Road surface recognition method, controller, vehicle, storage medium and program product

Abstract

The invention relates to a road surface recognition method, a controller, a vehicle, a storage medium and a program product. The method comprises the following steps: acquiring a signal of a vehicle; and determining first road surface information by using a first strategy and / or a second strategy according to the signal. Through the first strategy, accurate determination of the first road surface information can be realized based on vehicle dynamics, and high interpretability is achieved; through a second strategy, capturing of complex nonlinear features can be realized based on the model so as to realize accurate determination of the first road surface information; the two complementations can further improve the accuracy of road surface recognition. In this way, the sensor and computing power cost is reduced, and meanwhile accurate determination of the first road surface information can be achieved.

Description

Technical Field

[0001] This disclosure relates to the field of vehicles, and more specifically, to a road surface recognition method, controller, vehicle, storage medium, and program product. Background Technology

[0002] The safety and comfort of intelligent driving heavily rely on accurate perception of road conditions, but related technologies face bottlenecks. Traditional vision-based solutions are limited by lighting conditions, lack robustness, and struggle to ensure the accuracy of road recognition results. While LiDAR offers excellent accuracy, its high cost and excessive computing power requirements hinder its large-scale application. Therefore, achieving accurate road condition recognition at a low cost has become a pressing issue. Summary of the Invention

[0003] The purpose of this disclosure is to provide a road surface recognition method, controller, vehicle, storage medium, and program product device to achieve accurate recognition of road conditions at a low cost.

[0004] To achieve the above objectives, the first aspect of this disclosure provides a road surface recognition method, comprising: Acquire vehicle signals; Based on the signal, first road surface information is determined using a first strategy and / or a second strategy, wherein the first strategy is based on vehicle dynamics to determine the first road surface information, and the second strategy is based on a model to determine the first road surface information.

[0005] Optionally, the first road surface information includes a first road surface type; determining the first road surface type based on the signal and using a first strategy includes: Based on the signal, first reference data of the vehicle is determined, the first reference data including at least one of the following: vehicle speed, slip ratio and at least one of a first road surface feature; The first road surface type is determined based on the first reference data.

[0006] Optionally, determining the first road surface type based on the first reference data includes: Based on the first reference data, determine the probability that the current road surface is one of each preset road surface type; Based on the probability, determine the first road surface type of the current road surface.

[0007] Optionally, determining the probability that the current road surface is one of each preset road surface type based on the first reference data includes: Determine the sub-probability corresponding to each of the first reference data and each of the preset road surface types; For each of the preset road surface types, the probability of that preset road surface type is determined based on the sub-probability corresponding to that preset road surface type.

[0008] Optionally, determining the sub-probability corresponding to each of the first reference data and each of the preset road surface types includes: Based on the pre-defined correspondence, the sub-probability under each preset road surface type corresponding to the current first reference data is determined, wherein the correspondence is the relationship between the first reference data, the preset road surface type and the corresponding sub-probability.

[0009] Optionally, determining the probability of the preset road surface type based on the sub-probability corresponding to the preset road surface type includes: The sum of the subprobabilities corresponding to the preset road surface type is determined as the probability of the preset road surface type.

[0010] Optionally, determining the first road surface type of the current road surface based on the probability includes: The preset road surface type corresponding to the highest probability is determined as the first road surface type.

[0011] Optionally, the first road surface information includes a first road surface feature, the first road surface feature including smoothness; based on the signal, the smoothness is determined using a first strategy: Based on the signal, a second reference data is determined, which is data that can reflect the degree of vehicle bumps; The flatness is determined based on the second reference data.

[0012] Optionally, determining the flatness based on the second reference data includes: Determine the degree of change of the second reference data within the first target time period, where the first target time period is a historical time period with the current time as the end time and a first preset duration; The flatness is determined based on the degree of change.

[0013] Optionally, determining the degree of change of the second reference data within the first target time period includes: Determine the integral value of the absolute value of each of the second reference data points within the first target time period; The degree of change is determined based on the integral value and the weighting coefficient corresponding to each of the second reference data.

[0014] Optionally, determining the flatness based on the degree of change includes: The smoothness is determined based on the degree of change and the distance traveled by the vehicle during the first target time period.

[0015] Optionally, determining the smoothness based on the degree of change and the vehicle's travel distance within the first target time period includes: The flatness is determined by the following formula. :

[0016] in, The degree of change, For the first Second reference data, For the first target time period, For the first The weighting coefficients of the second reference data, This represents the distance traveled by the vehicle during the first target time period. The number of second reference data types Optionally, the first road surface information includes a first road surface type and / or a first road surface feature. Based on the signal, the first road surface information is determined using a second strategy, including: Based on the signal, construct the target matrix; The first road surface information is obtained based on the target matrix and the model.

[0017] Optionally, constructing the target matrix based on the signal includes: Based on the signal, a third reference data is obtained, which is data that can reflect the vehicle's operating status; The target matrix is ​​constructed based on the values ​​of third reference data at multiple preset time points within the second target time period, where the second target time period is a historical time period with the current time as the end time and a second preset duration.

[0018] Optionally, the values ​​of the elements in the target matrix are within a preset range and are obtained by amplitude processing of the third reference data, with the same amplitude processing method for the same type of third reference data.

[0019] Optionally, obtaining the first road surface information based on the target matrix and the model includes: The target matrix is ​​input into the model to obtain the first road surface information and the confidence level of the first road surface information.

[0020] Optionally, the first road surface information is determined using a first strategy and a second strategy, including: The first road surface information is determined based on the road surface information and corresponding confidence level obtained using the second strategy, and the road surface information obtained using the first strategy.

[0021] Optionally, the first road surface information includes a first road surface type; determining the first road surface type based on the road surface information obtained using the second strategy and the corresponding confidence level, and the road surface information obtained using the first strategy, includes: The third road surface type is determined to be the first road surface type if any of the following conditions are met: The second road surface type is the same as the third road surface type; The second road surface type is inconsistent with the third road surface type, and the confidence level of the third road surface type is greater than a preset first threshold. Wherein, the second road surface type is the road surface type determined based on the first strategy, and the third road surface type is the road surface type determined based on the second strategy.

[0022] Optionally, the first road surface information includes a first road surface type; determining the first road surface type based on the road surface information obtained using the second strategy and the corresponding confidence level, and the road surface information obtained using the first strategy, includes: If the second road surface type is inconsistent with the third road surface type, and the confidence level of the third road surface type is less than or equal to the preset first threshold, then the first road surface type of the previous moment is determined as the first road surface type of the current road surface. Wherein, the second road surface type is the road surface type determined based on the first strategy, and the third road surface type is the road surface type determined based on the second strategy.

[0023] Optionally, the first road surface information includes a first road surface feature; determining the first road surface feature based on the road surface information obtained using the second strategy and the corresponding confidence level, and the road surface information obtained using the first strategy, includes: The second road surface feature is determined to be the first road surface feature if any of the following conditions are met: The second road surface features are consistent with the third road surface features; The second road surface feature is inconsistent with the third road surface feature, and the confidence level of the third road surface feature is less than a preset second threshold. The second road surface feature is the road surface feature determined based on the first strategy, and the third road surface feature is the road surface feature determined based on the second strategy.

[0024] Optionally, the first road surface information includes a first road surface feature; determining the first road surface feature based on the road surface information obtained using the second strategy and the corresponding confidence level, and the road surface information obtained using the first strategy, includes: If the second road surface feature is inconsistent with the third road surface feature, and the confidence level of the third road surface feature is greater than or equal to the preset second threshold, then the first road surface feature of the previous moment is determined as the first road surface feature of the current road surface. The second road surface feature is the road surface feature determined based on the first strategy, and the third road surface feature is the road surface feature determined based on the second strategy.

[0025] Optionally, the first road surface information includes a first road surface feature, which includes at least one of the following: adhesion coefficient, slope, rolling resistance coefficient, and smoothness.

[0026] Optionally, the acquired signal is the vehicle's bus signal.

[0027] A second aspect of this disclosure provides a controller, comprising: processor; Memory used to store processor-executable instructions; The processor is configured to execute the executable instructions in the memory to implement the steps of the road surface recognition method provided in the first aspect of this disclosure.

[0028] A third aspect of this disclosure provides a vehicle including the controller provided in the second aspect of this disclosure.

[0029] The fourth aspect of this disclosure provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the road surface recognition method provided in the first aspect of this disclosure.

[0030] The fifth aspect of this disclosure provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the road surface recognition method provided in the first aspect of this disclosure.

[0031] In the above technical solution, the first strategy, based on vehicle dynamics, enables accurate determination of the first road surface information and possesses high interpretability; the second strategy, based on a model, enables the capture of complex nonlinear features to achieve accurate determination of the first road surface information; the complementarity of the two strategies further improves the accuracy of road surface recognition. Thus, while reducing sensor and computing power costs, accurate determination of the first road surface information can be achieved, providing a reliable basis for rapid adaptation of subsequent control strategies, thereby enhancing driving safety and comfort.

[0032] Other features and advantages of this disclosure will be described in detail in the following detailed description section. Attached Figure Description

[0033] The accompanying drawings are provided to further illustrate the present disclosure and form part of the specification. They are used together with the following detailed description to explain the present disclosure, but do not constitute a limitation thereof. In the drawings: Figure 1 This is a flowchart of a road surface recognition method provided in an exemplary embodiment of this disclosure.

[0034] Figure 2 This is a flowchart of road surface type identification using a first strategy provided in an exemplary embodiment of this disclosure.

[0035] Figure 3 This is a flowchart of a flatness determination using a first strategy provided in an exemplary embodiment of this disclosure.

[0036] Figure 4 This is a flowchart of a flatness determination using a first strategy provided in an exemplary embodiment of this disclosure.

[0037] Figure 5 This is a flowchart of road surface recognition using a first strategy provided in an exemplary embodiment of this disclosure.

[0038] Figure 6 This is a schematic diagram of a road surface recognition method provided in an exemplary embodiment of this disclosure.

[0039] Figure 7 This is a flowchart of road surface recognition using a second strategy provided in an exemplary embodiment of this disclosure.

[0040] Figure 8 This is a flowchart of road surface recognition using a second strategy provided in an exemplary embodiment of this disclosure.

[0041] Figure 9 This is a schematic diagram of a road surface recognition method provided in an exemplary embodiment of this disclosure.

[0042] Figure 10 This is a schematic diagram illustrating the determination of road surface information using a first strategy and a second strategy, provided by an exemplary embodiment of this disclosure.

[0043] Figure 11 This is a flowchart illustrating the determination of road surface type using a first strategy and a second strategy, provided by an exemplary embodiment of this disclosure.

[0044] Figure 12 This is a flowchart illustrating the determination of road surface features using a first strategy and a second strategy, provided by an exemplary embodiment of this disclosure.

[0045] Figure 13 This is a flowchart of a road surface recognition device provided in an exemplary embodiment of this disclosure. Detailed Implementation

[0046] The specific embodiments of this disclosure will be described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are for illustration and explanation only and are not intended to limit this disclosure.

[0047] In the following description, the words "first" and "second" are used only to distinguish the purpose of the description and should not be interpreted as indicating or implying relative importance or order.

[0048] Figure 1 This is a flowchart illustrating an exemplary embodiment of the present disclosure of a road surface recognition method. This method can be applied to a controller installed on a vehicle. Figure 1 As shown, the method may include steps S101 and S102.

[0049] In step S101, the vehicle's signal is acquired.

[0050] In step S102, based on the signal, the first road surface information is determined using a first strategy and / or a second strategy.

[0051] The first strategy is to determine the first road surface information based on vehicle dynamics, and the second strategy is to determine the first road surface information based on a model.

[0052] For example, the signal may be preprocessed to improve its reliability. The preprocessing may include at least one of the following: unit conversion, first-order low-pass filtering, limiting, and correction processing.

[0053] In one embodiment, the signal can be a vehicle bus signal, meaning the signal is acquired via the vehicle bus. In another embodiment, the signal can be directly transmitted from existing vehicle sensors.

[0054] In this way, existing vehicle sensors (such as wheel speed sensors and steering angle sensors) can be reused to obtain the required data, eliminating the need to deploy dedicated road surface detection sensors (such as LiDAR or cameras), effectively reducing the complexity and cost of the vehicle system. Furthermore, the bus signal possesses strong anti-interference capabilities and high real-time performance. Therefore, the bus signal can provide a high-real-time, low-cost, and highly reliable data foundation for road surface information identification.

[0055] Furthermore, the road surface recognition method provided in this disclosure uses signals to confirm the first road surface information. Therefore, the road surface recognition method has simple functional integration, relatively low computing power requirements, and can be integrated into controllers such as ADAS (Advanced Driving Assistance System) domain controllers and VCU (Vehicle Control Unit). It has high adaptability and strong applicability in various vehicle models.

[0056] In one embodiment, the determination of the first road surface information can be achieved using a first strategy based on vehicle dynamics.

[0057] For example, road surface information such as the road adhesion coefficient and slope can be determined using vehicle dynamics based on data such as vehicle acceleration, angular velocity, and wheel speed. In this way, road surface information can be determined using signals through vehicle dynamics. This process has high interpretability and can ensure high accuracy of the determined initial road surface information.

[0058] In one embodiment, a second strategy based on model construction can be used to determine the first road surface information.

[0059] For example, this model could be an artificial intelligence (AI) recognition model. An AI recognition model refers to a function or network structure that learns patterns from data and makes predictions, classifications, or decisions using machine learning or deep learning techniques. For instance, this AI recognition model could be a pre-trained neural network recognition model, which can be trained using historical data. Thus, using this second strategy, complex nonlinear features can be identified to achieve efficient and reliable perception of road surface information in complex environments.

[0060] In one embodiment, the identification results of the first strategy and the identification results of the second strategy can be fused to obtain the first road surface information.

[0061] For example, the recognition results of the first strategy and the second strategy can be weighted and summed based on the vehicle's operating scenario. The vehicle's location can be used to determine whether the operating scenario is highway cruising or complex urban road conditions, allowing for the assignment of different weights to the recognition results of different strategies. In this way, the recognition results of the first strategy and the second strategy complement each other, further improving the accuracy of road surface recognition.

[0062] In such Figure 1In the described technical solution, the first strategy, based on vehicle dynamics, enables accurate determination of the first road surface information and possesses high interpretability; the second strategy, based on a model, enables the capture of complex nonlinear features to achieve accurate determination of the first road surface information; the complementarity of these two strategies further improves the accuracy of road surface recognition. Thus, while reducing sensor and computing power costs, accurate determination of the first road surface information is achieved, providing a reliable basis for rapid adaptation of subsequent control strategies, thereby enhancing driving safety and comfort.

[0063] In an optional implementation, the first road surface information may include a first road surface type, which may be any of the following: snow, mud, grass, sand, mountain, rock, wading, and ordinary road. The first road surface information may include first road surface features, which may include at least one of the following: coefficient of adhesion, gradient, rolling resistance coefficient, and smoothness.

[0064] The technical solution for determining the first road surface information based on the signal and using the first strategy will be described below. Figure 2 This is a flowchart illustrating road surface type identification using a first strategy, provided in an exemplary embodiment of this disclosure. Figure 2 As shown, the method may include steps S201 and S202.

[0065] In step S201, the first reference data of the vehicle is determined based on the signal.

[0066] The first reference data includes at least one of the following: vehicle speed, slip ratio, and at least one of the first road surface characteristics.

[0067] For example, based on the signals, vehicle speed, slip ratio, coefficient of adhesion, gradient, and rolling resistance coefficient can be determined using vehicle dynamics through relevant technologies; this will not be elaborated upon here. Using the first strategy, this can be achieved through, as... Figure 3 Determine flatness as shown: In step S301, the second reference data is determined based on the signal.

[0068] Road surface smoothness measures the deviation of the longitudinal unevenness of the road surface. During vehicle operation, the worse the road surface smoothness or the higher the vehicle speed, the more pronounced the vehicle's bumps and jolts. Therefore, a second reference data reflecting the degree of vehicle bumps and jolts can be determined based on signals, and this second reference data can be used to determine the smoothness. For example, the second reference data may include at least one of the following: vertical acceleration change, pitch rate, roll rate, and suspension height change of each wheel. The vertical acceleration change, pitch rate, and roll rate can be acquired using the vehicle's inertial measurement unit (IMU).

[0069] In step S302, the flatness is determined based on the second reference data.

[0070] In one embodiment, it can be achieved through Figure 4 Steps S3021 and S3032 shown implement the step of determining the flatness based on the second reference data: In step S3021, the degree of change of the second reference data within the first target time period is determined.

[0071] The first target time period is a historical time period ending at the current time and having a first preset duration. For example, the first preset duration could be 2 seconds. For example, the degree of change of the second reference data within the first target time period can be determined in the following way: Determine the integral value of the absolute value of each second reference data point within the first target time period; The degree of change is determined based on the integral value and the weighting coefficient corresponding to each second reference data.

[0072] For example, the absolute values ​​of the changes in vertical acceleration, pitch rate, roll rate, and suspension height of each wheel can be taken. These absolute values ​​are then integrated over a period from the past 2 seconds to the current time. Each integrated value is multiplied by its corresponding weighting coefficient and then summed to represent the degree of change. The weighting coefficients corresponding to each second reference data point are shown in Table 1, and these coefficients can be pre-calibrated based on actual needs.

[0073] Table 1

[0074] In step S3022, the flatness is determined based on the degree of change.

[0075] For example, the flatness can be determined based on the degree of change in the following ways: The flatness is determined based on the degree of change and the distance traveled by vehicles during the first target period.

[0076] The distance traveled by the vehicle during the first target time period can be determined by integrating the vehicle's longitudinal speed. The smoothness can be determined based on the ratio of the degree of change to the vehicle's distance traveled during the corresponding time period; that is, the smoothness can be determined based on the ratio of the degree of change in the vehicle's posture and the vertical position of each wheel during the first target time period to the vehicle's distance traveled. For example, the smoothness can be determined using the following formula. Determination:

[0077] in, To indicate the degree of change, For the first Second reference data, For the first target period, For the first The weighting coefficients of the second reference data, The distance traveled by the vehicle within the first target time period. The number of the second reference data types. ,in, This refers to the longitudinal vehicle speed. The second reference data includes changes in vertical acceleration, pitch rate, roll rate, and changes in suspension height at each wheel, and the vehicle comprises four wheels. The value is 7.

[0078] Turn back Figure 2 In step S202, the first road surface type is determined based on the first reference data.

[0079] In one embodiment, it can be achieved through Figure 5 Steps S2021 and S2022 shown implement the step of determining the first road surface type based on the first reference data: In step S2021, the probability of the current road surface being any of the preset road surface types is determined based on the first reference data.

[0080] For example, the probability of the current road surface being any of the preset road surface types can be determined in the following ways: Determine the sub-probability corresponding to each first reference data point and each preset road surface type; For each preset road surface type, the probability of that preset road surface type is determined based on the sub-probability corresponding to that preset road surface type.

[0081] For example, based on a pre-defined correspondence, the sub-probability for each preset road surface type corresponding to the current first reference data can be determined. Here, the correspondence is the relationship between the first reference data, the preset road surface type, and the corresponding sub-probability. Thus, by looking up the pre-defined correspondence, the sub-probability can be determined simply and quickly.

[0082] Taking the first reference data as a and b, and the preset road surface types as A and B as an example, the sub-probabilities corresponding to the first reference data a and the preset road surface type A can be determined by querying the correspondence. The sub-probabilities corresponding to the first reference data a and the preset road surface type B The sub-probabilities corresponding to the first reference data b and the preset road surface type B The subprobabilities corresponding to the first reference data b and the preset road surface type A The sub-probabilities corresponding to the preset road surface type A include: and The sub-probabilities corresponding to the preset road surface type B include and .

[0083] For example, the probability of a road surface being the preset road surface type can be determined by summing the sub-probabilities corresponding to that preset road surface type. That is, the probability that the current road surface is the preset road surface type A is... The probability that the current road surface is the preset road surface type B is: .

[0084] In step S2022, the first road surface type of the current road surface is determined based on the probability.

[0085] For example, the preset road surface type corresponding to the highest probability can be determined as the first road surface type. That is, if Then, the first road surface type can be determined to be the preset road surface type A.

[0086] In such Figure 2 The road surface type identification method shown can determine the first road surface type based on the signal and by using vehicle dynamics. This process has high interpretability, requires low computing power, and the determined first road surface information has high accuracy.

[0087] For ease of understanding, the following explanation will use the first reference data, including vehicle speed, slip ratio, adhesion coefficient, gradient, rolling resistance coefficient, and smoothness, as an example. To simplify data processing, the first reference data can be categorized, and the processing standards are shown in Table 2. The processing standards in Table 2 are for reference only and can be calibrated according to actual conditions.

[0088] Table 2

[0089] Taking vehicle speed as an example, the probability of the road surface type corresponding to the vehicle speed can be divided into four levels: level 0 is almost impossible, level 1 is low probability, level 2 is medium probability, and level 3 is high probability. Based on big data analysis and testing experience, Table 3 can be obtained. The specific values ​​in this table are for reference only and can be calibrated according to the actual situation.

[0090] Table 3

[0091] Based on Table 3, the probability table shown in Table 4 can be obtained by dividing the probability of road type at a certain speed by the sum of the probability of road type at that speed. Table 4 shows the correspondence between the pre-calibrated speed, the preset road type and the corresponding sub-probability.

[0092] Table 4

[0093] Using the same method, probability tables corresponding to slip ratio, adhesion coefficient, slope, rolling resistance coefficient, and smoothness can be obtained, which will not be repeated here. The probability value (i.e., sub-probability) for each preset pavement type under each first reference data can be obtained by looking up the table. Then, the sub-probabilities of all first reference data corresponding to that preset pavement type can be added together to obtain the probability that the current pavement is of that preset pavement type. The pavement type result with the highest probability among all preset pavement types is taken as the matching pavement type result, i.e., the first pavement type. For example, the following formula can be used to determine the first pavement type. The probability of a preset road surface type :

[0094] in, For the first Subprobabilities corresponding to vehicle speed under a preset road surface type For the first Sub-probabilities corresponding to slip ratios under a preset road surface type For the first Sub-probabilities corresponding to the adhesion coefficient under a preset road surface type For the first Sub-probabilities corresponding to slope under a preset road surface type For the first Sub-probabilities corresponding to rolling resistance coefficients under a preset road surface type For the first Sub-probabilities corresponding to smoothness under a preset road surface type.

[0095] Figure 6 This is a schematic diagram of a road surface recognition method provided in an exemplary embodiment of this disclosure. Figure 6 This allows for a clearer understanding of the implementation process of road surface recognition using the first strategy provided in this disclosure. For example... Figure 6 As shown, the system acquires the vehicle's bus signal and preprocesses it. Then, parameter estimation is performed. Based on the preprocessed bus signal and vehicle dynamics, the system determines the vehicle speed, slip ratio, coefficient of adhesion, gradient, rolling resistance coefficient, and smoothness. Next, the parameters are categorized to determine the corresponding gears for vehicle speed, slip ratio, coefficient of adhesion, gradient, rolling resistance coefficient, and smoothness. A pre-defined road surface type matching probability table is then used for road surface type matching. Finally, the system outputs the gear identification results for the road surface type and road surface features. The road surface features include the coefficient of adhesion, gradient, rolling resistance coefficient, and smoothness. Figure 6 The specific implementation of each step has been described in detail above, and the repeated content will not be repeated here.

[0096] The technical solution for determining the first road surface information based on the signal and using the second strategy will be described below. Figure 7This is a flowchart illustrating road surface recognition using a second strategy, provided in an exemplary embodiment of this disclosure. Figure 7 As shown, the method may include steps S401 and S402.

[0097] In step S401, the target matrix is ​​constructed based on the signal.

[0098] In one embodiment, it can be achieved through, as follows Figure 8 The steps S4011 and S4012 shown implement the step of constructing the target matrix based on the signal.

[0099] In step S4011, third reference data is obtained based on the signal.

[0100] The third reference data refers to data that reflects the vehicle's operating status. This third reference data may include at least one of the following: wheel speeds, wheel accelerations, motor torques, pitch angles, roll angles, IMU six-axis data, wheel suspension heights, steering wheel angles, steering wheel angular velocity, battery current, traction control system (TCS) activation signal, anti-lock braking system (ABS) activation signal, throttle depth, and braking depth. Wheel accelerations can be determined based on wheel speeds.

[0101] In step S4012, a target matrix is ​​constructed based on the values ​​of third reference data at multiple preset time points within the second target time period.

[0102] The second target time period is a historical time period ending at the current time and having a second preset duration. For example, the second preset duration could be 5 seconds. Preset time points can be determined based on a pre-set time step; for example, this time step could be 0.01 seconds, meaning each 0.01-second interval of the second target time period could be defined as a preset time point. The data values ​​for multiple preset time points corresponding to all third reference data can be calculated through interpolation to generate the target matrix.

[0103] For example, the values ​​of the elements in the target matrix are within a preset range, obtained by amplitude processing of the third reference data. For instance, the preset range could be [-1, 1]. Thus, by performing amplitude processing on the data values ​​of the third reference data to ensure that the values ​​of the elements in the target matrix are within the preset range, the standardization requirements can be met, facilitating subsequent uniform quantization, digital storage, or normalization analysis.

[0104] Data of the same type share similar statistical characteristics; therefore, the amplitudes of the same type of third reference data should be processed in the same way to maintain consistency through uniform processing. Data of different types may have significantly different characteristics; therefore, the amplitudes of different types of third reference data can be processed differently to preserve their respective characteristics and avoid information loss or distortion through targeted processing.

[0105] For example, the third reference data can be windowed with a time width of 5 seconds (i.e., the values ​​of the third reference data from the past 5 seconds to the current time can be stored), and the third reference data can be interpolated. The resulting data values ​​can be arranged in the manner shown in Table 5 to obtain the reference matrix.

[0106] Table 5

[0107] As shown in Table 5, the row vectors represent the data values ​​of the third reference data from the past 5 seconds to the current time, with a time step of 0.01s. The column vectors are as follows: 1. Front left wheel speed, 2. Rear left wheel speed, 3. Front right wheel speed, 4. Rear right wheel speed, 5. Front left wheel speed acceleration, 6. Rear left wheel speed acceleration, 7. Front right wheel speed acceleration, 8. Rear right wheel speed acceleration, 9. Front left motor torque, 10. Rear left motor torque, 11. Front right motor torque, 12. Rear right motor torque, 13. Pitch angle, 14. Roll angle, 15. IMU accelerometer x-axis acceleration, 16. IMU accelerometer y-axis acceleration, 17. I... 18. IMU accelerometer z-axis acceleration, 19. IMU gyroscope x-axis angular velocity, 20. IMU gyroscope y-axis angular velocity, 21. IMU gyroscope z-axis angular velocity, 22. Left front suspension height, 23. Left rear suspension height, 24. Right front suspension height, 25. Right rear suspension height, 26. Steering wheel angle, 27. Steering wheel angular velocity, 28. Battery current, 29. TCS activation signal, 30. ABS activation signal, 31. Throttle depth, 32. Brake depth.

[0108] Different amplitude processing methods can be applied to different third reference data in Table 5. For example, the left front wheel speed, left rear wheel speed, right front wheel speed, and right rear wheel speed can be divided by a first preset value; the left front wheel speed acceleration, left rear wheel speed acceleration, right front wheel speed acceleration, and right rear wheel speed acceleration can be divided by a second preset value; the IMU accelerometer z-axis acceleration can be subtracted from a third preset value; and no amplitude processing is required for the TCS activation flag signal and the ABS activation flag signal. In this way, amplitude processing can be performed on each element in the reference matrix to obtain the target matrix. In step S402, the first road surface information is obtained based on the target matrix and the model.

[0109] In one embodiment, the target matrix can be input into the model to obtain first road surface information and the confidence level of the first road surface information.

[0110] As mentioned earlier, this model can be an artificial intelligence model, and this AI recognition model can be trained based on historical data. Training samples can be generated based on historical data, and the historical target matrix in the training samples can be used as input data for model training. The pre-labeled historical road surface information corresponding to the historical target matrix can be used as target input data for model training. The training termination condition of the model can include at least one of the following: the output value of the model's loss function is less than or equal to a preset threshold, or the number of iterations reaches a preset threshold.

[0111] For example, the first road surface feature in the first road surface information output by the artificial intelligence recognition model can be in the form of a gear. The gear can be determined based on the feature range in which the value of the determined road surface feature falls. For example, if the slip ratio is <0.1, the slip ratio can be determined as low gear; if the slip ratio is between 0.1 and 0.4, the slip ratio can be determined as medium gear; if the slip ratio is between 0.4 and 0.7, the slip ratio can be determined as high gear; and if the slip ratio is >0.9, the slip ratio can be determined as extra high gear.

[0112] In such Figure 7 The road surface recognition method shown can identify complex nonlinear features in the signal by using a model based on the signal, so as to achieve efficient and reliable perception of the first road surface information.

[0113] Figure 9 This is a schematic diagram of a road surface recognition method provided in an exemplary embodiment of this disclosure. Figure 9 This allows for a clearer understanding of the implementation process of road surface recognition using the second strategy provided in this disclosure. For example... Figure 9 As shown, the vehicle's bus signal is acquired and preprocessed; then, a target matrix is ​​generated; next, the target matrix is ​​input into the artificial intelligence recognition model to obtain the gear recognition results of road surface type and road surface features, as well as the confidence level. The artificial intelligence recognition model can be pre-built and trained. Figure 9 The specific implementation of each step has been described in detail above, and the repeated content will not be repeated here.

[0114] The following describes the technical solution for determining the first road surface information based on signals using the first and second strategies.

[0115] In an optional implementation, in step S102, determining the first road surface information based on the signal using a first strategy and a second strategy includes: The first road surface information is determined based on the road surface information and corresponding confidence level obtained using the second strategy, and the road surface information obtained using the first strategy.

[0116] Figure 10This is a schematic diagram illustrating the determination of road surface information using a first strategy and a second strategy, provided in an exemplary embodiment of this disclosure. Figure 10 As shown, the road surface information obtained using the second strategy and the road surface information obtained using the first strategy are fused and arbitrated to obtain the final output first road surface information.

[0117] Figure 11 This is a flowchart illustrating road surface type determination using a first strategy and a second strategy, provided in an exemplary embodiment of this disclosure. Figure 11 As shown, the process may include steps S501 to S505.

[0118] In step S501, it is determined whether the second road surface type and the third road surface type are consistent. If yes, then step S503 is executed; if no, then step S502 is executed.

[0119] The second road surface type is the road surface type determined based on the first strategy, and the third road surface type is the road surface type determined based on the second strategy.

[0120] In step S502, it is determined whether the confidence level of the third road surface type is greater than a preset first threshold. If yes, then step S503 is executed; otherwise, step S504 is executed.

[0121] In step S503, the third road surface type is determined as the first road surface type.

[0122] In step S504, the first road surface type of the previous moment is determined as the first road surface type of the current road surface.

[0123] In step S505, the first road surface type of the current road surface is output.

[0124] For example, if the second road surface type is inconsistent with the third road surface type, the confidence level of the third road surface type can be used to determine the final output recognition result.

[0125] For example, the first threshold can be preset based on actual needs, for instance, it can be set to 0.9. If the confidence level of the third road surface type is greater than the preset first threshold, it can be determined that the reliability of the road surface type determined based on the second strategy is high enough, that is, the reliability of the third road surface type is high enough. Determining the third road surface type as the first road surface type can ensure the reliability of the output first road surface type.

[0126] Conversely, if the confidence level of the third road surface type is less than or equal to the preset first threshold, it can be determined that the reliability of the third road surface type determined based on the second strategy is generally low, while the second road surface type determined based on the first strategy also has a certain degree of reliability. However, the second road surface type and the third road surface type are inconsistent at this time, making it difficult to determine which strategy yields a more reliable road surface type. Since the probability of a vehicle traveling to a new road surface type in a short period is low, the first road surface type from the previous moment can be determined as the first road surface type for the current road surface to ensure the reliability of the output first road surface type.

[0127] In such Figure 11 The road surface type recognition method shown can improve the accuracy and robustness of road surface type recognition by using a dual-strategy complementary fusion.

[0128] Figure 12 This is a flowchart illustrating road surface feature determination using a first strategy and a second strategy, provided in an exemplary embodiment of this disclosure. Figure 12 As shown, the process may include steps S601 to S605.

[0129] In step S601, it is determined whether the second road surface feature and the third road surface feature are consistent. If yes, then step S603 is executed; if no, then step S602 is executed.

[0130] The second road surface feature is the road surface feature determined based on the first strategy, and the third road surface feature is the road surface feature determined based on the second strategy.

[0131] For example, it can be determined whether the second road surface feature and the third road surface feature are consistent by comparing whether the value of the second road surface feature and the value of the third road surface feature are within the same feature range, that is, by comparing whether the gear position recognition result of the second road surface feature and the gear position recognition result of the third road surface feature are consistent.

[0132] If the value of the second road surface feature is within the same feature range as the value of the third road surface feature, that is, the gear identification result of the second road surface feature is consistent with the gear identification result of the third road surface feature, then it can be determined that the second road surface feature is consistent with the third road surface feature.

[0133] Conversely, if the value of the second road surface feature and the value of the third road surface feature are not in the same feature range, that is, the gear identification results of the second road surface feature and the gear identification results of the third road surface feature are not in the same range, then it can be determined that the second road surface feature and the third road surface feature are not in the same range.

[0134] If they match, the second road surface feature can be identified as the first road surface feature. If they do not match, the confidence level of the third road surface feature can be used to determine the final output recognition result.

[0135] In step S602, is the confidence level of the third road surface feature less than a preset second threshold? If yes, proceed to step S603; otherwise, proceed to step S604.

[0136] For example, if the second threshold can be preset based on actual needs, for instance, it can be set to 0.6. If the confidence level of the third road surface feature is less than the preset first threshold, it can be determined that the reliability of the road surface feature determined based on the second strategy is insufficient. In contrast, the reliability of the second road surface feature determined based on the first strategy is higher. In this case, determining the second road surface feature as the first road surface feature can ensure the reliability of the output second road surface feature.

[0137] In short, if the confidence level of the third road surface feature is greater than or equal to the preset first threshold, then the third road surface feature determined based on the second strategy can be considered to have a certain degree of reliability. The second road surface feature determined based on the first strategy also has a certain degree of reliability, but at this point, the second road surface feature is inconsistent with the third road surface feature, making it difficult to determine which strategy yields a more reliable road surface feature. Furthermore, the probability of a vehicle traveling to a road surface with new features in a short period is low. Therefore, the first road surface feature from the previous moment can be used as the first road surface feature for the current road surface to ensure the reliability of the output first road surface feature.

[0138] In step S603, the second road surface feature is determined as the first road surface feature.

[0139] In step S604, the first road surface feature of the previous moment is determined as the first road surface feature of the current road surface.

[0140] In step S605, the first road surface feature of the current road surface is output.

[0141] In such Figure 12 The road feature recognition method shown can improve the accuracy and robustness of road feature recognition by using a dual-strategy complementary fusion.

[0142] Based on the same concept, embodiments of this disclosure also provide a road surface recognition device. Figure 13 This is a flowchart of a road surface recognition device 700 provided in an exemplary embodiment of this disclosure. Figure 13 As shown, the road surface recognition device 700 may include: Acquisition module 701 is used to acquire vehicle signals; The determination module 702 is used to determine first road surface information based on the signal using a first strategy and / or a second strategy, wherein the first strategy is based on vehicle dynamics to determine the first road surface information, and the second strategy is based on a model to determine the first road surface information.

[0143] In the above technical solution, the first strategy, based on vehicle dynamics, enables accurate determination of the first road surface information and possesses high interpretability; the second strategy, based on a model, enables the capture of complex nonlinear features to achieve accurate determination of the first road surface information; the complementarity of the two strategies further improves the accuracy of road surface recognition. Thus, while reducing sensor and computing power costs, accurate determination of the first road surface information can be achieved, providing a reliable basis for rapid adaptation of subsequent control strategies, thereby enhancing driving safety and comfort.

[0144] Optionally, the first road surface information includes a first road surface type; the determining module 702 includes: A first determining submodule is configured to determine first reference data of the vehicle based on the signal, wherein the first reference data includes at least one of the following: vehicle speed, slip ratio, and at least one of a first road surface feature; The second determining submodule is used to determine the first road surface type based on the first reference data.

[0145] Optionally, the second determining submodule is used to determine the first road surface type in the following manner: Based on the first reference data, determine the probability that the current road surface is one of each preset road surface type; Based on the probability, determine the first road surface type of the current road surface.

[0146] Optionally, the second determining submodule is used to determine the probability that the current road surface is each preset road surface type by means of: Determine the sub-probability corresponding to each of the first reference data and each of the preset road surface types; For each of the preset road surface types, the probability of that preset road surface type is determined based on the sub-probability corresponding to that preset road surface type.

[0147] Optionally, the second determining submodule is used to determine the sub-probability corresponding to each of the first reference data and each of the preset road surface types in the following manner: Based on the pre-defined correspondence, the sub-probability under each preset road surface type corresponding to the current first reference data is determined, wherein the correspondence is the relationship between the first reference data, the preset road surface type and the corresponding sub-probability.

[0148] Optionally, the second determining submodule is used to determine the probability of the preset road surface type based on the sub-probability corresponding to the preset road surface type in the following manner: The sum of the subprobabilities corresponding to the preset road surface type is determined as the probability of the preset road surface type.

[0149] Optionally, the second determining submodule is used to determine the first road surface type of the current road surface based on the probability in the following manner: The preset road surface type corresponding to the highest probability is determined as the first road surface type.

[0150] Optionally, the first road surface information includes a first road surface feature, which includes smoothness; the determining module 702 includes: The third determining submodule is used to determine the second reference data based on the signal, wherein the second reference data is data that can reflect the degree of vehicle bumps; The fourth determining submodule is used to determine the flatness based on the second reference data.

[0151] Optionally, the fourth determining submodule is used to determine the flatness in the following manner: Determine the degree of change of the second reference data within the first target time period, where the first target time period is a historical time period with the current time as the end time and a first preset duration; The flatness is determined based on the degree of change.

[0152] Optionally, the fourth determining submodule is used to determine the degree of change of the second reference data during the first target time period by means of: Determine the integral value of the absolute value of each of the second reference data points within the first target time period; The degree of change is determined based on the integral value and the weighting coefficient corresponding to each of the second reference data.

[0153] Optionally, the fourth determining submodule is used to determine the flatness based on the degree of change in the following manner: The smoothness is determined based on the degree of change and the distance traveled by the vehicle during the first target time period.

[0154] Optionally, the fourth determining submodule is used to determine the flatness using the following formula. :

[0155] in, The degree of change, For the first Second reference data, For the first target time period, For the first The weighting coefficients of the second reference data, This represents the distance traveled by the vehicle during the first target time period. The number of the second reference data types.

[0156] Optionally, the first road surface information includes a first road surface type and / or a first road surface feature, and the determining module 702 includes: The fifth determining submodule is used to construct the target matrix based on the signal; The sixth determining submodule is used to obtain the first road surface information based on the target matrix and the model.

[0157] Optionally, the fifth determining submodule is used to construct the target matrix based on the signal in the following manner: Based on the signal, a third reference data is obtained, which is data that can reflect the vehicle's operating status; The target matrix is ​​constructed based on the values ​​of third reference data at multiple preset time points within the second target time period, where the second target time period is a historical time period with the current time as the end time and a second preset duration.

[0158] Optionally, the values ​​of the elements in the target matrix are within a preset range and are obtained by amplitude processing of the third reference data, with the same amplitude processing method for the same type of third reference data.

[0159] Optionally, the sixth determining submodule is used to obtain the first road surface information based on the target matrix and the model in the following manner: The target matrix is ​​input into the model to obtain the first road surface information and the confidence level of the first road surface information.

[0160] Optionally, the determining module 702 includes: The seventh determining submodule is used to determine the first road surface information based on the road surface information and corresponding confidence level obtained using the second strategy, and the road surface information obtained using the first strategy.

[0161] Optionally, the first road surface information includes a first road surface type; the seventh determining submodule is used to determine the third road surface type as the first road surface type if any of the following conditions are met: The second road surface type is the same as the third road surface type; The second road surface type is inconsistent with the third road surface type, and the confidence level of the third road surface type is greater than a preset first threshold. Wherein, the second road surface type is the road surface type determined based on the first strategy, and the third road surface type is the road surface type determined based on the second strategy.

[0162] Optionally, the first road surface information includes a first road surface type; the seventh determining submodule is used to determine the first road surface type of the previous moment as the first road surface type of the current road surface if the second road surface type is inconsistent with the third road surface type and the confidence level of the third road surface type is less than or equal to a preset first threshold. Wherein, the second road surface type is the road surface type determined based on the first strategy, and the third road surface type is the road surface type determined based on the second strategy.

[0163] Optionally, the first road surface information includes a first road surface feature; the seventh determining submodule is used to determine the second road surface feature as the first road surface feature if any of the following conditions are met: The second road surface features are consistent with the third road surface features; The second road surface feature is inconsistent with the third road surface feature, and the confidence level of the third road surface feature is less than a preset second threshold. The second road surface feature is the road surface feature determined based on the first strategy, and the third road surface feature is the road surface feature determined based on the second strategy.

[0164] Optionally, the first road surface information includes a first road surface feature; the seventh determining submodule is used to determine the first road surface feature of the previous moment as the first road surface feature of the current road surface if the second road surface feature is inconsistent with the third road surface feature and the confidence level of the third road surface feature is greater than or equal to a preset second threshold. The second road surface feature is the road surface feature determined based on the first strategy, and the third road surface feature is the road surface feature determined based on the second strategy.

[0165] Optionally, the first road surface information includes a first road surface feature, which includes at least one of the following: adhesion coefficient, slope, rolling resistance coefficient, and smoothness.

[0166] Optionally, the acquired signal is the vehicle's bus signal.

[0167] Based on the same concept, this disclosure also provides a controller, which includes: processor; Memory used to store processor-executable instructions; The processor is configured to execute the steps of the road surface recognition method described above.

[0168] Based on the same concept, embodiments of this disclosure also provide a vehicle including the controller described above.

[0169] Based on the same concept, embodiments of this disclosure also provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the road surface recognition method described above.

[0170] Based on the same concept, this disclosure also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the road surface recognition method described above.

[0171] The preferred embodiments of this disclosure have been described in detail above with reference to the accompanying drawings. However, this disclosure is not limited to the specific details of the above embodiments. Within the scope of the technical concept of this disclosure, various simple modifications can be made to the technical solutions of this disclosure, and these simple modifications all fall within the protection scope of this disclosure.

[0172] It should also be noted that the various specific technical features described in the above specific embodiments can be combined in any suitable manner without contradiction. In order to avoid unnecessary repetition, this disclosure will not describe the various possible combinations separately.

[0173] Furthermore, various different embodiments of this disclosure can be combined in any way, as long as they do not violate the spirit of this disclosure, they should also be regarded as the content disclosed in this disclosure.

Claims

1. A road surface recognition method, characterized in that, include: Acquire vehicle signals; Based on the signal, first road surface information is determined using a first strategy and / or a second strategy, wherein the first strategy is based on vehicle dynamics to determine the first road surface information, and the second strategy is based on a model to determine the first road surface information.

2. The road surface recognition method according to claim 1, characterized in that, The first road surface information includes the first road surface type; Based on the signal, the first road surface type is determined using a first strategy, including: Based on the signal, first reference data of the vehicle is determined, the first reference data including at least one of the following: vehicle speed, slip ratio and at least one of a first road surface feature; The first road surface type is determined based on the first reference data.

3. The road surface recognition method according to claim 2, characterized in that, Determining the first road surface type based on the first reference data includes: Based on the first reference data, determine the probability that the current road surface is one of each preset road surface type; Based on the probability, determine the first road surface type of the current road surface.

4. The road surface recognition method according to claim 3, characterized in that, The step of determining the probability that the current road surface belongs to each preset road surface type based on the first reference data includes: Determine the sub-probability corresponding to each of the first reference data and each of the preset road surface types; For each of the preset road surface types, the probability of that preset road surface type is determined based on the sub-probability corresponding to that preset road surface type.

5. The road surface recognition method according to claim 4, characterized in that, Determining the sub-probability corresponding to each of the first reference data and each of the preset road surface types includes: Based on the pre-defined correspondence, the sub-probability under each preset road surface type corresponding to the current first reference data is determined, wherein the correspondence is the relationship between the first reference data, the preset road surface type and the corresponding sub-probability.

6. The road surface recognition method according to claim 4, characterized in that, The step of determining the probability of the preset road surface type based on the sub-probability corresponding to the preset road surface type includes: The sum of the subprobabilities corresponding to the preset road surface type is determined as the probability of the preset road surface type.

7. The road surface recognition method according to claim 3, characterized in that, Determining the first road surface type of the current road surface based on the probability includes: The preset road surface type corresponding to the highest probability is determined as the first road surface type.

8. The road surface recognition method according to claim 1, characterized in that, The first road surface information includes a first road surface feature, which includes smoothness; based on the signal, the smoothness is determined using a first strategy: Based on the signal, a second reference data is determined, which is data that can reflect the degree of vehicle bumps; The flatness is determined based on the second reference data.

9. The road surface recognition method according to claim 8, characterized in that, Determining the flatness based on the second reference data includes: Determine the degree of change of the second reference data within the first target time period, where the first target time period is a historical time period with the current time as the end time and a first preset duration; The flatness is determined based on the degree of change.

10. The road surface recognition method according to claim 9, characterized in that, Determining the degree of change of the second reference data within the first target time period includes: Determine the integral value of the absolute value of each of the second reference data points within the first target time period; The degree of change is determined based on the integral value and the weighting coefficient corresponding to each of the second reference data.

11. The road surface recognition method according to claim 9, characterized in that, Determining the smoothness based on the degree of change includes: The smoothness is determined based on the degree of change and the distance traveled by the vehicle during the first target time period.

12. The road surface recognition method according to claim 11, characterized in that, Determining the smoothness based on the degree of change and the vehicle's travel distance within the first target time period includes: The flatness is determined by the following formula. : in, The degree of change, For the first Second reference data, For the first target time period, For the first The weighting coefficients of the second reference data, This represents the distance traveled by the vehicle during the first target time period. The number of the second reference data types.

13. The road surface recognition method according to claim 1, characterized in that, The first road surface information includes a first road surface type and / or a first road surface feature. Based on the signal, the first road surface information is determined using a second strategy, including: Based on the signal, construct the target matrix; The first road surface information is obtained based on the target matrix and the model.

14. The road surface recognition method according to claim 13, characterized in that, The step of constructing the target matrix based on the signal includes: Based on the signal, a third reference data is obtained, which is data that can reflect the vehicle's operating status; The target matrix is ​​constructed based on the values ​​of third reference data at multiple preset time points within the second target time period, where the second target time period is a historical time period with the current time as the end time and a second preset duration.

15. The road surface recognition method according to claim 14, characterized in that, The values ​​of the elements in the target matrix are within a preset range and are obtained by amplitude processing of the third reference data. The amplitude processing method is the same for the same type of third reference data.

16. The road surface recognition method according to claim 13, characterized in that, The step of obtaining the first road surface information based on the target matrix and the model includes: The target matrix is ​​input into the model to obtain the first road surface information and the confidence level of the first road surface information.

17. The road surface recognition method according to claim 1, characterized in that, Determining the first road surface information using a first strategy and a second strategy includes: The first road surface information is determined based on the road surface information and corresponding confidence level obtained using the second strategy, and the road surface information obtained using the first strategy.

18. The road surface recognition method according to claim 17, characterized in that, The first road surface information includes a first road surface type; determining the first road surface type based on the road surface information obtained using the second strategy and the corresponding confidence level, and the road surface information obtained using the first strategy, includes: The third road surface type is determined to be the first road surface type if any of the following conditions are met: The second road surface type is the same as the third road surface type; The second road surface type is inconsistent with the third road surface type, and the confidence level of the third road surface type is greater than a preset first threshold. Wherein, the second road surface type is the road surface type determined based on the first strategy, and the third road surface type is the road surface type determined based on the second strategy.

19. The road surface recognition method according to claim 17, characterized in that, The first road surface information includes a first road surface type; determining the first road surface type based on the road surface information obtained using the second strategy and the corresponding confidence level, and the road surface information obtained using the first strategy, includes: If the second road surface type is inconsistent with the third road surface type, and the confidence level of the third road surface type is less than or equal to the preset first threshold, then the first road surface type of the previous moment is determined as the first road surface type of the current road surface. Wherein, the second road surface type is the road surface type determined based on the first strategy, and the third road surface type is the road surface type determined based on the second strategy.

20. The road surface recognition method according to claim 17, characterized in that, The first road surface information includes the first road surface features; Based on the road surface information and corresponding confidence level obtained using the second strategy, and the road surface information obtained using the first strategy, the first road surface feature is determined, including: The second road surface feature is determined to be the first road surface feature if any of the following conditions are met: The second road surface features are consistent with the third road surface features; The second road surface feature is inconsistent with the third road surface feature, and the confidence level of the third road surface feature is less than a preset second threshold. The second road surface feature is the road surface feature determined based on the first strategy, and the third road surface feature is the road surface feature determined based on the second strategy.

21. The road surface recognition method according to claim 17, characterized in that, The first road surface information includes the first road surface features; Based on the road surface information and corresponding confidence level obtained using the second strategy, and the road surface information obtained using the first strategy, the first road surface feature is determined, including: If the second road surface feature is inconsistent with the third road surface feature, and the confidence level of the third road surface feature is greater than or equal to the preset second threshold, then the first road surface feature of the previous moment is determined as the first road surface feature of the current road surface. The second road surface feature is the road surface feature determined based on the first strategy, and the third road surface feature is the road surface feature determined based on the second strategy.

22. The road surface recognition method according to any one of claims 1-21, characterized in that, The first road surface information includes a first road surface feature, which includes at least one of the following: adhesion coefficient, slope, rolling resistance coefficient, and smoothness.

23. The road surface recognition method according to any one of claims 1-21, characterized in that, The acquired signal is the vehicle's bus signal.

24. A controller, characterized in that, include: processor; Memory used to store processor-executable instructions; The processor is configured to execute the executable instructions in the memory to implement the steps of the road surface recognition method according to any one of claims 1-23.

25. A vehicle, characterized in that, Includes the controller as described in claim 24.

26. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the road surface recognition method according to any one of claims 1-23.

27. A computer program product, characterized in that, It includes a computer program that, when executed by a processor, implements the steps of the road surface recognition method according to any one of claims 1-23.

Images