Car window anti-pinch detection method and device and computer storage medium
Through the anti-clip detection method of windows based on machine learning, the motor current and position data are used to adapt to changes in window state, and the adaptation problem of the anti-clip function of windows under assembly differences and aging is solved, and the consistency of anti-clip force is achieved Sex and stability improve user experience.
Patent Information
- Application Number
- CN202411200406.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-28
- Publication Date
- 2024-12-31
AI Technical Summary
Under the influence of window assembly differences, aging and climate change, existing window anti-clip functions are prone to error prevention of clamping or anti-clip force exceeding the regulatory standards, making it difficult to adapt to windows in different states.
Using the anti-clip detection method of the vehicle window based on the machine learning model, the normal working current threshold is obtained through the motor current information and window position data, and the model is established using the random forest algorithm to adapt to different window states and ambient temperatures, and the characteristic parameters of the decision tree are adjusted. , generate anti-pinch signal to control the reversal or holding of the window.
It improves the adaptability and stability of the anti-clip function of the window, reduces the probability of mis-clip attack, ensures the close consistency of the anti-clip resistance at different temperatures and positions, and improves the user experience.
Smart Images
Figure CN119221788A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of automobile window control, and in particular to a window anti-pinch detection method, device and computer storage medium. Background Art
[0002] With the advancement of automotive electronic technology, the level of intelligence of vehicles is getting higher and higher. Most of the electric windows of cars on the market are equipped with one-button lifting and anti-pinch functions. The anti-pinch function is realized based on the controller detecting the degree of change in the working state of the motor caused by the window glass encountering obstacles during the rising process, to confirm the occurrence of clamping, and immediately the controller stops the window from automatically rising and reverses.
[0003] The window lifters, guide grooves and other components will wear and age as the windows are used. At the same time, during use, the windows will be affected by differences in window assembly, changes in the vehicle's climate, etc., causing changes in the working state of the windows. This can cause the anti-pinch function to fail to prevent pinching or the anti-pinch force to exceed regulatory standards under fixed anti-pinch parameters. Therefore, the window anti-pinch function is also required to be able to adapt to windows in different states. Summary of the invention
[0004] The present application provides a vehicle window anti-pinch detection method, device and computer storage medium.
[0005] In order to solve the above technical problems, the present application proposes a vehicle window anti-pinch detection method, which includes: responding to a vehicle window rising command, obtaining the current working current and the current window position based on the motor current information; inputting the current window position into the vehicle window machine learning model to obtain the normal working current threshold; judging whether the current working current is greater than the normal working current threshold; if so, generating a vehicle window anti-pinch signal.
[0006] Wherein, the motor current information is a current waveform; the obtaining of the current working current and the current window position based on the motor current information includes: obtaining the current waveform of the window rising process based on the window rising instruction; obtaining the current working current based on the position of the current timestamp at the current waveform; obtaining the complete timestamp of the current waveform, and the initial window position that triggers the window rising instruction; obtaining the current window position according to the ratio of the current timestamp to the complete timestamp, and the initial window position.
[0007] Wherein, the determining whether the current working current is greater than the normal working current threshold comprises: determining whether the current working current is greater than the normal working current threshold, and the current window position is located in the anti-pinch zone.
[0008] Among them, the step of inputting the current window position into a window machine learning model to obtain a normal operating current threshold includes: inputting the current ambient temperature and the current window position into a window machine learning model to obtain a normal operating current threshold.
[0009] Among them, the vehicle window anti-pinch detection method also includes: establishing the vehicle window machine learning model based on the random forest algorithm; forming a vehicle window training data set with different vehicle window positions and normal working current data of each vehicle window position; inputting the vehicle window training data set into the vehicle window machine learning model to obtain training parameters; using the training parameters to adjust the characteristic parameters of the random forest decision tree to obtain the final vehicle window machine learning model; the characteristic parameters of the random forest decision tree include the influencing parameters of any one or more of the following factors: vehicle window assembly differences, motor performance degradation, window rubber strip aging, and lifter gear deformation.
[0010] The vehicle window training data set includes different vehicle window positions of vehicle windows in the same state, different vehicle window positions of vehicle windows in different states, and different vehicle window positions of vehicle windows of different types.
[0011] The forming of the window training data set with normal working current data of different window positions and each window position includes: forming the window training data set with normal working current data of different window positions and each window position under different ambient temperatures.
[0012] In order to solve the above technical problems, the present application proposes a window anti-pinch detection method, which includes: the window controller responds to the window rising command, and obtains the current working current and the current window position based on the motor current information; the central controller inputs the current window position into the window machine learning model to obtain the normal working current threshold; in response to the current working current being greater than the normal working current threshold, the central controller returns the window anti-pinch information to the window controller; the window controller controls the window to reverse or maintain the current window position based on the window anti-pinch information.
[0013] In order to solve the above technical problems, the present application proposes a vehicle window anti-pinch device, which includes a memory and a processor coupled to the memory; wherein the memory is used to store program data, and the processor is used to execute the program data to implement the above-mentioned vehicle window anti-pinch detection method.
[0014] In order to solve the above technical problems, the present application proposes a computer storage medium, wherein the computer storage medium is used to store program data, and when the program data is executed by a computer, it is used to implement the above vehicle window anti-pinch detection method.
[0015] Different from the prior art, the beneficial effects of the present application are: the window anti-pinch device responds to the window rising instruction, obtains the current working current and the current window position based on the motor current information; inputs the current window position into the window machine learning model to obtain the normal working current threshold; determines whether the current working current is greater than the normal working current threshold; if so, generates a window anti-pinch signal. In the above manner, the corresponding current threshold is calculated according to the real-time working condition of the window, ensuring that the anti-pinch force of the window at different temperatures and different positions is close, and ensuring the stability of the anti-pinch function. BRIEF DESCRIPTION OF THE DRAWINGS
[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.
[0017] Figure 1 It is a flowchart of the first embodiment of the vehicle window anti-pinch detection method provided by the present application;
[0018] Figure 2 It is a flow chart of a second embodiment of the vehicle window anti-pinch detection method provided by the present application;
[0019] Figure 3 is a flow chart of a third embodiment of a vehicle window anti-pinch detection method provided by the present application;
[0020] Figure 4 It is a structural schematic diagram of an embodiment of a vehicle window anti-pinch device provided by the present application;
[0021] Figure 5 It is a structural diagram of an embodiment of a computer storage medium provided by the present application. DETAILED DESCRIPTION
[0022] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.
[0023] The vehicle window anti-pinch detection method of the present application is applied to a vehicle window anti-pinch device, wherein the vehicle window anti-pinch device of the present application can be a server, or a system composed of a server and a local terminal. Accordingly, the various parts of the vehicle window anti-pinch device, such as various units, sub-units, modules, and sub-modules, can all be set in the server, or can be set in the server and the local terminal respectively.
[0024] Furthermore, the above-mentioned server can be hardware or software. When the server is hardware, it can be implemented as a distributed server cluster consisting of multiple servers, or it can be implemented as a single server. When the server is software, it can be implemented as multiple software or software modules, such as software or software modules used to provide distributed servers, or it can be implemented as a single software or software module, which is not specifically limited here. In some possible implementations, the vehicle window anti-pinch detection method of the embodiment of the present application can be implemented by a processor calling computer-readable instructions stored in a memory.
[0025] In order to solve the problem that differences in window assembly and aging of the window state lead to poor adaptation of the anti-pinch function and high probability of false anti-pinch, which affects the user experience, the present application provides a method for window anti-pinch detection, which improves the adaptability of the anti-pinch function, can adapt to the state differences of different windows, learn the original state of the same window and adapt to the changes in the window state, so that the window anti-pinch can achieve accurate and timely effects.
[0026] Please refer to Figure 1 , Figure 1 It is a flow chart of the first embodiment of the vehicle window anti-pinch detection method provided in the present application.
[0027] like Figure 1 As shown, the specific steps are as follows:
[0028] Step S11: In response to the window up instruction, the current operating current and the current window position are obtained based on the motor current information.
[0029] The window raising command is a command for controlling the window to rise. When the window anti-pinch device controls the window to rise, it collects the motor current information, and at the same time identifies the current ripple by detecting the rising and falling edges of the ripple in the current, and converts the current ripple into the window position information.
[0030] In one embodiment of the present application, the motor current information is a current waveform. The present application provides a specific embodiment for obtaining the current window position: the window anti-pinch device obtains the current waveform of the window rising process based on the window rising instruction; obtains the current working current based on the position of the current timestamp in the current waveform; obtains the complete timestamp of the current waveform, and the initial window position that triggers the window rising instruction; obtains the current window position according to the ratio of the current timestamp to the complete timestamp, and the initial window position.
[0031] In other embodiments of the present application, the window position information may also be determined by current intensity.
[0032] Step S12: Input the current window position into the window machine learning model to obtain the normal operating current threshold.
[0033] In one embodiment of the present application, the window anti-pinch device inputs the current ambient temperature and the current window position into the window machine learning model to obtain a normal operating current threshold.
[0034] Among them, in the embodiment of the present application, the vehicle window machine learning model is trained offline to ensure that the anti-pinch force of the vehicle window at different temperatures and different positions is similar. At the same time, the model can adjust parameters through self-learning training to adapt to the state differences of different vehicle windows and adapt to the state changes of the same vehicle window under different circumstances.
[0035] The data set required for offline training of the car window machine learning model is obtained in advance by performing several complete lifting and lowering of multiple car windows at different temperatures in a thermostat. Therefore, the model can learn the normal operating current I of the car window lifting system at different temperatures and different window positions.
[0036] The car window learning model will be updated at any time based on the recognition results, and the learning effect will be further improved based on offline training to increase the accuracy.
[0037] Step S13: Determine whether the current operating current is greater than the normal operating current threshold.
[0038] In one embodiment of the present application, the vehicle window anti-pinch device determines whether the current working current is greater than the normal working current threshold, and the current vehicle window position is located in the anti-pinch area.
[0039] Among them, the anti-pinch zone is an area where an anti-pinch signal will be triggered when in this area. The design of the anti-pinch zone can be designed based on historical mis-pinch signals, and the scope of the anti-pinch zone can also be determined based on the window height, width and vehicle model.
[0040] Specifically, the window machine learning model can calculate and output the current threshold for normal operation of the window based on the ambient temperature of the window and the current window position, and compare the window operating current with the window current threshold. If the current operating current is greater than the normal operating current threshold, step S14 is executed; if the current operating current is less than or equal to the normal operating current threshold, the original operation is kept unchanged.
[0041] In the embodiment of the present application, on the basis of determining whether the anti-pinch signal is triggered by the working current, a limitation is added that the window anti-pinch signal is only output when the window position is in the anti-pinch zone, which can avoid misrecognition and reduce the impact of misrecognition on the normal window usage experience.
[0042] Step S14: Generate a window anti-pinch signal.
[0043] If the working current is greater than the threshold and the window position is in the anti-pinch zone at this time, it is considered that the window is pinched. The window controller executes the next action based on the anti-pinch judgment result. The window anti-pinch device controls the window to reverse or keeps the current output unchanged and repeats the anti-pinch judgment process.
[0044] Furthermore, this application proposes an embodiment for training a vehicle window machine learning model. For details, please refer to Figure 2 , Figure 2 It is a flow chart of the second embodiment of the vehicle window anti-pinch detection method provided in the present application.
[0045] like Figure 2 As shown, the specific steps are as follows:
[0046] Step S21: Establish a vehicle window machine learning model based on the random forest algorithm.
[0047] The window anti-pinch device establishes an initial model of window machine learning based on the random forest algorithm, and divides the data of different temperatures, different window positions, and corresponding normal working current into window training data sets and verification data sets, so as to perform offline training and verification on the window machine learning model. The K-fold cross-validation method is adopted to repeatedly train and verify the model, and the characteristic parameters of the random forest decision tree are adjusted according to the verification results, and finally the window machine learning model is obtained. The model can output the normal working current threshold of the window at the current position according to the input ambient temperature and window position information.
[0048] Step S22: different vehicle window positions and the normal operating current data of each vehicle window position are combined into a vehicle window training data set.
[0049] The window training data set includes different window positions for windows in the same state, different window positions for windows in different states, and different window positions for windows of different types.
[0050] Step S23: Input the vehicle window training data set into the vehicle window machine learning model to obtain training parameters.
[0051] After adjusting the decision tree feature parameters of the model, the random forest-based machine learning model is trained offline on the PC using the window training data set. The random forest-based window machine model can adapt to different types of windows and has good adaptability. It does not need to calibrate the anti-pinch parameter at different stages for different types of windows and their state changes as in the traditional way. During normal use, the model parameters can be adjusted through self-learning to adapt to changes in window states, thereby ensuring the stability of the anti-pinch force and reducing the probability of false anti-pinch.
[0052] Step S24: Use the training parameters to adjust the feature parameters of the random forest decision tree to obtain the final vehicle window machine learning model.
[0053] In an embodiment of the present application, the characteristic parameters of the random forest decision tree include influencing parameters of any one or more of the following factors: differences in vehicle window assembly, degradation of motor performance, aging of window rubber strips, and deformation of lifter gears.
[0054] In the embodiment of the present application, the vehicle window machine learning model based on random forest can adapt to the assembly differences between different vehicle windows and the aging status of the same vehicle window, ensure the stability of the anti-pinch force of the vehicle window, and improve the user experience.
[0055] In order to implement the vehicle window anti-pinch detection method of the above embodiment, the present application also provides a vehicle window anti-pinch detection method. Figure 3 , Figure 3 It is a flow chart of the third embodiment of the vehicle window anti-pinch detection method provided in the present application.
[0056] like Figure 3 As shown, the specific steps are as follows:
[0057] Step S31: The window controller responds to the window up instruction and obtains the current operating current and the current window position based on the motor current information.
[0058] Step S32: The central controller inputs the current window position into the window machine learning model to obtain the normal operating current threshold.
[0059] Step S33: In response to the current operating current being greater than the normal operating current threshold, the central controller returns the window anti-pinch information to the window controller.
[0060] Step S34: The window controller controls the window to reverse or maintain the current window position based on the window anti-pinch information.
[0061] Step S31 to step S34 are the same as step S11 to step S14, and will not be described in detail here.
[0062] The anti-pinch detection algorithm based on fixed anti-pinch parameters has difficulty in ensuring the anti-pinch force is close at different positions of the car window, and the effect is average. Moreover, the anti-pinch force is even more difficult to maintain stability due to the influence of temperature.
[0063] This application trains the machine learning model offline. The model can learn the normal working current I of the window lifting system at different temperatures and different window positions, and calculate the corresponding current threshold I according to the real-time working conditions of the window. K , ensuring that the anti-pinch force of the car window at different temperatures and different positions is similar, ensuring the stability of the anti-pinch function.
[0064] Due to the influence of the assembly differences between each window, the current characteristics between the windows are not consistent. In this case, the anti-pinch detection algorithm based on fixed anti-pinch parameters cannot adapt to various types of windows, and is prone to problems such as false anti-pinch and anti-pinch force exceeding regulatory requirements.
[0065] In addition, the same car window may have motor performance degradation, window rubber strip aging, lifter gear deformation, etc. as the window is used, which will make the anti-pinch detection algorithm based on fixed anti-pinch parameters unable to meet the usage requirements. It is also easy to cause problems such as false anti-pinch and anti-pinch force exceeding regulatory requirements.
[0066] This application uses a machine learning model based on random forests to detect window anti-pinch. Through self-learning training, the decision tree feature parameters can be adjusted to adapt to the state differences of different windows and the state changes of the same window, ensuring the consistency of anti-pinch force and avoiding false anti-pinch phenomenon. This application is applicable to algorithms that need to realize anti-pinch function, including but not limited to: ripple anti-pinch and / or Hall anti-pinch, etc.
[0067] In order to implement the vehicle window anti-pinch detection method of the above embodiment, the present application also provides a vehicle window anti-pinch device. Figure 4 , Figure 4 It is a structural schematic diagram of an embodiment of a vehicle window anti-pinch device provided in the present application.
[0068] like Figure 4 As shown, the vehicle window anti-pinch device 600 of this embodiment includes a processor 61 , a memory 62 , an input / output device 63 and a bus 64 .
[0069] The processor 61 , the memory 62 , and the input / output device 63 are respectively connected to the bus 64 . The memory 62 stores a computer program, and the processor 61 is used to execute the computer program to implement the vehicle window anti-pinch detection method of the above embodiment.
[0070] In this embodiment, the processor 61 may also be referred to as a CPU (Central Processing Unit). The processor 61 may be an integrated circuit chip having the ability to process signals. The processor 61 may also be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The processor 61 may also be a GPU (Graphics Processing Unit), also known as a display core, a visual processor, or a display chip, which is a microprocessor that is specifically used for image computing on computers, workstations, game consoles, and some mobile devices (such as tablet computers, smart phones, etc.). The purpose of the GPU is to convert and drive the display information required by the computer system, and to provide a line scan signal to the display to control the correct display of the display. It is an important component that connects the display and the computer motherboard. As an important component of the computer host, the graphics card is responsible for outputting display graphics. The general-purpose processor may be a microprocessor or the processor 61 may also be any conventional processor, etc.
[0071] The present application also provides a computer storage medium, such as Figure 5 As shown, the computer storage medium 700 is used to store a computer program 71. When the computer program 71 is executed by a processor, it is used to implement the method described in the embodiment of the vehicle window anti-pinch detection method of the present application.
[0072] The method involved in the embodiment of the vehicle window anti-pinch detection method of the present application, when implemented in the form of a software functional unit and sold or used as an independent product, can be stored in a device, such as a computer-readable storage medium. Based on such an understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor (processor) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.
[0073] The above description is only an implementation method of the present application, and does not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly used in other related technical fields, are also included in the patent protection scope of the present application.
Claims
1. A vehicle window anti-pinch detection method, characterized in that: The vehicle window anti-pinch detection method comprises: In response to the window raising instruction, obtaining the current working current and the current window position based on the motor current information; Inputting the current window position into a window machine learning model to obtain a normal operating current threshold; Determine whether the current operating current is greater than the normal operating current threshold; If so, a window anti-pinch signal is generated.
2. The vehicle window anti-pinch detection method according to claim 1, characterized in that: The motor current information is a current waveform; The obtaining of the current operating current and the current window position based on the motor current information includes: Based on the window raising instruction, obtaining a current waveform of the window raising process; Based on the current timestamp at the position of the current waveform, obtain the current operating current; Obtaining a complete timestamp of the current waveform and an initial window position that triggers the window raising instruction; The current vehicle window position is obtained according to the ratio of the current timestamp to the complete timestamp and the initial vehicle window position.
3. The vehicle window anti-pinch detection method according to claim 1 or 2, characterized in that: The determining whether the current operating current is greater than the normal operating current threshold comprises: It is determined whether the current working current is greater than the normal working current threshold and the current window position is located in the anti-pinch zone.
4. The vehicle window anti-pinch detection method according to claim 1, characterized in that: The inputting the current window position into the window machine learning model to obtain the normal operating current threshold comprises: The current ambient temperature and the current window position are input into the window machine learning model to obtain the normal operating current threshold.
5. The vehicle window anti-pinch detection method according to claim 1, characterized in that: The vehicle window anti-pinch detection method further includes: Establishing the vehicle window machine learning model based on random forest algorithm; The different window positions and the normal working current data of each window position are combined into a window training data set; Inputting the vehicle window training data set into the vehicle window machine learning model to obtain training parameters; Using the training parameters to adjust the feature parameters of the random forest decision tree to obtain a final vehicle window machine learning model; The characteristic parameters of the random forest decision tree include influencing parameters of any one or more of the following factors: window assembly differences, motor performance degradation, window rubber strip aging, and lifter gear deformation.
6. The vehicle window anti-pinch detection method according to claim 5, characterized in that: The vehicle window training data set includes different vehicle window positions of vehicle windows in the same state, different vehicle window positions of vehicle windows in different states, and different vehicle window positions of vehicle windows of different types.
7. The vehicle window anti-pinch detection method according to claim 5 or 6, characterized in that: The window training data set is composed of different window positions and normal operating current data of each window position, including: The normal operating current data of each window position at different window positions and different ambient temperatures are used to form a window training data set.
8. A vehicle window anti-pinch detection method, characterized in that: The vehicle window anti-pinch detection method comprises: The window controller responds to the window raising instruction and obtains the current working current and the current window position based on the motor current information; The central controller inputs the current window position into the window machine learning model to obtain a normal operating current threshold; In response to the current operating current being greater than the normal operating current threshold, the central controller returns window anti-pinch information to the window controller; The window controller controls the window to reverse or maintain the current window position based on the window anti-pinch information.
9. A vehicle window anti-pinch device, characterized in that: The vehicle window anti-pinch device includes a memory and a processor coupled to the memory; The memory is used to store program data, and the processor is used to execute the program data to implement the vehicle window anti-pinch detection method as described in any one of claims 1 to 7.
10. A computer storage medium, characterized in that: The computer storage medium is used to store program data, and when the program data is executed by a computer, it is used to implement the vehicle window anti-pinch detection method as described in any one of claims 1 to 7.