A road roughness level identification method, device, vehicle and storage medium

Abstract

This invention relates to the field of vehicle control technology and discloses a method, device, vehicle, and storage medium for identifying road surface roughness levels. The method includes: acquiring vertical acceleration sensing signals of each wheel during vehicle operation, wherein the vertical acceleration sensing signals are collected by unsprung mass vertical acceleration sensors; calculating road surface elevation information from the vertical acceleration sensing signals of each wheel based on a vehicle dynamics model; and calculating the road surface roughness level by fusing the road surface elevation information of each wheel. This invention improves the accuracy of road surface roughness level identification.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and specifically to a method, device, vehicle, and storage medium for identifying road surface roughness levels. Background Technology

[0002] With the rapid development of intelligent driving and vehicle comfort control technologies, real-time and accurate identification of road surface roughness has become crucial for improving vehicle adaptive suspension systems and ride comfort. Road surface roughness directly affects suspension damping adjustment, vehicle attitude control, and energy recovery efficiency. Currently, mainstream road surface roughness identification methods mainly rely on visual sensors, vehicle acceleration sensors, or road surface elevation estimation models. For example, cameras capture road surface texture images, and then image processing algorithms assess roughness; or vehicle sensors detect vibration characteristics to indirectly infer road conditions. However, visual systems are greatly affected by factors such as lighting, weather, and occlusion, resulting in low reliability; vehicle sensors are easily affected by vehicle vibration and suspension system hysteresis, making it impossible to accurately capture high-frequency road surface excitations. Therefore, there is an urgent need for a road surface roughness identification method with strong anti-interference capabilities and rapid response to support the precise optimization of intelligent suspension and vehicle comfort control systems. Summary of the Invention

[0003] This invention provides a method, apparatus, vehicle, and storage medium for identifying road surface roughness levels, in order to solve the problem of inaccurate road surface roughness level identification.

[0004] In a first aspect, the present invention provides a method for identifying road surface roughness level, the method comprising: during vehicle operation, acquiring vertical acceleration sensing signals of at least one wheel collected by an unsprung mass vertical acceleration sensor; acquiring road surface elevation information from the vertical acceleration sensing signals based on a vehicle dynamics model; and calculating the road surface roughness level using the road surface elevation information.

[0005] Secondly, the present invention provides a road surface roughness level identification device, the device comprising: a signal acquisition module for acquiring vertical acceleration sensing signals of at least one wheel obtained by an unsprung mass vertical acceleration sensor during vehicle operation; an elevation back-calculation module for acquiring road surface elevation information from the vertical acceleration sensing signals based on a vehicle dynamics model; and a level discrimination module for calculating the road surface roughness level by fusing the road surface elevation information of each wheel.

[0006] Thirdly, the present invention provides a vehicle comprising: a memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, and the processor executing the computer instructions to perform the method provided in the first aspect.

[0007] Fourthly, the present invention provides a computer-readable storage medium storing computer instructions, the computer instructions being used to cause a computer to perform any of the methods provided in the first aspect. The technical solution provided by this invention has the following advantages: The road surface roughness level identification method of this invention directly acquires the vertical acceleration sensing signal of the unsprung mass, capturing raw data directly related to road surface excitation. This avoids the problems of traditional visual sensors being affected by lighting and weather, and sprung mass sensors being easily affected by vehicle vibration, ensuring signal authenticity. It accurately calculates road surface elevation information by combining a vehicle dynamics model, and then integrates the elevation data of each wheel to calculate the roughness level, simplifying the data processing link, increasing computational efficiency, and enabling real-time identification to meet the timeliness requirements of scenarios such as intelligent suspension dynamic adjustment. This invention requires no additional expensive hardware, is compatible with existing vehicle sensor networks, is easy to integrate, applicable to various road conditions, has strong anti-interference capabilities, and can provide accurate and reliable road surface input support for vehicle comfort control and intelligent driving function optimization. Attached Figure Description

[0008] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0009] Figure 1 This is a flowchart illustrating a first embodiment of a road surface roughness level identification method according to an embodiment of the present invention; Figure 2 This is a flowchart illustrating a second embodiment of a road surface roughness level identification method according to an embodiment of the present invention; Figure 3 This is a flowchart illustrating a third embodiment of a road surface roughness level identification method according to an embodiment of the present invention; Figure 4 This is another schematic flowchart of a third embodiment of a road surface roughness level identification method according to an embodiment of the present invention; Figure 5 This is a flowchart illustrating a fourth embodiment of a road surface roughness level identification method according to an embodiment of the present invention; Figure 6 This is a schematic diagram of a road surface roughness level identification device according to an embodiment of the present invention; Figure 7 This is a schematic diagram of the hardware structure of a vehicle according to an embodiment of the present invention. Detailed Implementation

[0010] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, 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, 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.

[0011] It is understood that before using the technical solutions disclosed in the various embodiments of the present invention, users should be informed of the types, scope of use, and usage scenarios of the personal information involved in the present invention and their authorization should be obtained in accordance with relevant laws and regulations through appropriate means.

[0012] Exemplary embodiments of the invention will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the invention are shown in the drawings, it should be understood that the invention can be implemented in various forms and should not be limited to the embodiments set forth herein.

[0013] Figure 1 A flowchart of a first embodiment of a road surface roughness level identification method according to the present invention is shown, the method being performed by a vehicle. Figure 1 As shown, the method includes the following steps: Step S101: During vehicle operation, acquire the vertical acceleration sensing signal of at least one wheel obtained by the unsprung mass vertical acceleration sensor. Step S102: Obtain road surface elevation information from vertical acceleration sensing signals based on the vehicle dynamics model; Step S103: Calculate the road surface roughness grade using road surface elevation information.

[0014] Specifically, unsprung mass refers to the total number of components such as wheels, hubs, and tires that are in direct contact with the road surface and are not supported by suspension springs. Vertical acceleration sensing signal is an electrical signal that characterizes the acceleration of unsprung mass in the direction perpendicular to the road surface. Road surface elevation information refers to the vertical displacement data of the road surface relative to the reference plane. Road surface roughness grade is a grade label that characterizes the smoothness of the road surface based on the road surface elevation fluctuation characteristics (e.g., grade 1 is a smooth road, grade 2 is a medium rough road, and grade 3 is an extremely rough road).

[0015] In practice, once the vehicle is started and in driving mode, the vertical acceleration sensors installed on each wheel bracket (a key component of unsprung mass) collect vertical acceleration sensing signals in real time at a preset sampling frequency (e.g., 100Hz). For example, when the vehicle is driving on a paved road, the acceleration signal collected by the sensor fluctuates less, with the peak value usually within ±0.5m / s². However, when driving on a gravel road, the signal fluctuates violently, with the peak value reaching more than ±3m / s². The sensor transmits the collected raw electrical signal to the vehicle controller via the CAN bus.

[0016] The vehicle controller then calls the pre-stored vehicle dynamics model, substitutes the vertical acceleration sensor signals of each wheel into the state space equation of the model, and calculates the road surface elevation information through integral calculation and dynamic back-calculation logic. For example, the left front wheel sensor collects a continuous vertical acceleration of 0.3 m / s², and after model calculation, the corresponding road surface elevation displacement is calculated to be approximately ±0.2 m. The right rear wheel collects an instantaneous acceleration of 1.8 m / s² due to the potholes, and the elevation displacement is calculated to be approximately ±1.5 m.

[0017] Finally, the controller calculates the road surface roughness level by combining the road surface elevation information of at least one wheel with the preset level classification rules.

[0018] For example, the controller finally fuses the road surface elevation information of the four wheels. By statistically analyzing the peak value, root mean square, and other characteristics of the elevation data of each wheel, and combining them with the preset classification rules, the road surface roughness level is calculated. For example, if the elevation displacements of the four wheels are ±0.2m, ±0.3m, ±0.25m, and ±0.35m respectively, all within the threshold range of level 1, the road surface roughness level is determined to be 1. If the elevation displacements of two wheels are ±1.6m and ±1.8m, and the other two are ±0.4m and ±0.5m, the combined fusion is determined to be level 2.

[0019] This invention directly acquires the vertical acceleration sensing signal of the unsprung mass, fully utilizing the characteristic of direct contact between the unsprung mass and the road surface. The acquired signal contains the most direct road excitation information, avoiding interference from vehicle vibration and suspension hysteresis response, which is common with traditional sprung mass sensors. It also eliminates the dependence of visual sensors on external environments such as lighting, weather, and occlusion, significantly improving the authenticity and purity of the signal and providing a reliable data foundation for subsequent identification. The elevation inversion is performed using a vehicle dynamics model. The model structure is simple and the computational logic is clear, eliminating the need for complex multi-sensor fusion algorithms and greatly reducing the computational load of data processing. This ensures a rapid response from signal acquisition to elevation calculation, meeting the needs of real-time identification during vehicle operation and providing timely road input for adaptive suspension adjustment and vehicle attitude control. The fusion of road elevation information from all four wheels for grade calculation comprehensively reflects the overall road condition of the entire vehicle, avoiding misjudgments caused by local road anomalies in a single wheel, and improving the accuracy and stability of road roughness grade identification. Finally, this invention does not require additional expensive hardware; it can directly utilize the vertical acceleration sensor and onboard controller in the vehicle's existing chassis sensor network. System integration is simple, reducing the cost and difficulty of technology implementation, and it is applicable to various vehicle types, including sedans, SUVs, and commercial vehicles. Based on the road roughness level determined by this invention, the results are accurate and real-time, providing crucial references for speed planning and energy recovery strategy optimization in intelligent driving. Simultaneously, it provides precise road condition input to the vehicle comfort control system, effectively mitigating the impact of uneven road surfaces on the driving experience and improving vehicle ride comfort and driving safety through functions such as suspension damping adjustment.

[0020] In some optional implementations, the vehicle dynamics model is a state-space equation of a quarter-vehicle model, and step S102 above includes: Step d1: Based on the coupling relationship between the vertical forces on the sprung mass and the unsprung mass, determine the mapping relationship between vertical acceleration and road surface elevation. Step d2: Use the mapping relationship to determine the road surface elevation information corresponding to the vertical acceleration sensing signal.

[0021] Specifically, the quarter-vehicle model focuses on the vertical vibration characteristics of the wheels, simplifying the entire vehicle into a dynamic coupling system of sprung mass, unsprung mass, suspension system, and tire corresponding to a single wheel. In this model, the net vertical force of the sprung mass is determined by the suspension elastic restoring force and damping damping force, while the net vertical force of the unsprung mass is the resultant force of the suspension elastic force, damping force, and tire elastic reaction force. The vertical forces of these two components are coupled and mutually influential. Based on this coupling relationship, dynamic differential equations characterizing the vertical motion of the sprung and unsprung masses are constructed using Newton's second law. By performing state-space modeling and solving these differential equations, the mathematical correspondence between the vertical acceleration of the unsprung mass and the road surface elevation caused by road excitation is derived, i.e., the mapping relationship between vertical acceleration and road surface elevation is obtained. This mapping relationship directly reflects the quantitative correlation between the vertical acceleration sensing signal and the road surface elevation information, providing a precise mathematical basis for subsequent elevation back-calculation.

[0022] Subsequently, the pre-processed unsprung mass vertical acceleration sensing signal from the vehicle controller is substituted into the determined mapping relationship between vertical acceleration and road surface elevation. The road surface elevation information corresponding to the sensing signal is directly calculated through numerical calculation. After completing the preliminary mapping calculation, the obtained road surface elevation information can be smoothed to remove residual minor fluctuations and accidental interference in the data, further ensuring the accuracy of the road surface elevation information and providing reliable basic data for subsequent determination of road surface roughness level.

[0023] This invention establishes a direct mapping relationship between vertical acceleration and road surface elevation through the state-space equations of a quarter-vehicle model. The model structure is simple and the computational logic is clear, which greatly reduces the amount of computation in data processing. It can quickly complete the conversion from vertical acceleration sensing signals to road surface elevation information, meeting the need for real-time identification of road surface roughness level during vehicle operation. At the same time, this mapping relationship is constructed based on the force coupling relationship between sprung and unsprung masses, which conforms to the dynamic characteristics of actual vehicle operation and effectively ensures the accuracy of road surface elevation information back-calculation.

[0024] In some optional implementations, step d1 above includes: Step a1: Based on the fact that the net vertical force on the sprung mass is equal to the resultant force of the elastic restoring force and the damping force of the sprung mass in vertical motion, a first mapping relationship is generated. The elastic restoring force is determined based on the vertical displacement difference between the sprung mass and the unsprung mass, and the damping force is determined based on the vertical velocity difference between the sprung mass and the unsprung mass. The vertical displacement difference and the vertical velocity difference are calculated through the vertical acceleration sensing signal. Step a2: Based on the fact that the net vertical force on the unsprung mass is equal to the resultant force of the elastic force, damping force and elastic reaction force of the unsprung mass in vertical motion, a second mapping relationship is generated. The elastic force is determined based on the vertical displacement difference between the sprung mass and the unsprung mass, the damping force is determined based on the vertical velocity difference between the sprung mass and the unsprung mass, and the elastic reaction force of the tire is determined based on the vertical displacement difference between the unsprung mass and the road surface excitation. The vertical displacement difference and the vertical velocity difference are calculated through the vertical acceleration sensing signal. Step a3: Merge the first and second mapping relationships to obtain the mapping relationship between vertical acceleration and road surface elevation.

[0025] Specifically, in this embodiment, the vertical acceleration sensing signal of the current wheel is input into the mapping relationship between the vertical acceleration and road surface elevation of the quarter-vehicle model, and the current road surface elevation information corresponding to the current wheel is calculated. The mapping relationship between the vertical acceleration and road surface elevation of the quarter-vehicle model is as follows: First mapping relationship:

[0026] Second mapping relationship:

[0027] In the formula, It's the car body quality. It is the vertical displacement of the sprung mass. It is the unsprung mass. It is the vertical displacement of the unsprung mass. It is the current road surface elevation information of the road surface excitation. It refers to the spring stiffness of the suspension. It is the damping coefficient of the suspension shock absorbers. It refers to the stiffness of the tire.

[0028] Specifically, the mapping relationship between the vertical acceleration and road surface elevation of the quarter-vehicle model simplifies the entire vehicle into a single-degree-of-freedom system consisting of a single wheel, suspension, and body. Focusing only on vertical vibration characteristics, it can accurately reflect the dynamic relationship between unsprung mass, sprung mass, and road excitation.

[0029] The vehicle body mass is the total mass of the sprung components (including the vehicle body, occupants, and load), which is preset and stored in the controller based on the vehicle's factory parameters. The vertical displacement of the sprung mass... It refers to the vertical position of the vehicle body relative to a reference plane, and the unsprung mass. The unsprung mass is the total mass of components such as wheels, hubs, and tires that are in direct contact with the road surface and are not supported by suspension springs, and its vertical displacement. This refers to the vertical position of the unsprung portion relative to the reference plane, both of which are obtained in real-time through vehicle dynamics model calculations; the spring stiffness of the suspension. Damping coefficient of vibration damper and tire stiffness All of these are based on the design parameters of the vehicle's suspension system and tires, and can be dynamically corrected according to changes in the vehicle's load.

[0030] Specifically, the vehicle controller first receives the pre-processed current wheel vertical acceleration sensor signal (i.e., unsprung mass vertical acceleration). This signal, having been filtered to remove high-frequency noise and pulse interference, accurately reflects the effect of road surface excitation on the unsprung mass. Subsequently, the controller calls the pre-stored state-space equations of a quarter of the vehicle model, combining the remaining known parameters with the real-time acquired data. Substituting into the equation, the second-order differential equation is solved using a numerical integration algorithm (such as the Runge-Kutta algorithm), first based on... Integrating yields the vertical velocity of the unsprung mass. Then, by integration, the vertical displacement of the unsprung mass is obtained. Simultaneously, by combining the dynamic coupling relationship between the sprung mass and the unsprung mass in the equation, the vertical acceleration of the sprung mass can be calculated. Vertical velocity and vertical displacement Finally, the current road surface elevation information of the road surface excitation is derived by reverse derivation through the second mapping relationship. ,Right now

[0031] This invention establishes a precise mapping relationship between vertical acceleration and road surface elevation using the state-space equations of a quarter-vehicle model. The model has a simple structure and low computational load, enabling rapid completion of the solution process from signal input to elevation output, thus meeting real-time recognition requirements. At the same time, the model parameters can be dynamically corrected according to the actual vehicle configuration and operating conditions, ensuring the accuracy of elevation back-calculation under different vehicle types and load conditions, and providing reliable basic data support for the subsequent accurate determination of road surface roughness level.

[0032] Finally, the road surface elevation information of each wheel is smoothed to remove residual minor fluctuations and random interference in the elevation data. In a specific embodiment, a moving average algorithm is used. For example, the sliding window length is set to 10 sampling points (sampling frequency 100Hz, corresponding to a window time of 100ms). The road surface elevation information of the left front wheel is 0.04m, 0.05m, 0.03m, 0.06m, 0.04m and other continuous data. After moving average processing, the output data is smoothed to 0.044m, 0.046m, 0.042m, etc., which not only preserves the true elevation characteristics of the road surface, but also suppresses irrelevant interference.

[0033] Figure 2A flowchart of a second embodiment of a road surface roughness level identification method according to the present invention is shown, the method being performed by a vehicle. Figure 2 As shown, the method includes the following steps: Step S201: During vehicle operation, acquire the vertical acceleration sensing signal of at least one wheel obtained by the unsprung mass vertical acceleration sensor. Step S201 includes: Step S2011: Collect vehicle status signals during vehicle operation, including vehicle speed signals. Step S2012: Determine whether the road surface roughness level identification conditions are met by using the vehicle status signal; Step S2013: When the road surface roughness level identification condition is met, receive the vertical acceleration sensing signal of each wheel. Step S2014: Preprocess the vertical acceleration sensing signal; Step S2015: Record the preprocessed vertical acceleration sensing signal.

[0034] Specifically, during vehicle operation, vehicle status signals are first collected through an onboard sensor network. The core of this data includes vehicle speed signals, but includes, but is not limited to, vehicle speed, vehicle load, and vehicle pitch angle signals. The vehicle status signals are a set of parameters reflecting the real-time operating conditions of the vehicle. The vehicle speed signal is calculated by the wheel speed sensors using an algorithm, and the acquisition frequency is synchronized with the subsequent vertical acceleration sensor signal at 100Hz to ensure time matching between the operating conditions and the signals. For example, when the vehicle is driving on urban roads, the vehicle speed signal provides real-time feedback of a stable driving state of 30km / h.

[0035] The system then uses vehicle status signals to determine whether the road surface roughness level recognition conditions are met. These conditions are preset thresholds to avoid invalid calculations. The core of the system is based on vehicle speed signals, with a preset speed threshold range of 10 km / h to 120 km / h. When the vehicle speed is within this range and remains stable for 100 ms without drastic fluctuations, the conditions are considered met. For example, if the vehicle speed is 50 km / h and remains stable, the recognition requirements are met. If the vehicle speed is below 10 km / h or above 120 km / h, the signal may be distorted due to unstable vehicle motion or changes in tire contact characteristics, and the conditions are not met.

[0036] Once the identification conditions are met, the vehicle controller receives the vertical acceleration sensing signals from each wheel via the CAN (Controller Area Network) bus. These signals are acquired by vertical acceleration sensors mounted on the unsprung mass. The vertical acceleration sensing signals are then preprocessed, including filtering and denoising. Filtering can use a Butterworth low-pass filter with a cutoff frequency calibrated to 10Hz to remove irrelevant high-frequency interference such as sensor electronic noise and tire tread friction. Denoising can use a sliding window median filter with 5 sampling points to eliminate impulse noise. For example, occasional outliers in the original signal are replaced with the median within the window after preprocessing, retaining the effective signal related to road excitation.

[0037] Finally, the preprocessed vertical acceleration sensing signal is stored in the cache unit of the vehicle MCU, and the corresponding timestamp and current vehicle speed, load and other status information are marked. The stored data is automatically cleared at a preset cycle of 1 minute after the subsequent elevation calculation is completed to avoid occupying too much storage resources.

[0038] This invention effectively filters unstable operating conditions such as low speed and high speed by collecting vehicle status signals and determining recognition conditions, avoiding invalid calculations, improving system operating efficiency, and ensuring the relevance of subsequent signal acquisition and processing. A vertical acceleration sensor for unsprung mass is used to collect signals. Utilizing the characteristic of direct contact between the unsprung mass and the road surface, the acquired vertical acceleration sensor signal contains the most direct road excitation information, avoiding interference from vehicle vibration and suspension hysteresis response, which is a problem with traditional sprung mass sensors, resulting in more accurate signals. Through filtering and denoising operations in the preprocessing stage, irrelevant interference such as high-frequency noise and pulse anomalies is effectively eliminated, purifying the effective signal and laying a reliable data foundation for subsequent road elevation calculation and roughness level identification.

[0039] Step S202: Obtain road surface elevation information from the vertical acceleration sensor signal based on the vehicle dynamics model. For details, please refer to [link to details]. Figure 1 Step S102 of the illustrated embodiment will not be described again here.

[0040] Step S203: Calculate the road surface roughness grade using road surface elevation information. For details, please refer to [link to relevant documentation]. Figure 1 Step S103 of the illustrated embodiment will not be described again here.

[0041] Figure 3 A flowchart of a third embodiment of a road surface roughness level identification method according to the present invention is shown, the method being performed by a vehicle. Figure 3 As shown, the method includes the following steps: Step S301: During vehicle operation, acquire the vertical acceleration sensing signal of at least one wheel obtained from the unsprung mass vertical acceleration sensor. For details, please refer to [link to relevant documentation]. Figure 2 Step S201 of the illustrated embodiment will not be described again here.

[0042] Step S302: Obtain road surface elevation information from the vertical acceleration sensor signal based on the vehicle dynamics model. For details, please refer to [link to relevant documentation]. Figure 2 Step S202 of the illustrated embodiment will not be described again here.

[0043] Step S303: Calculate the road surface roughness grade using road surface elevation information.

[0044] Step S303 includes: determining the road surface roughness level from the sub-road surface roughness levels corresponding to each wheel based on the voting principle.

[0045] Specifically, in this embodiment of the invention, after multiple wheels have identified their corresponding sub-road surface roughness levels, the road surface roughness level is arbitrated from the sub-road surface roughness levels corresponding to each wheel according to the majority voting principle.

[0046] Specifically, it includes: Step S3031: Obtain the road surface grade discrimination threshold; Step S3032: Extract the elevation index of each wheel through the road surface elevation information of each wheel. The elevation index is a numerical value used to represent the characteristics of the road surface elevation information data. Step S3033: Determine the sub-road roughness level corresponding to each wheel by using the relationship between the elevation index and the road surface grade discrimination threshold. Step S3034: Arbitrate the road surface roughness level corresponding to each wheel to obtain the road surface roughness level.

[0047] Step S303 aims to accurately calculate the roughness grade of the road surface on which the vehicle travels by integrating the road surface elevation information of each wheel. The specific implementation is as follows: Specifically, the road surface grade discrimination threshold is a critical numerical standard used to classify different roughness grades. It needs to be obtained based on the initial road surface grade discrimination threshold calibrated at the vehicle factory, and then combined with the threshold correction amount obtained from the analysis of vehicle state signals to generate the target road surface grade discrimination threshold. For example, in the preset initial threshold, the elevation threshold corresponding to grade 1 (smooth road) can be ±0.05m, grade 2 (medium rough road) can be ±0.15m, and grade 3 (extremely rough road) can be ±0.3m.

[0048] Elevation indicators are core parameters that characterize road surface roughness. They are extracted from the smoothed road surface elevation information and include peak elevation, root mean square elevation, and cumulative elevation deviation. The peak elevation is the maximum absolute value of the elevation data within the sliding window, reflecting the height of a single road surface protrusion or depression. The root mean square elevation is the root mean square value of the data within the window, reflecting the average roughness of the continuous road surface. The cumulative elevation deviation is the sum of the absolute values ​​of the deviations of each elevation data within the window from the benchmark value (such as the average road surface elevation of 0m), reflecting the total amplitude of road surface elevation changes over a period of time. For example, in the smoothed elevation data of the right rear wheel, the peak elevation is 0.12m, the root mean square elevation is 0.08m, and the cumulative elevation deviation is 0.5m. These indicators together constitute the road surface roughness characteristics corresponding to that wheel.

[0049] Subsequently, based on the relationship between the extracted elevation indices and the target road surface grade discrimination threshold, this embodiment of the invention determines the sub-road surface roughness grade of each wheel. For example, the judgment rule is set as follows: when the peak elevation < grade 1 threshold, the root mean square elevation < grade 1 threshold, and the cumulative elevation deviation < the cumulative standard corresponding to the grade 1 threshold, it is judged as sub-grade 1; when any elevation index is between grade 1 and grade 2 thresholds, it is judged as sub-grade 2; when any elevation index exceeds grade 2 threshold but does not reach grade 3 threshold, it is judged as sub-grade 2; when any elevation index exceeds grade 3 threshold, it is judged as sub-grade 3. Assuming that the peak elevation of the left front wheel is 0.06m, the root mean square elevation is 0.03m, and the cumulative elevation deviation is 0.2m, all of which are less than the grade 1 target threshold ±0.07m and the corresponding cumulative standard, the sub-road surface roughness grade of the left front wheel is 1; the peak elevation of the right rear wheel is 0.12m, which is between the grade 1 (0.07m) and grade 2 (0.17m) thresholds, and is judged as sub-grade 2.

[0050] When the sub-road surface roughness level corresponding to each wheel has been calculated, the embodiments of the present invention arbitrate the sub-road surface roughness level of each wheel. For example, the embodiments of the present invention adopt the arbitration logic of majority voting: when the number of wheels corresponding to a certain sub-road surface roughness level (the first sub-road surface roughness level) is greater than that of other sub-levels, the level is determined as the overall road surface roughness level. For example, the four wheel sub-levels are 1, 1, 2, and 1 respectively. Level 1 corresponds to 3 wheels, which is more than the 1 wheel of level 2. Therefore, the final road surface roughness level is 1.

[0051] This invention improves the flexibility and accuracy of road surface grade determination by acquiring a target road surface grade discrimination threshold. The smoothing process effectively eliminates interference components in the elevation data, providing a clean and reliable data foundation for subsequent elevation index extraction and grade determination. By extracting multi-dimensional elevation indicators, the road surface roughness characteristics are comprehensively characterized, reflecting the true road surface condition better than single-indicator determination and reducing the probability of misjudgment. Furthermore, the sub-grade determination logic based on threshold relationships is clear and computationally efficient, meeting real-time recognition requirements. Finally, an arbitration rule combining majority voting and maximum roughness priority is adopted to avoid misjudgment of the entire vehicle grade due to local road surface anomalies in a single wheel, while also promptly identifying extremely rough road surface areas, providing accurate road surface condition input for functions such as vehicle suspension adjustment and speed planning.

[0052] In some optional implementations, step S3031 above includes: Step b1: Obtain vehicle status signals and initial thresholds; Step b2: Analyze the threshold correction amount based on the vehicle status signal; Step b3: Overlay the threshold correction amount and the initial threshold to obtain the target road surface grade discrimination threshold.

[0053] Specifically, such as Figure 4 As shown, the road surface grade discrimination threshold provided in this embodiment of the invention is a dynamic value. The initial threshold is a basic critical value pre-calibrated based on ISO standard rough road surface test data, used as an initial standard for classifying different roughness grades. It is stored in the local memory of the vehicle controller. For example, the preset initial thresholds are: ±0.05m for grade 1 (smooth road), ±0.15m for grade 2 (medium rough road), and ±0.3m for grade 3 (extremely rough road). This initial threshold provides a benchmark for subsequent dynamic correction.

[0054] The threshold correction amount is a parameter used to fine-tune the initial threshold based on the vehicle's real-time driving status. Its analysis process is based on vehicle status signals. First, key status parameters are extracted from the vehicle status signals, including vehicle speed, vehicle load, and vehicle pitch angle. Vehicle speed is obtained from wheel speed sensors through algorithmic conversion; vehicle load is derived from suspension compression data collected by suspension displacement sensors; and vehicle pitch angle is detected in real-time by an IMU (Inertial Measurement Unit). Then, based on the extracted multi-dimensional key status parameters, a preset working condition-correction amount calibration rule is matched (this rule is established through extensive test data of real vehicles under different combinations of vehicle speed, load, and pitch angle, recording the optimal mapping relationship between various working conditions and corresponding threshold correction amounts). This directly determines the threshold correction amount suitable for the current driving state, ensuring that the correction amount can accurately offset the impact of different working conditions on road elevation recognition and improving the working condition adaptability of the threshold.

[0055] In some alternative implementations, step b2 includes: Step b21: Obtain key status parameters from the vehicle status signal. The key status parameters include at least one of the following: vehicle speed, vehicle load, and vehicle pitch angle. Step b22: Use the key state parameters to query the preset working condition-correction mapping table to obtain the threshold correction amount.

[0056] Specifically, the aforementioned key state parameters form a set of multi-dimensional operating condition indexes. For example, "vehicle speed 50km / h + full vehicle load + pitch angle 0.5°" is a specific operating condition index. Finally, the multi-dimensional operating condition indexes are used to query a preset operating condition-correction mapping table. This mapping table is generated through calibration using a large amount of test data from real vehicles under different operating conditions. It records the correspondence between various operating condition combinations and corresponding threshold correction values. Different threshold correction values ​​correspond to different ranges of vehicle state signal values, thus corresponding to different operating conditions. For example, when the operating condition is "vehicle speed 80km / h + half vehicle load + pitch angle 0°", the threshold correction value obtained from the mapping table is -0.02m. When the operating condition is "vehicle speed 30km / h + full vehicle load + pitch angle 1°", the threshold correction value obtained is +0.03m. The creation of the condition-correction mapping table needs to follow objective physical laws. For example, when the vehicle speed is higher, the bump amplitude is more obvious, so the threshold needs to be increased; when the vehicle speed is slower, the bump amplitude is weaker, so the threshold needs to be decreased. The corresponding threshold correction amount is then determined based on this law.

[0057] The obtained threshold correction is then superimposed on the initial threshold to obtain the target pavement grade discrimination threshold adapted to the current working conditions. For example, if the initial threshold is ±0.05m for grade 1, ±0.15m for grade 2, and ±0.3m for grade 3, and the threshold correction obtained under the current working conditions is +0.02m, then the target thresholds after superposition are adjusted to ±0.07m for grade 1, ±0.17m for grade 2, and ±0.32m for grade 3; if the correction is -0.02m, then the target thresholds are adjusted to ±0.03m for grade 1, ±0.13m for grade 2, and ±0.28m for grade 3.

[0058] This invention provides a stable benchmark through an initial threshold and dynamically analyzes and corrects the threshold based on vehicle status signals. This enables adaptive adjustment of the threshold under different operating conditions, avoiding recognition deviations caused by fixed thresholds under varying vehicle speeds and loads, and significantly improving the accuracy of road surface roughness level determination. The multi-dimensional fusion of key status parameters into an operating condition index comprehensively reflects the dynamic characteristics of vehicle driving, making the correction analysis more closely aligned with actual driving scenarios. The operating condition-correction mapping table is generated based on real-vehicle calibration, ensuring high data reliability and a simple query operation logic that does not increase the system's computational burden, thus guaranteeing the real-time performance of the recognition process.

[0059] In some optional implementations, step S3034 above includes: Step c1: When the number of wheels corresponding to the first sub-road surface roughness level is greater than the number of wheels corresponding to other sub-road surface roughness levels, the road surface roughness level is determined to be the first sub-road surface roughness level. Step c2: When there are at least two sub-road surface roughness levels with the same number of wheels, determine the road surface roughness level as the second sub-road surface roughness level. The second sub-road surface roughness level is the level with the highest road surface roughness among the sub-road surface roughness levels with the same number of wheels.

[0060] Specifically, in this embodiment of the invention, the final road surface roughness level of the entire vehicle is obtained by combining the road surface roughness levels of the four wheels with an arbitration logic that prioritizes the largest roughness. The first road surface roughness level refers to the sub-level with the largest number of wheels among the four wheels, and the second road surface roughness level refers to the level that represents the most severe road surface roughness when there are multiple sub-levels with the largest number of wheels (e.g., if the sub-levels are divided into level 1, level 2, and level 3, with roughness increasing sequentially, then level 3 is the second road surface roughness level).

[0061] During arbitration, the vehicle controller first counts the number of wheels corresponding to each sub-road surface roughness level. For example, if the four wheel sub-levels are 1, 1, 1, and 2, then the number of wheels corresponding to level 1 is 3, which is more than the 1 wheel corresponding to level 2. Therefore, level 1 is the first sub-road surface roughness level, and the overall vehicle road surface roughness level is directly determined to be level 1. Similarly, if the four wheel sub-levels are 2, 2, 3, and 2, then the number of wheels corresponding to level 2 is 3, which is more than the 1 wheel corresponding to level 3. Therefore, the overall vehicle level is determined to be level 2.

[0062] When statistics show that at least two sub-levels have the most and the same number of wheels, for example, if the four-wheel sub-levels are level 1, level 1, level 2, and level 2, and level 1 and level 2 each have 2 wheels, then the level with the highest roughness needs to be selected from these two sub-levels with the most wheels. Since level 2 has a higher roughness than level 1, the second sub-road surface roughness level is level 2, and the final road surface roughness level of the whole vehicle is determined to be level 2. Similarly, if the four-wheel sub-levels are level 2, level 3, level 2, and level 3, and level 2 and level 3 each have 2 wheels, and level 3 has a higher roughness, then the overall vehicle level is determined to be level 3.

[0063] This invention employs majority voting arbitration logic, fully integrating road surface information collected from all four wheels. This avoids misjudgments of the overall vehicle roughness level caused by a single wheel running over manhole covers, potholes, or other localized road surface anomalies, ensuring that the recognition results reflect the overall road surface condition and improving recognition accuracy. When multiple sub-levels have the same number of wheels, the level with the highest roughness is prioritized as the final result. This allows for the timely detection of high-risk rough areas on the road surface, providing a more conservative and safer control input for the vehicle's suspension system and preventing a decrease in comfort or component wear due to underestimating road roughness. The arbitration logic rules are clear and computationally intensive, requiring only a simple comparison of the number of wheels to obtain the result, without increasing the computational burden on the onboard controller. This ensures the real-time performance of road roughness level recognition, meeting the response speed requirements of intelligent driving and active suspension control. It also ensures the consistency and stability of recognition results under different operating conditions, reducing recognition deviations caused by local road surface differences or signal fluctuations. This provides reliable road condition support for functions such as vehicle comfort control, speed planning, and energy recovery strategy optimization.

[0064] In some alternative implementations, such as Figure 5 As shown, a complete process of an embodiment of the present invention is as follows: 1. Sensor data acquisition: Vertical acceleration sensors installed on the unsprung mass acquire the raw vertical acceleration signals of each wheel in real time. Auxiliary sensors such as IMU synchronously acquire vehicle status signals such as vehicle speed, vehicle load, and vehicle pitch angle. All acquired raw data are transmitted to the vehicle controller in real time via the CAN bus to ensure the real-time performance and stability of data transmission.

[0065] 2. Input Signal Processing: After receiving the raw data, the vehicle controller performs filtering and noise reduction preprocessing on the raw vertical acceleration signal to remove high-frequency interference such as sensor electronic noise and tire tread friction. Then, pulse anomalies are eliminated by mid-range filtering in a sliding window. At the same time, based on the vehicle speed signal in the collected vehicle status signal, it verifies whether the road roughness level recognition conditions are met (e.g., vehicle speed is between 10km / h and 120km / h and continuously stable for 100ms). Only when the conditions are met is the preprocessed vertical acceleration sensing signal retained and the process proceeded to the next step. Otherwise, the current round of recognition is paused and the system waits for a valid working condition.

[0066] 3. Elevation data calculation: The on-board controller calls the pre-stored state space equations of a quarter-vehicle model, substitutes the pre-processed vertical acceleration sensor signals into the equations, and combines preset calibration parameters such as vehicle mass, unsprung mass, suspension spring stiffness, shock absorber damping coefficient, and tire stiffness. It then uses a numerical integration algorithm to calculate the vertical displacement and velocity of the sprung and unsprung masses, and then reversely derives the road elevation information corresponding to each wheel, accurately restoring the vertical displacement state of the road surface.

[0067] 4. Logical parameter pre-calculation: Based on the vehicle status signals collected by sensors, a multi-dimensional working condition index is constructed, and the preset working condition-correction amount mapping table is queried to calculate the logical parameters such as the threshold correction amount required for subsequent roughness level determination in real time, providing a basis for dynamically adjusting the judgment criteria.

[0068] 5. Roughness Level Determination: The road surface elevation information obtained by reverse calculation is smoothed by moving average to remove residual minor interference. Then, elevation indicators that characterize the roughness of the road surface, such as peak elevation, root mean square elevation, and cumulative elevation deviation, are extracted. Combined with the preset initial threshold and the calculated threshold correction, the target road surface level discrimination threshold is obtained. By comparing the relationship between the elevation indicators and the target threshold, the sub-road surface roughness level corresponding to each wheel is determined, and four independent sub-level signals for each wheel are generated.

[0069] 6. Arbitration Output: The vehicle controller performs arbitration logic on the sub-road surface roughness level signals of the four wheels. When the number of wheels corresponding to a certain sub-road surface roughness level is greater than that of other sub-levels, that level is determined as the overall road surface roughness level. When at least two sub-levels have the most wheels and are the same, the level with the highest road surface roughness is selected as the final result. Finally, the overall road surface roughness level signal is sent to the suspension control unit or the vehicle controller via the CAN bus to provide accurate road surface input for functions such as suspension damping adjustment and vehicle attitude control.

[0070] Figure 6 A schematic diagram of an embodiment of a road surface roughness grade identification device according to the present invention is shown. The device 600 includes: The signal acquisition module 601 is used to acquire the vertical acceleration sensing signal of at least one wheel obtained by the unsprung mass vertical acceleration sensor during vehicle operation. The elevation reverse calculation module 602 is used to obtain road surface elevation information from vertical acceleration sensing signals based on the vehicle dynamics model; The grade discrimination module 603 is used to calculate the road surface roughness grade based on the road surface elevation information.

[0071] In one alternative embodiment, the signal acquisition module 601 includes: The status signal acquisition unit is used to acquire vehicle status signals during vehicle operation, including vehicle speed signals. The condition determination unit is used to determine whether the road surface roughness level identification conditions are met based on the vehicle status signal. The unsprung acceleration recognition unit is used to receive the vertical acceleration sensing signals of each wheel when the road roughness level recognition conditions are met. The preprocessing unit is used to preprocess the vertical acceleration sensing signal; The data storage unit is used to record the preprocessed vertical acceleration sensing signal.

[0072] In one alternative embodiment, the level discrimination module 603 includes: Threshold unit, used to obtain the road surface grade discrimination threshold; The index calculation unit is used to extract the elevation index of each wheel from the road surface elevation information after smoothing of each wheel. The single-wheel roughness determination unit is used to determine the sub-road roughness level corresponding to each wheel by utilizing the relationship between the elevation index and the road surface grade discrimination threshold. The arbitration unit is used to arbitrate the road surface roughness level corresponding to each wheel to obtain the road surface roughness level.

[0073] Figure 7 The diagram shows a structural schematic of an embodiment of the vehicle of the present invention. The specific embodiments of the present invention do not limit the specific implementation of the vehicle.

[0074] like Figure 7 As shown, the vehicle may include: a processor 402, a communications interface 404, a memory 406, and a communications bus 408.

[0075] The processor 402, communication interface 404, and memory 406 communicate with each other via communication bus 408. Communication interface 404 is used to communicate with other network elements such as clients or other servers. The processor 402 executes program 410, specifically performing the relevant steps described above in the method embodiment.

[0076] Specifically, program 410 may include program code, which includes computer-executable instructions.

[0077] Processor 402 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. The vehicle may include one or more processors of the same type, such as one or more CPUs; or processors of different types, such as one or more CPUs and one or more ASICs.

[0078] Memory 406 is used to store program 410. Memory 406 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk storage device.

[0079] Specifically, program 410 can be invoked by processor 402 to execute the relevant steps described above in the method embodiment.

[0080] A portion of this invention can be applied as a computer program product, such as computer program instructions, which, when executed by a computer, can invoke or provide the methods and / or technical solutions according to the invention through the operation of the computer. Those skilled in the art will understand that the forms in which computer program instructions exist in a computer-readable medium include, but are not limited to, source files, executable files, installation package files, etc. Correspondingly, the ways in which computer program instructions are executed by a computer include, but are not limited to: the computer directly executing the instructions, or the computer compiling the instructions and then executing the corresponding compiled program, or the computer reading and executing the instructions, or the computer reading and installing the instructions and then executing the corresponding installed program. Here, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible to a computer.

[0081] The algorithms or displays provided herein are not inherently related to any particular computer, virtual system, or other device. Furthermore, the embodiments of this invention are not directed to any particular programming language.

[0082] Numerous specific details are set forth in the specification provided herein. However, it will be understood that embodiments of the invention may be practiced without these specific details. Similarly, for the sake of brevity and to aid in understanding one or more aspects of the invention, in the description of exemplary embodiments of the invention above, various features of the embodiments are sometimes grouped together in a single embodiment, figure, or description thereof. The claims, which follow the detailed description, are hereby expressly incorporated into that detailed description, wherein each claim itself is a separate embodiment of the invention.

[0083] Those skilled in the art will understand that the modules in the device of the embodiment can be adaptively changed and placed in one or more devices different from that embodiment. Modules, units, or components in the embodiment can be combined into a single module, unit, or component, and further, they can be divided into multiple sub-modules, sub-units, or sub-components, except that at least some of such features and / or processes or units are mutually exclusive.

[0084] It should be noted that the above embodiments are illustrative of the invention and not restrictive, and that those skilled in the art can devise alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses should not be construed as limiting the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention can be implemented by means of hardware comprising several different elements and by means of a suitably programmed computer. In the unit claims enumerating several means, several of these means may be embodied by the same item of hardware. The use of the words first, second, and third, etc., does not indicate any order. These words can be interpreted as names. The steps in the above embodiments, unless otherwise specified, should not be construed as limiting the order of execution.

Claims

1. A method for identifying road surface roughness levels, characterized in that, The method includes: During vehicle operation, the vertical acceleration sensing signal of at least one wheel is acquired by the unsprung mass vertical acceleration sensor. Road surface elevation information is obtained from the vertical acceleration sensing signal based on the vehicle dynamics model; The road surface roughness grade is calculated using the road surface elevation information.

2. The method according to claim 1, characterized in that, The process of obtaining road surface elevation information from the vertical acceleration sensing signal based on the vehicle dynamics model includes: Based on the coupling relationship between the vertical forces on the sprung mass and the unsprung mass, the mapping relationship between vertical acceleration and road surface elevation is determined. The mapping relationship is used to determine the road surface elevation information corresponding to the vertical acceleration sensing signal.

3. The method according to claim 1, characterized in that, The determination of the mapping relationship between vertical acceleration and road surface elevation based on the coupling relationship between the vertical forces on the sprung mass and the unsprung mass includes: Obtain a first mapping relationship, which indicates that the net vertical force on the sprung mass is equal to the resultant force of the elastic restoring force and the damping force of the sprung mass in vertical motion. The elastic restoring force is determined based on the vertical displacement difference between the sprung mass and the unsprung mass, and the damping force is determined based on the vertical velocity difference between the sprung mass and the unsprung mass. The vertical displacement difference and the vertical velocity difference are calculated through the vertical acceleration sensing signal. Obtain a second mapping relationship, which indicates that the net vertical force on the unsprung mass is equal to the resultant force of the elastic force, damping force, and tire elastic reaction force of the unsprung mass's vertical motion. The elastic force is determined based on the vertical displacement difference between the sprung and unsprung masses, the damping force is determined based on the vertical velocity difference between the sprung and unsprung masses, and the tire elastic reaction force is determined based on the vertical displacement difference between the unsprung mass and the road surface excitation. The vertical displacement difference and the vertical velocity difference are calculated using the vertical acceleration sensing signal. By combining the first mapping relationship and the second mapping relationship, the mapping relationship between the vertical acceleration and the road surface elevation is obtained.

4. The method according to claim 1, characterized in that, When vertical acceleration sensor signals from multiple wheels are collected, the calculation of road surface roughness grade using the road surface elevation information includes: Obtain the road surface grade discrimination threshold; The elevation index of each wheel is extracted from the road surface elevation information of each wheel. The elevation index is a numerical value used to represent the characteristics of the road surface elevation information data. The roughness level of each wheel's sub-road surface is determined by using the relationship between the elevation index and the road surface grade discrimination threshold. Arbitrate the road surface roughness level corresponding to each wheel to obtain the road surface roughness level.

5. The method according to claim 4, characterized in that, The threshold for obtaining the road surface grade discrimination includes: Acquire vehicle status signals; Obtain the initial threshold; Threshold correction amount based on vehicle status signal analysis; The road surface grade discrimination threshold is obtained by superimposing the threshold correction amount and the initial threshold.

6. The method according to claim 5, characterized in that, The threshold correction amount based on vehicle state signal analysis includes: Acquire key status parameters from the vehicle status signal, wherein the key status parameters include at least one of vehicle speed, vehicle load, and vehicle pitch angle; The threshold correction amount is obtained by querying a preset working condition-correction amount mapping table using the key state parameters.

7. The method according to claim 4, characterized in that, The arbitration of the sub-road surface roughness level corresponding to each wheel to obtain the road surface roughness level includes: determining the road surface roughness level from the sub-road surface roughness levels corresponding to each wheel based on a voting principle.

8. The method according to claim 7, characterized in that, The determination of the road surface roughness level from the sub-road surface roughness levels corresponding to each wheel based on the voting principle includes: When the number of wheels corresponding to the first sub-road surface roughness level is greater than the number of wheels corresponding to other sub-road surface roughness levels, the road surface roughness level is determined to be the first sub-road surface roughness level. When there are at least two sub-road surface roughness levels corresponding to the highest number of wheels, the road surface roughness level is determined as the second sub-road surface roughness level. The second sub-road surface roughness level is the level that represents the highest road surface roughness among the sub-road surface roughness levels with the highest number of wheels.

9. The method according to claim 1, characterized in that, Acceleration sensor signals from each wheel are acquired during vehicle operation, including: Vehicle status signals are collected during vehicle operation, including vehicle speed signals; The vehicle status signal is used to determine whether the road surface roughness level identification conditions are met. When the road surface roughness level identification condition is met, the vertical acceleration sensing signal of each wheel is received. The vertical acceleration sensing signal is preprocessed; Record the preprocessed vertical acceleration sensor signal.

10. A road surface roughness grade identification device, characterized in that, The device includes: The signal acquisition module is used to acquire the vertical acceleration sensing signal of at least one wheel, which is collected by the unsprung mass vertical acceleration sensor, during vehicle operation. The elevation reverse calculation module is used to obtain road surface elevation information from the vertical acceleration sensing signal based on the vehicle dynamics model; The grade discrimination module is used to calculate the road surface roughness grade based on the road surface elevation information.

11. A vehicle, characterized in that, include: A memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, the processor executing the computer instructions to perform the method of any one of claims 1 to 9.

Images