, based on the statistical difference in the distribution of adhesion coefficient differences between split surfaces and chessboard surfaces, or it can be calibrated based on simulation test results of different vehicle models and tire characteristics. Through multi-dimensional logical judgment, chessboard surfaces can be accurately distinguished from stable left and right split surfaces.

[0061] Step S104: In response to recognizing the chessboard road surface distribution state, suppress the output of the left and right side attachment separation state.

[0062] The output of the left and right side adhesion separation status can refer to signals or parameters characterizing a significant difference in the adhesion coefficients between the left and right wheels of the vehicle, such as a split-path marking, split direction, or percentage of split. In response to recognizing a checkerboard-like path distribution, the system will perform a suppression operation, blocking or correcting the output of such signals. This is because a checkerboard path may momentarily appear as if it is higher on the left or right, and directly outputting the left and right side adhesion separation status could easily mislead the downstream controller into executing incorrect torque distribution or yaw control. Suppression operations can be implemented by forcibly setting the percentage of split-path to zero, freezing it at the previous value, or outputting a specific checkerboard-like path flag to instruct the downstream controller to ignore the conventional split-path judgment logic. This processing structurally prevents the checkerboard path from being misjudged as a stable split-path, ensuring the correctness of the vehicle control strategy.

[0063] In summary, this invention discloses a vehicle road surface condition recognition method. It acquires the operating state parameters of each wheel and calculates the current road surface adhesion coefficient and slip ratio of each wheel based on these parameters. Furthermore, it performs asymmetric temporal updates on the current road surface adhesion coefficient according to a preset slip ratio range. This suppresses the decrease in output road surface adhesion coefficient under weak excitation conditions to prevent underestimation of high-adhesion road surfaces, and suppresses the increase in output road surface adhesion coefficient under high-slip conditions to filter out excessive slippage. This yields the output road surface adhesion coefficient for each wheel, overcoming the problem of inaccurate adhesion coefficient estimation caused by insignificant changes in tire longitudinal force and slip ratio, wheel speed fluctuations, torque impacts, and load estimation errors. This provides a stable and reliable data foundation for subsequent road surface type recognition. Based on this, it identifies the chessboard road surface distribution state based on the output road surface adhesion coefficient of each wheel, and suppresses the output of left and right side adhesion separation states when the chessboard road surface distribution state is identified. This effectively distinguishes the chessboard road surface from stable left and right side adhesion separation roads, solving the problem of misjudgment of chessboard conditions caused by the limitations of a single feature index and avoiding directional errors in the vehicle control strategy.

[0064] Furthermore, for new energy vehicles such as pure electric vehicles and hybrid vehicles, the requirements for the accuracy and real-time performance of road surface adhesion coefficient recognition are even more stringent due to the fast response speed and high torque control precision of the motor. To address this, this invention provides a vehicle road surface state recognition method, which can be widely applied to new energy vehicles such as pure electric vehicles, hybrid vehicles, and fuel cell vehicles. It provides reliable road surface state input for their four-wheel drive torque distribution and chassis stability control, thereby effectively solving the problems of directional errors in vehicle control strategies caused by underestimation of road surface adhesion under weak excitation conditions, instantaneous artificially high adhesion coefficient under high slip conditions, and difficulty in effectively distinguishing between checkerboard road surfaces and stable split-type road surfaces.

[0065] In one embodiment, step S102 may specifically include:

[0066] (1) In response to the absolute value of the slip ratio being less than or equal to the first slip ratio threshold, determining that the slip ratio is in the low slip ratio interval, and taking the larger value of the current road surface adhesion coefficient and the road surface adhesion coefficient output at the previous moment as the unfiltered original value of the output road surface adhesion coefficient at the current moment.

[0067] When the vehicle is under weak excitation working conditions such as small driving torque or constant-speed driving, the tire fails to fully stimulate its adhesion potential, which may lead to the current road surface adhesion coefficient calculated based on the real-time torque being on the low side. For this reason, the first slip ratio threshold N1 is set as a calibration boundary parameter for defining the low slip ratio interval. The physical meaning of the first slip ratio threshold N1 corresponds to the critical point before the relation curve between the tire longitudinal force and the slip ratio enters the linear region, and usually takes a small value, for example, 2% or 3%, which is used to judge whether the tire is in a slight slip or steady rolling state.

[0068] When the slip ratio has an absolute value less than or equal to the first slip ratio threshold N1, it is determined that the current working condition belongs to the low slip ratio interval. At this time, the contact characteristics between the tire and the road surface are not fully stimulated, and if the currently calculated adhesion coefficient is directly adopted, a high-adhesion road surface is likely to be misjudged as a low-adhesion road surface. Therefore, the maximum-taking strategy is adopted as one of the logics for asymmetric time sequence update, that is, the larger value is selected between the current road surface adhesion coefficient and the road surface adhesion coefficient output at the previous moment.

[0069] Specifically, let N1 be the first slip ratio threshold, N2 be the second slip ratio threshold, and 0≤N1<N2. For each wheel, according to the interval where the absolute value of the slip ratio lies, for the current road surface adhesion coefficient of the i-th wheel and the i-th wheel at the previous moment road surface adhesion coefficient performs the following time sequence update, where is the unfiltered original value of the road surface adhesion coefficient output value , that is, the unfiltered original value of the output road surface adhesion coefficient at the current moment.

[0070] If | |≤N1: =max , .

[0071] The maximum-taking strategy is adopted for the low slip region. The reason is: at this time, the tire is not fully stimulated, and what is usually calculated according to the wheel-end driving torque, the effective tire radius and the wheel-end vertical load is only a small current utilized adhesion, and direct output is prone to misestimating a high-adhesion road surface as a low-adhesion road surface. Taking the larger value from the previous output prevents the μ output from being dragged down by low-excitation conditions. This approach preserves the memory of higher adhesion coefficients in historical records, preventing the output value from being lowered due to insufficient instantaneous driving force, thus avoiding underestimation of the adhesion coefficient on high-adhesion surfaces. The original value of the road adhesion coefficient generated by the above calculation process serves as the input data for subsequent filtering stages, achieving monotonically non-decreasing characteristics of adhesion coefficient estimation in the time dimension, enhancing the robustness of the system in the weak-excitation stage.

[0072] (2) In response to the absolute value of the slip ratio being greater than or equal to the second slip ratio threshold, the slip ratio is determined to be in the high slip ratio range, and the smaller value between the current road surface adhesion coefficient and the road surface adhesion coefficient output at the previous moment is taken as the original value of the output road surface adhesion coefficient.

[0073] For vehicles experiencing excessive slippage during rapid acceleration or on certain low-friction surfaces, the tire's operating point may have entered a declining or unstable region after peak adhesion. A second slip ratio threshold N2 is set as another key calibration parameter to define the high slip ratio range. Its value is typically set within the slip ratio range after the tire reaches its peak adhesion coefficient, for example, between 15% and 20%. The specific value needs to be calibrated based on the tire model and road characteristics. When the absolute value of the slip ratio reaches or exceeds this second slip ratio threshold, the slip ratio is determined to be in the high slip ratio range. Within this range, due to drastic wheel speed fluctuations, torque intervention in the control system, and errors in vertical load estimation, the directly calculated current road adhesion coefficient often contains a large positive deviation, exhibiting a momentary artificially high value. To suppress this abnormal rise, a smaller value strategy is implemented: the current road adhesion coefficient is compared with the road adhesion coefficient output at the previous moment, and the smaller value is selected as the original output road adhesion coefficient value.

[0074] If | |≥N2: =min[ , ].

[0075] A strategy of minimizing slippage is adopted for the high-slip zone. This is because the tire may enter a non-linear or overslip region at this point, where wheel speed fluctuations, torque surges, torque reduction in control, and load estimation errors can cause… Instantaneous high, by By reducing the output value compared to the previous moment, abnormal lift under overslip conditions can be suppressed. This method limits the increase in the coefficient of adhesion during the high-slip phase, effectively filtering out noise interference caused by excessive slip, and ensuring that the output coefficient of adhesion is not artificially high due to instantaneous anomalies. This provides a safer and more conservative decision-making basis for subsequent vehicle stability control.

[0076] (3) In response to the absolute value of the slip ratio being greater than the first slip ratio threshold and less than the second slip ratio threshold, the slip ratio is determined to be in the middle slip ratio range, and the road surface adhesion coefficient output at the previous moment is used as the original value of the output road surface adhesion coefficient.

[0077] For the transitional region between the low and high slip ratios, i.e., the intermediate slip ratio range, tire adhesion utilization is in a relatively sensitive dynamic change phase. This range is jointly defined by the first slip ratio threshold N1 and the second slip ratio threshold N2. Within this range, the calculated current road adhesion coefficient may be slightly affected by load transfer, or it may exhibit unstable characteristics due to approaching the saturation region. To avoid frequent jumps in the adhesion coefficient output value due to small fluctuations in the input signal within this critical region, a hold strategy is adopted, that is, the road adhesion coefficient output at the previous moment is directly used as the original value of the output road adhesion coefficient at the current moment.

[0078] That is, N1<| | <N2: = .

[0079] A holding strategy is adopted for the intermediate slip zone. The reason is that this zone is a transitional state between low excitation and overslip, and the raw value is sensitive to torque fluctuations and wheel-end vertical load errors. Maintaining the output from the previous moment helps reduce frequent jumps in the critical region. The holding scheme disclosed in this step essentially establishes a hysteresis characteristic in the intermediate transition zone, shielding the jitter of the original calculated value within this zone, so that the output of the adhesion coefficient remains stable when the slip ratio crosses this zone. By maintaining the output state from the previous moment, the temporal inertia of the adhesion coefficient estimation is achieved, effectively reducing the false trigger rate of the system under critical conditions and improving the stability of the recognition results.

[0080] (4) The original value of the output road surface adhesion coefficient is filtered to obtain the output road surface adhesion coefficient at the current time.

[0081] After generating the original output pavement adhesion coefficient based on the slip ratio range characteristics, high-frequency noise in the data is further eliminated through filtering to obtain a smooth final output pavement adhesion coefficient that meets engineering application requirements. The filtering process employs a first-order low-pass filter algorithm, which physically smooths the original value over time, suppressing rapid fluctuations caused by sensor sampling errors, wheel speed pulsations, or torque disturbances. The input to this filtering process is the original output pavement adhesion coefficient value generated in the preceding steps, and the output is the output pavement adhesion coefficient at the current moment. This filtering logic can be formally expressed as:

[0082] = +Kμ·[ - ].

[0083] In the formula, Indicates the first Each wheel at the current moment The filtered output road surface adhesion coefficient; This indicates that the wheel was at the previous moment. The output road surface adhesion coefficient; This represents the original, unfiltered output road surface adhesion coefficient obtained from the asymmetric update logic described above at the current moment; Kμ represents the first-order filter coefficient, with a value range of... This coefficient determines the time constant of the filter. The smaller Kμ is, the stronger the filtering smoothing effect, but the slower the response speed, and vice versa. For example, it can be set to a value between 0.1 and 0.5.

[0084] After filtering and calculation using the above formula, the estimated values ​​of the four-wheel adhesion coefficients are obtained, which are the output adhesion coefficients of the left front wheel. Right front wheel output adhesion coefficient Left rear wheel output adhesion coefficient Right rear wheel output adhesion coefficient Depending on actual needs, strategies such as rate of change limiting, invalid signal holding, and power-on initial value setting can be superimposed to improve strategy stability.

[0085] By introducing filtering, the original values ​​generated by asymmetric updates are transformed into continuously changing, gently trending adhesion coefficient curves, thus providing a high-quality data foundation for subsequent chessboard road surface recognition and split road surface judgment.

[0086] In summary, this invention employs a strategy of maximizing the slip ratio in the low slip ratio range to prevent underestimation of the adhesion coefficient, a strategy of minimizing the slip ratio in the high slip ratio range to suppress artificially high adhesion coefficients, and a strategy of maintaining the slip ratio in the intermediate slip ratio range to avoid critical jitter. Finally, the output is smoothed through filtering. This differentiated processing strategy based on slip ratio states enables the estimation results of the road adhesion coefficient to balance dynamic response speed and steady-state accuracy, effectively solving the problem that the single update logic in existing technologies cannot adapt to changes in slip state under all working conditions, and significantly improving the vehicle's state recognition capability under different road adhesion conditions.

[0087] In one embodiment, the process of calculating the current road adhesion coefficient and slip ratio of each wheel based on operating state parameters is as follows:

[0088] The current road adhesion coefficient of each wheel is determined based on the wheel-end driving torque, effective tire radius, and wheel-end vertical load at the current moment.

[0089] Specifically, This reflects the degree to which the longitudinal force of the tire utilizes the vertical load. For each wheel, based on the wheel-end drive torque, effective tire radius, and wheel-end vertical load, the road adhesion coefficient raw value at the current time k is calculated, which is also the wheel's current road adhesion coefficient. ,as follows:

[0090] ;

[0091] In the formula, For wheel-end drive torque, For the effective radius of the tire and, This is the vertical load on the wheel end.

[0092] This indicates the current longitudinal drive at the wheel end. If compatibility between drive and braking is required, the longitudinal force direction at the wheel end can be considered separately, or the absolute value of the force can be taken and then combined with the operating condition indicator for judgment. For the drive-opening judgment that this embodiment focuses on, the current adhesion coefficient corresponding to the drive torque is preferentially used.

[0093] To avoid torque burrs, abnormal radius or Calculation errors cause output distortion. Set physical limit, for example ∈[0, ]. This is the calibrable peak adhesion coefficient of the road surface. When the input signal is invalid, Not participating in this update cycle.

[0094] The slip ratio of each wheel is determined based on the reference wheel speed, actual wheel speed, and preset low-speed protection threshold of each wheel.

[0095] The i-th wheel's position at the current time is calculated using the following formula. slip ratio :

[0096] ;

[0097] In the formula, For the first Each wheel at the current moment Reference wheel speed, For the first The current moment of each wheel The actual wheel speed, The preset low-speed protection threshold can be 0.5 m / s to 2.0 m / s. This represents the function that takes the maximum value.

[0098] It should be noted that when referring to wheel speed Too small or below the preset low-speed protection threshold In this case, you can choose to disable the slip ratio calculation for that wheel, or use the reference wheel speed. With low speed protection threshold The larger value in the formula is used as the denominator to prevent division by zero errors under low-speed conditions and to avoid amplifying wheel speed noise. When the wheel end is under vertical load... When the load is below the preset effective threshold for vertical load, the adhesion coefficient output of the wheel can remain at the value of the previous moment, or the state of the wheel can be set to invalid, so as to avoid abnormal amplification of the current road surface adhesion coefficient due to small load division operation.

[0099] Furthermore, in subsequent processing, the absolute value of the slip ratio is usually used. Interval judgment is performed to accommodate the differences in sign definition under both driving and braking conditions. The determined slip ratio provides accurate state variables for subsequent judgment of slip ratio intervals (low, medium, and high) and execution of asymmetric timing updates, which is a prerequisite for realizing differentiated processing of different slip regions.

[0100] In summary, in this embodiment, the wheel-end driving torque, effective tire radius, and wheel-end vertical load constitute the physical triangle for calculating the adhesion coefficient. These three parameters provide the driving force input, geometric transformation reference, and force-bearing support surface information, respectively, and are indispensable, collectively transforming rotational motion into translational adhesion characteristics. Meanwhile, the reference wheel speed, actual wheel speed, and low-speed protection threshold construct a robust slip ratio observation system. The intervention of the low-speed protection threshold changes the vulnerability of traditional slip ratio calculation in the low-speed region, ensuring numerical stability across the entire speed domain. The coordinated calculation of these two sets of parameters ensures that the road condition recognition system can obtain reliable feature inputs under different vehicle speeds, loads, and driving intensities.

[0101] In one embodiment, the chessboard road surface distribution state is identified based on the output road surface adhesion coefficient of each wheel, including at least one of the following judgment conditions:

[0102] Judgment condition 1: The difference in road surface adhesion coefficient output by the left and right sides of the front axle has the opposite sign to the difference in road surface adhesion coefficient output by the left and right sides of the rear axle, and the absolute values ​​of both are greater than the preset chessboard threshold.

[0103] Specifically, if the difference in road surface adhesion coefficient between the left and right sides of the front axle is ( The difference between the road surface adhesion coefficient output on the left and right sides of the rear axle () If the signs are opposite and both absolute values ​​are greater than the preset chessboard threshold, then it is determined that it has chessboard features.

[0104] This embodiment calculates the difference between the road surface adhesion coefficients of the left and right front wheels of the front axle, based on the obtained coefficients. This difference represents the degree of difference in road surface adhesion between the left and right wheels of the front axle, and its sign reflects the relative level of adhesion on both sides: a positive sign indicates the left side is higher than the right, and a negative sign indicates the right side is higher than the left. Similarly, based on the road surface adhesion coefficients of the left and right rear wheels of the rear axle, the difference in road surface adhesion coefficients between the left and right rear wheels is calculated to represent the difference in road surface adhesion between the left and right wheels of the rear axle.

[0105] The above judgment logic aims to capture the contradictory phenomenon of left-right adhesion differences between the front and rear axles of a vehicle. This is a typical characteristic that distinguishes checkerboard road surfaces from stable split-type road surfaces. On stable split-type road surfaces, such as a fully high-adhesion left-side and fully low-adhesion right-side surface, the left-right adhesion differences between the front and rear axles usually have the same sign, both pointing to the adhesion advantage on the same side. However, on checkerboard road surfaces, the left-right adhesion differences between the front and rear axles are in opposite directions, resulting in opposite signs for the two differences. When the signs of the difference in road surface adhesion coefficients output from the left and right sides of the front axle are identified as opposite to those output from the left and right sides of the rear axle, it indicates that the front and rear axles of the vehicle are in opposite adhesion tilt states, exhibiting the potential characteristics of a checkerboard road surface.

[0106] To eliminate misjudgments caused by measurement noise or minor road surface unevenness, a preset checkerboard threshold is introduced as a judgment threshold. The preset checkerboard threshold is a calibrated value set to filter out minor adhesion fluctuations. Its value can be configured according to the vehicle's sensor accuracy, tire characteristics, and the sensitivity requirements of the control system, for example, set to a value between 0.1 and 0.3. Only when the absolute values ​​of the differences in road surface adhesion coefficients output from the left and right sides of the front axle and the left and right sides of the rear axle are both greater than the preset checkerboard threshold is the contradictory state determined to be engineering significant, thus confirming it as a checkerboard road surface distribution. Through this dual judgment mechanism based on front and rear axle sign comparison and amplitude filtering, it is possible to effectively distinguish between spatially alternating checkerboard road surfaces and spatially stable, separated open road surfaces.

[0107] Judgment condition two: The difference between the sum of the output road surface adhesion coefficients of the diagonal wheel combinations is greater than the difference between the sum of the output road surface adhesion coefficients of the left and right wheel combinations.

[0108] Specifically, if the difference between diagonal combinations |( + )-( + )| Greater than the difference between left and right combinations| ( + )-( + If |, then it is determined to have chessboard characteristics.

[0109] A diagonal wheel combination refers to a set of two wheels located diagonally opposite each other in the vehicle chassis structure. Specifically, it includes a first diagonal combination consisting of the left front wheel and the right rear wheel, and a second diagonal combination consisting of the right front wheel and the left rear wheel. The sum of the output road surface adhesion coefficients of the left front wheel and the right rear wheel in the first diagonal combination, and the sum of the output road surface adhesion coefficients of the right front wheel and the left rear wheel in the second diagonal combination, are calculated separately. The difference between the sums of the output road surface adhesion coefficients of the diagonal wheel combinations is the absolute value of the difference between the calculated results of the two diagonal combinations. This value quantifies the degree of unevenness in the adhesion distribution along the diagonal direction.

[0110] The left and right wheel combinations refer to the set of two wheels on the same side of the vehicle's chassis structure. Specifically, this includes the left combination consisting of the left front wheel and the left rear wheel, and the right combination consisting of the right front wheel and the right rear wheel. The sum of the output road surface adhesion coefficients of the left front and left rear wheels in the left combination, and the sum of the output road surface adhesion coefficients of the right front and right rear wheels in the right combination, are calculated separately. The difference between the sums of the output road surface adhesion coefficients of the left and right wheel combinations is the absolute value of the difference between the calculated results for the two left and right combinations. This value quantifies the degree of adhesion separation in the left and right directions of the vehicle.

[0111] This step utilizes the geometric characteristic of a chessboard-like surface where the adhesion difference along the diagonal is typically greater than the adhesion difference along the left and right sides. In a chessboard-like surface distribution, the sum of the wheel adhesion coefficients along one diagonal is significantly higher than that along the other diagonal; this difference amplifies the spatial alternation characteristics. For surfaces with separate adhesion along the left and right sides, the difference in the sum of the adhesion coefficients on both sides is larger, while the adhesion distribution along the diagonal is relatively balanced with a smaller difference. Therefore, when the difference in the sum of the output surface adhesion coefficients of the diagonal wheel combinations is greater than the difference in the sum of the output surface adhesion coefficients of the left and right wheel combinations, it indicates that the unevenness of the surface adhesion distribution along the diagonal dimension exceeds that along the left and right dimensions; this is a typical mathematical characteristic of a chessboard-like surface. This comparative judgment allows for further confirmation of the chessboard-like surface state from the perspective of spatial geometric distribution, avoiding potential omissions that might result from relying solely on a single axial difference.

[0112] Judgment condition 3: The difference in the output road surface adhesion coefficient of the front and rear wheels on the same side exceeds the preset alternation threshold, and the difference in the average output road surface adhesion coefficient between the left and right sides is less than the preset opening threshold.

[0113] Specifically, if the difference in road surface adhesion coefficient between the front and rear wheels on the same side is | - | and | - If all values ​​exceed the preset threshold and the difference in the average value of the road surface adhesion coefficients on the left and right sides is insufficient to form a stable opposite direction, then it is determined that it has chessboard or alternating adhesion characteristics.

[0114] When the chessboard surface is established, the percentage of the open surface is set to zero, or the open surface output from the previous moment is frozen, and a chessboard surface flag is output. This prevents subsequent controllers from incorrectly compensating according to the direction of the stable open surface.

[0115] The front and rear wheels on the same side refer to the front and rear wheels located on the same side of the vehicle, i.e., the front and rear wheel combination on the left and the front and rear wheel combination on the right. The absolute value of the difference in the output road surface adhesion coefficient between the left front wheel and the rear wheel, and the absolute value of the difference in the output road surface adhesion coefficient between the right front wheel and the rear wheel, are calculated to obtain the difference in output road surface adhesion coefficient between the front and rear wheels on the same side. This index reflects the consistency of adhesion of the wheels on the same side in the longitudinal direction. In checkerboard road conditions, the front and rear wheels on the same side of the vehicle often contact road surfaces with different adhesion characteristics; for example, the left front wheel contacts a high-adhesion surface, and the left rear wheel contacts a low-adhesion surface. Therefore, the difference in output road surface adhesion coefficient between the front and rear wheels on the same side will be relatively large.

[0116] The preset alternation threshold is a threshold parameter used to determine whether the difference in adhesion between the front and rear wheels on the same side reaches the characteristics of an alternating road surface. Its setting is based on the typical fluctuation range of the adhesion coefficient exhibited by the front and rear wheels when a vehicle is driving on an alternating adhesion road surface. This threshold can be obtained through statistical analysis of historical data under a large number of checkerboard road surface conditions, or determined based on vehicle dynamics simulations; for example, it can be set to values ​​such as 0.2 or 0.4. When the difference in adhesion between the front and rear wheels on the left and right sides both exceed this preset alternation threshold, it indicates that there is a significant longitudinal adhesion separation phenomenon on both sides of the vehicle, which highly matches the alternating distribution pattern of a checkerboard road surface.

[0117] The difference in the average road surface adhesion coefficient between the left and right sides can be the absolute value of the difference between the average road surface adhesion coefficient of all wheels on the left and the average road surface adhesion coefficient of all wheels on the right. It is used to measure whether there is a significant trend of left-right adhesion separation on the vehicle as a whole. Under stable left-right adhesion separation road surfaces, this average difference is usually large; while under checkerboard road surfaces, because the adhesion coefficients of the four wheels are relatively evenly distributed on the left and right sides, this average difference is often small. A preset split threshold is used to define whether a stable left-right adhesion separation road surface has been formed. When the difference in the average road surface adhesion coefficient between the left and right sides is less than the preset split threshold, it indicates that the vehicle as a whole has not formed a stable unidirectional adhesion advantage distribution. By performing a logical AND operation on the two conditions—a large difference in the adhesion coefficient between the front and rear wheels on the same side and a small difference in the average difference between the left and right sides—it is possible to accurately capture checkerboard road surface characteristics that are generally balanced on the left and right sides but exhibit significant alternation between the front and rear wheels in certain areas, while effectively excluding left-right adhesion separation road surface conditions where the overall left and right sides are separated but the front and rear wheels on the opposite side are consistent.

[0118] In summary, the embodiments of this application can comprehensively identify the chessboard road surface distribution state from multiple dimensions, such as the contradiction between the front and rear axle signs, the comparison of diagonal and left-right side differences, and the joint verification of same-side alternation and overall stability. This multi-dimensional identification logic significantly improves the system's ability to perceive complex road surface distribution patterns, reduces the probability of misjudgment caused by interference from single-dimensional signals, and provides a reliable logical basis for subsequent suppression of erroneous left-right side attachment separation state outputs.

[0119] In one embodiment, the vehicle road surface condition recognition method may further include:

[0120] In response to the failure to identify the chessboard road surface distribution state, the normalized values ​​of the difference in road surface adhesion coefficient between the left and right sides and the difference in torque between the left and right wheel ends are calculated.

[0121] The normalized values ​​of the difference in road surface adhesion coefficient between the left and right sides and the difference in torque at the left and right wheel ends are normalized and then weighted and fused to obtain the percentage of split road surface.

[0122] The normalized value of the torque difference between the left and right wheel ends is calculated by dividing the difference between the sum of the driving torques at the left wheel end and the sum of the driving torques at the right wheel end by the maximum value among the sum of the driving torques at the left wheel end, the sum of the driving torques at the right wheel end, and the normalized protection threshold.

[0123] Furthermore, the difference in road surface adhesion coefficients between the left and right sides was calculated separately. Normalized value of torque difference between left and right wheel ends Then, the two were normalized separately and weighted and fused to obtain the percentage of the road surface in both directions:

[0124] Original value of split road surface: SplitRaw = Wμ·norm( )+WT·norm( );

[0125] SplitPct = saturate(100) SplitRaw, 0, 100).

[0126] Where Wμ and WT are calibration weights, satisfying Wμ+WT=1; norm() is a normalization function used to calculate the difference in road surface adhesion coefficients between the left and right sides. Normalized value of torque difference between left and right wheel ends Mapping to the same numerical range eliminates dimensional differences, making it weighted and comparable; saturate() is a limiting function used to restrict the weighted fusion result to the range of 0 to 100, ensuring that the output percentage of the split road surface does not exceed the effective range, thereby avoiding control misjudgment caused by numerical anomalies.

[0127] When determining the split direction, priority should be given to the direction with the difference in adhesion coefficients between the left and right sides. Decide. Direction of torque difference between left and right wheel ends In the event of a conflict, the previous direction is maintained and timing begins. If the conflict time exceeds the calibrated maximum confirmation time, the SplitPct unreliable flag is output. By setting direction consistency verification logic, the misjudgment of split direction caused by a single attachment estimation error or abnormal torque response is further reduced, improving the reliability of the split road surface percentage.

[0128] When it is determined that the vehicle is not in a checkerboard road surface distribution state, it indicates that the road surface state of the vehicle tends to be a stable single adhesion or split adhesion. At this time, the quantitative calculation of the split road surface degree is initiated. By comparing the adhesion capabilities of the left and right wheels after asymmetric time-series updates, the difference in the road surface adhesion coefficient between the left and right sides is calculated. This difference is used to quantify the intrinsic difference of road surface adhesion in the lateral direction, directly reflecting the degree of unevenness of the road surface friction coefficient between the left and right sides. At the same time, based on the actual dynamic response of the vehicle, the normalized value of the torque difference between the left and right wheel ends is calculated by comparing the actual driving torque borne by the two wheels, so as to quantify the degree of torque separation of the drive system in the left and right transmission paths. This feature serves as the second judgment criterion for split road surface identification, and can provide redundant verification based on the actual control response of the vehicle when there is a deviation in the adhesion coefficient estimation, thereby improving the robustness of the identification.

[0129] This embodiment maps the normalized values ​​of the difference in road surface adhesion coefficients and the difference in wheel-end torque between the left and right sides to the same numerical range, effectively reducing the impact of differences in units and numerical ranges on the fusion results and making them weighted and comparable. Subsequently, a weighted fusion algorithm is used to synthesize the processed feature values, where the weighting coefficients can be calibrated according to the actual vehicle type, sensor accuracy, and control strategy requirements. The weighted fusion result is then subjected to amplitude limiting or proportional mapping to finally generate the percentage of the split road surface. This percentage represents the degree of left and right road surface adhesion separation in the form of continuous numerical values ​​(e.g., 0% to 100%). The larger the value, the more severe the separation, thus providing the vehicle controller with a graded and quantitative decision-making basis, avoiding abrupt changes in control strategy caused by simply outputting Boolean states.

[0130] In this embodiment, the sum of the driving torques at the left and right wheel ends includes the sum of the driving torques at the left wheel end and the sum of the driving torques at the right wheel end.

[0131] Sum of drive torques at the left wheel end The calculation formula is as follows:

[0132] =| |+| |;

[0133] In the formula, This refers to the drive torque at the left front wheel end. This is the driving torque at the left rear wheel end.

[0134] Sum of drive torques at the right wheel end The calculation formula is as follows:

[0135] =| |+| |;

[0136] In the formula, This refers to the drive torque at the right front wheel end. Right rear wheel drive torque.

[0137] After obtaining the sum of the drive torques at the left and right wheel ends, the torque difference between the left and right wheel ends is normalized. First, the difference between the sum of the drive torques at the left wheel end and the sum of the drive torques at the right wheel end is calculated, and then divided by the maximum value among the sum of the drive torques at the left wheel end, the sum of the drive torques at the right wheel end, and the normalization protection threshold, to obtain the original value of the normalized torque difference between the left and right wheel ends. Then, the absolute value of this original value is taken to obtain the final normalized torque difference between the left and right wheel ends.

[0138] Normalized original value of torque difference between left and right wheel ends The calculation formula is as follows:

[0139] ;

[0140] In the formula, This is a normalized protection threshold used to prevent division by zero errors under low torque conditions.

[0141] Normalized value of torque difference between left and right wheel ends =| Normalized value of torque difference between left and right wheel ends. It is used to reflect the actual degree of separation of torque between the left and right wheel ends during drive control or transmission. Compared with relying solely on the difference in road surface adhesion coefficient, torque difference can serve as another criterion, helping to reduce misjudgments caused by single adhesion estimation errors.

[0142] Furthermore, based on the normalized original value of the torque difference between the left and right wheel ends... The sign of the symbol determines the direction of the torque difference between the left and right wheel ends, i.e., the direction of the torque difference between the left and right wheel ends. =sign( ).

[0143] The torque at the left wheel end is greater than that at the right wheel end. =1, the torque at the right wheel end is greater than the torque at the left wheel end. =-1.

[0144] In summary, this embodiment constructs a multi-dimensional split-road identification logic by fusing features based on road adhesion estimation and features based on actual torque response. When there are errors in the adhesion coefficient estimation, the torque difference feature can provide effective correction and verification; when torque distribution is disturbed by the control strategy, the adhesion difference feature can still reflect the essential differences in the road surface. This synergistic mechanism effectively suppresses misjudgments caused by anomalies in a single signal source, ensuring the accuracy and reliability of the split-road percentage output.

[0145] In one embodiment, calculating the difference in adhesion coefficients between the left and right road surfaces includes using at least one of the following methods:

[0146] Method 1: In response to the mean output road surface adhesion coefficient of the left side being greater than or equal to the mean output road surface adhesion coefficient of the right side, the smaller value of the output road surface adhesion coefficient of the two left wheels is subtracted from the larger value of the output road surface adhesion coefficient of the two right wheels to obtain the difference in road surface adhesion coefficients between the left and right sides; in response to the mean output road surface adhesion coefficient of the right side being greater than the mean output road surface adhesion coefficient of the left side, the smaller value of the output road surface adhesion coefficient of the two right wheels is subtracted from the larger value of the output road surface adhesion coefficient of the two left wheels to obtain the difference in road surface adhesion coefficients between the left and right sides.

[0147] Method 2: Take half of the difference between the sum of the output road surface adhesion coefficients of the two left wheels and the sum of the output road surface adhesion coefficients of the two right wheels as the original value of the left and right side adhesion difference, and take the absolute value of the original value of the left and right side adhesion difference as the difference of the left and right side road surface adhesion coefficients.

[0148] Based on the obtained road surface adhesion coefficients output by each wheel, two optional calculation logics can be used to quantitatively evaluate the degree of separation between the adhesion levels on the left and right sides. In engineering applications, one calculation method can be selected through calibration parameters, or different calculation logics can be switched according to the vehicle's operating conditions.

[0149] In practical applications, the first method in this embodiment is the min / max robust method. This method requires that the lower μ value of the two wheels on the higher attachment side is still higher than the higher μ value of the two wheels on the lower attachment side, which can reduce misjudgments caused by outliers in a single wheel. This represents the difference in adhesion between the left and right sides calculated using method A.

[0150] If the left mean is greater than or equal to the right mean:

[0151] =max{0,min( , )-max( )}.

[0152] If the right-hand mean is greater than the left-hand mean:

[0153] =max{0,min( )-max( , )}.

[0154] In the formula, The coefficient of friction is the output of the left front wheel. The coefficient of friction is the output of the right front wheel. The coefficient of adhesion is the output of the left rear wheel. This is the adhesion coefficient output by the right rear wheel.

[0155] Simultaneously record the direction of the difference in adhesion coefficients between the left and right sides. When the adhesion coefficient is high on the left side and low on the right side, =1, when the adhesion coefficient is high on the right side and low on the left side. =-1.

[0156] Method two in this embodiment is the left-right summation / mean difference method. This method has high sensitivity and is suitable for calibration scenarios where it is desirable to identify left-right adhesion differences earlier.

[0157] in The difference in adhesion between the left and right sides was calculated using method B.

[0158] for The original value is obtained by taking the absolute value of the original value. ;

[0159] The direction of the difference in adhesion coefficients between the left and right sides, calculated using method B, is obtained by using the sign function (sign) on the original value.

[0160] Original value of left and right side adhesion difference =0.5·[( + )-( + )).

[0161] Left and right adhesion differences calculated using method two =| |

[0162] Direction of the difference in adhesion coefficients between the left and right sides =sign( ).

[0163] In different vehicle operating scenarios, you can choose either method one or method two to calculate the difference in the left and right adhesion coefficients, or use weighted calculation to calculate the difference in the left and right adhesion coefficients.

[0164] This embodiment allows for flexible adjustment based on the actual vehicle control strategy requirements by configuring the two calculation methods described above: in mass production control scenarios with high robustness requirements, method one is preferred to reduce false triggering; in development and debugging scenarios or scenarios with high early warning requirements, method two is preferred to capture weak signals. This configurable differentiated calculation logic improves the adaptability of the road condition recognition system in different application scenarios.

[0165] In one embodiment, the vehicle road surface condition recognition method further includes post-processing of the output road surface adhesion coefficient, wherein the post-processing includes at least one of the following:

[0166] The original value of the output road surface adhesion coefficient is subjected to a first-order low-pass filter to obtain the output road surface adhesion coefficient at the current moment.

[0167] The change in the output road surface adhesion coefficient at the current moment relative to the output road surface adhesion coefficient at the previous moment is limited so that the change is between the maximum allowable decrease in a single step and the maximum allowable increase in a single step.

[0168] The original output value of the road surface adhesion coefficient is intermediate data generated without smoothing after performing asymmetric time-series updates based on the wheel slip rate interval. While this original value reflects the road surface adhesion characteristics at the current instant, it is susceptible to high-frequency interference such as wheel speed sensor noise, road surface micro-irregularities, and vertical load fluctuations caused by vehicle suspension vibrations, leading to short-term data jitter. To effectively reduce these high-frequency interferences, a first-order low-pass filtering algorithm is used to smooth the original value over time. This filtering process uses a weighted fusion of the filtered output value from the previous time step and the original value from the current time step, assigning a certain inertial weight to recent historical data.

[0169] Although data jitter is suppressed after filtering, under certain extreme conditions, such as when the wheels pass over road joints with abrupt changes in adhesion coefficient, sensor signal jumps, or numerical divergence in the calculation model, the filtered output value may still exhibit significant jumps within a single control cycle. If the magnitude of such a jump exceeds the response bandwidth of the vehicle dynamics control system, it can easily lead to aggressive erroneous operations by downstream torque distribution or braking control modules, affecting the vehicle's ride comfort.

[0170] To address this issue, the embodiments may also perform one or more of the following post-processing steps on the output road surface adhesion coefficient:

[0171] This embodiment introduces a rate-of-change limitation method, setting a maximum allowable increase in the adhesion coefficient of a single wheel for each step length. and the maximum allowable decrease in the coefficient of adhesion per wheel for each step length The change in the single-wheel adhesion coefficient for each step length is Restriction relationship: This constraint ensures that the variation of the single-wheel adhesion coefficient within each step does not exceed the preset physical allowable range, preventing drastic fluctuations in the output value due to abnormal signals, thereby improving the stability and reliability of the system.

[0172] Invalid signal hold: When critical input signals such as vehicle speed signal, vertical load signal, or wheel-end drive torque signal are invalid, The current value is maintained and not updated in this cycle. This holding method ensures that the output value does not jump due to invalid input when the input signal is briefly lost or abnormal. It guarantees that the system can still output a reliable adhesion coefficient during signal abnormalities, avoiding control malfunctions caused by momentary signal loss and improving the system's fault tolerance and output stability.

[0173] Initial power-on setting: When the controller is initially powered on, output the adhesion coefficient to the left front wheel. Right front wheel output adhesion coefficient Left rear wheel output adhesion coefficient Right rear wheel output adhesion coefficient The initial value is assigned to the calibration value. , , By properly calibrating the initial values, we can ensure that the controller can quickly enter the normal working state after power-on, avoid system overshoot or response delay caused by improper initial value settings, and ensure the stability of the system output during the startup phase.

[0174] The above post-processing methods can be used individually or in combination to further enhance the robustness and engineering applicability of the output pavement adhesion coefficient.

[0175] In summary, this embodiment effectively suppresses high-frequency noise and irrational jumps in road surface adhesion coefficient estimation by using two post-processing methods: first-order low-pass filtering and change limiting. This results in an adhesion coefficient that can both track real road surface changes and exhibit good dynamic stability.

[0176] In one embodiment, before identifying the chessboard surface distribution state, an enabling condition judgment is performed, the enabling condition judgment including:

[0177] If the absolute value of the vehicle's lateral acceleration is greater than a preset lateral acceleration threshold, or the vehicle's required driving torque is less than a preset required torque threshold, or the vehicle speed is lower than a preset low speed threshold, the update of the left and right side attachment separation state is prohibited, and the output result of the previous moment is maintained.

[0178] Specifically, before identifying the chessboard surface distribution, it also includes performing enable condition checks:

[0179] Before proceeding with the split-road assessment, it is first determined whether the current operating condition has sufficient recognition reliability. If any of the following conditions are met, no new split-road assessment will be performed, or the split-road percentage from the previous moment will be maintained:

[0180] Condition 1: Absolute value of vehicle lateral acceleration | | Greater than the absolute value threshold of vehicle lateral acceleration At this point, lateral load transfer and tire lateral force account for a larger proportion, while longitudinal load transfer is less significant. The correlation between the surface adhesion and the road surface becomes weaker.

[0181] Condition 2: Vehicle demand drives torque Less than the threshold At this point, the longitudinal excitation of the tire is insufficient, and the confidence levels of the torque difference between the left and right wheel ends and the difference in road surface adhesion coefficient are low.

[0182] Condition 3: The vehicle speed is below the low-speed threshold, or the wheel speed, wheel-end torque, or wheel-end vertical load are below the threshold. Key input signals are invalid.

[0183] When any of the above conditions are met, the left and right side attachment separation states are not updated, and the output results of the previous time step are retained.

[0184] The absolute value of vehicle lateral acceleration is a key parameter for assessing the intensity of vehicle lateral motion. This physical quantity directly reflects the level of centrifugal force generated when the vehicle is cornering or performing obstacle avoidance maneuvers, and the resulting degree of lateral force and load transfer on the tires. When the absolute value of vehicle lateral acceleration exceeds a preset lateral acceleration threshold, it indicates that the vehicle is under intense lateral dynamic conditions. At this time, the main force characteristics of the tires shift from longitudinal adhesion to lateral adhesion, resulting in a weakening of the linear correspondence between the road adhesion coefficient derived from longitudinal driving force and the actual peak road adhesion capacity.

[0185] The preset lateral acceleration threshold is a threshold value determined based on the vehicle chassis stability control strategy and tire mechanical characteristics boundary. The determination process includes: analyzing the steady-state rotation characteristic curves of the vehicle under different road surface adhesion coefficients, extracting the value corresponding to the critical point of the lateral acceleration reaching the tire lateral force saturation zone; at the same time, combining the vehicle roll stability constraints and occupant comfort requirements, setting the maximum tolerable lateral acceleration upper limit, and comprehensively considering the margin brought by sensor measurement noise and signal filtering delay, finally determining the threshold. For example, it can be set to a value between 0.3g and 0.5g, and the specific value depends on the vehicle's target performance indicators.

[0186] Vehicle demand torque characterizes the driver's intent and the level of longitudinal driving excitation applied to the wheels by the vehicle's powertrain system, serving as the active input source for activating tire-road adhesion characteristic testing. When the vehicle demand torque is less than the preset demand torque threshold, it indicates that the vehicle is currently in a low-drive demand or coasting condition, and the tires are not receiving sufficient longitudinal force excitation. Under this weak excitation state, even if the road surface is in a high-adhesion state, the changes in wheel slip ratio and wheel-end reaction force are extremely insignificant, resulting in the calculated adhesion coefficient being mainly affected by system noise and zero drift, failing to truly reflect the road surface adhesion potential. The preset demand torque threshold is determined by: statistically analyzing the minimum driving torque required to induce tire slip ratio into the linear sensitive zone on roads with different adhesion coefficients; and combining the powertrain torque response accuracy and transmission efficiency loss to determine the minimum torque excitation level that can effectively distinguish differences in road surface adhesion; for example, it can be set to 5% to 10% of the engine's maximum torque, or 50 Nm to 100 Nm based on the actual wheel-end driving torque, to eliminate invalid or extremely low signal-to-noise ratio driving signals.

[0187] Vehicle speed is a fundamental condition determining the quality of wheel speed signals and the effectiveness of the dynamic model, directly affecting the numerical stability and physical meaning of the denominator in the slip ratio calculation formula. When the vehicle speed is below a preset low-speed threshold, the reference wheel speed value is extremely small. At this speed, quantization errors in the wheel speed sensor, installation deviations, and the non-rigid deformation effect of the tires during low-speed rolling can cause severe fluctuations or even distortions in the calculated slip ratio, thereby rendering the various adhesion coefficient estimation results based on the slip ratio unreliable. The preset low-speed threshold is determined comprehensively based on the measurement accuracy of the wheel speed sensor, low-speed protection logic, and vehicle start-up control strategy. For example, it can be set to 5 km / h to 10 km / h. Below this speed, the dynamic model is no longer applicable, and other strategies must be relied upon.

[0188] In summary, this embodiment constructs multi-dimensional enabling conditions to monitor the vehicle's current lateral motion state, longitudinal excitation intensity, and speed level in real time before identifying the chessboard-like road surface distribution. Once any condition that does not meet the recognition confidence level is detected, a protection mechanism is immediately triggered, prohibiting the updating of the left and right side attachment separation states. This process freezes the current output, keeping the previous output unchanged, thereby avoiding misjudgments caused by low signal-to-noise ratio or model mismatch under conditions such as strong lateral interference, weak drive excitation, or low speed. This ensures that the final road surface state result is updated only within a time window with high confidence, significantly improving the robustness and reliability of the vehicle road surface state recognition system.

[0189] In one embodiment, the present invention also provides the definition of the output signal and a subsequent control interface.

[0190] The module outputs include, but are not limited to, the following signals:

[0191] (1) Estimated values ​​of the adhesion coefficients of the four wheels: , , These correspond to the output road surface adhesion coefficients for the left front wheel, right front wheel, left rear wheel, and right rear wheel, respectively, and are used to characterize the adhesion level of each wheel on the road surface.

[0192] (2) Road surface type markings: including non-split road surface, road surface with high adhesion coefficient on the left and low adhesion coefficient on the right, road surface with high adhesion coefficient on the right and low adhesion coefficient on the left, checkerboard road surface, and prohibited. These markings are used to provide the vehicle controller with the classification results of the current road surface so that the controller can execute the corresponding control strategy according to different road surface types.

[0193] (3) SplitPct percentage: 0% indicates no effective split feature, and 100% indicates strong split feature. This percentage quantitatively describes the degree of separation of the left and right road surfaces in the form of continuous numerical values, which facilitates the downstream controller to perform graded control according to the degree of separation.

[0194] (4) Confidence level or validity indicator: This is obtained by combining vehicle speed, torque excitation, lateral acceleration, chessboard characteristics, and consistency of the opening direction. This indicator is used to indicate the confidence level of the current output. When the confidence level is low, the downstream controller can adopt a conservative strategy or ignore the current output.

[0195] The aforementioned output signals can be used for drive torque limiting, left-right torque distribution, brake slip prevention, yaw moment control, rear-wheel steering assist control, or vehicle stability control strategies, providing accurate road condition input to the vehicle chassis control system, thereby improving vehicle driving safety and handling stability under complex road conditions. Based on this, the present invention is particularly suitable for pure electric four-wheel drive vehicles and hybrid four-wheel drive vehicles, fully leveraging the advantage of rapid motor torque response to achieve high-precision real-time perception of road adhesion status.

[0196] Corresponding to the above method embodiments, the present invention also discloses a vehicle road surface condition recognition device.

[0197] See Figure 2 The present invention discloses a schematic diagram of a vehicle road surface condition recognition device, which includes:

[0198] The parameter acquisition unit 201 is used to acquire the operating status parameters of each wheel of the vehicle, and calculate the current road surface adhesion coefficient and slip ratio of each wheel based on the operating status parameters.

[0199] The adhesion coefficient update unit 202 is used to perform asymmetric time-series update on the current road surface adhesion coefficient according to the preset slip rate range in which the slip rate is located, so as to obtain the output road surface adhesion coefficient of each wheel.

[0200] The asymmetric time-series update includes: suppressing the decrease of the output road surface adhesion coefficient when the slip ratio is in the low slip ratio range, and suppressing the increase of the output road surface adhesion coefficient when the slip ratio is in the high slip ratio range.

[0201] The road surface recognition unit 203 is used to identify the chessboard road surface distribution state based on the output road surface adhesion coefficient of each of the wheels.

[0202] The output suppression unit 204 is used to suppress the output of the left and right side attachment separation state in response to the identification of the chessboard road surface distribution state.

[0203] In summary, this invention discloses a vehicle road surface condition recognition device. It acquires the operating state parameters of each wheel of the vehicle and calculates the current road surface adhesion coefficient and slip ratio of each wheel based on these parameters. Furthermore, it performs asymmetric temporal updates on the current road surface adhesion coefficient according to the preset slip ratio range. This suppresses the decrease in output road surface adhesion coefficient under weak excitation conditions to prevent underestimation of high-adhesion road surfaces, and suppresses the increase in output road surface adhesion coefficient under high-slip conditions to filter out excessive slippage. This yields the output road surface adhesion coefficient of each wheel, overcoming the problem of inaccurate adhesion coefficient estimation caused by insignificant changes in tire longitudinal force and slip ratio, wheel speed fluctuations, torque impacts, and load estimation errors. This provides a stable and reliable data foundation for subsequent road surface type recognition. Based on this, it identifies the chessboard road surface distribution state based on the output road surface adhesion coefficient of each wheel, and suppresses the output of left and right side adhesion separation states when the chessboard road surface distribution state is identified. This effectively distinguishes the chessboard road surface from stable left and right side adhesion separation roads, solving the problem of misjudgment of chessboard conditions caused by the limitations of a single feature index and avoiding directional errors in the overall vehicle control strategy.

[0204] In one embodiment, the adhesion coefficient update unit 202 can be specifically used for:

[0205] In response to the absolute value of the slip ratio being less than or equal to a first slip ratio threshold, it is determined that the slip ratio is in the low slip ratio range, and the larger value between the current road surface adhesion coefficient and the road surface adhesion coefficient output at the previous moment is taken as the original value of the unfiltered output road surface adhesion coefficient at the current moment.

[0206] In response to the absolute value of the slip ratio being greater than or equal to a second slip ratio threshold, it is determined that the slip ratio is in the high slip ratio range, and the smaller value between the current road surface adhesion coefficient and the road surface adhesion coefficient output at the previous moment is taken as the original value of the output road surface adhesion coefficient.

[0207] In response to the absolute value of the slip ratio being greater than the first slip ratio threshold and less than the second slip ratio threshold, it is determined that the slip ratio is in the intermediate slip ratio range, and the road surface adhesion coefficient output at the previous moment is used as the original value of the output road surface adhesion coefficient.

[0208] The original value of the output road surface adhesion coefficient is filtered to obtain the output road surface adhesion coefficient at the current time.

[0209] In one embodiment, the parameter acquisition unit 201 can be specifically used for:

[0210] The current road adhesion coefficient of each wheel is determined based on the wheel-end driving torque, effective tire radius, and wheel-end vertical load of each wheel at the current moment.

[0211] The slip ratio of each wheel is determined based on the reference wheel speed, actual wheel speed, and preset low-speed protection threshold of each wheel.

[0212] In one embodiment, the road surface recognition unit 203 can be specifically used for:

[0213] The difference in road surface adhesion coefficient output by the left and right sides of the front axle has the opposite sign to the difference in road surface adhesion coefficient output by the left and right sides of the rear axle, and the absolute values ​​of both are greater than the preset chessboard threshold.

[0214] The difference between the sum of the output road surface adhesion coefficients of the diagonal wheel combinations is greater than the difference between the sum of the output road surface adhesion coefficients of the left and right wheel combinations.

[0215] The difference in the output road surface adhesion coefficient of the front and rear wheels on the same side both exceed the preset alternation threshold, and the difference in the average output road surface adhesion coefficient between the left and right sides is less than the preset opening threshold.

[0216] In one embodiment, the vehicle road surface condition recognition device may further include:

[0217] The calculation unit is used to calculate the difference in road surface adhesion coefficient between the left and right sides and the normalized value of the difference in torque between the left and right wheel ends in response to the failure to identify the chessboard road surface distribution state;

[0218] The normalization processing unit is used to normalize the normalized values ​​of the difference in road surface adhesion coefficient between the left and right sides and the difference in torque between the left and right wheel ends, and then perform weighted fusion to obtain the percentage of split road surface.

[0219] The normalized value of the torque difference between the left and right wheel ends is calculated by dividing the difference between the sum of the driving torques at the left wheel end and the sum of the driving torques at the right wheel end by the maximum value among the sum of the driving torques at the left wheel end, the sum of the driving torques at the right wheel end, and the normalized protection threshold.

[0220] The calculation of the difference in adhesion coefficients between the left and right sides of the road surface includes at least one of the following methods:

[0221] Method 1: In response to the mean output road surface adhesion coefficient of the left side being greater than or equal to the mean output road surface adhesion coefficient of the right side, the smaller value of the output road surface adhesion coefficient of the two left wheels is subtracted from the larger value of the output road surface adhesion coefficient of the two right wheels to obtain the difference in road surface adhesion coefficients between the left and right sides; In response to the mean output road surface adhesion coefficient of the right side being greater than the mean output road surface adhesion coefficient of the left side, the smaller value of the output road surface adhesion coefficient of the two right wheels is subtracted from the larger value of the output road surface adhesion coefficient of the two left wheels to obtain the difference in road surface adhesion coefficients between the left and right sides.

[0222] Method 2: Take half of the difference between the sum of the output road surface adhesion coefficients of the two left wheels and the sum of the output road surface adhesion coefficients of the two right wheels as the original value of the left and right side adhesion difference, and take the absolute value of the original value of the left and right side adhesion difference as the difference of the left and right side road surface adhesion coefficients.

[0223] In one embodiment, the vehicle road surface condition recognition device further includes:

[0224] The post-processing unit is used to perform post-processing on the output road surface adhesion coefficient.

[0225] The post-processing unit is used to perform at least one of the following:

[0226] The original value of the output road surface adhesion coefficient is subjected to a first-order low-pass filter to obtain the output road surface adhesion coefficient at the current moment.

[0227] The change in the output road surface adhesion coefficient at the current moment relative to the output road surface adhesion coefficient at the previous moment is limited so that the change is between the maximum allowable decrease in a single step and the maximum allowable increase in a single step.

[0228] In one embodiment, the vehicle road surface condition recognition device may further include:

[0229] The enable condition judgment unit is used to perform enable condition judgment before identifying the chessboard surface distribution state.

[0230] The enable condition judgment unit can be specifically used for:

[0231] If the absolute value of the vehicle's lateral acceleration is greater than a preset lateral acceleration threshold, or the vehicle's required driving torque is less than a preset required torque threshold, or the vehicle speed is lower than a preset low speed threshold, the update of the left and right side attachment separation state is prohibited, and the output result of the previous moment is maintained.

[0232] It should be noted that for the specific working principles of each component in the device embodiment, please refer to the corresponding section of the method embodiment, which will not be repeated here.

[0233] Corresponding to the above embodiments, such as Figure 3 As shown, the present invention also provides a structural schematic diagram of a vehicle controller, which may include: a processor 1 and a memory 2;

[0234] The processor 1 and memory 2 communicate with each other via communication bus 3.

[0235] Processor 1, for executing at least one instruction;

[0236] Memory 2 is used to store at least one instruction;

[0237] Processor 1 may be a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement embodiments of the present invention.

[0238] Memory 2 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk storage device.

[0239] In this embodiment, processor 1 executes at least one instruction to implement the steps shown in the vehicle road condition recognition method.

[0240] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0241] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0242] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for vehicle road surface condition recognition, characterized in that, include: The operating status parameters of each wheel of the vehicle are obtained, and the current road adhesion coefficient and slip ratio of each wheel are calculated based on the operating status parameters. Based on the preset slip ratio range in which the slip ratio is located, an asymmetric time-series update is performed on the current road surface adhesion coefficient to obtain the output road surface adhesion coefficient of each wheel; wherein, the asymmetric time-series update includes: suppressing the decrease of the output road surface adhesion coefficient when the slip ratio is in the low slip ratio range, and suppressing the increase of the output road surface adhesion coefficient when the slip ratio is in the high slip ratio range. Based on the output road surface adhesion coefficient of each wheel, the chessboard road surface distribution state is identified; In response to the identification of the chessboard surface distribution state, the output of the left and right side attachment separation state is suppressed.

2. The vehicle road condition recognition method according to claim 1, characterized in that, The step of performing asymmetric temporal updates on the current road surface adhesion coefficient based on the preset slip ratio range to obtain the output road surface adhesion coefficient for each wheel includes: In response to the absolute value of the slip ratio being less than or equal to a first slip ratio threshold, it is determined that the slip ratio is in the low slip ratio range, and the larger value between the current road surface adhesion coefficient and the road surface adhesion coefficient output at the previous moment is taken as the original value of the unfiltered output road surface adhesion coefficient at the current moment. In response to the absolute value of the slip ratio being greater than or equal to a second slip ratio threshold, it is determined that the slip ratio is in the high slip ratio range, and the smaller value between the current road surface adhesion coefficient and the road surface adhesion coefficient output at the previous moment is taken as the original value of the output road surface adhesion coefficient. In response to the absolute value of the slip ratio being greater than the first slip ratio threshold and less than the second slip ratio threshold, it is determined that the slip ratio is in the intermediate slip ratio range, and the road surface adhesion coefficient output at the previous moment is used as the original value of the output road surface adhesion coefficient. The original value of the output road surface adhesion coefficient is filtered to obtain the output road surface adhesion coefficient at the current time.

3. The vehicle road surface condition recognition method according to claim 2, characterized in that, The calculation of the current road adhesion coefficient and slip ratio of each wheel based on the operating state parameters includes: The current road adhesion coefficient of each wheel is determined based on the wheel-end driving torque, effective tire radius, and wheel-end vertical load of each wheel at the current moment. The slip ratio of each wheel is determined based on the reference wheel speed, actual wheel speed, and preset low-speed protection threshold of each wheel.

4. The vehicle road condition recognition method according to claim 1, characterized in that, The identification of the chessboard road surface distribution state based on the output road surface adhesion coefficient of each of the wheels includes at least one of the following judgment conditions: The difference in road surface adhesion coefficient output by the left and right sides of the front axle has the opposite sign to the difference in road surface adhesion coefficient output by the left and right sides of the rear axle, and the absolute values ​​of both are greater than the preset chessboard threshold. The difference between the sum of the output road surface adhesion coefficients of the diagonal wheel combinations is greater than the difference between the sum of the output road surface adhesion coefficients of the left and right wheel combinations. The difference in the output road surface adhesion coefficient of the front and rear wheels on the same side both exceed the preset alternation threshold, and the difference in the average output road surface adhesion coefficient between the left and right sides is less than the preset opening threshold.

5. The vehicle road surface condition recognition method according to claim 4, characterized in that, Also includes: In response to the failure to identify the chessboard road surface distribution state, the normalized values ​​of the difference in road surface adhesion coefficient between the left and right sides and the difference in torque between the left and right wheel ends are calculated. The normalized values ​​of the difference in road surface adhesion coefficient between the left and right sides and the difference in torque at the left and right wheel ends are normalized and then weighted and fused to obtain the percentage of split road surface. The normalized value of the torque difference between the left and right wheel ends is calculated by dividing the difference between the sum of the driving torques at the left wheel end and the sum of the driving torques at the right wheel end by the maximum value among the sum of the driving torques at the left wheel end, the sum of the driving torques at the right wheel end, and the normalized protection threshold.

6. The vehicle road surface condition recognition method according to claim 5, characterized in that, The calculation of the difference in adhesion coefficients between the left and right sides of the road surface includes at least one of the following methods: Method 1: In response to the mean output road surface adhesion coefficient of the left side being greater than or equal to the mean output road surface adhesion coefficient of the right side, the smaller value of the output road surface adhesion coefficient of the two left wheels is subtracted from the larger value of the output road surface adhesion coefficient of the two right wheels to obtain the difference in road surface adhesion coefficients between the left and right sides; In response to the mean output road surface adhesion coefficient of the right side being greater than the mean output road surface adhesion coefficient of the left side, the smaller value of the output road surface adhesion coefficient of the two right wheels is subtracted from the larger value of the output road surface adhesion coefficient of the two left wheels to obtain the difference in road surface adhesion coefficients between the left and right sides. Method 2: Take half of the difference between the sum of the output road surface adhesion coefficients of the two left wheels and the sum of the output road surface adhesion coefficients of the two right wheels as the original value of the left and right side adhesion difference, and take the absolute value of the original value of the left and right side adhesion difference as the difference of the left and right side road surface adhesion coefficients.

7. The vehicle road condition recognition method according to claim 1, characterized in that, The method further includes post-processing the output road surface adhesion coefficient, the post-processing including at least one of the following: The original value of the output road surface adhesion coefficient is subjected to a first-order low-pass filter to obtain the output road surface adhesion coefficient at the current moment. The change in the output road surface adhesion coefficient at the current moment relative to the output road surface adhesion coefficient at the previous moment is limited so that the change is between the maximum allowable decrease in a single step and the maximum allowable increase in a single step.

8. The vehicle road surface condition recognition method according to claim 1, characterized in that, Before identifying the chessboard surface distribution state, the method further includes performing an enable condition judgment, which includes: If the absolute value of the vehicle's lateral acceleration is greater than a preset lateral acceleration threshold, or the vehicle's required driving torque is less than a preset required torque threshold, or the vehicle speed is lower than a preset low speed threshold, the update of the left and right side attachment separation state is prohibited, and the output result of the previous moment is maintained.

9. A vehicle road surface condition recognition device, characterized in that, include: The parameter acquisition unit is used to acquire the operating status parameters of each wheel of the vehicle, and calculate the current road surface adhesion coefficient and slip ratio of each wheel based on the operating status parameters. An adhesion coefficient update unit is used to perform asymmetric temporal update on the current road surface adhesion coefficient according to the preset slip ratio range in which the slip ratio is located, so as to obtain the output road surface adhesion coefficient of each wheel; wherein, the asymmetric temporal update includes: suppressing the decrease of the output road surface adhesion coefficient when the slip ratio is in the low slip ratio range, and suppressing the increase of the output road surface adhesion coefficient when the slip ratio is in the high slip ratio range. A road surface recognition unit is used to identify the chessboard road surface distribution state based on the output road surface adhesion coefficient of each of the wheels. The output suppression unit is used to suppress the output of the left and right side attachment separation state in response to the identification of the chessboard road surface distribution state.

10. A vehicle controller, characterized in that, The vehicle controller includes: a memory and a processor; The memory is used to store at least one instruction; The processor is used to execute the at least one instruction to implement the vehicle road surface condition recognition method as described in any one of claims 1 to 8.

Images