Data-intelligent driven adaptive federated observation method and device, vehicle and medium

Abstract

The application relates to the field of vehicle dynamics and control technology, in particular to a data intelligent driving adaptive joint observation method and device, a vehicle and a medium, wherein the method comprises the following steps: compensating wheel speed according to yaw rate and rotation angle; calculating a longitudinal reference vehicle speed measurement value according to the compensated wheel speed and longitudinal acceleration; respectively estimating noise covariance of a lateral reference vehicle speed measurement value and a lateral-longitudinal reference vehicle speed measurement value according to a first input feature, a second input feature and a third input feature; determining a joint observation weight of the lateral-longitudinal reference vehicle speed measurement value based on the noise covariance, and performing joint observation on a lateral-longitudinal reference vehicle speed estimation value based on the joint observation weight, so as to obtain a vehicle body side slip angle estimation value. Therefore, the problems that the MBD method and the DDD method in the related art have limitations, the applicability and robustness are poor, the measurement value is inaccurate, the accuracy of the observation result is low, a large amount of data is required, and the application is limited due to the black box feature are solved.

Description

Technical Field

[0001] This application relates to the field of automotive dynamics and control technology, and in particular to a data-driven intelligent adaptive joint observation method, device, vehicle, and medium. Background Technology

[0002] Key states in vehicle dynamics mainly include longitudinal reference vehicle speed and body slip angle. Based on different principles, the observation methods for key states in vehicle dynamics are mainly divided into two categories: MBD (Model-Based Design) and DDD (Data-Driven Design).

[0003] MBD mainly builds key state observers based on vehicle dynamics or kinematic models; DDD mainly uses real vehicle test data to train neural networks for key state observation, and the input signals can be directly output as observation results after being processed by the neural network.

[0004] In related technologies, the MBD method based on vehicle dynamics typically requires known additional parameters such as tire forces and demands high accuracy in vehicle dynamics and tire mechanics models, resulting in poor applicability. The MBD method based on vehicle kinematics can use relatively simple kinematic equations, but its design rules often heavily rely on human experience, leading to poor robustness. The DDD method overcomes, to some extent, the limitations of the MBD method based on vehicle dynamics in the nonlinear region, but it cannot handle random noise in sensor signals, resulting in low accuracy of observations. Furthermore, the DDD method typically requires large amounts of data and exhibits black-box characteristics, limiting its application scope. Summary of the Invention

[0005] This application provides a data intelligence-driven adaptive joint observation method, device, vehicle, and medium to address the limitations of MBD and DDD methods in related technologies, which result in poor applicability and robustness, inaccurate measurement values ​​leading to low accuracy of observation results, and the requirement for large amounts of data and black-box characteristics that limit their application.

[0006] The first aspect of this application provides a data-driven adaptive joint observation method, comprising the following steps: acquiring vehicle yaw rate signals, wheel speed signals, steering angle signals, lateral and longitudinal acceleration signals, and driving and braking torque signals, and using one or more signals as data features; forming first to third input features by combining data features with preset time steps; compensating the wheel speed based on the yaw rate and the steering angle; calculating a longitudinal reference vehicle speed measurement based on the compensated wheel speed and longitudinal acceleration; estimating a lateral reference vehicle speed measurement based on the first input feature, and estimating the noise covariance of the longitudinal reference vehicle speed measurement and the lateral reference vehicle speed measurement based on the second and third input features, respectively; determining a joint observation weight for the longitudinal reference vehicle speed measurement and the lateral reference vehicle speed measurement based on the noise covariance; and jointly observing the estimated longitudinal reference vehicle speed and the estimated lateral reference vehicle speed based on the joint observation weight to obtain an estimated vehicle body slip angle.

[0007] Optionally, the wheel speed compensation formula is:

[0008]

[0009] Among them, W ij It is the wheel speed signal, ij∈{fl,fr,rl,rr}; W ij c It is the compensated wheel speed signal; It is the yaw rate signal; δ is the front wheel steering angle signal; k f k r These are the track widths of the front and rear axles, respectively.

[0010] Optionally, the step of calculating the longitudinal reference vehicle speed measurement based on the compensated wheel speed and longitudinal acceleration includes: obtaining an upper threshold and a lower threshold for the longitudinal acceleration; if the longitudinal acceleration is greater than or equal to the lower threshold and less than or equal to the upper threshold, then the average of the compensated wheel speed is used as the longitudinal reference vehicle speed measurement; if the longitudinal acceleration is less than the lower threshold, then when the longitudinal acceleration is greater than the derivative of the estimated longitudinal reference vehicle speed at the current moment, the maximum value between the compensated wheel speed and the longitudinal reference vehicle speed measurement at the current moment is used as the longitudinal reference vehicle speed measurement; in the longitudinal... When the acceleration is less than or equal to the derivative of the estimated longitudinal reference vehicle speed at the current moment, the maximum value among the compensated wheel speeds is taken as the measured longitudinal reference vehicle speed. If the longitudinal acceleration is greater than the upper limit threshold, when the longitudinal acceleration is less than the derivative of the estimated longitudinal reference vehicle speed at the current moment, the minimum value among the compensated wheel speed and the measured longitudinal reference vehicle speed at the current moment is taken as the measured longitudinal reference vehicle speed. When the longitudinal acceleration is greater than or equal to the derivative of the estimated longitudinal reference vehicle speed at the current moment, the minimum value among the compensated wheel speeds is taken as the measured longitudinal reference vehicle speed.

[0011] Optionally, before calculating the longitudinal reference vehicle speed measurement based on the compensated wheel speed and longitudinal acceleration, the method further includes: obtaining an estimated value of the longitudinal reference vehicle speed at the previous moment; calculating the change in the estimated value based on the time difference between the current moment and the previous moment and the longitudinal acceleration; and calculating the longitudinal reference vehicle speed measurement at the current moment based on the estimated value of the longitudinal reference vehicle speed at the previous moment and the change in the estimated value.

[0012] Optionally, the formula for calculating the lateral reference vehicle speed and noise covariance is:

[0013]

[0014] Among them, R x N R represents the noise covariance of the longitudinal reference vehicle speed measurement. y N V represents the noise covariance of the lateral reference vehicle speed measurement. y N The value represents the lateral reference vehicle speed measurement, k represents the current sampling time, and X1, X2, and X3 represent different input features. The input features refer to the time series of data features within a preset historical time step. y,vel The nonlinear fitting function f of the lateral reference vehicle speed estimation module. x,cov The nonlinear fitting function f of the noise covariance estimation module for longitudinal reference vehicle speed measurements. y,covThis refers to the nonlinear fitting function of the noise covariance estimation module for the lateral reference vehicle speed measurement.

[0015] Optionally, the lateral reference vehicle speed and noise covariance are fitted using a structurally similar neural network structure, wherein the neural network structure includes a long short-time neural network layer, a fully connected layer, a batch normalization layer, and a multilayer perceptron processing layer.

[0016] Optionally, determining the joint observation weight of the longitudinal reference vehicle speed measurement and the lateral reference vehicle speed measurement based on the noise covariance includes: if the noise covariance increases, then decreasing the joint observation weight of the longitudinal reference vehicle speed measurement; if the noise covariance decreases, then increasing the joint observation weight of the longitudinal reference vehicle speed measurement.

[0017] A second aspect of this application provides a data-driven adaptive joint observation device, comprising: a data storage module for acquiring vehicle yaw rate signals, wheel speed signals, steering angle signals, lateral and longitudinal acceleration signals, and driving and braking torque signals, and using one or more signals as data features, and forming first to third input features by combining data features with preset time steps; a yaw compensation module for compensating the wheel speed based on the yaw rate and the steering angle; a wheel speed signal processing module for calculating a longitudinal reference vehicle speed measurement based on the compensated wheel speed and longitudinal acceleration; a data-driven estimation module for estimating a lateral reference vehicle speed measurement based on a first input feature, and estimating the noise covariance of the longitudinal reference vehicle speed measurement and the lateral reference vehicle speed measurement based on a second input feature and a third input feature, respectively; and an adaptive joint observation module for determining a joint observation weight for the longitudinal reference vehicle speed measurement and the lateral reference vehicle speed measurement based on the noise covariance, and performing joint observation of the longitudinal reference vehicle speed estimate and the lateral reference vehicle speed estimate based on the joint observation weight, thereby obtaining a vehicle body slip angle estimate.

[0018] Optionally, the wheel speed compensation formula is:

[0019]

[0020] Among them, W ij It is the wheel speed signal, ij∈{fl,fr,rl,rr}; W ij c It is the compensated wheel speed signal; It is the yaw rate signal; δ is the front wheel steering angle signal; k f k r These are the track widths of the front and rear axles, respectively.

[0021] Optionally, the wheel speed signal processing module is further configured to: obtain an upper threshold and a lower threshold for the longitudinal acceleration; if the longitudinal acceleration is greater than or equal to the lower threshold and less than or equal to the upper threshold, then use the average value of the compensated wheel speed as the longitudinal reference vehicle speed measurement value; if the longitudinal acceleration is less than the lower threshold, then when the longitudinal acceleration is greater than the derivative of the estimated longitudinal reference vehicle speed at the current moment, use the maximum value between the compensated wheel speed and the longitudinal reference vehicle speed measurement value at the current moment as the longitudinal reference vehicle speed measurement value; when the longitudinal acceleration is less than or equal to ... or equal to the lower threshold, use the maximum value between the compensated wheel speed and the longitudinal reference vehicle speed measurement value at the current moment as the longitudinal reference vehicle speed measurement value; when the longitudinal acceleration is less than or equal to the lower threshold, then when the longitudinal acceleration is greater than or equal to the lower threshold, use the maximum value between the compensated wheel speed and the longitudinal reference vehicle speed measurement value at the current moment as the longitudinal reference vehicle speed measurement value. When the derivative of the estimated longitudinal reference vehicle speed at the current moment is taken, the maximum value among the compensated wheel speeds is used as the measured longitudinal reference vehicle speed. If the longitudinal acceleration is greater than the upper limit threshold, then when the longitudinal acceleration is less than the derivative of the estimated longitudinal reference vehicle speed at the current moment, the minimum value among the compensated wheel speed and the measured longitudinal reference vehicle speed at the current moment is used as the measured longitudinal reference vehicle speed. When the longitudinal acceleration is greater than or equal to the derivative of the estimated longitudinal reference vehicle speed at the current moment, the minimum value among the compensated wheel speeds is used as the measured longitudinal reference vehicle speed.

[0022] Optionally, it further includes: an acquisition module, configured to acquire an estimated value of the longitudinal reference vehicle speed at the previous moment before calculating the longitudinal reference vehicle speed measurement value based on the compensated wheel speed and longitudinal acceleration; calculate the change in the estimated value based on the time difference between the current moment and the previous moment and the longitudinal acceleration; and calculate the longitudinal reference vehicle speed measurement value at the current moment based on the estimated value of the longitudinal reference vehicle speed at the previous moment and the change in the estimated value.

[0023] Optionally, the formula for calculating the lateral reference vehicle speed and noise covariance is:

[0024]

[0025] Among them, R x N R is expressed as the noise covariance of the longitudinal reference vehicle speed measurement. y N V is represented as the lateral reference vehicle speed noise covariance. y N This represents the estimated value of the lateral reference vehicle speed measurement, k represents the current sampling time, X1, X2, and X3 represent different input features, and the input features refer to the time series of data features within a preset historical time step, f y,vel The nonlinear fitting function f of the lateral reference vehicle speed estimation module. x,cov The nonlinear fitting function f of the longitudinal reference vehicle speed measurement noise covariance estimation module. y,cov This refers to the nonlinear fitting function of the noise covariance estimation module for lateral reference vehicle speed measurements.

[0026] Optionally, the lateral reference vehicle speed and noise covariance are fitted using a structurally similar neural network structure, wherein the neural network structure includes a long short-time neural network layer, a fully connected layer, a batch normalization layer, and a multilayer perceptron processing layer.

[0027] Optionally, the adaptive joint observation module is further configured to: decrease the joint observation weight of the longitudinal reference vehicle speed if the noise covariance increases; and increase the joint observation weight of the longitudinal reference vehicle speed if the noise covariance decreases.

[0028] A third aspect of this application provides a vehicle comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the data intelligence-driven adaptive joint observation method as described in the above embodiments.

[0029] A fourth aspect of this application provides a computer-readable storage medium having a computer program stored thereon, which is executed by a processor to implement the data intelligence-driven adaptive joint observation method as described in the above embodiments.

[0030] Therefore, this application has at least the following beneficial effects:

[0031] (1) The embodiments of this application can quickly and conveniently obtain the measurement value of longitudinal reference vehicle speed, and can ensure that the estimation results have good accuracy under acceleration, deceleration and steady-state conditions.

[0032] (2) The embodiments of this application can directly fit the measurement value of the lateral reference vehicle speed according to the data feature time series, avoiding the problem of solving complex nonlinear vehicle dynamics models, with high applicability and good accuracy of the estimation results; the noise covariance of the measurement values ​​of the longitudinal reference vehicle speed and the lateral reference vehicle speed can be fitted according to the data feature time series respectively, which intuitively reflects the accuracy of the measurement values.

[0033] (3) In this embodiment of the application, considering the uncertainty of the measurement values, the noise covariance of the measurement values ​​of the longitudinal reference vehicle speed and the lateral reference vehicle speed is estimated by a data-driven intelligent algorithm, which can realize the adaptive adjustment of the weights of key state quantities of vehicle dynamics, thereby enhancing the robustness of the joint observation method.

[0034] (4) The data-driven key state adaptive joint observation method constructed in this application embodiment can be applied to various powertrain models and has the characteristics of data-driven, adaptive and high-precision compared with traditional observation algorithms.

[0035] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0036] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein:

[0037] Figure 1 This is a flowchart of a data intelligence-driven adaptive joint observation method provided according to an embodiment of this application;

[0038] Figure 2 This is a schematic diagram of the wheel speed signal processing module provided according to an embodiment of this application;

[0039] Figure 3 This is a diagram of a data-driven, intelligently-driven, adaptive joint observation architecture for critical states of vehicle dynamics, provided in the embodiments of this application.

[0040] Figure 4 This is a block diagram of a data-driven adaptive joint observation device according to an embodiment of this application;

[0041] Figure 5 This is a structural schematic diagram of a vehicle according to an embodiment of this application. Detailed Implementation

[0042] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.

[0043] Currently, the methods for observing critical states in vehicle dynamics are mainly divided into two categories: MBD and DDD. Specifically:

[0044] (1) MBD mainly constructs observers for key states based on vehicle dynamics or kinematic models. Vehicle dynamics-based MBD methods can use high-precision observation methods such as KF (Kalman Filter), EKF (Extended Kalman Filter), CKF (Cubature Kalman Filter), and SCKF (Squared-root Cubature Kalman Filter), but usually require additional parameters such as tire forces to be known, and have high accuracy requirements for vehicle dynamics and tire mechanics models.

[0045] The MBD method based on vehicle kinematics can use relatively simple kinematic equations, and usually only wheel speed signals and acceleration signals are needed to achieve high-precision observation. If it is necessary to improve the robustness of the observation method, KF-based MMSM (Multi-mode Switching Method) and MSFM (Multi-sensor Fusion Method) can be used, but the design of related rules depends heavily on experience.

[0046] (2) DDD mainly uses real vehicle test data to train a neural network for key state observation. The input signal can be directly output as the observation result after being processed by the neural network. The neural network can achieve high-precision fitting of complex nonlinear systems, which to some extent overcomes the limitations of MBD based on vehicle dynamics in the nonlinear region.

[0047] Commonly used neural networks include Fully Connected Neural Networks (FCNN), Radial Basis Neural Networks (RBNN), Feed-forward Neural Networks (FFNN), Gate Recurrent Unit Neural Networks (GRUNN), and Long Short-Term Memory Neural Networks (LSTMNN). FFNNs can reflect the nonlinear relationship between input and output, but they cannot represent temporal information. RNNs, because they can map temporal inputs to temporal outputs, are more suitable for modeling inertial systems. Although neural networks have general fitting characteristics, their large data requirements and black-box nature limit the widespread application of data-driven methods. Furthermore, while these methods avoid the difficulty of directly solving nonlinear models, they cannot handle random noise in sensor signals.

[0048] The following description, with reference to the accompanying drawings, describes a data-driven adaptive joint observation method, apparatus, vehicle, and medium according to embodiments of this application. Specifically, Figure 1 This is a flowchart illustrating a data intelligence-driven adaptive joint observation method provided in an embodiment of this application.

[0049] like Figure 1 As shown, this data-driven adaptive joint observation method includes the following steps:

[0050] In step S101, the vehicle's yaw rate signal, wheel speed signal, steering angle signal, lateral and longitudinal acceleration signal, and driving and braking torque signal are acquired, and one or more signals are used as data features. The data features with preset time steps are respectively composed of the first to third input features.

[0051] The preset time step can be set according to the actual situation, for example, 10 time steps, without specific limitation.

[0052] It is understood that the embodiments of this application can acquire vehicle yaw rate signals, wheel speed signals, steering angle signals, lateral and longitudinal acceleration signals and driving and braking torque signals through various sensors, etc., use the signals as data features, and use the data features with a set time step to form the first to third input features. Based on the above signals, the input variables for joint observation are used to facilitate the subsequent calculation of longitudinal reference vehicle speed and vehicle body slip angle.

[0053] It should be noted that the first input feature may include the wheel speed of all wheels, longitudinal acceleration of the vehicle body, driving torque of all wheels, braking torque of all wheels, front wheel steering angle, and yaw rate of the vehicle body. The second and third input features may each include the longitudinal acceleration of the vehicle body, driving torque of all wheels, braking torque of all wheels, front wheel steering angle, and yaw rate of the vehicle body. The three estimation modules, namely lateral reference vehicle speed, longitudinal reference vehicle speed noise covariance, and lateral reference vehicle speed noise covariance, can adopt different design methods according to actual needs. The input features selected by different design methods are also different, and no specific limitation is made.

[0054] In step S102, the wheel speed is compensated based on the yaw rate and the turning angle.

[0055] The formula for wheel speed compensation is as follows:

[0056]

[0057] Among them, W ij It is the wheel speed signal, ij∈{fl,fr,rl,rr}; W ij c It is the compensated wheel speed signal; It is the yaw rate signal; δ is the front wheel steering angle signal; k f k r These are the track widths of the front and rear axles, respectively.

[0058] It is understood that the embodiments of this application can compensate for wheel speed by yaw rate and turning angle, and calculate the wheel speed values ​​of different wheels of the vehicle, so as to facilitate the subsequent calculation of longitudinal reference vehicle speed measurement values.

[0059] In step S103, the longitudinal reference vehicle speed measurement value is calculated based on the compensated wheel speed and longitudinal acceleration.

[0060] It is understood that the embodiments of this application can calculate the longitudinal reference vehicle speed measurement value based on the compensated wheel speed and longitudinal acceleration, thereby ensuring the accuracy of the estimated value of the longitudinal reference vehicle speed.

[0061] In the embodiments of this application, such as Figure 2 As shown, the longitudinal reference vehicle speed measurement is calculated based on the compensated wheel speed and longitudinal acceleration, including: obtaining the upper and lower thresholds of the longitudinal acceleration; if the longitudinal acceleration is greater than or equal to the lower threshold and less than or equal to the upper threshold, the average of the compensated wheel speeds is used as the longitudinal reference vehicle speed measurement; if the longitudinal acceleration is less than the lower threshold, when the longitudinal acceleration is greater than the derivative of the estimated longitudinal reference vehicle speed at the current moment, the maximum value between the compensated wheel speed and the current longitudinal reference vehicle speed measurement is used as the longitudinal reference vehicle speed measurement; when the longitudinal acceleration is less than the derivative of the estimated longitudinal reference vehicle speed at the current moment, the maximum value among the compensated wheel speeds is used as the longitudinal reference vehicle speed measurement; if the longitudinal acceleration is greater than the upper threshold, when the longitudinal acceleration is less than the derivative of the estimated longitudinal reference vehicle speed at the current moment, the minimum value among the compensated wheel speed and the current longitudinal reference vehicle speed measurement is used as the longitudinal reference vehicle speed measurement; when the longitudinal acceleration is greater than or equal to the derivative of the estimated longitudinal reference vehicle speed at the current moment, the minimum value among the compensated wheel speeds is used as the longitudinal reference vehicle speed measurement.

[0062] The upper and lower thresholds can be set according to actual needs, without specific limitations.

[0063] It is understood that the embodiments of this application analyze and calculate three types of working conditions, namely acceleration, deceleration and steady state, based on kinematic rules, to improve the problem of inaccurate measurement of longitudinal reference vehicle speed under large slip rate and improve the accuracy of longitudinal reference vehicle speed.

[0064] It should be noted that, under steady-state conditions, W ij c It closely approximates the actual longitudinal vehicle speed, but during rapid acceleration or deceleration, the tires experience a significant slippage rate, which is even more pronounced on low-traction surfaces. Therefore, this method, based on kinematic rules, is specifically designed for acceleration (A... x >T Hup ), deceleration (A) x <TH low ) and steady state (TH low ≤A x ≤TH up The system is specifically designed for three types of operating conditions, such as ( ), which can improve the inaccuracy of longitudinal reference vehicle speed measurements under large slip rates. Figure 2 As shown, Ax This represents the longitudinal acceleration signal output by the IMU; T Hup TH low These represent the upper and lower threshold values ​​for longitudinal acceleration, respectively; V -1 This represents the estimated longitudinal reference speed at the previous moment; This represents the estimated longitudinal reference speed at the current moment, and is also used as the measured value of the longitudinal reference speed in the adaptive joint observation module. express The derivative of is the estimated value of the longitudinal acceleration at the current moment.

[0065] Specifically, the current vehicle's operating condition information is obtained to determine whether the current vehicle's operating condition belongs to an acceleration condition. If it belongs to an acceleration condition (A... x >T Hup If the derivative of the estimated longitudinal reference vehicle speed at the current moment is greater than the longitudinal acceleration signal output by the IMU, then the compensated wheel speed (W) will be adjusted accordingly. ij c ) and the estimated value of the longitudinal reference vehicle speed measurement at the current time (V -1 +A x The minimum value in Δt) is used as the longitudinal reference vehicle speed. If the derivative of the estimated value of the longitudinal reference vehicle speed at the current moment is less than or equal to the longitudinal acceleration signal output by the IMU, then the minimum value of the compensated wheel speed is used as the measured value of the longitudinal reference vehicle speed.

[0066] If it is not a machining condition, then it is further determined whether it is a deceleration condition. If it is a deceleration condition, then it is further determined whether the derivative of the estimated value of the longitudinal reference vehicle speed at the current moment is less than the longitudinal acceleration signal output by the IMU. If the derivative of the estimated value of the longitudinal reference vehicle speed at the current moment is less than the longitudinal acceleration signal output by the IMU, then the maximum value between the compensated wheel speed and the estimated value of the longitudinal reference vehicle speed at the current moment is taken as the measured value of the longitudinal reference vehicle speed. If the derivative of the estimated value of the longitudinal reference vehicle speed at the current moment is greater than or equal to the longitudinal acceleration signal output by the IMU, then the maximum value of the compensated wheel speed is taken as the measured value of the longitudinal reference vehicle speed.

[0067] If it is neither a machining condition nor a deceleration condition, then it is a steady-state condition. In this case, W ij c The mean value is very close to the actual longitudinal vehicle speed and is basically unaffected by the cumulative error of the sensors, which can ensure the accuracy of the estimated longitudinal reference vehicle speed.

[0068] In this embodiment of the application, before calculating the longitudinal reference vehicle speed measurement value based on the compensated wheel speed and longitudinal acceleration, the method further includes: obtaining the estimated value of the longitudinal reference vehicle speed at the previous moment; calculating the change in the estimated value based on the time difference between the current moment and the previous moment and the longitudinal acceleration; and calculating the longitudinal reference vehicle speed measurement value at the current moment based on the estimated value of the longitudinal reference vehicle speed at the previous moment and the change in the estimated value.

[0069] It is understood that the embodiments of this application can calculate the change in the estimated value based on the time difference between the previous moment and the current moment and the longitudinal acceleration, and calculate the measured value of the longitudinal reference vehicle speed at the current moment through the estimated value of the longitudinal reference vehicle speed at the previous moment and the change in the estimated value.

[0070] It should be noted that the estimated longitudinal reference vehicle speed at the current moment is calculated as follows: (V -1 +A x Δt), where V -1 A represents the estimated longitudinal reference speed at the previous moment. x A represents the longitudinal acceleration at the current moment, Δt represents the time difference between the previous moment and the current moment, and A represents the longitudinal acceleration at the current moment. x Δt represents the change in the estimated longitudinal acceleration between the previous moment and the current moment.

[0071] In step S104, the lateral reference vehicle speed measurement is estimated based on the first input feature, and the noise covariance of the longitudinal reference vehicle speed measurement and the lateral reference vehicle speed measurement is estimated based on the second input feature and the third input feature, respectively.

[0072] The formula for calculating the lateral reference vehicle speed and noise covariance is as follows:

[0073]

[0074] Among them, R x N R is expressed as the noise covariance of the longitudinal reference vehicle speed measurement. y N V represents the noise covariance of the lateral reference vehicle speed measurement. y N The value represents the lateral reference vehicle speed measurement, k represents the current sampling time, X1, X2, and X3 represent different input features, and the input features refer to the time series of data features within a preset historical time step, f y,vel The nonlinear fitting function f of the lateral reference vehicle speed estimation module. x,cov The nonlinear fitting function f of the longitudinal reference vehicle speed measurement noise covariance estimation module. y,cov This refers to the nonlinear fitting function of the noise covariance estimation module for lateral reference vehicle speed measurements.

[0075] It is understood that the embodiments of this application can determine the uncertainty of the estimated values ​​of the longitudinal reference vehicle speed and the lateral reference vehicle speed by calculating the noise covariance of the lateral reference vehicle speed measurement and the noise covariance of the longitudinal reference vehicle speed measurement. This constitutes a key component of the noise covariance matrix of the relevant measurements of the adaptive joint observation, which facilitates subsequent joint observation.

[0076] It should be noted that the noise covariance includes the noise covariance of the lateral reference vehicle speed measurement and the noise covariance of the longitudinal reference vehicle speed measurement; the historical preset time step can be set according to the actual situation, for example: 10 historical time steps, without specific limitation.

[0077] In this embodiment, the lateral reference vehicle speed and noise covariance are fitted using a structurally similar neural network structure, wherein the neural network structure includes a long short-time neural network layer, a fully connected layer, a batch normalization layer, and a multilayer perceptron processing layer.

[0078] It is understood that in the embodiments of this application, the noise covariance of the lateral reference vehicle speed and the longitudinal reference vehicle speed is fitted using a structurally similar neural network structure, the input data feature time series is extracted, the inherent features are determined based on the differences of the corresponding estimation modules and combined with experience, and the correspondence between input and output is established. Furthermore, the current state can be predicted based on historical data, thereby improving efficiency.

[0079] Specifically, the noise covariance of the lateral and longitudinal reference vehicle speed measurements can employ similar neural network structures, both comprising two parts: an LSTMNN-based feature extraction algorithm and an MLP (Multi-layer Perceptron) post-processing algorithm. The LSTMNN is used to extract the input data feature time series (input feature 1, input feature 2, and input feature 3), determining inherent features based on the differences between the corresponding estimation modules and combining experience to establish the correspondence between input and output.

[0080] Since cars are a typical inertial system, the reference speed does not change abruptly, and the current state information shows a strong correlation with previous times. Therefore, LSTMNN can predict the current state based on historical data. After the LSTMNN layer, there is an MLP post-processing layer consisting of two fully connected layers (FC layers), one batch normalization layer, and one Tanh activation layer. The fully connected layers can reduce the dimensionality of the multiple hidden units contained in the LSTMNN layer output, further integrating the output characteristics of the LSTMNN layer; the batch normalization layer can accelerate the training process of the neural network and significantly reduce the network's sensitivity to initial values.

[0081] In step S105, the joint observation weight of the longitudinal reference vehicle speed measurement and the lateral vehicle speed measurement is determined based on the noise covariance. Based on the joint observation weight, the longitudinal reference vehicle speed estimate and the lateral reference vehicle speed estimate are jointly observed to obtain the vehicle body slip angle estimate.

[0082] It is understood that the embodiments of this application determine the joint observation weight of the longitudinal reference vehicle speed measurement and the lateral vehicle speed measurement based on the noise covariance, and perform joint observation of the longitudinal reference vehicle speed estimate and the lateral reference vehicle speed estimate according to the joint observation weight, thereby obtaining the vehicle body slip angle estimate. This enables the joint observation weight to be adaptively adjusted when the accuracy of the input signal value is unstable, reducing the impact on the overall joint observation method output result, improving the accuracy of the observation result, and enhancing the robustness of the joint observation method.

[0083] In this embodiment of the application, the joint observation weight of the longitudinal reference vehicle speed measurement and the vehicle body slip angle is determined based on the noise covariance, including: if the noise covariance increases, the joint observation weight of the longitudinal reference vehicle speed measurement is decreased; if the noise covariance decreases, the joint observation weight of the longitudinal reference vehicle speed measurement is increased.

[0084] It is understood that the embodiments of this application can adaptively adjust the joint observation weights of the longitudinal reference vehicle speed according to the noise covariance, thereby enhancing the robustness of the joint observation method.

[0085] The data intelligence-driven adaptive joint observation method proposed in this application calculates the longitudinal reference vehicle speed by acquiring one or more signal values, calculates the lateral reference vehicle speed based on the lateral acceleration, estimates the noise covariance of the lateral and longitudinal reference vehicle speed measurements, determines the joint observation weights of the longitudinal and lateral reference vehicle speed measurements based on the noise covariance, and performs joint observations on the estimated longitudinal and lateral reference vehicle speeds based on the joint observation weights. This allows for adaptive adjustment of the joint observation weights when the accuracy of the input signal values ​​is unstable, reducing the impact on the overall output of the joint observation method, improving the accuracy of the observation results, and enhancing the robustness of the joint observation method.

[0086] The following will combine Figure 3This paper elaborates on the data-driven adaptive joint observation method of this application. The data-driven adaptive joint observation architecture for key states of vehicle dynamics includes five modules: YC (Yaw Compensation), WSP (Wheel Speed ​​Processing), DS (Data Storage), DDE (Data-driven Estimation), and AJO (Adaptive Joint Observation), as detailed below:

[0087] (1) DS module

[0088] The output signals from the steering angle sensor, wheel speed sensor, IMU, drive system, and braking system are acquired as input variables for the joint observation architecture, including but not limited to: vehicle yaw rate, wheel speed, steering angle, lateral and longitudinal acceleration, and drive and braking torque.

[0089] (2) YC module

[0090] This invention constructs a two-layer signal processing architecture to obtain an estimate of the longitudinal reference vehicle speed. The first layer is the YC module, whose function is to transfer the four wheel speed signals to the vehicle's center of gravity. The relevant equations are as follows:

[0091]

[0092] Among them, W ij It is the wheel speed signal, ij∈{fl,fr,rl,rr}; W ij c It is the compensated wheel speed signal; It is the yaw rate signal; δ is the front wheel steering angle signal; k f k r These are the track widths of the front and rear axles, respectively.

[0093] (3) WSP module

[0094] The second layer of the two-layer signal processing architecture is the WSP module. Under steady-state conditions, W... ij c It closely approximates the actual longitudinal vehicle speed, but during rapid acceleration or deceleration, the tires experience a significant slippage rate, which is even more pronounced on low-traction surfaces. Therefore, this method, based on kinematic rules, is specifically designed for acceleration (A... x >T Hup ), deceleration (A) x <TH low ) and steady state (TH low ≤A x ≤TH upThe system is specifically designed for three types of operating conditions, such as [list of conditions], to improve the inaccuracy of longitudinal reference speed estimation under high slip rates. Figure 2 As shown, A x This represents the longitudinal acceleration signal output by the IMU (Inertial Measurement Unit); T Hup TH low These represent the upper and lower threshold values ​​for longitudinal acceleration, respectively; V -1 This represents the estimated longitudinal reference speed at the previous moment; This represents the estimated longitudinal reference speed at the current moment, and is also used as the measured value of the longitudinal reference speed in the adaptive joint observation module. express The derivative of W is the estimated value of the longitudinal acceleration at the current moment. Under steady-state conditions, W... ij c The mean value is very close to the actual longitudinal vehicle speed and is basically unaffected by the cumulative error of the sensors, which can ensure the accuracy of the estimated longitudinal reference vehicle speed.

[0095] (5) DDE module

[0096] The DDE module includes three sub-modules: longitudinal reference vehicle speed noise covariance estimation (RxNet), lateral reference vehicle speed estimation (VyNet), and lateral reference vehicle speed noise covariance estimation (RyNet). VyNet is used to obtain the estimated value of the lateral reference vehicle speed. RxNet and RyNet are used to obtain the uncertainties of the estimated values ​​of longitudinal and lateral reference vehicle speeds, respectively, including the longitudinal reference vehicle speed noise covariance RxN and the lateral reference vehicle speed noise covariance RyN. RxN and RyN will serve as key components of the noise covariance matrix of the relevant measurements in the AJO module.

[0097] The three sub-modules can adopt similar neural network structures, each including two parts: a feature extraction algorithm based on LSTMNN and a post-processing algorithm based on MLP. LSTMNN is used to extract the inherent features in the input data feature time series (input feature 1, input feature 2, and input feature 3, which are determined based on the differences between the corresponding estimation modules and experience) and establish the correspondence between input and output.

[0098] Since cars are a typical inertial system, the reference speed does not change abruptly, and the current state information shows a strong correlation with previous times. Therefore, LSTMNN can predict the current state based on historical data. After the LSTMNN layer, there is an MLP post-processing layer consisting of two fully connected layers (FCLayer), one BatchNormalization layer, and one Tanh activation layer. Among them, the fully connected layer can reduce the dimensionality of the multiple hidden units contained in the LSTMNN layer output, further integrating the output characteristics of the LSTMNN layer; the BatchNormalization layer can accelerate the training process of the neural network and significantly reduce the network's sensitivity to initial values.

[0099] Therefore, the three sub-modules involved in the DDE module can be represented by nonlinear functions as follows:

[0100]

[0101] Among them, R x N R is expressed as the noise covariance of the longitudinal reference vehicle speed measurement. y N V represents the noise covariance of the lateral reference vehicle speed measurement. y N The value represents the lateral reference vehicle speed measurement, k represents the current sampling time, X1, X2, and X3 represent different input features, and the input features refer to the time series of data features within a preset historical time step, f y,vel The nonlinear fitting function f of the lateral reference vehicle speed estimation module. x,cov The nonlinear fitting function f of the longitudinal reference vehicle speed measurement noise covariance estimation module. y,cov This refers to the nonlinear fitting function of the noise covariance estimation module for lateral reference vehicle speed measurements.

[0102] (5) AJO module

[0103] The design of the AJO module is based on the equations of motion of a vehicle.

[0104]

[0105] in, and V represents the derivatives of the longitudinal and lateral velocities at the vehicle's center of gravity, respectively. x and V y Let A represent the longitudinal velocity and lateral velocity at the vehicle's center of gravity, respectively. x and A y These represent the longitudinal acceleration signal and the lateral acceleration signal output by the IMU, respectively; b x and b yThese represent the longitudinal and lateral acceleration signal deviations, respectively, and are generally constants; ε x and ε y These represent random noise in the longitudinal and lateral acceleration signals, respectively.

[0106] Based on formula (1), the discrete state-space equation with sampling time Δt can be obtained:

[0107] x(k)=Ax(k-1)+Bu(k)+e(k) (2)

[0108] Where k represents the current sampling time, x = (V x ,b x V y ,b y ) T , u=(A x A y ) T e = (ε x ,0,ε y ,0) T ,

[0109] Based on formulas (3) and (4) and the KF algorithm, a joint observation method of AKF (Adaptive Kalman Filter) based on LSTMNN is proposed. The algorithm steps are shown in Table 1 below.

[0110] Table 1. AKF Joint Observation Method Based on LSTMNN

[0111]

[0112] The above-mentioned AKF joint observation method based on LSTMNN (such as...) Figure 3 Data-driven estimation module and vehicle dynamics

[0113] (Mechanical critical state adaptive joint observation module), if the data features X1, X2, and X3 are known, input state After initializing the system matrix A, control matrix B, measurement matrix C, estimated noise covariance P, and process noise covariance Q, V can be obtained through the three estimation modules of lateral reference vehicle speed, longitudinal reference vehicle speed noise covariance, and lateral reference vehicle speed noise covariance in the data-driven estimation module. y N (k), R x N (k) and R y N (k), (corresponding to steps 2, 3, and 4 in Table 1), based on R x N (k) and Ry N R(k) can be obtained from (k).

[0114] Assuming x(k-1|k-1) and u(k) are known, x(k|k-1) can be obtained according to step 6, P(k|k-1) can be obtained according to steps 1 and 7, K(k) can be obtained according to steps 1, 5 and 7, x(k|k) can be obtained according to steps 1, 6 and 8, and P(k|k) can be obtained according to steps 1, 7 and 8. Thus, the current sampling time k, the state estimate x(k|k) of the longitudinal reference vehicle speed and the vehicle body sideslip angle and its covariance matrix P(k|k) are obtained.

[0115] As shown in Table 1, in the AKF joint observation method based on LSTMNN, the process noise covariance Q is generally taken as a fixed value. Since the AKF gain K is directly related to Q and the measurement noise covariance R, adjusting R will directly affect the weight of state prediction and measurement. Therefore, the weight of key state quantities of vehicle dynamics can be adaptively adjusted, thereby enhancing the robustness of the joint observation method.

[0116] For example, when observing longitudinal reference vehicle speed, the WSP module acquires data when significant tire slippage occurs. Accuracy decreases, RxNet's output value RxN increases, and the joint observation method will... The weight of the variable is reduced accordingly to minimize its impact on the output of the joint observation method.

[0117] In summary, the data intelligence-driven adaptive joint observation architecture and application method proposed in this application can overcome the limitations of MBD and DDD to achieve high-precision, adaptive joint observation of key vehicle dynamic states (including longitudinal reference vehicle speed and vehicle body slip angle).

[0118] Next, referring to the accompanying drawings, we describe the data-driven adaptive joint observation device proposed according to the embodiments of this application.

[0119] Figure 4 This is a block diagram of a data intelligence-driven adaptive joint observation device according to an embodiment of this application.

[0120] like Figure 4 As shown, the data-driven adaptive joint observation device 10 includes: a data storage module 100, a yaw compensation module 200, a wheel speed signal processing module 300, a data-driven estimation module 400, and an adaptive joint observation module 500.

[0121] The data storage module 100 is used to acquire the vehicle's yaw rate signal, wheel speed signal, steering angle signal, lateral and longitudinal acceleration signals, and driving and braking torque signals, and uses one or more signals as data features, and the data features with preset time steps are respectively composed of the first to third input features; the yaw compensation module 200 is used to compensate the wheel speed according to the yaw rate and steering angle; the wheel speed signal processing module 300 is used to calculate the longitudinal reference vehicle speed according to the compensated wheel speed and longitudinal acceleration; the data-driven estimation module 400 is used to estimate the lateral reference vehicle speed measurement value according to the first input feature, and estimate the noise covariance of the longitudinal reference vehicle speed measurement value and the lateral reference vehicle speed measurement value according to the second and third input features respectively; the adaptive joint observation module 500 is used to determine the joint observation weight of the longitudinal reference vehicle speed measurement value and the lateral vehicle speed measurement value based on the noise covariance, and perform joint observation of the longitudinal reference vehicle speed estimation value and the lateral reference vehicle speed estimation value based on the joint observation weight, thereby obtaining the vehicle body slip angle estimation value.

[0122] In this embodiment, the wheel speed compensation formula is:

[0123]

[0124] Among them, W ij It is the wheel speed signal, ij∈{fl,fr,rl,rr}; W ij c It is the compensated wheel speed signal; It is the yaw rate signal; δ is the front wheel steering angle signal; k f k r These are the track widths of the front and rear axles, respectively.

[0125] In this embodiment, the wheel speed signal processing module 300 is further configured to: obtain an upper limit threshold and a lower limit threshold for longitudinal acceleration; if the longitudinal acceleration is greater than or equal to the lower limit threshold and less than or equal to the upper limit threshold, then the average value of the compensated wheel speed is used as the longitudinal reference vehicle speed measurement value; if the longitudinal acceleration is less than the lower limit threshold, then when the longitudinal acceleration is greater than the derivative of the estimated longitudinal reference vehicle speed at the current moment, the maximum value between the compensated wheel speed and the measured longitudinal reference vehicle speed at the current moment is used as the longitudinal reference vehicle speed measurement value; when the longitudinal acceleration is less than or equal to the derivative of the estimated longitudinal reference vehicle speed at the current moment, the maximum value among the compensated wheel speeds is used as the longitudinal reference vehicle speed measurement value; if the longitudinal acceleration is greater than the upper limit threshold, then when the longitudinal acceleration is less than the derivative of the estimated longitudinal reference vehicle speed at the current moment, the minimum value among the compensated wheel speed and the measured longitudinal reference vehicle speed at the current moment is used as the longitudinal reference vehicle speed measurement value; when the longitudinal acceleration is less than or equal to the derivative of the estimated longitudinal reference vehicle speed at the current moment, the minimum value among the compensated wheel speeds is used as the longitudinal reference vehicle speed measurement value.

[0126] In this embodiment of the application, it further includes: an acquisition module, which is used to acquire the estimated value of the longitudinal reference vehicle speed at the previous moment before calculating the longitudinal reference vehicle speed measurement value based on the compensated wheel speed and longitudinal acceleration; calculate the change in the estimated value based on the time difference between the current moment and the previous moment and the longitudinal acceleration; and calculate the longitudinal reference vehicle speed measurement value at the current moment based on the estimated value of the longitudinal reference vehicle speed at the previous moment and the change in the estimated value.

[0127] In this embodiment of the application, the formula for calculating the lateral reference vehicle speed and noise covariance is as follows:

[0128]

[0129] Among them, R x N R is expressed as the noise covariance of the longitudinal reference vehicle speed measurement. y N V represents the noise covariance of the lateral reference vehicle speed measurement. y N The value represents the lateral reference vehicle speed measurement, k represents the current sampling time, X1, X2, and X3 represent different input features, and the input features refer to the time series of data features within a preset historical time step, f y,vel The nonlinear fitting function f of the lateral reference vehicle speed estimation module. x,cov The nonlinear fitting function f of the longitudinal reference vehicle speed measurement noise covariance estimation module. y,cov This refers to the nonlinear fitting function of the noise covariance estimation module for lateral reference vehicle speed measurements.

[0130] In this embodiment, the lateral reference vehicle speed and noise covariance are fitted using a structurally similar neural network structure, wherein the neural network structure includes a long short-time neural network layer, a fully connected layer, a batch normalization layer, and a multilayer perceptron processing layer.

[0131] In this embodiment of the application, the adaptive joint observation module 500 is further configured to: reduce the joint observation weight of the longitudinal reference vehicle speed measurement if the noise covariance increases; and increase the joint observation weight of the longitudinal reference vehicle speed measurement if the noise covariance decreases.

[0132] It should be noted that the foregoing explanation of the data intelligence-driven adaptive joint observation method embodiment also applies to the data intelligence-driven adaptive joint observation device of this embodiment, and will not be repeated here.

[0133] The data intelligence-driven adaptive joint observation device proposed in this application calculates the longitudinal reference vehicle speed by acquiring one or more signal values, calculates the lateral reference vehicle speed based on the lateral acceleration, estimates the noise covariance of the lateral and longitudinal reference vehicle speed measurements, determines the joint observation weights of the longitudinal and lateral reference vehicle speed measurements based on the noise covariance, and performs joint observations on the estimated longitudinal and lateral reference vehicle speeds based on the joint observation weights. This allows for adaptive adjustment of the joint observation weights when the accuracy of the input signal values ​​is unstable, reducing the impact on the overall joint observation method output, improving the accuracy of the observation results, and enhancing the robustness of the joint observation method.

[0134] Figure 5 A schematic diagram of the structure of a vehicle provided in an embodiment of this application. The vehicle may include:

[0135] The memory 501, the processor 502, and the computer program stored on the memory 501 and capable of running on the processor 502.

[0136] When the processor 502 executes the program, it implements the data-driven adaptive joint observation method provided in the above embodiments.

[0137] Furthermore, the vehicle also includes:

[0138] Communication interface 503 is used for communication between memory 501 and processor 502.

[0139] The memory 501 is used to store computer programs that can run on the processor 502.

[0140] The memory 501 may include high-speed RAM (Random Access Memory) memory, and may also include non-volatile memory, such as at least one disk storage.

[0141] If the memory 501, processor 502, and communication interface 503 are implemented independently, then the communication interface 503, memory 501, and processor 502 can be interconnected via a bus to complete communication between them. The bus can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus, or an EISA (Extended Industry Standard Architecture) bus, etc. The bus can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 5 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0142] Optionally, in a specific implementation, if the memory 501, processor 502, and communication interface 503 are integrated on a single chip, then the memory 501, processor 502, and communication interface 503 can communicate with each other through an internal interface.

[0143] Processor 502 may be a CPU (Central Processing Unit), an ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement embodiments of this application.

[0144] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described data-driven adaptive joint observation method.

[0145] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0146] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "N" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0147] Any process or method described in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or N executable instructions for implementing custom logic functions or processes, and the scope of the preferred embodiments of this application includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as should be understood by those skilled in the art to which embodiments of this application pertain.

[0148] It should be understood that the various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, the N steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (FPGAs), field-programmable gate arrays (FPGAs), etc.

[0149] Those skilled in the art will understand that all or part of the steps of the methods in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, the program includes one or a combination of the steps of the method embodiments.

[0150] Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of this application.

Claims

1. A data-driven intelligent adaptive joint observation method, characterized in that, Includes the following steps: Acquire the vehicle's yaw rate signal, wheel speed signal, steering angle signal, lateral and longitudinal acceleration signal, and driving and braking torque signal, and use one or more signals as data features. Combine the data features with preset time steps to form the first to third input features respectively. The wheel speed is compensated based on the yaw rate and the turning angle; Calculate the longitudinal reference vehicle speed based on the compensated wheel speed and longitudinal acceleration; The lateral reference vehicle speed measurement is estimated based on the first input feature, and the noise covariance of the longitudinal reference vehicle speed measurement and the lateral reference vehicle speed measurement is estimated based on the second input feature and the third input feature, respectively. Based on the noise covariance, the joint observation weights of the longitudinal reference vehicle speed measurement and the lateral vehicle speed measurement are determined. Based on the joint observation weights, the longitudinal reference vehicle speed estimate and the lateral reference vehicle speed estimate are jointly observed to obtain the vehicle body slip angle estimate.

2. The data-driven adaptive joint observation method according to claim 1, characterized in that, The formula for compensating for wheel speed is: Among them, W ij It is the wheel speed signal, ij∈{fl,fr,rl,rr}; W ij c It is the compensated wheel speed signal; It is the yaw rate signal; δ is the front wheel steering angle signal; k f k r These are the track widths of the front and rear axles, respectively.

3. The data-driven adaptive joint observation method according to claim 1, characterized in that, The calculation of the longitudinal reference vehicle speed measurement based on the compensated wheel speed and longitudinal acceleration includes: Obtain the upper and lower threshold values ​​of the longitudinal acceleration; If the longitudinal acceleration is greater than or equal to the lower threshold and less than or equal to the upper threshold, then the average value of the compensated wheel speed is used as the longitudinal reference vehicle speed measurement value. If the longitudinal acceleration is less than the lower threshold, then when the longitudinal acceleration is greater than the derivative of the estimated longitudinal reference vehicle speed at the current moment, the maximum value between the compensated wheel speed and the measured longitudinal reference vehicle speed at the current moment shall be used as the measured longitudinal reference vehicle speed; when the longitudinal acceleration is less than or equal to the derivative of the estimated longitudinal reference vehicle speed at the current moment, the maximum value among the compensated wheel speeds shall be used as the measured longitudinal reference vehicle speed. If the longitudinal acceleration is greater than the upper limit threshold, then when the longitudinal acceleration is less than the derivative of the estimated longitudinal reference vehicle speed at the current moment, the minimum value between the compensated wheel speed and the measured longitudinal reference vehicle speed at the current moment shall be used as the measured longitudinal reference vehicle speed; when the longitudinal acceleration is greater than or equal to the derivative of the estimated longitudinal reference vehicle speed at the current moment, the minimum value among the compensated wheel speeds shall be used as the measured longitudinal reference vehicle speed.

4. The data-driven adaptive joint observation method according to claim 3, characterized in that, Before calculating the longitudinal reference vehicle speed measurement based on the compensated wheel speed and longitudinal acceleration, the following steps are also included: Obtain the estimated value of the longitudinal reference vehicle speed at the previous moment; The estimated change in the longitudinal acceleration is calculated based on the time difference between the current moment and the previous moment. The measured value of the longitudinal reference vehicle speed at the current moment is calculated based on the estimated value of the longitudinal reference vehicle speed at the previous moment and the change in the estimated value.

5. The data-driven adaptive joint observation method according to claim 1, characterized in that, in, The formula for calculating the lateral reference speed and noise covariance is as follows: Among them, R x N R is expressed as the noise covariance of the longitudinal reference vehicle speed measurement. y N V represents the noise covariance of the lateral reference vehicle speed measurement. y N The value is represented as the lateral reference vehicle speed measurement, k represents the current sampling time, and X1, X2, and X3 represent different input features. The input features refer to the time series of data features within a historical preset time step. y,vel The nonlinear fitting function f of the lateral reference vehicle speed estimation module. x,cov The nonlinear fitting function for estimating the noise covariance of longitudinal reference vehicle speed measurements, f y,cov This refers to the nonlinear fitting function for estimating the noise covariance of the lateral reference vehicle speed measurement.

6. The data-driven adaptive joint observation method according to claim 1 or 5, characterized in that, The lateral reference vehicle speed and noise covariance are fitted using a structurally similar neural network structure, wherein the neural network structure includes a long short-time neural network layer, a fully connected layer, a batch normalization layer, and a multilayer perceptron processing layer.

7. The data-driven adaptive joint observation method according to claim 1 or 5, characterized in that, The determination of the joint observation weights for the longitudinal reference vehicle speed measurement and the lateral reference vehicle speed measurement based on the noise covariance includes: If the noise covariance increases, then the joint observation weight of the longitudinal reference vehicle speed measurement is reduced; If the noise covariance decreases, then the joint observation weight of the longitudinal reference vehicle speed measurement is increased.

8. A data-driven intelligent adaptive joint observation device, characterized in that, include: The data storage module is used to acquire the vehicle's yaw rate signal, wheel speed signal, steering angle signal, lateral and longitudinal acceleration signal, and driving and braking torque signal, and to use one or more signals as data features, and to combine the data features with preset time steps to form the first to third input features respectively; A yaw compensation module is used to compensate the wheel speed based on the yaw angular velocity and the turning angle; The wheel speed signal processing module is used to calculate the longitudinal reference vehicle speed measurement value based on the compensated wheel speed and longitudinal acceleration; A data-driven estimation module is used to estimate the lateral reference vehicle speed measurement value based on the first input feature, and to estimate the noise covariance of the longitudinal reference vehicle speed measurement value and the lateral reference vehicle speed measurement value based on the second input feature and the third input feature, respectively. An adaptive joint observation module is used to determine the joint observation weight of the longitudinal reference vehicle speed measurement and the lateral vehicle speed measurement based on the noise covariance, and to perform joint observation of the longitudinal reference vehicle speed estimate and the lateral reference vehicle speed estimate based on the joint observation weight, thereby obtaining the vehicle body slip angle estimate.

9. A vehicle, characterized in that, include: A memory, a processor, and a computer program stored in the memory and executable on the processor, the processor executing the program to implement the data intelligence-driven adaptive joint observation method as described in any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, The program is executed by the processor to implement the data intelligence-driven adaptive joint observation method as described in any one of claims 1-7.

Images