Steering wheel torque determination method and apparatus, computer device, and readable storage medium

Abstract

The application relates to a steering wheel torque determination method and device, computer equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: acquiring a vehicle body pitch angle, a real-time steering angle, a real-time steering angle rate and an expected steering angle of a steering wheel; determining an expected steering angle rate and a basic torque of the steering wheel according to the real-time steering angle, the real-time steering angle rate and the expected steering angle; predicting a torque compensation value of the steering wheel according to the vehicle body pitch angle, the expected steering angle and the expected steering angle rate; and determining a target torque of the steering wheel based on the torque compensation value and the basic torque, wherein the target torque is used to adjust the steering wheel to the expected steering angle. According to the method, the target torque of the steering wheel can be more accurate without repeatedly calibrating control parameters in different driving environments, and more accurate steering control can be realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of vehicle control, and in particular to a steering wheel torque determination method and device, computer equipment, computer readable storage medium and computer program product. BACKGROUND

[0002] In the system of assisting the driving of a vehicle, a driving controller plays a vital role, which realizes the precise control of an EPS (Electric Power Steering) by sending precise steering angle or torque control instructions. As a core component of the steering system of a vehicle, the EPS outputs a torque signal to an electronic control unit in real time through a built-in torque sensor. The electronic control unit sends precise control instructions to the steering motor controller according to the received torque signal, the angle signal of the steering sensor and the vehicle speed signal and other multi-dimensional information, so as to ensure the precision and stability of the steering of the vehicle.

[0003] The design characteristics of the EPS controller cause significant differences in the gain effect of torque control at different vehicle speeds. Since the system parameters such as the moment of inertia, damping and stiffness of the steering wheel are often not accurately estimated or measured in actual application, it is difficult to accurately determine the actual required torque of the steering wheel, which further affects the precise control of the expected steering angle of the EPS. SUMMARY

[0004] Therefore, it is necessary to provide a more accurate steering wheel torque determination method, device, computer equipment, computer readable storage medium and computer program product in view of the above technical problems.

[0005] In a first aspect, the present application provides a steering wheel torque determination method, comprising:

[0006] obtaining a body pitch angle, a real-time steering angle, a real-time steering angle rate and an expected steering angle of a steering wheel;

[0007] determining an expected steering angle rate and a basic torque of the steering wheel according to the real-time steering angle, the real-time steering angle rate and the expected steering angle;

[0008] predicting a torque compensation value of the steering wheel according to the body pitch angle, the expected steering angle and the expected steering angle rate;

[0009] determining a target torque of the steering wheel based on the torque compensation value and the basic torque, the target torque being used to adjust the steering wheel to the expected steering angle.

[0010] In one embodiment, the torque compensation value of the steering wheel is predicted according to the body pitch angle, the expected steering angle and the expected steering angle rate, comprising:

[0011] determine a target nonlinear regression model to be called, the nonlinear regression model being trained based on body pitch angles, steering wheel turning angles and turning rate of the steering wheel in different vehicle speed intervals at historical time points;

[0012] call the target nonlinear regression model with the body pitch angle, the expected turning angle and the expected turning rate as inputs, and predict a torque compensation value of the steering wheel.

[0013] In one of the embodiments, the determination of the target nonlinear regression model to be called comprises:

[0014] when the number of the trained nonlinear regression models is one, the trained nonlinear regression model is determined as the target nonlinear regression model to be called;

[0015] when the number of the trained nonlinear regression models is multiple, a real-time vehicle speed of the vehicle is obtained, and based on the real-time vehicle speed, the target nonlinear regression model to be called is selected from the trained multiple nonlinear regression models, the nonlinear regression model corresponding to a vehicle speed interval one by one.

[0016] In one of the embodiments, the selection of the target nonlinear regression model to be called from the trained multiple nonlinear regression models based on the real-time vehicle speed comprises:

[0017] comparing the real-time vehicle speed with preset multiple vehicle speed intervals to determine a target vehicle speed interval where the real-time vehicle speed is located;

[0018] from the trained multiple nonlinear regression models, a nonlinear regression model matched with the target vehicle speed interval is found out, and the found nonlinear regression model is determined as the target nonlinear regression model to be called.

[0019] In one of the embodiments, the determination of the expected turning rate and the basic torque of the steering wheel according to the real-time turning angle, the real-time turning rate and the expected turning angle comprises:

[0020] determining a turning angle deviation according to the real-time turning angle and the expected turning angle;

[0021] determining the expected turning rate of the steering wheel based on the turning angle deviation;

[0022] determining a turning rate deviation according to the expected turning rate and the real-time turning rate;

[0023] determining the basic torque of the steering wheel based on the turning rate deviation.

[0024] In one of the embodiments, the determining the desired steering wheel angular velocity and the base torque of the steering wheel according to the real-time steering angle, the real-time steering angular velocity and the desired steering angle comprises:

[0025] The desired steering wheel angular velocity and the base torque are determined by a preset PID (Proportional-Integral-Derivative) controller according to the real-time steering angle, the real-time steering angular velocity and the desired steering angle.

[0026] In one of the embodiments, the preset PID controller comprises a position loop PID controller and a velocity loop PID controller; the determining the desired steering wheel angular velocity and the base torque according to the real-time steering angle, the real-time steering angular velocity and the desired steering angle comprises:

[0027] The desired steering wheel angular velocity is obtained by taking the real-time steering angle and the desired steering angle as inputs of the position loop PID controller;

[0028] The base torque of the steering wheel is obtained by taking the desired steering wheel angular velocity and the real-time steering angular velocity as inputs of the velocity loop PID controller.

[0029] In one of the embodiments, the obtaining the desired steering angle of the steering wheel comprises:

[0030] The initial desired steering angle of the steering wheel is obtained;

[0031] The parameters of the steepest descent tracker are initialized;

[0032] The initial desired steering angle is taken as an input of the initialized steepest descent tracker; the initialized steepest descent tracker updates internal state variables by a preset steepest control function, the real-time steering angle and the real-time steering angular velocity to obtain the optimized desired steering angle of the steering wheel.

[0033] In a second aspect, the application further provides a steering wheel torque determination device, comprising:

[0034] A data acquisition module is configured to acquire a vehicle body pitch angle, a real-time steering angle, a real-time steering angular velocity and a desired steering angle of a steering wheel;

[0035] A base torque determination module is configured to determine a desired steering wheel angular velocity and a base torque of the steering wheel according to the real-time steering angle, the real-time steering angular velocity and the desired steering angle;

[0036] A compensation torque prediction module is configured to predict a torque compensation value of the steering wheel according to the vehicle body pitch angle, the desired steering angle and the desired steering angular velocity;

[0037] a target torque determination module configured to determine a target torque of the steering wheel based on the torque compensation value and the base torque, the target torque being used to adjust the steering wheel to the desired steering angle.

[0038] In a third aspect, the present application also provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps in any one of the above-mentioned steering wheel torque determination method embodiments when executing the computer program.

[0039] In a fourth aspect, the present application also provides a computer-readable storage medium, which stores a computer program, and the computer program, when executed by a processor, implements the steps in any one of the above-mentioned steering wheel torque determination method embodiments.

[0040] In a fifth aspect, the present application also provides a computer program product, comprising a computer program, and the computer program, when executed by a processor, implements the steps in any one of the above-mentioned steering wheel torque determination method embodiments.

[0041] The above-mentioned steering wheel torque control method, device, computer device, computer-readable storage medium and computer program product fully consider factors influencing steering performance such as vehicle body posture, steering angle rate and steering angle, introduce a torque feed-forward compensation mechanism suitable for different scenarios, on the one hand, determine the desired steering angle rate and the base torque of the steering wheel through the real-time steering angle, real-time steering angle rate and desired steering angle of the vehicle, on the other hand, based on the torque feed-forward compensation mechanism, predict the torque compensation value of the steering wheel according to the desired steering angle rate, vehicle body pitch angle and desired steering angle, and then compensate the torque based on the torque compensation value and the base torque, i.e. dynamically adjust the target torque of the steering wheel, effectively reduce the steering error caused by factors such as changes in vehicle dynamic characteristics, external environmental interference, and achieve precise control of the desired steering angle. Moreover, the whole process does not need to repeatedly calibrate the control parameters in different driving environments, reduces the complexity of the control system, improves the control adaptability and robustness, makes the predicted torque compensation value of the steering wheel more accurate, and further improves the control precision of the desired steering angle. BRIEF DESCRIPTION OF DRAWINGS

[0042] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the related art, the following will briefly introduce the drawings needed to be used in the description of the embodiments of the present application or the related art. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other related drawings can also be obtained without creative labor based on these drawings.

[0043] Figure 1 a flowchart of the steering wheel torque determination method in one embodiment;

[0044] Figure 2 Flowchart of the method for determining the steering wheel torque in another embodiment;

[0045] Figure 3 Flowchart of the method for determining the steering wheel torque in another embodiment;

[0046] Figure 4 Flowchart of the method for determining the steering wheel torque in another embodiment;

[0047] Figure 5 Flowchart of the method for determining the steering wheel torque in another embodiment;

[0048] Figure 6 Flowchart of the method for determining the steering wheel torque in another embodiment;

[0049] Figure 7 Flowchart of the method for determining the steering wheel torque in another embodiment;

[0050] Figure 8 Flowchart of the method for determining the steering wheel torque in another embodiment;

[0051] Figure 9 Block diagram of the device for determining the steering wheel torque in another embodiment;

[0052] Figure 10 Block diagram of the device for determining the steering wheel torque in another embodiment;

[0053] Figure 11 Internal structure diagram of the computer device in an embodiment. DETAILED DESCRIPTION

[0054] In order to make the purpose, technical scheme and advantages of the present application clearer, the present application will be further described in detail below with reference to the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and do not limit the present application.

[0055] In the application scenario of assisted driving, the EPS torque control mode is relatively common, especially in complex and changeable driving environments, and the control effect is directly related to the driving safety of the vehicle. The design characteristics of the EPS controller make the gain effect of torque control at different vehicle speeds significantly different. However, the traditional EPS torque control method, such as directly outputting the EPS control torque based on the vehicle alignment error, the heading error, the reference line curvature and other information, although simple and direct, it is often difficult to fully reflect the complex relationship of vehicle motion.

[0056] To make up for this deficiency, some people have proposed a control strategy that converts vehicle motion information into front wheel motion angle, and then realizes precise tracking control of the front wheel turning angle through EPS torque control. However, most of the schemes use a simple single-loop PID control algorithm, and require tedious parameter calibration for different vehicle speed environments, which not only increases the complexity of the control system, but also reduces its adaptability and robustness. Moreover, since the system parameters such as the moment of inertia, damping and stiffness of the steering wheel are often not accurately estimated or measured in actual applications, it is difficult to accurately determine the actual required torque of the steering wheel, which in turn affects the precise control of the EPS expected turning angle.

[0057] To solve the above problems, the present application provides a steering wheel torque determination method. Specifically, factors that affect steering performance such as vehicle body posture, turning angle rate and turning angle are fully considered, and a torque feedforward compensation mechanism suitable for different scenarios is designed based on the vehicle body pitch angle, expected turning angle rate and expected turning angle. In actual applications, the vehicle body pitch angle, as well as the real-time turning angle, real-time turning angle rate and expected turning angle of the steering wheel are obtained. On the one hand, the expected turning angle rate and the basic torque of the steering wheel are determined based on the real-time turning angle, real-time turning angle rate and expected turning angle of the vehicle. On the other hand, based on the torque feedforward compensation mechanism, the torque compensation value of the steering wheel is predicted according to the expected turning angle rate, vehicle body pitch angle and expected turning angle. Then, torque compensation is performed based on the torque compensation value and the basic torque, i.e. the target torque of the steering wheel is dynamically adjusted, effectively reducing the steering error caused by changes in vehicle dynamic characteristics, external environmental disturbances and other factors, and realizing precise control of the EPS expected turning angle. Moreover, the entire process does not require repeated calibration of control parameters under different driving environments, reducing the complexity of the control system, improving control adaptability and robustness, making the predicted torque compensation value of the steering wheel more accurate, and further improving the control accuracy of the EPS expected turning angle.

[0058] In one embodiment, as shown in Figure 1 A steering wheel torque determination method is provided. In this embodiment, the method is applied to a controller in a vehicle. It should be understood that the method can also be applied to a server, and can also be applied to a system including a vehicle and a server, and can be implemented through the interaction of the vehicle and the server. In this embodiment, the method includes the following steps (hereinafter referred to as S) S200 to S800. Wherein:

[0059] S200, obtaining the vehicle body pitch angle, as well as the real-time turning angle, real-time turning angle rate and expected turning angle of the steering wheel.

[0060] The body pitch angle refers to the angle between the longitudinal axis of the vehicle and the horizontal plane. The real-time steering angle refers to the angle offset of the steering wheel from its initial position at the current time. The real-time steering rate refers to the angle of rotation of the steering wheel per unit time. The desired steering angle is the target value of the steering angle that the vehicle control system or the driver hopes to achieve. In an automatic driving system, the desired steering angle is usually calculated by a lateral control module based on the planned path and surrounding environment information.

[0061] In practical applications, the collection and processing of body pitch angle, real-time steering angle, real-time steering rate, and desired steering angle data usually involve the cooperative work of multiple sensors and controllers. For example, the vehicle can be equipped with multiple sensors, including a steering angle sensor, an angular velocity sensor, an accelerometer, and a gyroscope. The steering angle sensor can be installed in the steering column or the steering machine to detect the rotation angle of the steering wheel and provide the rotation angle of the steering wheel relative to the initial position. The angular velocity sensor can be installed at different positions of the vehicle to accurately measure the angular velocity of the vehicle rotating around the vertical axis. The accelerometer and the gyroscope can be installed at key positions of the chassis or the vehicle body to measure the inclination angle of the vehicle, including the pitch angle (forward and backward inclination) and the roll angle. In other embodiments, the pitch angle can also be obtained by an inertial measurement unit. The inertial measurement unit usually contains an accelerometer and a gyroscope, which can provide three-axis acceleration and angular velocity information, and the body pitch angle can be calculated by a fusion algorithm. The desired steering angle can be calculated by a lateral control module based on map data, road conditions, vehicle state, and planned target path. It can be understood that in other embodiments, the desired steering angle can also be a value set by a person in advance. In practical applications, the data collected or measured by all the above sensors are transmitted to the electronic control unit (i.e., the controller) in the EPS system, and the controller determines the size and direction of the electric power assistance based on the data collected by the sensors and other states of the vehicle (such as vehicle speed, road conditions, etc.), including determining the desired steering rate, the base torque, and the target torque of the steering wheel.

[0062] In other embodiments, the vehicle attitude information can also be provided by a global positioning system and an inertial navigation system, and the vehicle attitude information contains the pitch angle.

[0063] S400, determining the desired steering rate and the base torque of the steering wheel according to the real-time steering angle, the real-time steering rate, and the desired steering angle.

[0064] The desired steering rate refers to the speed of the desired change in the steering wheel steering angle. The base torque, which can also be referred to as the base assist torque, refers to the torque that needs to be applied on the steering wheel to achieve the desired steering action without considering the compensation damping torque. The base torque reflects the basic assist torque that is needed based on the driver's steering input only, without additional compensation requirements. It can be understood that in the assisted driving scenario, the base torque can be a value calculated by the controller of the EPS system according to the real-time steering angle, the real-time steering rate, and the desired steering angle, and the like.

[0065] In actual applications, there are multiple feasible ways to determine the desired steering rate and the base torque of the steering wheel. Exemplarily, after obtaining the real-time steering angle, the real-time steering rate, and the desired steering angle, the controller can first determine the desired steering rate of the steering wheel based on a feedback control strategy, and then determine the base torque of the steering wheel based on the desired steering rate.

[0066] In other embodiments, an adaptive controller or a fuzzy logic controller can also be called to determine the desired steering rate and the base torque of the steering wheel with the real-time steering angle, the real-time steering rate, and the desired steering angle as inputs. It can be understood that for the adaptive controller, if the vehicle operating conditions (such as tire grip, road conditions) change, the adaptive controller can adjust the control parameters in real time to maintain the best performance. If it is determined through the fuzzy logic controller, the desired steering rate and the base torque can be output by fuzzy rule reasoning according to the membership degree of the input variables (such as the steering angle and the steering rate) by using the fuzzy logic principle.

[0067] S600, a torque compensation value of the steering wheel is predicted according to the body pitch angle, the desired steering angle, and the desired steering rate.

[0068] The torque compensation value of the steering wheel refers to a compensation value of the torque applied on the steering wheel by the driver to improve the steering feel, improve the steering accuracy and responsiveness. The torque compensation value can be used to solve the inherent "dead zone" problem of the EPS system, and dynamically provide better steering assist according to the driving state of the vehicle under certain driving conditions (such as high-speed driving, road condition change, body posture change, and the like). The dead zone in the EPS system refers to that when the steering wheel is turned at a small angle, the system cannot immediately respond to the steering intention of the driver due to the sensor resolution or the control strategy. In the "dead zone" case, even if the driver has turned the steering wheel, the steering angle of the vehicle does not change immediately. At this time, the EPS system can automatically add a small torque compensation so that the vehicle can quickly respond to the steering instruction and eliminate the dead zone phenomenon.

[0069] In practical applications, there can be multiple ways to predict the steering wheel torque compensation value. Exemplarily, a vehicle dynamics model can be constructed in advance, in practical applications, the body pitch angle, the desired steering angle and the desired steering angle rate are taken as inputs of the vehicle dynamics model, the vehicle response in a future period of time is predicted through the model, and based on the predicted vehicle response result, the optimal torque compensation value is determined. In some other embodiments, a neural network model for predicting the torque compensation value can also be trained through a large amount of training data (historical driving data), so that the neural network model learns the mapping relationship between the body pitch angle, the desired steering angle and the desired steering angle rate and the required torque compensation value, and a trained torque compensation value prediction model is obtained. Subsequently, the body pitch angle, the desired steering angle and the desired steering angle rate are input into the trained torque compensation value prediction model to obtain the torque compensation value.

[0070] S800, based on the torque compensation value and the base torque, determining a target torque of the steering wheel, the target torque being used to adjust the steering wheel to the desired steering angle.

[0071] The target torque can also be referred to as a control torque, which refers to the torque actually output by the assist motor to the steering wheel, and is used to adjust the steering wheel to the desired steering angle. In other words, the target torque is the result of the combination of the base torque and the torque compensation value, and represents the total assist torque required by the EPS system to achieve the desired steering angle.

[0072] After the controller determines the torque compensation value and the base torque, the base torque and the torque compensation value can be combined, and the sum of the torque compensation value and the base torque is determined as the target torque of the steering wheel. Further, based on the target torque, the controller can adjust the output of the electric assist steering motor using a closed-loop control strategy, the assist motor outputs the target torque to form assist force for the driver to turn the steering wheel, and assist the driver to control the steering of the vehicle.

[0073] In the above-mentioned steering wheel torque determination method, the factors influencing the steering performance such as the body posture, the steering angle rate and the steering angle are fully considered, and a torque feedforward compensation mechanism suitable for different scenarios is introduced. On the one hand, the desired steering angle rate and the base torque of the steering wheel are determined through the real-time steering angle, the real-time steering angle rate and the desired steering angle of the vehicle. On the other hand, based on the torque feedforward compensation mechanism, the torque compensation value of the steering wheel is predicted according to the desired steering angle rate, the body pitch angle and the desired steering angle, and then the torque compensation is performed based on the torque compensation value and the base torque, i.e. the target torque of the steering wheel is dynamically adjusted, which effectively reduces the steering error caused by the change of the vehicle dynamic characteristics, the external environmental disturbance and other factors, and realizes the precise control of the EPS desired steering angle. Moreover, the whole process does not need to repeatedly calibrate the control parameters in different driving environments, which reduces the complexity of the control system, improves the control adaptability and robustness, makes the predicted torque compensation value of the steering wheel more accurate, and further improves the control precision of the EPS desired steering angle.

[0074] In practical applications, the predicted value of the torque compensation value can be obtained in a model prediction manner. In an exemplary embodiment, as shown in FIG. 6, S600 includes S620-S640. Among them: Figure 2

[0075] S620, determining a target nonlinear regression model to be called.

[0076] S640, calling the target nonlinear regression model with the body pitch angle, the desired steering angle and the desired steering angle rate as inputs, to predict the torque compensation value of the steering wheel.

[0077] Among them, the nonlinear regression model is trained based on the body pitch angle, the steering angle and the steering angle rate in different vehicle speed intervals at historical time. The nonlinear regression model is a mathematical model used to describe the nonlinear relationship between the dependent variable and one or more independent variables in statistics and machine learning.

[0078] In this embodiment, the torque compensation value can be predicted by a nonlinear regression model (hereinafter referred to as a model). Since the steering performance is affected by various factors under different driving conditions, a general nonlinear regression model or multiple nonlinear regression models can be trained for different driving conditions during the model training stage.

[0079] Specifically, the vehicle speed, the body pitch angle, the steering angle, the steering angle rate and the torque torque applied to the steering wheel by the driver under various driving conditions at historical time can be collected to ensure that the data covers all possible vehicle speed intervals. Then, the collected data is preprocessed, including but not limited to removing outliers and missing values, and standardizing the data. Then, based on the preprocessed data, a training set is constructed.

[0080] Next, the body pitch angle, the steering angle and the steering angle rate are confirmed as input features, and a model suitable for fitting a nonlinear relationship is selected, such as a neural network, a support vector regression, a Gaussian process regression or other nonlinear regression model. Then, the training data set is used to fit the model that can predict the torque compensation value of the steering wheel based on the body pitch angle, the steering angle and the steering angle rate. Specifically, for a neural network, multiple network layers and multiple neurons can be used to capture the complex relationship between input features; for a support vector regression model, a suitable kernel function can be selected; for a Gaussian process regression model, a suitable covariance function can be selected.

[0081] ​In a specific implementation, in order to obtain a more accurate torque compensation value and make the steering wheel accurately adjust to the desired steering angle, the body pitch angle, the desired steering angle and the desired steering angle rate can be used as model inputs. After the controller determines the target nonlinear regression model, the target nonlinear regression model to be called can be obtained, the real-time body pitch angle, the desired steering angle and the desired steering angle rate are converted into a data format and a range conforming to the model input, and then the converted body pitch angle, the desired steering angle and the desired steering angle rate are used as inputs to call the target nonlinear regression model. The target nonlinear regression model can predict the torque compensation value of the steering wheel in real time, and further, the controller can apply the torque compensation value to the EPS system. In this way, the change of the driving condition can be responded in time, and the vehicle steering performance can be optimized. Further, with the accumulation of more data, the model can be further optimized and updated.

[0082] In the embodiment, the torque compensation value is predicted in real time by the nonlinear regression model, which can capture the complex relationship between the body pitch angle, the desired steering angle and the desired steering angle rate, provide a more accurate torque compensation value, and improve the driving experience.

[0083] As shown in FIG. 6, in one embodiment, S620 includes: Figure 3

[0084] S622, in the case where the number of trained nonlinear regression models is one, the trained nonlinear regression model is determined as the target nonlinear regression model to be called.

[0085] S624, in the case where the number of trained nonlinear regression models is multiple, the real-time vehicle speed of the vehicle is obtained, and based on the real-time vehicle speed, the target nonlinear regression model to be called is selected from the trained multiple nonlinear regression models.

[0086] Since one or more nonlinear regression models can be trained in actual application, the most suitable target nonlinear regression model can be selected as needed when the model is selected. For example, if there is only one model, no additional model selection or comparison process is needed, and the deployed trained nonlinear regression model is directly determined as the target nonlinear regression model to be called. If there are multiple trained models, each model can be optimized for a specific input range, and the most suitable model for the current scene can be selected according to specific conditions, such as road conditions and driving conditions. For example, the current driving condition can be determined according to the current body pitch angle, the desired steering angle and the desired steering angle rate, and then the nonlinear regression model suitable for the current driving condition is selected from the multiple trained nonlinear regression models, and the selected nonlinear regression model is determined as the target nonlinear regression model to be called.

[0087] ​In this embodiment, the target nonlinear regression model to be called can be selected based on the vehicle speed. For example, the model switching can be performed based on a preset rule. For example, a certain type of model is preferentially selected at a certain vehicle speed, or the target nonlinear regression model is switched to when a certain driving mode is detected based on the vehicle speed. Alternatively, the vehicle speed and driving conditions can be monitored in real time, and a suitable model is selected according to the rule.

[0088] In this embodiment, when there is only one trained nonlinear regression model, the model comparison is not needed, the selection of the single model ensures the consistency and predictability of the system, and reduces the computational complexity. When there are multiple trained nonlinear regression models, the target nonlinear regression model to be called is selected based on the real-time vehicle speed, which improves the pertinence and accuracy of model prediction.

[0089] It should be noted that the manner of implementing S624 is not unique. As shown in FIG. 6B, in one embodiment, S624 includes S6242. S6242 compares the real-time vehicle speed with a plurality of preset vehicle speed intervals, determines a target vehicle speed interval in which the real-time vehicle speed is located, and finds a nonlinear regression model matched with the target vehicle speed interval from the plurality of trained nonlinear regression models, and determines the found nonlinear regression model as the target nonlinear regression model. Figure 4

[0090] In this embodiment, for each vehicle speed interval, a corresponding nonlinear regression model can be trained based on the training data under the vehicle speed interval. In the data collection stage, after the key information such as the torsion bar torque of the EPS, the steering wheel angle, the steering wheel angle rate, the vehicle speed, and the vehicle pitch angle (used to reflect the road slope) under different vehicle speeds at historical time points is collected, the collected data is classified and stored according to different vehicle speed intervals, and a training set corresponding to the vehicle speed interval is constructed. For example, the speed interval is divided into a low speed interval (0-30km / h), a medium speed interval (30-80km / h), and a high speed interval (80-150km / h). In the model training stage, for each vehicle speed interval, a nonlinear polynomial regression model for the vehicle speed interval is trained based on the training set under the vehicle speed interval. Taking the nonlinear regression model as a nonlinear polynomial regression model as an example, the nonlinear polynomial regression model can be trained by using a supervised learning algorithm based on the training set, so that the model captures the complex nonlinear relationship between the torque and the steering wheel angle, the steering wheel angle rate, the vehicle speed, and the vehicle pitch angle. The nonlinear polynomial regression model can be expressed as:

[0091]

[0092] where y represents the torque compensation value required by the EPS system ,​ Steering wheel angle representing the EPS system , The steering wheel angle rate representing the EPS system , Represents the vehicle's pitch angle , ~ The parameters to be estimated are obtained by minimizing the sum of squared errors between the predicted and actual values. It is understandable that during the actual prediction process of the model, Corresponding to the desired steering wheel angle, Corresponding steering wheel desired angle rate Furthermore, by simplifying the above expression and dividing the model according to different speed ranges, the expression for the nonlinear polynomial regression model is obtained as follows:

[0093]

[0094] in, It represents vehicle speed.

[0095] By training the aforementioned nonlinear polynomial regression model, it is possible to predict the torque compensation value required by the EPS system in the current scenario based on the acquired desired steering wheel angle, desired steering rate, and vehicle pitch angle. Specifically, this can be achieved by comparing the real-time vehicle speed with multiple preset speed ranges to determine the target speed range. Then, from multiple nonlinear regression models corresponding to different speed ranges, the nonlinear polynomial regression model corresponding to the target speed range is selected and designated as the target nonlinear polynomial regression model. The acquired vehicle pitch angle, desired steering wheel angle, and desired steering rate are then processed into a format matching the model input. The processed vehicle pitch angle, desired steering wheel angle, and desired steering rate are input into the target nonlinear polynomial regression model, which then predicts the steering wheel torque compensation value in real time.

[0096] In this embodiment, the dynamic characteristics of the vehicle under different speed ranges are fully considered. Different nonlinear regression models are trained for different speed ranges. The more suitable model can be flexibly selected to predict the torque compensation value under different speeds and driving conditions, and the obtained torque compensation value is more accurate.

[0097] It's understandable that there are multiple ways to achieve the basic torque of a steering wheel. For example... Figure 5 As shown, in one embodiment, S400 includes:

[0098] S420 determines the angle deviation based on the real-time angle and the desired angle.

[0099] S422 determines the desired steering wheel rate based on the steering angle deviation.

[0100] S424 determines the turning rate deviation based on the desired turning rate and the real-time turning rate.

[0101] S426 determines the basic torque of the steering wheel based on the angular rate deviation.

[0102] In this embodiment, after acquiring the real-time steering angle and the desired steering angle, the controller can calculate the difference between the real-time steering angle and the desired steering angle to obtain the steering angle deviation. Then, using adaptive fuzzy logic or an adaptive neural network, the controller parameters are dynamically adjusted based on the current steering angle deviation to determine the most suitable desired steering angle rate for the current driving conditions. Similarly, the steering angle rate deviation between the real-time steering angle rate and the desired steering angle rate is calculated. Based on the steering angle rate deviation, the basic torque of the steering wheel is calculated using adaptive control.

[0103] In other embodiments, the difference between the real-time steering angle and the desired steering angle can be calculated to obtain the steering angle deviation. Subsequently, based on the vehicle dynamics model and the current steering angle deviation, a model predictive control algorithm is used to predict the steering angle response over a future period, thereby determining the optimal desired steering angle rate. Similarly, the steering angle rate deviation between the real-time steering angle rate and the desired steering angle rate is calculated. Based on the steering angle rate deviation, an MPC (Model Predictive Control) algorithm is used to calculate the base torque of the steering wheel to minimize future prediction errors.

[0104] In this embodiment, by providing a closed-loop control method to offer continuous feedback, the controller can automatically adjust the vehicle steering according to different driving conditions and vehicle status, thereby improving the driver's experience.

[0105] like Figure 6 As shown, in some other exemplary embodiments, S400 includes: S440, determining the desired steering angle rate and base torque of the steering wheel through a preset PID controller based on the real-time steering angle, the real-time steering angle rate and the desired steering angle.

[0106] A PID controller is a closed-loop control mechanism that calculates the deviation between the system output and the desired value (setpoint). It combines proportional, integral, and derivative control terms to generate a control signal that adjusts the controlled system's output to approach and maintain it near the setpoint. Specifically, the proportional term in the PID controller responds to deviations instantly, the integral term eliminates steady-state errors by accumulating past errors, and the derivative term predicts the rate of change of error, reducing overshoot and oscillations through the differential error. These three terms work together to quickly and accurately adjust the steering wheel's base torque. The output of the PID controller... It can be represented as:

[0107]

[0108] in, For the output of the PID controller, This represents the error of the current system, i.e., the difference between the setpoint and the actual output. For system sampling time, The change in error of the current system ( ), This is the proportional gain coefficient. This is the integral gain coefficient. The differential gain coefficient is used to adjust the gain coefficient. , , It can control the system's tracking effect.

[0109] Following the previous embodiment, to obtain a more accurate base torque for the steering wheel and adjust it to the desired angle, a PID controller can be used to achieve higher precision steering control. Specifically, the difference between the real-time steering angle and the desired steering angle can be calculated first to obtain the steering angle deviation. Then, based on the current steering angle deviation, the proportional term of the PID controller is used to calculate the desired steering angle rate. Subsequently, the real-time steering angle is compared with the calculated desired steering angle rate, and the difference between the two is calculated to obtain the steering angle rate deviation. Based on the steering angle rate deviation, the PID controller determines the base torque of the steering wheel.

[0110] In this embodiment, the introduction of a PID controller can effectively improve the system's response speed and control accuracy, achieving more precise steering control.

[0111] like Figure 7 As shown, in one exemplary embodiment, S440 includes:

[0112] S442, taking the real-time steering angle and the desired steering angle as inputs of the position loop PID controller, obtaining the desired steering angle rate of the steering wheel.

[0113] S444, taking the desired steering angle rate and the real-time steering angle rate as inputs of the preset speed loop PID controller, obtaining the basic torque of the steering wheel.

[0114] In this embodiment, by introducing a cascade double-loop controller, the extreme precision of steering angle tracking is improved by the design principle of "the differential of the outer loop is the differential of the inner loop". The position loop PID controller, i.e. the outer loop controller, focuses on tracking the target steering angle. The goal of the outer loop controller is to ensure that the vehicle can turn according to the desired steering angle while maintaining the smoothness and stability of the steering process. The speed loop PID controller, i.e. the inner loop controller, is mainly responsible for accurate tracking of the target steering angle rate. The main goal of the inner loop controller is to ensure that the EPS system can quickly and accurately respond to changes in the steering angle rate, thereby achieving stable tracking of the steering angle rate.

[0115] In specific implementation, the desired steering angle and the real-time steering angle can be taken as inputs of the position loop PID control. The position loop PID control determines the deviation between the desired steering angle and the real-time steering angle to obtain the steering angle deviation. Then, based on the steering angle deviation, the desired steering angle rate is calculated through the PID control algorithm, and the desired steering angle rate is taken as the input of the speed loop PID controller. The speed loop PID controller obtains the steering angle rate deviation based on the input desired steering angle rate and real-time steering angle rate, and then calculates the corresponding basic torque through the PID control algorithm. In this process, the position loop PID controller can achieve accurate tracking of the target steering angle by continuously adjusting the desired steering angle rate, thereby improving the vehicle's handling performance and driving safety. It should be noted that the output of the position loop PID controller can be limited according to the actual performance of the system response, The maximum steering wheel steering angle rate of the system in the maximum torque control state is:

[0116]

[0117] In this embodiment, by introducing a cascade double-loop PID controller, the vehicle can turn according to the desired steering angle while maintaining the smoothness and stability of the steering process through the outer loop controller, and the steering responsiveness and handling stability of the vehicle are improved through the inner loop controller. Compared with the traditional single-loop PID architecture, the tracking effect is greatly improved, and the robustness of the system is significantly enhanced.

[0118] It can be understood that after the desired steering angle is obtained, the desired steering angle can also be pre-processed by the tracking differentiator to quickly track and differentiate the desired steering angle. In one of the exemplary embodiments, obtaining the desired steering angle of the steering wheel comprises: obtaining an initial desired steering angle of the steering wheel, initializing parameters of the tracking differentiator, taking the initial desired steering angle as an input of the initialized tracking differentiator, and the initialized tracking differentiator updating internal state variables through a preset tracking differentiator function, a real-time steering angle and a real-time steering angle rate to obtain an optimized desired steering angle of the steering wheel.

[0119] The tracking differentiator (TD, hereinafter referred to as TD tracker) is a special dynamic system model, through which fast tracking and differentiation of input signals can be achieved. Specifically, the TD tracker can track and smooth the input signal based on a set of state equations and a tracking differentiator function (such as fhan function). The TD tracker specifically includes the following aspects of processing:

[0120] 1. State equation description: TD tracker usually describes its dynamic behavior through a set of state equations. For example, the state equation of linear TD can be expressed in the form of a second-order system, while nonlinear TD may adopt a more complex dynamic model.

[0121] 2. Tracking differentiator function: In order to achieve the fastest tracking and differentiation, a tracking differentiator function (such as fhan function) is introduced in TD, which calculates the optimal control input according to the current state and target value.

[0122] 3. Parameter adjustment: The performance of TD can be optimized by adjusting its parameters (such as velocity factor r, filter factor h, etc.). Among them, the velocity factor r determines the tracking speed, and the filter factor h affects the noise suppression ability. Its discrete expression is:

[0123]

[0124] where, is the steering angle of the steering wheel at the kth step; is the steering angle rate of the steering wheel at the kth step; is the steering angle of the steering wheel at the next step; is the steering angle rate of the steering wheel at the next step; is the optimal acceleration calculated; is the desired steering angle of the steering wheel; is the velocity factor, the larger the value, the faster the approximation speed, but the tracking effect cannot exceed the maximum bearing capacity of the system; is the integral step.

[0125] To reduce the influence of the step signal on the tracking effect, a TD tracker can be added in the front end of the system input to make the step signal transition smoothly and output more smoothly.

[0126] Exemplarily, the parameters of the TD tracker, such as the speed factor r and the filter factor h, can be initialized according to actual scenes and performance requirements. In each sampling period k, the TD tracker updates its internal state using the current state (such as the steering angle and the steering angle rate) and the initial desired steering angle. Finally, the fhan function obtains the optimized desired steering angle according to the current state (the steering angle and the steering angle rate) and the desired steering angle, in combination with the speed factor r and the filter factor h, and inputs the optimized desired steering angle into the position loop PID controller to further improve the tracking of the target position and the steering angle.

[0127] In this embodiment, the introduction of the steepest descent tracker for optimizing the initial desired steering angle can make the step signal transition smoothly, reduce the influence of the mutation on the control system, and improve the dynamic performance of the system.

[0128] To make the steering wheel torque determination method provided in the present application clearer, a specific embodiment is described below in combination with the accompanying drawings, which includes the following steps: Figure 8

[0129] S100, acquiring a body pitch angle, and a real-time steering angle, a real-time steering angle rate and a desired steering angle of a steering wheel.

[0130] In actual application, a TD tracker is added in the front end of the system input. Exemplarily, the vehicle can be installed with multiple sensors, including a steering angle sensor and an angular velocity sensor. The steering angle sensor can be installed in a steering column or inside a steering machine to detect the rotation angle of the steering wheel and provide the rotation angle of the steering wheel relative to the initial position. The number of angular velocity sensors can be multiple, which are installed at different positions of the vehicle to accurately measure the angular velocity of the vehicle rotating around the vertical axis. The pitch angle is obtained by an inertial measurement unit. The inertial measurement unit usually contains an accelerometer and a gyroscope, which can provide three-axis acceleration and angular velocity information, and the body pitch angle can be calculated through a fusion algorithm. The lateral control module plans the ideal steering angle of the front path according to a preset path planning algorithm, i.e., the initial desired steering angle, and sends the initial desired steering angle to the controller. The controller pre-processes the initial desired steering angle through the TD tracker to make the output more smooth.

[0131] ​Specifically, the parameters of the TD tracker, such as the speed factor r and the filtering factor h, can be initialized according to the actual scene and performance requirements. At each sampling period k, the TD tracker updates its internal state using the current state (such as the steering angle and the steering angle rate) and the initial desired steering angle. Finally, the fhan function combines the speed factor r and the filtering factor h to obtain the optimized desired steering angle based on the current state (the steering angle and the steering angle rate) and the initial desired steering angle, and then inputs the optimized desired steering angle into the position loop PID controller to further improve the tracking of the target position and steering angle.

[0132] S102, taking the real-time steering angle and the desired steering angle as inputs of the position loop PID controller, obtaining the desired steering angle rate of the steering wheel.

[0133] S104, taking the desired steering angle rate and the real-time steering angle rate as inputs of the speed loop PID controller, obtaining the basic torque of the steering wheel.

[0134] After obtaining the desired steering angle after the above steps, the desired steering angle and the real-time steering angle can be input into the position loop PID control. The position loop PID control determines the deviation between the desired steering angle and the real-time steering angle, and then calculates the desired steering angle rate based on the steering angle deviation through the PID control algorithm. The desired steering angle rate is input into the speed loop PID controller. The speed loop PID controller obtains the steering angle rate deviation based on the input desired steering angle rate and real-time steering angle rate, and then calculates the corresponding basic torque through the PID control algorithm. In this process, the position loop PID controller can realize accurate tracking of the target steering angle by continuously adjusting the desired steering angle rate, thereby improving the control performance and driving safety of the vehicle. It should be noted that the output of the position loop PID controller can be limited according to the actual performance of the system response, The maximum steering wheel steering angle rate of the system in the maximum torque control state is:

[0135]

[0136] S106, obtaining the real-time vehicle speed of the vehicle, comparing the real-time vehicle speed with the preset plurality of vehicle speed intervals to determine the target vehicle speed interval in which the real-time vehicle speed is located, and searching for a nonlinear regression model matched with the target vehicle speed interval from the plurality of trained nonlinear regression models to determine the searched nonlinear regression model as the target nonlinear regression model.

[0137] S108, taking the body pitch angle, the desired steering angle and the desired steering angle rate as inputs, calling the target nonlinear regression model to predict the torque compensation value of the steering wheel.

[0138] Specifically, for each vehicle speed interval, a corresponding nonlinear regression model can be trained based on the training data under the vehicle speed interval. In the data collection phase, after collecting key information such as the torsion bar torque of the EPS, the steering wheel angle, the steering wheel angle rate, the vehicle speed, and the vehicle pitch angle (used to reflect the road slope) under different vehicle speeds at historical time points, the speed interval is divided into a low speed interval (0-30km / h), a medium speed interval (30-80km / h), and a high speed interval (80-150km / h). The collected data is classified and stored according to different vehicle speed intervals, and a training set corresponding to the vehicle speed interval is constructed. In the model training phase, for each vehicle speed interval, a nonlinear polynomial regression model for the vehicle speed interval is trained based on the training set under the vehicle speed interval. Taking the nonlinear regression model as a nonlinear polynomial regression model as an example, based on the training set, the nonlinear polynomial regression model can be trained using a supervised learning algorithm, so that the model captures the complex nonlinear relationship between the torque and the steering wheel angle, the steering wheel angle rate, the vehicle speed, and the vehicle pitch angle. The nonlinear polynomial regression model can be expressed as:

[0139]

[0140] where y represents the torque compensation value required by the EPS system , represents the steering wheel angle of the EPS system , represents the steering wheel angle rate of the EPS system , represents the vehicle pitch angle , ~ is a parameter to be estimated, obtained by minimizing the sum of squared errors between the predicted value and the actual value. It can be understood that in the actual prediction process of the model, corresponds to the expected steering wheel angle, corresponds to the expected steering wheel angle rate . Further, by simplifying the above expression and dividing the model according to different speed intervals, the nonlinear polynomial regression model expression is obtained as:

[0141]

[0142] where represents the vehicle speed.

[0143] By training the above nonlinear polynomial regression model, the torque compensation value required by the EPS system in the current scene can be predicted based on the obtained steering wheel expected steering angle, expected steering angle rate and vehicle body pitch angle in actual application. Specifically, the real-time vehicle speed can be compared with the preset multiple vehicle speed intervals to determine the target vehicle speed interval in which the real-time vehicle speed is located, and then a nonlinear polynomial regression model corresponding to the target vehicle speed interval is selected from multiple nonlinear regression models corresponding to different vehicle speed intervals, and the selected nonlinear polynomial regression model is determined as the target nonlinear polynomial regression model. The obtained vehicle body pitch angle, expected steering angle and expected steering angle rate are processed into data matching the format of the model input, and the processed vehicle body pitch angle, expected steering angle and expected steering angle rate are input into the target nonlinear polynomial regression model, and the model predicts the torque compensation value of the steering wheel in real time.

[0144] In S110, the target torque of the steering wheel is determined based on the torque compensation value and the base torque, and the target torque is used to adjust the steering wheel to the expected steering angle.

[0145] After the controller determines the torque compensation value and the base torque, the base torque and the torque compensation value can be synthesized, and the sum of the torque compensation value and the base torque is determined as the target torque of the steering wheel. Further, the controller can adjust the output of the electric power steering motor based on the target torque using a closed-loop control strategy, and the output of the electric power steering motor targets the torque to provide assistance to the driver turning the steering wheel and assist the driver in controlling the steering of the vehicle.

[0146] The above-mentioned steering wheel torque determination method fully considers factors such as vehicle body posture, steering angle rate and steering angle that affect steering performance, introduces a torque feedforward compensation mechanism suitable for different scenarios, and based on the torque feedforward compensation mechanism, predicts the torque compensation value of the steering wheel according to the expected steering angle rate, vehicle body pitch angle and expected steering angle, and then compensates the torque based on the torque compensation value and the base torque, i.e. dynamically adjusts the target torque of the steering wheel, effectively reduces the steering error caused by factors such as changes in vehicle dynamic characteristics and external environmental disturbances, and realizes precise control of the EPS expected steering angle. Moreover, the entire process does not need to repeatedly calibrate control parameters in different driving environments, reduces the complexity of the control system, and improves the control adaptability and robustness.

[0147] It should be understood that, although the steps in the flowcharts involved in the embodiments described above are shown in sequence according to the arrows, the steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, the execution of the steps is not strictly limited in sequence, and the steps can be executed in other sequences. Moreover, at least some of the steps in the flowcharts involved in the embodiments described above can include multiple steps or multiple stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution of the steps or stages is not necessarily sequential, but can be executed alternately or alternately with at least part of other steps or steps or stages in other steps.

[0148] Based on the same inventive concept, the embodiments of the present application also provide a steering wheel torque determination device for implementing the above-mentioned steering wheel torque determination method. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme described in the above method, so the specific limitations in one or more steering wheel torque determination device embodiments provided below can refer to the limitations of the steering wheel torque determination method described above, and will not be repeated here.

[0149] In an exemplary embodiment, as shown in Figure 9 A steering wheel torque determination device 900 is provided, comprising a data acquisition module 910, a basic torque determination module 920, a compensation torque prediction module 930 and a target torque determination module 940, wherein:

[0150] The data acquisition module 910 is configured to acquire the body pitch angle, and the real-time steering angle, the real-time steering angle rate and the expected steering angle of the steering wheel.

[0151] The basic torque determination module 920 is configured to determine the expected steering angle rate and the basic torque of the steering wheel according to the real-time steering angle, the real-time steering angle rate and the expected steering angle.

[0152] The compensation torque prediction module 930 is configured to predict the torque compensation value of the steering wheel according to the body pitch angle, the expected steering angle and the expected steering angle rate.

[0153] The target torque determination module 940 is configured to determine the target torque of the steering wheel based on the torque compensation value and the basic torque, and the target torque is used to adjust the steering wheel to the expected steering angle.

[0154] The steering wheel torque control device fully considers factors influencing steering performance such as a vehicle body posture, a steering angle rate, and a steering angle, and introduces a torque feedforward compensation mechanism suitable for different scenarios. On the one hand, the expected steering angle rate and the basic torque of the steering wheel are determined according to the real-time steering angle, the real-time steering angle rate, and the expected steering angle. On the other hand, the torque compensation value of the steering wheel is predicted according to the expected steering angle rate, the vehicle body pitch angle, and the expected steering angle based on the torque feedforward compensation mechanism. Then, the torque compensation is performed based on the torque compensation value and the basic torque, that is, the target torque of the steering wheel is dynamically adjusted. The steering error caused by factors such as changes in vehicle dynamic characteristics and external environmental disturbances is effectively reduced, and the accurate control of the EPS expected steering angle is realized. Moreover, the control parameters under different driving environments do not need to be repeatedly calibrated in the whole process, the complexity of the control system is reduced, the control adaptability and robustness are improved, the torque compensation value of the steering wheel is more accurate, and the control precision of the EPS expected steering angle is further improved.

[0155] In one of the example embodiments, the compensation torque prediction module 930 is further configured to determine a target nonlinear regression model to be called, and the nonlinear regression model is trained based on the vehicle body pitch angle, the steering angle, and the steering angle rate in different vehicle speed intervals at historical time points, and the target nonlinear regression model is called to predict the torque compensation value of the steering wheel with the vehicle body pitch angle, the expected steering angle, and the expected steering angle rate as inputs.

[0156] In one of the example embodiments, the compensation torque prediction module 930 is further configured to, in a case where the number of trained nonlinear regression models is one, determine the trained nonlinear regression model as the target nonlinear regression model to be called, and in a case where the number of trained nonlinear regression models is multiple, acquire a real-time vehicle speed, and based on the real-time vehicle speed, screen the target nonlinear regression model to be called from the trained multiple nonlinear regression models, and the nonlinear regression model corresponds to the vehicle speed interval one by one.

[0157] In one of the example embodiments, the compensation torque prediction module 930 is further configured to compare the real-time vehicle speed with the preset multiple vehicle speed intervals, determine a target vehicle speed interval where the real-time vehicle speed is located, find out a nonlinear regression model matched with the target vehicle speed interval from the trained multiple nonlinear regression models, and determine the found nonlinear regression model as the target nonlinear regression model to be called.

[0158] In one of the example embodiments, the basic torque determination module 920 is further configured to determine a steering angle deviation according to the real-time steering angle and the expected steering angle, determine the expected steering angle rate of the steering wheel based on the steering angle deviation, determine a steering angle rate deviation according to the expected steering angle rate and the real-time steering angle rate, and determine the basic torque of the steering wheel based on the steering angle rate deviation.

[0159] In one of the illustrative embodiments, the base torque determination module 920 is further configured to determine the desired steering wheel angular velocity and the base torque of the steering wheel according to the real-time steering wheel angle, the real-time steering wheel angular velocity and the desired steering wheel angle by a preset PID controller.

[0160] In one of the illustrative embodiments, the preset PID controller includes a position loop PID controller and a velocity loop PID controller; and the base torque determination module 920 is further configured to take the real-time steering wheel angle and the desired steering wheel angle as inputs of the position loop PID controller to obtain the desired steering wheel angular velocity, and take the desired steering wheel angular velocity and the real-time steering wheel angular velocity as inputs of the preset velocity loop PID controller to obtain the base torque of the steering wheel.

[0161] As shown in Figure 10 In one of the illustrative embodiments, the apparatus further includes a signal preprocessing module 902 configured to obtain an initial desired steering wheel angle, initialize parameters of the steepest descent tracker, and take the initial desired steering wheel angle as an input of the initialized steepest descent tracker, so that the initialized steepest descent tracker updates internal state variables by a preset steepest control function, the real-time steering wheel angle and the real-time steering wheel angular velocity to obtain an optimized desired steering wheel angle.

[0162] The above-mentioned modules of the steering wheel torque determination apparatus can be realized by software, hardware or a combination thereof. The above-mentioned modules can be embedded in or independent of a processor in a computer device in hardware form, or stored in a memory in a computer device in software form, so as to be called and executed by a processor to perform operations corresponding to the above-mentioned modules.

[0163] In one of the illustrative embodiments, a computer device is provided, which can be a terminal, and specifically can include but is not limited to a vehicle, a ship and other transportation tools. Its internal structure diagram can be as shown in Figure 11The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit and an input device. The processor, the memory and the input / output interface are connected through a system bus. The communication interface, the display unit and the input device are connected to the system bus through the input / output interface. The processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for running the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is configured to exchange information between the processor and external devices. The communication interface of the computer device is configured to perform wired or wireless communication with external terminals. The wireless communication can be achieved through WIFI, mobile cellular network, Near Field Communication (NFC) or other technologies. The computer program is executed by the processor to implement a steering wheel torque determination method. The display unit of the computer device is configured to form a visually visible picture, which can be a display screen, a projection device or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer overlaid on the display screen, or a key, a trackball or a touchpad arranged on the shell of the computer device, or an external keyboard, a touchpad or a mouse, etc.

[0164] Those skilled in the art can understand that Figure 11 The skilled in the art can understand that

[0165] In one example embodiment, a computer device is provided, including a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the steps in any one of the above steering wheel torque determination method embodiments.

[0166] In one embodiment, a computer readable storage medium is provided, storing a computer program, and the computer program is executed by a processor to implement the steps in any one of the above steering wheel torque determination method embodiments.

[0167] In one embodiment, a computer program product is provided, including a computer program, and the computer program is executed by a processor to implement the steps in any one of the above steering wheel torque determination method embodiments.

[0168] It should be noted that the data involved in the present application (including but not limited to data for analysis such as expected corner, vehicle speed and corner rate, stored data, displayed data, etc.) are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of related data need to comply with relevant regulations.

[0169] It can be understood by those skilled in the art that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing related hardware through a computer program. The computer program can be stored in a non-volatile computer readable storage medium, and when executed, can include the processes of the above-mentioned embodiments. Any reference to memory, database or other medium used in the embodiments provided by the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical storage, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration but not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The database involved in the embodiments provided by the present application can include at least one of a relational database and a non-relational database. The non-relational database can include a distributed database based on a block chain, etc., without being limited thereto. The processor involved in the embodiments provided by the present application can be a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., without being limited thereto.

[0170] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0171] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A steering wheel torque determination method characterized by, The method comprises: acquiring a real-time steering angle, a real-time steering angle rate and a desired steering angle of a steering wheel, and a body pitch angle of a vehicle; determining a desired steering angle rate and a basic torque of the steering wheel according to the real-time steering angle, the real-time steering angle rate and the desired steering angle; determining a target nonlinear regression model to be called, calling the target nonlinear regression model with the body pitch angle, the desired steering angle and the desired steering angle rate as inputs, and predicting a torque compensation value of the steering wheel, the target nonlinear regression model comprising a nonlinear polynomial regression model trained based on body pitch angles, steering angles and steering angle rates of the steering wheel in different vehicle speed intervals at historical time points, the nonlinear polynomial regression model being used to capture a nonlinear relationship between the torque and the steering angle, the steering angle rate, the vehicle speed and the vehicle pitch angle; determining a target torque of the steering wheel based on the torque compensation value and the basic torque, the target torque being used to adjust the steering wheel to the desired steering angle.

2. The method of claim 1, wherein, The determining of the target nonlinear regression model to be called comprises: in a case where the number of trained nonlinear regression models is one, determining the trained nonlinear regression model as the target nonlinear regression model to be called; in a case where the number of trained nonlinear regression models is multiple, acquiring a real-time vehicle speed, and screening the target nonlinear regression model to be called from the trained multiple nonlinear regression models based on the real-time vehicle speed, the nonlinear regression model corresponding to a vehicle speed interval one by one.

3. The method of claim 2, wherein, The screening of the target nonlinear regression model to be called from the trained multiple nonlinear regression models based on the real-time vehicle speed comprises: comparing the real-time vehicle speed with preset multiple vehicle speed intervals to determine a target vehicle speed interval where the real-time vehicle speed is located; finding out a nonlinear regression model matched with the target vehicle speed interval from the trained multiple nonlinear regression models, and determining the found nonlinear regression model as the target nonlinear regression model to be called.

4. The method according to any one of claims 1 to 3, characterized in that, The determining of the desired steering angle rate and the basic torque of the steering wheel according to the real-time steering angle, the real-time steering angle rate and the desired steering angle comprises: determining a steering angle deviation according to the real-time steering angle and the desired steering angle; determining the desired steering angle rate of the steering wheel based on the steering angle deviation; determining a steering angle rate deviation according to the desired steering angle rate and the real-time steering angle rate; determining the basic torque of the steering wheel based on the steering angle rate deviation.

5. The method according to any one of claims 1 to 3, characterized in that, The determining of the desired steering angle rate and the basic torque of the steering wheel according to the real-time steering angle, the real-time steering angle rate and the desired steering angle comprises: determining the desired steering angle rate and the basic torque of the steering wheel through a preset PID controller according to the real-time steering angle, the real-time steering angle rate and the desired steering angle.

6. The method of claim 5, wherein, The preset PID controller comprises a position loop PID controller and a speed loop PID controller; the determining of the desired steering angle rate and the basic torque of the steering wheel through the preset PID controller according to the real-time steering angle, the real-time steering angle rate and the desired steering angle comprises: The real-time steering angle and the expected steering angle are input to the position loop PID controller to obtain an expected steering angle rate of the steering wheel; The expected steering angle rate and the real-time steering angle rate are input to the speed loop PID controller to obtain a base torque of the steering wheel.

7. The method of claim 6, wherein, An expected steering angle of a steering wheel is obtained, including: An initial expected steering angle of a steering wheel is obtained; Parameters of a maximum derivative tracker are initialized; The initial expected steering angle is input to the initialized maximum derivative tracker, which updates internal state variables through a preset maximum control function, a real-time steering angle and a real-time steering angle rate to obtain an optimized expected steering angle of the steering wheel.

8. A steering wheel torque determination device characterized by comprising: The device includes: A data acquisition module is configured to acquire a body pitch angle, a real-time steering angle, a real-time steering angle rate and an expected steering angle of a steering wheel; A base torque determination module is configured to determine an expected steering angle rate and a base torque of the steering wheel according to the real-time steering angle, the real-time steering angle rate and the expected steering angle; A compensation torque prediction module is configured to determine a target nonlinear regression model to be called, and to call the target nonlinear regression model to predict a torque compensation value of the steering wheel with the body pitch angle, the expected steering angle and the expected steering angle rate as input, the target nonlinear regression model including a nonlinear polynomial regression model trained based on body pitch angles, steering angles and steering angle rates in different vehicle speed intervals at historical time points, the nonlinear polynomial regression model being used to capture a nonlinear relationship between torque and steering angle, steering angle rate, vehicle speed and vehicle pitch angle; A target torque determination module is configured to determine a target torque of the steering wheel based on the torque compensation value and the base torque, the target torque being used to adjust the steering wheel to the expected steering angle.

9. The apparatus of claim 8, wherein, The base torque determination module is further configured to determine a steering angle deviation based on the real-time steering angle and the expected steering angle, to determine an expected steering angle rate of the steering wheel based on the steering angle deviation, to determine a steering angle rate deviation based on the expected steering angle rate and the real-time steering angle rate, and to determine a base torque of the steering wheel based on the steering angle rate deviation. 10.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is configured to perform the method according to any one of claims 1-9. The processor executes the computer program to implement the steps of the method of any one of claims 1 to 7.

Images