A window ripple anti-pinch management system that intelligently identifies bumpy roads

Abstract

The invention relates to the technical field of vehicle window management, and discloses a vehicle window ripple anti-pinch management system for intelligently identifying bumpy roads, comprising a cabin, a linear slide rail, an elastic member and a slider; wherein the linear slide rails are symmetrically arranged on the inner walls of the cabin, and two groups of elastic members are arranged in a mirror image in the upper and lower directions in the reserved gap between the linear slide rails, and the central area of ​​the elastic member accommodates and constrains the slider, and the two ends of the slider are connected to the linear slide rail by a sliding coupling; the system is used to use the central section of the linear slide rail as a coordinate reference and obtain a displacement vector of the slider relative to the coordinate reference; and is also used to obtain operating status data of a target vehicle; the operating status data includes vehicle driving speed, vehicle acceleration, vehicle driving angle and suspension system telescopic parameters; and is used to establish a supplementary signal unit, and obtain an expected filter value based on the output of the supplementary signal unit; based on the expected filter value, the current parameter of the window motor is filtered to obtain a net parameter, and the operating status of the window is managed based on the net parameter.

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle window management, and more specifically, to a vehicle window ripple anti-pinch management system capable of intelligently identifying bumpy roads. Background Art

[0002] Existing window anti-pinch devices mainly rely on current monitoring technology to perform real-time sampling and threshold comparison of the driving motor current during the window closing process. Specifically, the system quantifies the motor current through preset parameters, and then quickly interrupts the closing command when the monitoring value exceeds the current design tolerance, in order to avoid accidental clamping. However, the existing technology has limitations under vehicle dynamic disturbances. Especially when the vehicle is subjected to bumps, its motion state can be divided into two stages: in the rising stage, the vehicle body is subjected to a downward pressure effect, causing the current signal of the driving motor to show abnormal sensitivity, thereby inducing the system to terminate the closing action prematurely; in the descending stage, due to the upward reaction force obtained by the vehicle body, the change amplitude of the motor current signal is significantly weakened, causing the anti-pinch mechanism to react slowly, making it difficult to identify the real obstacle in time, and ultimately leading to an increased risk of mis-clamping. Summary of the invention

[0003] The present invention provides a vehicle window ripple anti-pinch management system capable of intelligently identifying bumpy roads, and solves the technical problems existing in the background technology.

[0004] The present invention provides a window ripple anti-pinch management system capable of intelligently identifying bumpy roads, comprising:

[0005] The signal compensation unit comprises: a cabin, a linear slide rail, an elastic member and a slider; wherein the linear slide rails are symmetrically arranged on the two inner walls of the cabin, two sets of elastic members are arranged in a mirror image up and down in the reserved gap between the linear slide rails, the slider is accommodated and constrained in the central area of ​​the elastic member, and the two ends of the slider are connected to the linear slide rail by a sliding coupling;

[0006] The data acquisition module is used to use the central section of the linear slide as a coordinate reference and obtain the displacement vector of the slide relative to the coordinate reference; it is also used to obtain the running status data of the target vehicle; the running status data includes the vehicle running speed, vehicle acceleration, vehicle running angle and suspension system extension and contraction parameters;

[0007] A signal filtering module, used for establishing a supplementary signal unit based on a displacement vector of the slider and the running state data of the target vehicle, and obtaining an expected filter value based on an output of the supplementary signal unit;

[0008] The window management module is used to filter the current parameters of the window motor based on the expected filter value to obtain the net parameters within the target time period, and manage the operation status of the window based on the net parameters.

[0009] Furthermore, the linear slide rail is parallel to the lifting stroke of the vehicle window in its setting position.

[0010] Furthermore, based on the mapping of the contact point between the slider and the linear guide rail, the displacement vector of the slider relative to the coordinate reference is obtained; wherein the displacement above the coordinate reference represents the positive component of the displacement vector, and the displacement below the coordinate reference represents the negative component of the displacement vector.

[0011] Further, a supplementary signal unit is established, comprising:

[0012] Step 41, establishing a test path, and assigning a random bumpy obstacle to the test path in each test time period; by randomly changing the height of the bumpy obstacle and the angle between the edge of the bumpy obstacle and the horizontal test path, and manually opening or closing the window of the test vehicle when the test vehicle passes through the bumpy obstacle, the current parameters of the window motor, the operating status data of the target vehicle, and the displacement vector of the slider are obtained at fixed time intervals in each test time period;

[0013] Step 42, establishing a regression model of the running state data of the test vehicle and the displacement vector of the slider and the corresponding current parameters, as follows:

[0014] Step 421, obtaining expected current parameters, and calculating the current difference between the current parameters at K moments and the expected current parameters; wherein the expected current parameters represent the current parameters of the window motor working under an ideal state;

[0015] Step 422, the running state data of the target vehicle at K moments and each parameter in the displacement vector of the slider are arranged in order to obtain K initial regression sequences; a symbol filling sequence is manually defined for each initial regression sequence, and a symbol selection range is manually defined for each ranking position in the symbol filling sequence;

[0016] Step 423, establishing a symbol library, the symbol library includes several mathematical operation symbols; establishing N mathematical operation symbol sequences, and randomly selecting the mathematical operation symbol of each sequence position in each mathematical operation symbol sequence based on the symbol selection range of the sorting position corresponding to the symbol filling sequence;

[0017] Step 424, each mathematical operation symbol sequence is used to fill the symbol filling sequence of each initial regression sequence in the K initial regression sequences according to the sequence position, and the output values ​​of the K initial regression sequences are calculated;

[0018] Step 425, calculating the mean square error loss value of the output values ​​of the K initial regression sequences and the current differences at K moments to obtain the fitness value of the corresponding mathematical operation symbol sequence;

[0019] Step 426, sort the N mathematical operation symbol sequences from small to large based on the fitness value to obtain a genetic sorting, and sort the sequence positions in the genetic sorting from small to large. arrive The mathematical operation symbol sequence is crossed or mutated to obtain An updated sequence of mathematical operation symbols;

[0020] Step 427, loop step 426 for a preset number of times, output a mathematical operation symbol sequence with a sequence rank of 1 in the genetic order, build a regression model based on the corresponding mathematical operation symbol sequence, and configure the regression model as a supplementary signal unit.

[0021] Further, obtaining an expected filter value based on the output of the supplementary signal unit includes:

[0022] At the i-th moment in the target time period, the operating status data of the target vehicle and the displacement vector of the slider are obtained, and the output value of the i-th moment calculated by the regression model of the supplementary signal unit is obtained, and the output value is used as the expected filter value at the i-th moment.

[0023] Furthermore, the current parameters of the window motor are filtered based on the expected filter value to obtain the net parameters, including: obtaining the circuit parameters of the window motor of the target vehicle at the i-th moment, and calculating the difference between the expected filter value at the i-th moment and the circuit parameters of the window motor of the target vehicle at the i-th moment, to obtain the net parameters at the i-th moment.

[0024] Furthermore, the operating status of the windows is managed based on the net parameters, including:

[0025] If the net parameter at the i-th moment is less than or equal to the preset current tolerance, the operating state of the window motor of the target vehicle is maintained; if the net parameter at the i-th moment is greater than the preset current tolerance, the operating state of the window motor of the target vehicle is stopped.

[0026] Furthermore, the supplementary signal unit is communicatively connected with the window management module based on a CAN bus.

[0027] The beneficial effects of the present invention are as follows: through the fusion of mechanical displacement perception and multi-dimensional vehicle status data, a nonlinear compensation model is constructed in combination with symbolic regression, thereby achieving real-time and accurate filtering of interference in the window motor current caused by bumps, thereby significantly improving the response speed and accuracy of the anti-pinch system, effectively reducing the risk of misjudgment, ensuring passenger safety and comfort, and having high real-time and robustness, showing broad prospects for engineering applications. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 It is a module diagram of a vehicle window ripple anti-pinch management system for intelligently identifying bumpy roads according to the present invention;

[0029] Figure 2 It is a structural diagram of the signal compensation unit of the present invention. DETAILED DESCRIPTION

[0030] The subject matter described herein will now be discussed with reference to example embodiments. It should be understood that the discussion of these embodiments is only to enable those skilled in the art to better understand and implement the subject matter described herein, and the functions and arrangements of the elements discussed may be changed without departing from the scope of protection of the contents of this specification. Each example may omit, replace or add various processes or components as needed. In addition, the features described relative to some examples may also be combined in other examples.

[0031] like Figure 1 and Figure 2 As shown, a window ripple anti-pinch management system for intelligently identifying bumpy roads includes:

[0032] The signal compensation unit comprises: a cabin, a linear slide rail, an elastic member and a slider; wherein the linear slide rails are symmetrically arranged on the two inner walls of the cabin, two sets of elastic members are arranged in a mirror image up and down in the reserved gap between the linear slide rails, the slider is accommodated and constrained in the central area of ​​the elastic member, and the two ends of the slider are connected to the linear slide rail by a sliding coupling;

[0033] The data acquisition module is used to use the central section of the linear slide as a coordinate reference and obtain the displacement vector of the slide relative to the coordinate reference; it is also used to obtain the running status data of the target vehicle; the running status data includes the vehicle running speed, vehicle acceleration, vehicle running angle and suspension system extension and contraction parameters;

[0034] A signal filtering module, used for establishing a supplementary signal unit based on a displacement vector of the slider and the running state data of the target vehicle, and obtaining an expected filter value based on an output of the supplementary signal unit;

[0035] The window management module is used to filter the current parameters of the window motor based on the expected filter value to obtain the net parameters within the target time period, and manage the operation status of the window based on the net parameters.

[0036] In one embodiment of the present invention, the linear slide rail is parallel to the lifting stroke of the vehicle window in its setting orientation.

[0037] Specifically, by limiting the installation direction of the linear slide rail, that is, requiring that its setting position must be parallel to the lifting stroke of the window. This arrangement ensures that when the slider moves on the slide rail, its movement direction is consistent with the actual vertical movement direction of the window, so that the displacement vector of the slider relative to the reference point can truly and accurately reflect the disturbance caused to the window when the vehicle bumps. If the installation angle of the slide rail deviates from the direction of window movement, it may introduce additional errors, affect the subsequent signal compensation and accurate calculation of net parameters, and thus weaken the anti-pinch management system's ability to judge the state of the window.

[0038] In one embodiment of the present invention, based on the mapping of the contact point between the slider and the linear guide rail, the displacement vector of the slider relative to the coordinate reference is obtained; wherein the displacement above the coordinate reference represents the positive component of the displacement vector, and the displacement below the coordinate reference represents the negative component of the displacement vector.

[0039] Specifically, the displacement vector of the slider relative to the coordinate reference is clearly determined by mapping the contact points between the slider and the linear guide. In other words, the system first uses the central section of the linear slide as a fixed reference reference, and then calculates the actual displacement of the slider in the vertical direction based on the contact between the slider and the guide. In order to distinguish the direction of the slider's movement, when the slider moves upward relative to the reference, its displacement is defined as the positive component; and when the slider moves downward, the displacement is defined as the negative component. In this way, by distinguishing between the positive and negative components of the displacement vector, the system can accurately reflect the specific movement trend of the slider when the vehicle is in a bumpy state, and provide accurate input data for subsequent signal filtering and compensation, thereby ensuring that the anti-pinch management system can perform accurate judgment and control under different working conditions.

[0040] In one embodiment of the present invention, establishing a supplementary signal unit includes:

[0041] Step 41, establishing a test path, and assigning a random bump obstacle to the test path in each test time period; by randomly changing the height of the bump obstacle and the angle between the edge of the bump obstacle and the horizontal test path, and manually opening or closing the window of the test vehicle when the test vehicle passes through the bump obstacle, the current parameters of the window motor, the operating status data of the target vehicle and the displacement vector of the slider are obtained at fixed time intervals in each test time period;

[0042] Step 42, establishing a regression model of the running state data of the test vehicle and the displacement vector of the slider and the corresponding current parameters, as follows:

[0043] Step 421, obtaining expected current parameters, and calculating the current difference between the current parameters at K moments and the expected current parameters; wherein the expected current parameters represent the current parameters of the window motor working under an ideal state;

[0044] Step 422, the running state data of the target vehicle at K moments and each parameter in the displacement vector of the slider are arranged in order to obtain K initial regression sequences; a symbol filling sequence is manually defined for each initial regression sequence, and a symbol selection range is manually defined for each ranking position in the symbol filling sequence;

[0045] Step 423, establishing a symbol library, the symbol library includes several mathematical operation symbols; establishing N mathematical operation symbol sequences, and randomly selecting the mathematical operation symbol of each sequence position in each mathematical operation symbol sequence based on the symbol selection range of the sorting position corresponding to the symbol filling sequence;

[0046] Step 424, each mathematical operation symbol sequence is used to fill the symbol filling sequence of each initial regression sequence in the K initial regression sequences according to the sequence position, and the output values ​​of the K initial regression sequences are calculated;

[0047] Step 425, calculating the mean square error loss value of the output values ​​of the K initial regression sequences and the current differences at K moments to obtain the fitness value of the corresponding mathematical operation symbol sequence;

[0048] Step 426, sort the N mathematical operation symbol sequences from small to large based on the fitness value to obtain a genetic sorting, and sort the sequence positions in the genetic sorting from small to large. arrive The mathematical operation symbol sequence is crossed or mutated to obtain An updated sequence of mathematical operation symbols;

[0049] Step 427, loop step 426 for a preset number of times, output a mathematical operation symbol sequence with a sequence rank of 1 in the genetic order, build a regression model based on the corresponding mathematical operation symbol sequence, and configure the regression model as a supplementary signal unit.

[0050] Specifically, by constructing a "supplementary signal unit", the unit is used to establish a regression mapping between the vehicle operating state, the slider displacement and the window motor current parameter, so as to generate the current parameter for interference compensation in actual operation. The specific implementation steps are as follows: First, in step 41, a test path is constructed, and the bumpy obstacles are randomly arranged in each test time period. By randomly changing the height of the obstacle and the angle between the obstacle edge and the horizontal path, different road conditions are simulated; at the same time, during the driving process of the test vehicle, the window opening or closing state is manually controlled, and the current data of the window motor, the vehicle operating state data and the displacement vector of the slider are obtained at a fixed sampling interval. Next, in step 42, a regression model is established using the collected data. The specific process is: in step 421, the expected current parameter representing the window motor current under the ideal working state is first obtained, and the difference between the actual current and the expected current at K moments is calculated. In step 422, the vehicle running state data at K moments and the various parameters of the slider displacement vector are arranged in order to form K initial regression sequences, and each sequence is filled with symbols, and the symbol selection range is predefined for each ranking position in the filled sequence. In step 423, a symbol library containing various mathematical operation symbols is established, and N mathematical operation symbol sequences are randomly generated, wherein the symbols selected for each sequence at each ranking position are restricted by the selection range of the corresponding position of the symbol filling sequence. In step 424, the above-mentioned mathematical operation symbol sequence and each initial regression sequence are filled in accordance with the sequence position, and the output value of each initial regression sequence is calculated. In step 425, the fitness value of the mathematical operation symbol sequence is obtained by calculating the mean square error between the output value of each regression sequence and the current difference at the corresponding moment. In step 426, all mathematical operation symbol sequences are sorted from small to large according to the fitness value, and the top-ranked sequences are crossover and mutation operations are performed to generate a new symbol sequence. In step 427, the above steps are repeated until the preset number of times is reached, and finally the sequence of mathematical operation symbols with the best fitness is output, and a complete regression model is constructed based on this sequence. This regression model is configured in the system as a supplementary signal unit. The system can automatically construct a nonlinear regression model that can accurately predict the performance of the window motor current under ideal conditions, thereby generating a supplementary signal for filtering the current anomaly caused by bumps. In this way, the entire anti-pinch management system can compensate for interference in real time under variable road conditions to ensure accurate management of the window operation status.

[0051] In one embodiment of the present invention, obtaining an expected filter value based on the output of the supplementary signal unit includes:

[0052] At the i-th moment in the target time period, the operating status data of the target vehicle and the displacement vector of the slider are obtained, and the output value of the i-th moment calculated by the regression model of the supplementary signal unit is obtained, and the output value is used as the expected filter value at the i-th moment.

[0053] Specifically, the "expected filter value" corresponding to each moment is generated in real time through the regression model output of the supplementary signal unit. The specific explanation is as follows: within the set target time period, for the i-th moment, the system first collects the running status data of the target vehicle (such as vehicle speed, acceleration, driving angle and suspension telescopic parameters) and the displacement vector of the slider, which reflect the state of the vehicle under bumpy conditions. Then, these data are input into the previously established regression model of the supplementary signal unit, and the model calculates an output value, which is defined as the expected filter value at the i-th moment in this system. This expected filter value represents the characteristic value that the motor current should present under ideal conditions and when it is not disturbed by bumps, thereby providing a reference benchmark for subsequent signal filtering. In other words, by performing regression calculations on the multi-dimensional data collected in real time, an "expected" current value is generated to be compared with the actual current signal, which is used to accurately judge and filter out the transient interference signal caused by bumps in the subsequent steps, thereby ensuring the response accuracy and stability of the window anti-pinch management system.

[0054] In one embodiment of the present invention, the current parameters of the window motor are filtered based on the expected filter value to obtain the net parameters, including: obtaining the circuit parameters of the window motor of the target vehicle at the i-th moment, and calculating the difference between the expected filter value at the i-th moment and the circuit parameters of the window motor of the target vehicle at the i-th moment, to obtain the net parameters at the i-th moment.

[0055] Specifically, by comparing the expected filter value with the actual circuit parameters collected by the window motor at the same time, a "net parameter" after interference filtering is obtained. Specifically, at the i-th moment, the system first obtains the expected filter value based on the regression model of the supplementary signal unit. The filter value reflects the current characteristics of the motor under ideal working conditions; then, the system synchronously obtains the actual circuit parameters of the window motor at that moment. By calculating the difference between the expected filter value and the actual circuit parameter, the abnormal fluctuations caused by bumps and other dynamic interferences can be eliminated, thereby obtaining the net parameters that truly reflect the operating status of the window. In short, this step uses difference calculations to achieve real-time correction and filtering of the current signal, providing a more accurate data basis for subsequent anti-pinch judgments.

[0056] In one embodiment of the present invention, managing the operating state of the vehicle window based on the net parameter includes:

[0057] If the net parameter at the i-th moment is less than or equal to the preset current tolerance, the operating state of the window motor of the target vehicle is maintained; if the net parameter at the i-th moment is greater than the preset current tolerance, the operating state of the window motor of the target vehicle is stopped.

[0058] Specifically, the net parameters obtained by filtering are used to manage the operating state of the window motor, thereby realizing the control logic of anti-pinch protection. Specifically, the requirement stipulates that at the predetermined detection moment, if the net parameter obtained (i.e., the difference between the actual circuit parameter and the expected filter value after interference filtering) is less than or equal to the preset current tolerance, the motor operating state is considered to be within the safe range, and the window motor maintains the original operating state; conversely, if the net parameter exceeds the preset current tolerance, it indicates that there is an abnormal situation or potential clamping risk, and the system immediately stops the operation of the window motor. Through this threshold control method, the system can promptly identify and respond to current anomalies caused by bumps, thereby effectively preventing safety hazards caused by mis-clamping.

[0059] In one embodiment of the present invention, the core of the signal compensation unit is to convert dynamic disturbances such as vehicle bumps into quantifiable displacement signals through a set of precise mechanical configurations, providing real-time physical basis for subsequent signal filtering. Specifically, the unit is composed of the following parts and works in coordination: First, linear slide rails are symmetrically arranged on both sides of the fixed cabin, and the installation orientation of these slide rails is strictly parallel to the direction of the window lifting movement to ensure that the movement of the slider is consistent with the actual movement trajectory of the window. A uniform gap is reserved between the linear slide rails, and two sets of elastic members are mirrored up and down in the gap. These elastic members not only play a role in buffering vibrations, but also provide restoring force and stable support for the slider. The slider is precisely placed in the center area of ​​the elastic member, and its two ends are connected to the slide rail in a sliding coupling manner, allowing it to move back and forth with low friction and smoothly in the vertical direction. When the vehicle encounters bumps, vibrations or other dynamic disturbances during driving, the acceleration generated by the vehicle body will induce elastic deformation in the elastic member, thereby causing the slider to displace relative to the reference position of the linear slide rail. By measuring the relative motion of the contact point between the slider and the guide rail, the displacement vector of the slider relative to the fixed coordinate reference can be obtained, where the displacement above the reference is defined as the positive component and the displacement below the reference is defined as the negative component. This displacement vector objectively reflects the external force generated by the vehicle due to bumps and captures the disturbance information at the physical level. After that, the displacement signal will be collected by the data acquisition module and used together with other operating state parameters of the vehicle (vehicle speed, acceleration, driving angle and suspension system expansion and contraction parameters) to construct a nonlinear regression model to generate a supplementary signal. The supplementary signal is a correction value for the ideal current parameter, which is used to suppress the interference component in the motor current signal caused by vehicle bumps. In this way, the signal compensation unit not only realizes the direct physical perception of the impact of vehicle bumps, but also provides a reliable real-time compensation signal for the entire anti-pinch management system, thereby ensuring the safe operation of the window under complex road conditions.

[0060] In one embodiment of the present invention, the supplementary signal unit is communicatively connected with the vehicle window management module based on a CAN bus.

[0061] Specifically, the charging signal unit and the window management module are transmitted and interconnected via the CAN bus. The CAN bus has high-speed and stable data transmission capabilities, which can ensure the synchronization and consistency in time between the pseudo-signal or expected filter value generated by the supplementary signal unit and the actual current data received by the window management module, thereby ensuring the real-time responsiveness of the anti-pinch system. By adopting a standardized CAN bus communication protocol, the modules can interoperate with each other, and even between modules on different hardware platforms or provided by different suppliers, data sharing and collaborative control can be achieved through a unified communication interface, thereby improving the overall reliability and scalability of the system. The CAN bus is widely used in vehicle electronic systems, and its anti-interference ability and fault-tolerant mechanism provide guarantees for the stable operation of the present invention in complex electromagnetic environments, ensuring that under dynamic conditions such as bumps, the data interaction between modules will not affect the realization of the anti-pinch protection function due to communication errors.

[0062] It should be noted that by integrating the mechanical signal compensation unit with the multi-dimensional data fusion technology, real-time and accurate compensation of the window motor current signal under vehicle bumps is achieved. Compared with the existing solutions that rely on neural network prediction methods, the present invention avoids the uncertainty of the neural network model and can generate high-confidence supplementary signals in real time, thereby effectively eliminating transient fluctuations caused by vehicle bumps. This technology not only greatly reduces the risk of mis-pinching, but also outperforms traditional neural network prediction solutions in terms of response speed, calculation delay and robustness to external interference, providing a more efficient, stable and reliable solution for the automotive window anti-pinch management system.

[0063] The above describes an embodiment of the present embodiment, but the present embodiment is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present embodiment, ordinary technicians in this field can also make many forms, all of which are within the protection of the present embodiment.

Claims

1. A window ripple anti-pinch management system that intelligently identifies bumpy roads, characterized in that: include: The signal compensation unit comprises: a cabin, a linear slide rail, an elastic member and a slider; wherein the linear slide rails are symmetrically arranged on the two inner walls of the cabin, two sets of elastic members are arranged in a mirror image up and down in the reserved gap between the linear slide rails, the slider is accommodated and constrained in the central area of ​​the elastic member, and the two ends of the slider are connected to the linear slide rail by a sliding coupling; The data acquisition module is used to use the central section of the linear slide as a coordinate reference and obtain the displacement vector of the slide relative to the coordinate reference; it is also used to obtain the running status data of the target vehicle; the running status data includes the vehicle running speed, vehicle acceleration, vehicle running angle and suspension system extension and contraction parameters; A signal filtering module, used for establishing a supplementary signal unit based on a displacement vector of the slider and the running state data of the target vehicle, and obtaining an expected filter value based on an output of the supplementary signal unit; The window management module is used to filter the current parameters of the window motor based on the expected filter value to obtain the net parameters within the target time period, and manage the operation status of the window based on the net parameters.

2. According to claim 1, a vehicle window ripple anti-pinch management system for intelligently identifying bumpy roads is characterized in that: The linear guide rail is parallel to the lifting stroke of the window in its setting position.

3. The system for preventing window ripples and pinching according to claim 1 is characterized in that: The displacement vector of the slider relative to the coordinate reference based on the mapping of the contact points between the slider and the linear guide; wherein the displacement above the coordinate reference represents the positive component of the displacement vector, and the displacement below the coordinate reference represents the negative component of the displacement vector.

4. The system for preventing window ripples and pinching according to claim 1 is characterized in that: Establishment of a supplementary signal unit, including: Step 41, establishing a test path, and assigning a random bump obstacle to the test path in each test time period; by randomly changing the height of the bump obstacle and the angle between the edge of the bump obstacle and the horizontal test path, and manually opening or closing the window of the test vehicle when the test vehicle passes through the bump obstacle, the current parameters of the window motor, the operating status data of the target vehicle and the displacement vector of the slider are obtained at fixed time intervals in each test time period; Step 42, establishing a regression model of the running state data of the test vehicle and the displacement vector of the slider and the corresponding current parameters, as follows: Step 421, obtaining expected current parameters, and calculating the current difference between the current parameters at K moments and the expected current parameters; wherein the expected current parameters represent the current parameters of the window motor working under an ideal state; Step 422, the running state data of the target vehicle at K moments and each parameter in the displacement vector of the slider are arranged in order to obtain K initial regression sequences; a symbol filling sequence is manually defined for each initial regression sequence, and a symbol selection range is manually defined for each ranking position in the symbol filling sequence; Step 423, establishing a symbol library, the symbol library includes several mathematical operation symbols; establishing N mathematical operation symbol sequences, and randomly selecting the mathematical operation symbol of each sequence position in each mathematical operation symbol sequence based on the symbol selection range of the sorting position corresponding to the symbol filling sequence; Step 424, each mathematical operation symbol sequence is used to fill the symbol filling sequence of each initial regression sequence in the K initial regression sequences according to the sequence position, and the output values ​​of the K initial regression sequences are calculated; Step 425, calculating the mean square error loss value of the output values ​​of the K initial regression sequences and the current differences at K moments to obtain the fitness value of the corresponding mathematical operation symbol sequence; Step 426, sort the N mathematical operation symbol sequences from small to large based on the fitness value to obtain a genetic sorting, and sort the sequence positions in the genetic sorting from small to large. To N mathematical operation symbol sequence crossover or mutation, get An updated sequence of mathematical operation symbols; Step 427, loop step 426 for a preset number of times, output a mathematical operation symbol sequence with a sequence rank of 1 in the genetic order, build a regression model based on the corresponding mathematical operation symbol sequence, and configure the regression model as a supplementary signal unit.

5. The system for preventing window ripples and pinching according to claim 4 is characterized in that: The expected filter value is obtained based on the output of the supplementary signal unit, including: At the i-th moment in the target time period, the operating status data of the target vehicle and the displacement vector of the slider are obtained, and the output value of the i-th moment calculated by the regression model of the supplementary signal unit is obtained, and the output value is used as the expected filter value at the i-th moment.

6. The system for preventing window ripples and pinching according to claim 5 is characterized in that: The current parameters of the window motor are filtered based on the expected filter value to obtain the net parameters, including: obtaining the circuit parameters of the window motor of the target vehicle at the i-th moment, and calculating the difference between the expected filter value at the i-th moment and the circuit parameters of the window motor of the target vehicle at the i-th moment to obtain the net parameters at the i-th moment.

7. The system for preventing window ripples and pinching according to claim 6 is characterized in that: Manage the operating status of the windows based on net parameters, including: If the net parameter at the i-th moment is less than or equal to the preset current tolerance, the operating state of the window motor of the target vehicle is maintained; if the net parameter at the i-th moment is greater than the preset current tolerance, the operating state of the window motor of the target vehicle is stopped.

8. The system for preventing window ripples from being pinched and managed by intelligently identifying bumpy roads according to claim 1, characterized in that: The supplementary signal unit is communicatively connected with the window management module based on the CAN bus.

Images