Intelligent windscreen wiper system self-learning control method

Abstract

The invention relates to a self-learning control method for an intelligent windscreen wiper system, which makes up the precision error and performance degradation of a rainfall sensor through the self-learning control method, reduces the intervention behavior of a driver on the intelligent windscreen wiper system in the driving process, and practically improves the driving safety and comfort. And the adaptive capacity of the intelligent windscreen wiper system to special use scenes such as rain and snow is improved. A rain sensor collection value is corrected by recognizing a windscreen wiper movement speed adjusting control signal input by a driver, meanwhile, rain correction needs to be associated with an actual rain sensor collection value, a rain correction curve is obtained, and then a windscreen wiper speed control value is obtained by looking up a table according to the value obtained after rain correction. And the precision error of the rainfall sensor and the performance attenuation in the use process are compensated for self-learning correction. According to the invention, the driving safety and comfort are practically improved.

Description

Technical Field

[0001] The invention relates to a self-learning control method, and in particular to a self-learning control method for an intelligent wiper system. Background Art

[0002] In recent years, smart wiper systems have gradually begun to be used in the market, but basically they cannot fully meet the needs of intelligent work that does not require driver intervention. A proportional adjustment knob is generally added to the wiper operating handle for the driver to actively intervene to adjust the wiper speed of the smart wiper system. However, due to the accuracy of the rain sensor itself and the performance degradation during use, the proportional adjustment knob cannot meet actual usage needs. Sometimes when the rainfall is light, the proportional adjustment knob is adjusted to the appropriate position, but when the rainfall is heavy, it will be found that the wiper speed does not meet the requirements, resulting in the driver frequently having to intervene in the smart wiper system by operating the proportional adjustment knob according to the vehicle speed or in different scenarios such as rain and snow, reducing the comfort and safety of the driver in driving the vehicle. Summary of the invention

[0003] The present invention compensates for the accuracy error and performance attenuation of the rain sensor through a self-learning control method, reduces the driver's intervention in the intelligent wiper system during driving, effectively improves driving safety and comfort, and improves the adaptability of the intelligent wiper system to special usage scenarios such as rain and snow.

[0004] The present invention is achieved by the following technical solutions:

[0005] A self-learning control method for an intelligent wiper system corrects the value collected by a rain sensor by identifying a wiper movement speed adjustment control signal input by a driver. At the same time, the rain correction is associated with the actual rain sensor collection value to obtain a rain correction curve. The wiper speed control value is then obtained by looking up the table from the corrected rain value to compensate for the accuracy error of the rain sensor and the performance attenuation during use for self-learning correction.

[0006] The first step is to design the control signal, including: speed plus and speed minus two inching self-reset switch input signals as wiper acceleration and deceleration adjustment input signals; rain sensor acquisition signal, in the range of 0 to 100, with an accuracy of 1; wiper speed signal, in the range of 0 to 100, with an accuracy of 1;

[0007] The second step is to correct the collected value of the rain sensor. By identifying the driver's windshield wiper acceleration and deceleration adjustment signal, that is, the speed increase or speed decrease two inching switches are triggered each time, the corresponding value of the rain correction curve, that is, the rain correction record value, is increased or decreased by 1, and the collected value of the rain sensor is corrected. The rain correction record value is obtained by calibration to obtain a rain correction curve table with a horizontal coordinate of 0 to 100 and a vertical coordinate of -100 to +100 rain correction value. The rain correction update value is obtained by speed increase or speed decrease based on the rain correction record value.

[0008] The third step is to self-learn the rainfall correction value. The horizontal coordinate of the rainfall correction curve is 0-100. The difference of rainfall correction values ​​between adjacent points is designed to be no more than 1. If it exceeds 1, the rainfall correction value of the adjacent point will be automatically increased by 1.

[0009] The fourth step is to set upper and lower limits for the rainfall correction value. Through calibration, that is, theoretical calculation, it is determined that the upper limit of the rainfall correction value is 20 and the lower limit is -20;

[0010] The fifth step is to use the value collected by the rain sensor as the input of the rainfall correction curve table lookup, and the rainfall correction value is obtained by looking up the table. The value collected by the rain sensor is added to the rainfall correction value to obtain the rainfall correction value;

[0011] The sixth step is to perform limit processing on the corrected rainfall value to ensure that the corrected rainfall value is between 0 and 100, and the wiper speed is 0 to 100, corresponding to the corrected rainfall value between 0 and 100, that is, the corrected rainfall value after the limit processing is used as the wiper speed control signal;

[0012] The seventh step is to filter the wiper speed control signal and output it. After calibration testing, confirm that the value after using mean filtering (taking the average of 10 numbers within 100ms) is used as the final output wiper speed control signal.

[0013] The present invention can realize self-learning control of the intelligent wiper system, reduce the driver's intervention in the intelligent wiper system during driving, improve the adaptability of the intelligent wiper system to special usage scenarios such as rain and snow, and effectively improve driving safety and comfort. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 It is the self-learning control method framework of the present invention;

[0015] Figure 2 This is a rainfall correction value acquisition diagram of the present invention;

[0016] Figure 3 It is a calculation diagram of the wiper value after correction of the present invention. DETAILED DESCRIPTION

[0017] like Figure 1-3As shown, the present invention designs a self-learning control method for an intelligent wiper system, which corrects the collected value of the rain sensor by identifying the wiper movement speed adjustment control signal input by the driver. At the same time, the rain correction must be associated with the actual rain sensor collected value to obtain a rain correction curve, and then the wiper speed control value is obtained by looking up the table from the corrected rain value, compensating for the accuracy error of the rain sensor and the performance attenuation during use to perform self-learning correction, thereby improving the adaptability of the self-learning control method under different weather conditions such as different rainfall, vehicle speed, rain and snow.

[0018] The overall self-learning control method architecture is as follows: Figure 1 shown.

[0019] The first step is to design the control signal, including: two inching self-reset switch input signals of speed plus and speed minus, as the wiper acceleration and deceleration adjustment input signals; the rain sensor acquisition signal, in the range of 0 to 100, with an accuracy of 1; the wiper speed signal, in the range of 0 to 100, with an accuracy of 1.

[0020] The second step is to correct the collected value of the rain sensor. By identifying the driver's windshield wiper acceleration and deceleration adjustment signal, that is, the speed increase or speed decrease two inching switches are triggered once, the corresponding value of the rain correction curve, that is, the rain correction record value, is increased or decreased by 1, and the collected value of the rain sensor is corrected.

[0021] Through calibration, a rainfall correction curve table with a horizontal coordinate of 0 to 100 and a vertical coordinate of -100 to +100 is obtained by looking up the rainfall correction value. The rainfall correction update value is obtained by adding or subtracting the speed based on the rainfall correction record value to compensate for the problem of inconsistent accuracy or performance attenuation of the rainfall sensor in different collection ranges.

[0022] The third step is self-learning of rainfall correction value. The horizontal coordinate of rainfall correction curve is 0-100, and the difference of rainfall correction value between adjacent points is designed to be no more than 1. If it exceeds 1, the rainfall correction value of adjacent points will be automatically increased by 1. The smoothness of rainfall correction curve is improved through self-learning to avoid the problem of uneven linear adjustment of wiper speed with rainfall caused by discontinuous rainfall correction value points.

[0023] The fourth step is to set upper and lower limits for the rainfall correction value to prevent the rain sensor correction value from exceeding the upper limit, causing the wiper speed to lose its linear range adjustment ability. Through calibration, that is, theoretical calculation, it is determined that the upper limit of the rainfall correction value is 20 and the lower limit is -20.

[0024] The fifth step is to use the value collected by the rainfall sensor as the input of the rainfall correction curve table lookup, and the rainfall correction value is obtained by looking up the table. The value collected by the rainfall sensor is added to the rainfall correction value to obtain the rainfall corrected value.

[0025] The sixth step is to limit the corrected rainfall value to ensure that the corrected rainfall value is between 0 and 100. At the same time, the wiper speed is 0 to 100, corresponding to the corrected rainfall value between 0 and 100. That is, the corrected rainfall value after limit processing is used as the wiper speed control signal.

[0026] The seventh step is to improve the possibility of inconsistency at different points from 0 to 100 due to rainfall learning and rain sensor performance attenuation. The wiper speed control signal needs to be filtered and output. After calibration testing, the value after mean filtering (averaging 10 numbers within 100ms) is confirmed as the final output wiper speed control signal.

Claims

1. A self-learning control method for an intelligent wiper system, characterized in that: By identifying the wiper movement speed adjustment control signal input by the driver, the rain sensor collected value is corrected. At the same time, the rain correction must be associated with the actual rain sensor collected value to obtain a rain correction curve. The wiper speed control value is then obtained by looking up the table from the corrected rain value to compensate for the accuracy error of the rain sensor and the performance attenuation during use for self-learning correction.

2. The self-learning control method of an intelligent wiper system according to claim 1, characterized in that: The following steps are involved: The first step is to design the control signal, including: speed plus and speed minus two inching self-reset switch input signals as wiper acceleration and deceleration adjustment input signals; rain sensor acquisition signal, in the range of 0 to 100, with an accuracy of 1; wiper speed signal, in the range of 0 to 100, with an accuracy of 1; The second step is to correct the collected value of the rain sensor. By identifying the driver's windshield wiper acceleration and deceleration adjustment signal, that is, the speed increase or speed decrease two inching switches are triggered each time, the corresponding value of the rain correction curve, that is, the rain correction record value, is increased or decreased by 1, and the collected value of the rain sensor is corrected. The rain correction record value is obtained by calibration to obtain a rain correction curve table with a horizontal coordinate of 0 to 100 and a vertical coordinate of -100 to +100 rain correction value. The rain correction update value is obtained by speed increase or speed decrease based on the rain correction record value. The third step is to self-learn the rainfall correction value. The horizontal coordinate of the rainfall correction curve is 0-100. The difference of rainfall correction values ​​between adjacent points is designed to be no more than 1. If it exceeds 1, the rainfall correction value of the adjacent point will be automatically increased by 1. The fourth step is to set upper and lower limits on the rainfall correction value; The fifth step is to use the value collected by the rain sensor as the input of the rainfall correction curve table lookup, and the rainfall correction value is obtained by looking up the table. The value collected by the rain sensor is added to the rainfall correction value to obtain the rainfall correction value; The sixth step is to perform limit processing on the corrected rainfall value to ensure that the corrected rainfall value is between 0 and 100, and the wiper speed is 0 to 100, corresponding to the corrected rainfall value between 0 and 100, that is, the corrected rainfall value after the limit processing is used as the wiper speed control signal; The seventh step is to filter the wiper speed control signal and output it. After calibration testing, it is confirmed that the value after mean filtering is used as the final output wiper speed control signal.

3. The self-learning control method of an intelligent wiper system according to claim 2, characterized in that: In the fourth step, when setting upper and lower limits for the rainfall correction value, the upper limit of the rainfall correction value is determined to be 20 and the lower limit is -20 through calibration, i.e. theoretical calculation.

4. The self-learning control method of an intelligent wiper system according to claim 2, characterized in that: In the seventh step, when using mean filtering, 10 numbers are taken within 100ms to calculate the average value.

Images