A steering wheel angle self-learning method and system

Abstract

The application discloses a steering wheel angle self-learning method, comprising the following steps: step 1: judging whether to enter a self-learning step based on vehicle parameters; step 2: after entering the self-learning step, multiplying a difference value obtained by subtracting a real-time collected steering wheel rotation angle from a memorized steering wheel rotation angle deviation by a filtering coefficient, and recording the filtering coefficient as a filtering value; step 3: taking a sum of the filtering value and the memorized steering wheel rotation angle deviation as a real-time rotation angle deviation value; and step 4: adding the current real-time collected steering wheel rotation angle and the real-time rotation angle deviation value to obtain an actual steering wheel angle value. The application has the advantages that the steering wheel deviation and the corresponding actual steering wheel rotation angle can be accurately and quickly obtained, and the obtained steering wheel angle is more accurate by filtering the error caused by fluctuation through the filtering coefficient.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of auxiliary driving of automobile, in particular to a steering wheel angle self-learning method and system for auxiliary driving. BACKGROUND

[0002] ADAS auxiliary driving technology can recognize the road in front of the vehicle by using a front camera, and output lane line information to a control unit. The control unit controls the vehicle to drive in the middle of the road according to the received lane line information, which is the lane centering function. Current steering wheel control is generally divided into two types, one is torque control, and the other is angle control. In the steering wheel torque control, the final output torque is also obtained from the target angle of the steering wheel, so the actual steering wheel angle signal is very important in the lane lateral control function.

[0003] However, in actual vehicles, due to the installation angle deviation of the steering wheel or wear and other reasons, the signal often deviates. If the PD control is simply relied on in the control, the vehicle will deviate in one side of the lane in this case, and the driver is easy to feel uneasy. If the integral term is added in the PID control, the accumulation of error needs time, so in the case that the integral term is not enough at the beginning, the vehicle control will still be deviated. As a result, the vehicle will still deviate in one side at the initial moment when the function is started, and the deviation is constantly changing with time, so there is a deficiency in the existing technology in the steering wheel angle control, and the actual angle signal of the steering wheel needs to be accurately and quickly obtained. SUMMARY

[0004] The purpose of the present application is to overcome the deficiencies of the prior art, and provide a steering wheel angle self-learning method and system for accurately and quickly obtaining the deviation of the steering wheel and the corresponding control angle.

[0005] In order to achieve the above purpose, the technical scheme adopted by the present application is:

[0006] A steering wheel angle self-learning method, comprising,

[0007] Step 1: judging whether to enter the self-learning step based on the vehicle parameters;

[0008] Step 2: after entering the self-learning step, multiplying the difference value obtained by the difference between the real-time collected steering wheel rotation angle and the memorized steering wheel rotation angle deviation by a filtering coefficient, and recording it as a filtering value;

[0009] Step 3: taking the sum of the filtering value and the memorized steering wheel rotation angle deviation as the real-time rotation angle deviation value;

[0010] Step 4: adding the current real-time collected steering wheel rotation angle and the real-time rotation angle deviation value to obtain the actual angle value of the steering wheel.

[0011] In step 1, it is judged based on the parameters of the vehicle whether the vehicle is in a straight driving state, and if it is in a straight driving state, it enters a self-learning step.

[0012] The vehicle parameters for judging whether to enter the self-learning step include the yaw rate of the vehicle, the steering wheel angular velocity, the vehicle speed, the steering wheel angle, the torque of the driver applied to the steering wheel, and the yaw rate of the vehicle, the steering wheel angular velocity, the vehicle speed, the steering wheel angle, and the torque of the driver applied to the steering wheel are used to judge whether the vehicle is in a straight driving state.

[0013] When the absolute value of the yaw rate of the vehicle is less than the yaw rate threshold, the absolute value of the steering wheel angular velocity is less than the angular velocity threshold, the vehicle speed is greater than the vehicle speed threshold, the steering wheel angle is less than the steering wheel angle threshold, and the torque of the driver applied to the steering wheel is less than the torque threshold, it is judged that the vehicle is in a straight driving state at this time, and at this time, the angle self-learning step is entered.

[0014] In step 2, the filter coefficient is a real number less than 1.

[0015] The filter coefficient is positively correlated with the vehicle speed, and the faster the vehicle speed, the larger the corresponding filter coefficient.

[0016] In step 3, the real-time steering angle deviation value is compared with the memorized steering wheel angle deviation, and if they are different, the memorized steering wheel angle deviation is updated to the real-time steering angle deviation value.

[0017] A steering wheel angle self-learning system, comprising:

[0018] A vehicle data acquisition unit for acquiring parameters of the vehicle;

[0019] A self-learning starting unit receives the vehicle parameters sent by the vehicle data acquisition unit and judges whether the angle self-learning condition is met based on the vehicle parameters;

[0020] A processing unit connected to the vehicle data acquisition unit and the self-learning starting unit, after judging to enter the self-learning, the difference between the real-time collected steering wheel angle and the memorized steering wheel angle deviation is multiplied by the filter coefficient, which is recorded as the filter value; then the sum of the filter value and the memorized steering wheel angle deviation is taken as the real-time steering angle deviation value; and the current real-time collected steering wheel angle and the real-time steering angle deviation value are added to obtain the actual steering angle value of the steering wheel.

[0021] The output end of the processing unit is connected to an actuator for sending the actual steering angle value of the steering wheel obtained by the self-learning to the actuator for control operation based on the actual angle.

[0022] The present application has the advantages that the deviation of the steering wheel and the corresponding actual steering wheel rotation angle can be obtained accurately and quickly, the error caused by fluctuation is filtered by the filter coefficient, so that the obtained steering wheel angle is more accurate, the filter coefficient is adjusted based on the vehicle speed, so that the accuracy of the rotation angle self-learning is further ensured, and the memory rotation error is updated in real time, so that the actual rotation angle value calculated based on the actual error of the rotation angle is more accurate. BRIEF DESCRIPTION OF DRAWINGS

[0023] The content expressed by each drawing of the present application and the marks in the drawings are briefly described as follows.

[0024] Figure 1 The flowchart of the self-learning method of the present application. DETAILED DESCRIPTION

[0025] The specific embodiments of the present application are further described in detail by comparing the drawings and describing the optimal embodiments.

[0026] As shown in Figure 1 When the vehicle is started, whether the vehicle meets the self-learning condition is judged in real time,

[0027] When the vehicle state meets the following conditions:

[0028] (1) the absolute value of the yaw rate of the vehicle is less than a threshold;

[0029] (2) the absolute value of the rotation rate of the steering wheel is less than a threshold;

[0030] (3) the vehicle speed is greater than a threshold;

[0031] (4) the steering wheel rotation angle is less than a threshold;

[0032] (5) the torque applied to the steering wheel by the driver is less than a threshold.

[0033] When the above conditions are met, it indicates that the vehicle is in a straight-line driving state, and at this time, the straight-line driving state meets the basic requirement of entering the self-learning step, and the self-learning step is entered.

[0034] After entering the self-learning step, the filtered steering wheel angle at this time is the steering wheel angle error when the vehicle is straight running. The specific method is to subtract the real-time steering wheel angle at this time from the system-remembered steering wheel angle deviation, then multiply the difference by a coefficient (less than 1), and then add it to the current steering wheel angle deviation (actual-remembered deviation) * filter coefficient + remembered deviation. The actual value of the deviation is obtained by multiplying the difference obtained by the filter coefficient and then summing it with the remembered steering wheel angle deviation. In this way, the fluctuation of the steering angle sensor can be eliminated, and the final value will be infinitely close to the actual value of the steering wheel angle deviation. The coefficient multiplied by the deviation can be considered as a filter coefficient, which is positively correlated with the vehicle speed. If the vehicle speed is fast, the steering wheel angle changes less, and a larger filter coefficient is selected, so the time of the entire steering wheel angle self-learning will be smaller. Conversely, when the vehicle speed is slow, the steering wheel angle changes more, and a smaller filter coefficient is selected, so the self-learning time will be longer, otherwise the learning result will be inaccurate. The steering wheel angle input to the control is the result of subtracting the deviation from the steering angle sensor value. The self-learning principle will be described in detail below:

[0035] A steering wheel angle self-learning method, comprising the following steps:

[0036] Step 1: Determine whether to enter the self-learning step based on vehicle parameters; in this step, it is necessary to determine whether the vehicle is in a straight running state, and only in the straight running state can the self-learning step be started to self-learn the steering wheel angle. Based on the vehicle parameters, it is determined whether the vehicle is in a straight running state, and if it is in a straight running state, the self-learning step is entered. Here, straight running is used to start self-learning, and only when the vehicle is straight running, the angle of the steering wheel should be 0 without deviation. The steering wheel angle at this time is the deviation angle.

[0037] The vehicle parameters for determining whether to enter the self-learning step include the vehicle yaw rate, steering wheel angular velocity, vehicle speed, steering wheel angle, and driver torque applied to the steering wheel. Based on the vehicle yaw rate, steering wheel angular velocity, vehicle speed, steering wheel angle, and driver torque applied to the steering wheel, it is determined whether the vehicle is in a straight running state. When the absolute value of the vehicle yaw rate is less than the yaw rate threshold, the absolute value of the steering wheel angular velocity is less than the angular velocity threshold, the vehicle speed is greater than the vehicle speed threshold, the steering wheel angle is less than the steering wheel angle threshold, and the driver torque applied to the steering wheel is less than the torque threshold, it is determined that the vehicle is in a straight running state at this time, and the angle self-learning step is entered.

[0038] Step 2: After entering the self-learning step, multiply the difference obtained by subtracting the real-time steering wheel angle from the remembered steering wheel angle deviation by a filter coefficient, denoted as a filter value.

[0039] Step 3: the sum of the filtered value and the memorized steering wheel angle deviation is taken as the real-time steering wheel angle deviation value;

[0040] Step 4: the current real-time collected steering wheel angle is added to the real-time steering wheel angle deviation value to obtain the actual steering wheel angle value. When the vehicle is driving straight, the steering wheel angle is the deviation angle. The result calculated in the present patent is the filtering of the angle, but the filtering is relatively slow, and the result should be accurate.

[0041] In order to adjust in real time and meet the self-learning accuracy of the steering angle, after starting the self-learning step, steps 1-4 are periodically executed, and the real-time steering angle deviation value calculated in step 4 or 3 is compared with the memorized steering wheel angle deviation. If they are different, the memorized steering wheel angle deviation is updated to the real-time steering angle deviation value. Thus, the memorized steering wheel angle deviation used in each execution of steps 1-4 will be infinitely close to the true value, so that the self-learning is automatically close to the true value and the calculated steering angle is more accurate.

[0042] In step 2, the filtering coefficient is a real number less than 1. The filtering coefficient actually has a correcting effect, mainly correcting the deviation, so as to ensure that the true deviation value is more accurate, thereby realizing the accuracy of the calculation of the steering wheel angle. The filtering coefficient is positively correlated with the vehicle speed. The faster the vehicle speed, the larger the corresponding filtering coefficient. The corresponding relationship between the speed and the filtering coefficient is calibrated in advance, forming a comparison table or being fitted into a speed-filtering coefficient relationship curve according to an interpolation method such as least squares method, and then stored for calling and obtaining later.

[0043] In order to run the above self-learning method, the present application also provides a steering wheel angle self-learning system, comprising:

[0044] a vehicle data acquisition unit for acquiring parameters of the vehicle; including: the yaw rate of the vehicle, the steering wheel angular velocity, the vehicle speed, the steering wheel angle, and the torque applied by the driver to the steering wheel;

[0045] a self-learning starting unit receiving the vehicle parameters sent by the vehicle data acquisition unit and determining whether the angle self-learning condition is met based on the vehicle parameters; when the absolute value of the yaw rate of the vehicle is less than the yaw rate threshold, the absolute value of the steering wheel angular velocity is less than the steering wheel angular velocity threshold, the vehicle speed is greater than the vehicle speed threshold, the steering wheel angle is less than the steering wheel angle threshold, and the torque applied by the driver to the steering wheel is less than the torque threshold, it is determined that the vehicle is in a straight driving state at this time, and the angle self-learning step is entered at this time.

[0046] The processing unit is connected to the vehicle data acquisition unit and the self-learning starting unit respectively. After judging that the self-learning is entered, a difference value obtained by subtracting the real-time collected steering wheel angle from the memorized steering wheel angle deviation is multiplied by a filtering coefficient, which is recorded as a filtering value. Then, the filtering value and the memorized steering wheel angle deviation are added as a real-time steering angle deviation value. The current real-time collected steering wheel angle and the real-time steering angle deviation value are added as an actual angle value of the steering wheel. The processing unit can be generally integrated in some controllers in the vehicle, such as a vehicle controller VCU and some auxiliary driving controllers or a newly added special controller. The output end of the processing unit is connected to an executor for sending the actual angle value of the steering wheel obtained by the self-learning to the executor for performing a control operation based on the actual angle. The executor is an execution mechanism for the vehicle controller based on the steering angle.

[0047] Obviously, the specific implementation of the present application is not limited by the above-mentioned manner, and various non-essential improvements made by adopting the method concept and technical scheme of the present application are within the protection scope of the present application.

Claims

1. A self-learning method for steering wheel angle, characterized in that: include, Step 1: Determine whether to enter the self-learning step based on vehicle parameters; Step 2: After entering the self-learning step, the difference obtained by subtracting the real-time collected steering wheel angle and the memory of the steering wheel angle deviation is multiplied by the filtering coefficient and recorded as the filtering value; Step 3: Use the sum of the filtered value and the memorized steering wheel angle deviation as the real-time steering angle deviation value; Step 4: Add the currently collected real-time steering wheel angle to the real-time steering angle deviation value to obtain the actual steering wheel angle value; In step 1, the vehicle is determined to be in a straight-line driving state based on its parameters. If it is in a straight-line driving state, the self-learning step is initiated. In step 3, the calculated real-time steering angle deviation value is compared with the memorized steering wheel steering angle deviation. If they are different, the memorized steering wheel steering angle deviation is updated to the real-time steering angle deviation value. After initiating the self-learning process, steps 1-4 are executed periodically.

2. The steering wheel angle self-learning method as described in claim 1, characterized in that: The vehicle parameters used to determine whether to enter the self-learning step include: vehicle yaw rate, steering wheel angular velocity, vehicle speed, steering wheel angle, and torque applied to the steering wheel by the driver. Based on the vehicle yaw rate, steering wheel angular velocity, vehicle speed, steering wheel angle, and torque applied to the steering wheel by the driver, it is determined whether the vehicle is in a straight-line driving state.

3. The steering wheel angle self-learning method as described in claim 2, characterized in that: When the absolute value of the vehicle's yaw rate is less than the yaw rate threshold, the absolute value of the steering wheel's angular velocity is less than the angular velocity threshold, the vehicle speed is greater than the vehicle speed threshold, the steering wheel angle is less than the steering wheel angle threshold, and the torque applied by the driver to the steering wheel is less than the torque threshold, it is determined that the vehicle is in a straight-line driving state, and then the angle self-learning step is initiated.

4. A steering wheel angle self-learning method as described in any one of claims 1-3, characterized in that: In step 2, the filter coefficients are real numbers less than 1.

5. The steering wheel angle self-learning method as described in claim 4, characterized in that: The filtering coefficient is positively correlated with the vehicle speed; the faster the vehicle speed, the larger the corresponding filtering coefficient.

6. A steering wheel angle self-learning system, characterized in that: include: Vehicle data acquisition unit, used to collect vehicle parameters; The self-learning start unit receives vehicle parameters from the vehicle data acquisition unit and determines whether the steering wheel angle self-learning conditions are met based on the vehicle parameters. The processing unit, which is connected to the vehicle data acquisition unit and the self-learning start unit respectively, performs the following steps: Step 1: Determine whether to enter the self-learning step based on vehicle parameters; Step 2: After entering the self-learning step, the difference obtained by subtracting the real-time collected steering wheel angle and the memory of the steering wheel angle deviation is multiplied by the filtering coefficient and recorded as the filtering value; Step 3: Use the sum of the filtered value and the memorized steering wheel angle deviation as the real-time steering angle deviation value; Step 4: Add the currently collected real-time steering wheel angle to the real-time steering angle deviation value to obtain the actual steering wheel angle value; In step 1, the vehicle is determined to be in a straight-line driving state based on its parameters. If it is in a straight-line driving state, the self-learning step is initiated. In step 3, the calculated real-time steering angle deviation value is compared with the memorized steering wheel steering angle deviation. If they are different, the memorized steering wheel steering angle deviation is updated to the real-time steering angle deviation value. After initiating the self-learning process, steps 1-4 are executed periodically.

7. The steering wheel angle self-learning system as described in claim 6, characterized in that: The output of the processing unit is connected to the actuator, and is used to send the actual angle value of the steering wheel obtained by self-learning to the actuator so that the actuator can perform control operations based on the actual angle.

Images