Vehicle windscreen wiper service life detection method and device, vehicle and program product

Abstract

The invention discloses a vehicle windscreen wiper service life detection method and device, a vehicle and a program product, and the method comprises the steps: building a residual life prediction model of a vehicle windscreen wiper through an end-to-end deep neural network model, and enabling the model to take the state data of a windscreen wiper rubber strip and the parameter characteristics of the environment where the vehicle is located as the input; and outputting the predicted residual life of the windscreen wiper rubber strip. The input of the model comprises the multi-mode state data, obtained in the current detection period, of the windscreen wiper rubber strip of the vehicle and the parameter characteristics of the environment where the windscreen wiper rubber strip is located after the windscreen wiper rubber strip is replaced last time. When the vehicle is in different external environments, the aging speeds of the windscreen wiper rubber strip are different, the parameter characteristics of the environment where the vehicle is located serve as the input parameters of the model, and the model can be guided to dynamically predict the remaining life of the windscreen wiper rubber strip according to the parameter characteristics of the environment where the vehicle is located. Therefore, the accuracy of predicting the residual life of the windscreen wiper rubber strip is further improved.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and more specifically, to a vehicle wiper life detection method, device, vehicle, and program product. Background Art

[0002] The current automotive industry's understanding of wiper (i.e. wiper strip) maintenance is still at the traditional stage: car owners rely heavily on subjective experience to determine replacement timing, while car companies continue to use fixed-period reminder rules (such as every 10,000 kilometers or six months). This results in a serious disconnect between maintenance strategies and the actual aging status of wiper strips, which can easily lead to untimely replacement of wiper strips, scratching the glass, and even causing driving hazards. Summary of the Invention

[0003] In view of the above problems, this application is proposed to provide a vehicle wiper life detection method, device, vehicle and program product to accurately identify the wiper life and provide a basis for wiper maintenance. The specific solution is as follows:

[0004] In a first aspect, a vehicle wiper life detection method is provided, comprising:

[0005] Acquire state data of two or more modes of the wiper rubber strip of the vehicle in the current detection period, wherein the state data is a state parameter related to the life of the wiper rubber strip;

[0006] Obtaining environmental parameter characteristics of the vehicle since the wiper rubber strip was last replaced until the present;

[0007] The configured remaining life prediction model is called to predict the remaining life of the wiper rubber strip based on the status data and the environmental parameter characteristics. The remaining life prediction model is an end-to-end deep neural network model, which is trained with the status data of the sample wiper rubber strip and its environmental parameter characteristics as input samples and the remaining life of the sample wiper rubber strip as the output label.

[0008] In one possible design, in another implementation of the first aspect of the embodiments of the present application, the status data includes status data of at least two or more of the following modalities:

[0009] Wiper blade pressure information;

[0010] Audio information when the wipers are running;

[0011] Surface wear information of the wiper blades.

[0012] In one possible design, in another implementation of the first aspect of the embodiments of the present application, the status data further includes:

[0013] Material information of the wiper blade.

[0014] In a possible design, in another implementation manner of the first aspect of the embodiment of the present application, the process of acquiring the pressure information of the wiper blade includes:

[0015] The pressure information of the wiper blade is obtained by setting a pressure-sensitive resistor at the root of the wiper arm or embedding the pressure-sensitive resistor in the contact surface of the wiper blade.

[0016] In a possible design, in another implementation manner of the first aspect of the embodiment of the present application, the process of acquiring the audio information of the wiper during operation includes:

[0017] The audio signal generated by the wiper during operation and the friction with the vehicle window glass is collected by a microphone, and the audio feature of the audio signal is extracted.

[0018] In a possible design, in another implementation manner of the first aspect of the embodiment of the present application, the process of acquiring the surface wear information of the wiper blade includes:

[0019] The surface wear line depth of the wiper blade is detected by an optical sensor based on the infrared reflection principle.

[0020] In a possible design, in another implementation manner of the first aspect of the embodiment of the present application, the process of acquiring the environmental parameter feature of the vehicle from the time when the wiper blade was replaced last time to the present includes:

[0021] The environmental data detected by the vehicle-mounted environmental sensor and the meteorological data of the location of the vehicle from the time when the wiper blade was replaced last time to the present are acquired.

[0022] The environmental parameter feature is determined based on the environmental data and the meteorological data.

[0023] In a possible design, in another implementation manner of the first aspect of the embodiment of the present application, the process of determining the environmental parameter feature based on the environmental data and the meteorological data includes:

[0024] All the environmental data and the meteorological data from the time when the wiper blade was replaced last time to the present are taken as the environmental parameter feature.

[0025] Or,

[0026] All the environmental data and the meteorological data from the time when the wiper blade was replaced last time to the present are statistically summarized to obtain the environmental parameter feature.

[0027] In one possible design, in another implementation of the first aspect of the embodiments of the present application, the remaining life prediction model is a large model structure; the process of calling the configured remaining life prediction model to predict the remaining life of the wiper rubber strip based on the state data and the environmental parameter characteristics includes:

[0028] Obtain a prompt template, the prompt template including a task instruction and a data slot, the data slot being used to populate state data and environmental parameter characteristics, the task instruction being used to instruct the model to reference the data in the data slot to predict the remaining life of the wiper blade;

[0029] The acquired state data and the environmental parameter characteristics are filled into the data slot to obtain a first prompt instruction prompt, and the first prompt instruction prompt is sent to the remaining life prediction model to obtain the remaining life of the wiper rubber strip output by the model.

[0030] In one possible design, in another implementation of the first aspect of the embodiments of the present application, the method further includes:

[0031] Determining whether a set warning condition is met based on the remaining life of the wiper rubber strip and status data of at least one mode;

[0032] When it is determined that the target warning condition is met, a warning action corresponding to the target warning condition is triggered.

[0033] In one possible design, in another implementation of the first aspect of the embodiments of the present application, the warning condition includes:

[0034] The remaining life of the wiper rubber strip is in a first life range, and at least one of the following conditions exists:

[0035] The pressure value of the wiper rubber strip drops to a first pressure value range;

[0036] The abnormal sound frequency of the audio when the wiper is running is in a first frequency range, and the abnormal sound frequency is the energy proportion of the audio in the set abnormal sound frequency range;

[0037] The surface wear of the wiper rubber strip is within a first wear range.

[0038] In one possible design, in another implementation of the first aspect of the embodiments of the present application, the warning action includes at least one of the following:

[0039] Prompt the health status of the wiper blades through one or more channels;

[0040] In set extreme weather scenarios, when it is determined that the target warning conditions are met, the vehicle speed is controlled to be reduced;

[0041] In an automatic driving scenario, when it is determined that the target warning condition is met, the control starts the redundant camera compensation collection of the field of view information.

[0042] In a second aspect, a vehicle wiper service life detection device is provided, comprising:

[0043] A state data acquisition unit is configured to acquire state data of two or more modalities of a vehicle wiper rubber strip in a current detection period, the state data being a state parameter related to the service life of the wiper rubber strip;

[0044] An environmental data acquisition unit is configured to acquire environmental parameter features of a vehicle from the time of the last replacement of the wiper rubber strip to the present time;

[0045] A computing unit is configured to call a configured residual life prediction model, and predict the residual life of the wiper rubber strip based on the state data and the environmental parameter features, the residual life prediction model being an end-to-end deep neural network model, which takes the state data of a sample wiper rubber strip and its environmental parameter features as input samples, and takes the residual life of the sample wiper rubber strip as output labels to train.

[0046] In a third aspect, a vehicle is provided, comprising a memory and a processor;

[0047] The processor;

[0048] The memory for storing processor-executable instructions;

[0049] The processor is configured to perform each step of the vehicle wiper service life detection method described in any one of the preceding first aspects.

[0050] In a fourth aspect, a readable storage medium is provided, which stores a computer program, and the computer program is executed by a processor to implement each step of the vehicle wiper service life detection method described in any one of the preceding first aspects.

[0051] In a fifth aspect, a computer program product is provided, comprising a computer program, which is executed by a processor to implement each step of the vehicle wiper service life detection method described in any one of the preceding first aspects.

[0052] By leveraging the above technical solution, the present application constructs a vehicle wiper blade remaining life prediction model using an end-to-end deep neural network model. This model takes the wiper strip state data and the vehicle's environmental parameter characteristics as input and outputs a predicted remaining life of the wiper strip. The model input includes multimodal state data of the vehicle wiper strip acquired during the current detection cycle. This multimodal state data provides the model with more state parameters related to the wiper strip life, thereby guiding the model to more accurately predict the remaining life of the wiper strip. Furthermore, the model input also includes the vehicle's environmental parameter characteristics from the time the wiper strip was last replaced to the current time. It is understood that the aging rate of the wiper strip varies depending on the vehicle's external environment; for example, the aging rate is significantly accelerated in acid rain. Therefore, by using the vehicle's environmental parameter characteristics as model input parameters, the present application can guide the model to dynamically predict the remaining life of the wiper strip based on the vehicle's environmental parameter characteristics, thereby further improving the accuracy of the wiper strip remaining life prediction.

[0053] Furthermore, considering that the aging process of wiper strips is an extremely complex process, the remaining life of the wiper strips is highly nonlinearly related to their status data and environmental parameter characteristics, making it impossible to accurately calculate the remaining life through manually set mathematical formulas. This application uses an end-to-end deep neural network model, which is pre-trained using sample training data. This model can learn the difficult-to-express, highly nonlinear dependencies between input and output on the training data set, thereby providing more accurate and robust remaining life prediction results. This can then be used to guide the timely maintenance of wiper strips. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Various other advantages and benefits will become apparent to those skilled in the art upon reading the detailed description of the preferred embodiment below. The accompanying drawings are for illustration purposes only and are not to be considered as limiting the present application. The same reference symbols are used throughout the drawings to represent the same components. In the drawings:

[0055] Figure 1 A schematic diagram of a system architecture for implementing the vehicle wiper life detection method provided in an embodiment of the present application;

[0056] Figure 2 A schematic flow chart of a vehicle wiper life detection method provided in an embodiment of the present application;

[0057] Figure 3 A schematic structural diagram of a vehicle wiper life detection device provided in an embodiment of the present application;

[0058] Figure 4A schematic diagram of the structure of a vehicle provided in an embodiment of the present application. DETAILED DESCRIPTION

[0059] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0060] Traditional wiper blade replacement relies on manual inspection for visible faults like unusual noises or wiper marks. By then, the rubber strips have already suffered irreversible damage, such as cracking and hardening, which can cause scratches on the windshield or blind spots. For example, in hot climates, ultraviolet rays accelerate rubber aging, but car owners still replace them every six months. This can shorten the actual wiper lifespan to as little as three months, creating safety risks.

[0061] To this end, the present application provides a vehicle wiper life detection solution that can accurately identify the remaining life of the wiper, thereby guiding the replacement of the wiper rubber strips.

[0062] The vehicle wiper life detection method provided in this application can be applied to Figure 1 The system architecture shown in FIG. 1 may include a terminal 100 and a server 200. The server 200 may include one or more servers ( Figure 1 (This section includes a server as an example).

[0063] The terminal 100 or the server 200 can be used alone to execute the vehicle wiper life detection method provided in the embodiment of the present application. In addition, the terminal 100 and the server 200 can also be used in conjunction to execute the vehicle wiper life detection method provided in the embodiment of the present application.

[0064] Next describe Figure 1 The product form of the mid-terminal 100;

[0065] The terminal 100 in the embodiment of the present application may be a vehicle central control processor, or other vehicle-mounted terminal controller.

[0066] In some possible implementation schemes, the process of the terminal 100 and the server 200 collaboratively executing the vehicle wiper life detection method is taken as an example. The terminal 100 on the vehicle can obtain the status data of two or more modes of the vehicle wiper strips in the current detection period, and obtain the environmental parameter characteristics of the vehicle since the last replacement of the wiper strips to the present, such as detecting the external environmental parameter characteristics through the vehicle-mounted temperature sensor, rain sensor, etc. The terminal 100 can upload the obtained status data and environmental parameter characteristics to the server 200. The server 200 can call the configured remaining life prediction model and predict the remaining life of the wiper strips based on the status data and environmental parameter characteristics. The server 200 can further send the predicted remaining life of the wiper strips to the vehicle-mounted terminal 100, and the terminal 100 will perform the corresponding early warning action.

[0067] In the process of the terminal 100 and the server 200 cooperating to execute the vehicle wiper life detection method, the various steps involved in the method can be freely allocated to be executed by the terminal 100 or the server 200. The above only illustrates an optional implementation scheme.

[0068] In some other possible implementations, the terminal 100 may deploy a remaining life prediction model locally, so that the vehicle wiper remaining life prediction method can be completed by the terminal 100 alone.

[0069] The embodiment of the present application provides a vehicle wiper life detection method, which is illustrated by applying the method to a computer device. The computer device can be specifically Figure 1 The terminal 100 or the system consisting of the terminal 100 and the server 200. Figure 2 The vehicle wiper life detection method specifically includes the following steps:

[0070] Step S100: Acquire state data of two or more modes of the vehicle wiper rubber strip in the current detection period, where the state data is a state parameter related to the life of the wiper rubber strip.

[0071] This application can pre-configure the detection period, which can be in hours, days, weeks or other time periods.

[0072] The multimodal sensor can be used to obtain multimodal state data of the vehicle wiper rubber strip, wherein each modal state data is a state parameter related to the life of the wiper rubber strip.

[0073] Exemplarily, the multimodal status data may include one or more of the following:

[0074] The pressure information of the wiper strip, the audio information when the wiper is running, the surface wear information of the wiper strip, the material information of the wiper strip, etc.

[0075] Among them, the pressure information of the wiper strip can be detected by a pressure sensor. The pressure sensor can be set at the root of the wiper arm or embedded in the contact surface of the strip. The pressure sensor can be exemplarily a piezoresistor, such as a micro piezoresistor with a range of 0-20N and an accuracy of ±0.1N. The pressure value of the wiper strip can be monitored in real time through the pressure sensor. Generally, the pressure value of the wiper strip is 10N-15N during the normal life stage. When the pressure value drops by more than 30%, it can be determined that the wiper strip shows signs of aging. The more the pressure value drops, the more serious the aging of the wiper strip, that is, the shorter the remaining life.

[0076] The audio information of the wipers when in operation can be collected directionally by a microphone. That is, the microphone collects the audio signal generated by the friction between the wipers and the window glass when in operation, and then the audio features of the audio signal can be extracted as audio information.

[0077] It is understandable that the wiper strips become hard after aging, and the friction with the glass will produce high-frequency abnormal noises. Therefore, the audio features of the audio signal can be extracted as audio information for subsequent models to predict the remaining life of the strips.

[0078] An optional example is to convert the audio signal into the frequency domain through Fourier transform and count the energy proportion in the set abnormal sound frequency band. The higher the energy proportion, the more serious the aging of the rubber strip and the shorter the remaining life. The abnormal sound frequency band is the frequency band of the abnormal sound signal generated by the friction after the rubber strip ages. Generally, the abnormal sound frequency band is between 5kHZ and 7kHZ. When the energy proportion of the abnormal sound frequency band exceeds 30%, it can be determined that the wiper rubber strip shows signs of aging.

[0079] In other optional examples, other types of audio features can be extracted from the audio signal, such as the short-term energy entropy (typically 0.8-1.2 when the wiper blade is in a normal lifespan) and the zero-crossing rate (typically 80 Hz-120 Hz when the wiper blade is in a normal lifespan). These audio features will vary with the aging state of the wiper blade, thus guiding the model to predict the remaining lifespan of the blade.

[0080] Optionally, before extracting audio features from the audio signal, a step of denoising the audio signal can be added, such as using a wavelet threshold denoising algorithm or other denoising algorithm to eliminate environmental noise (such as wind noise, engine noise, etc.) in the audio signal and retain high-frequency abnormal sound features.

[0081] The present application can detect the surface wear of the wiper strips using an optical sensor based on the principle of infrared reflection. This information can include information such as the depth and width of the surface wear lines. The optical sensor, located inside the vehicle window, can detect the surface wear of the wiper strips based on the principle of infrared reflection.

[0082] As you can understand, a new rubber strip (with a smooth surface) primarily reflects specular light. Most of the incident light is concentrated in a direction where the reflection angle equals the incident angle. The sensor receives a higher intensity of light in the specular reflection direction.

[0083] The surface of the worn rubber strip (rough surface) contains micro-pits and scratches (wear lines), which mainly cause diffuse reflection. The incident light is scattered in all directions, and the light intensity received in the original specular reflection direction is significantly reduced.

[0084] Relationship between wear depth and reflected light intensity: The deeper and wider the lines, the rougher the surface, the weaker the specular reflection component, and the stronger the diffuse reflection component. This application measures the reflected light intensity or intensity distribution changes in a specific direction (such as the specular reflection direction) to infer the surface roughness of the rubber strip and thus characterize the surface wear of the rubber strip.

[0085] In some optional grading standards, when the depth of the wear lines on the rubber strip surface is <1mm, it can be judged as light wear; when the depth of the wear lines on the rubber strip surface is between 1mm-2mm, it can be judged as moderate wear; when the depth of the wear lines on the rubber strip surface is >2mm, it can be judged as heavy wear.

[0086] The material information of the wiper strips can be obtained by querying the vehicle manual. Alternatively, when the user replaces the non-original wiper strips, the user can input the material information of the newly replaced strips. The material information of the strips includes silicone rubber, fluororubber, etc. The aging rate and service life of strips of different materials may be different. For example, under the same environment, 1 hour of operation of a silicone rubber wiper strip can be equivalent to 0.7 hours of an ordinary fluororubber wiper strip. By feeding the material information of the wiper strips as status data into the remaining life prediction model, the model can dynamically adjust the predicted remaining life according to the wiper strips of different materials, thereby realizing a universal prediction of the remaining life of strips of different materials.

[0087] Step S110: Obtain the environmental parameter characteristics of the vehicle from the time the wiper rubber strips were last replaced to the present.

[0088] Considering that the lifespan of wiper blades is closely related to the vehicle's external environment, such as varying rainfall frequencies, light intensities, acid rain, and sandstorms, the aging rate of wiper blades is significantly affected, thereby affecting the remaining lifespan of the wiper blades. Therefore, this embodiment obtains the environmental parameter characteristics of the vehicle from the last time the wiper blades were replaced until the current moment, which serve as input parameters for the remaining lifespan prediction model in the next step. This allows the model to dynamically predict the remaining lifespan of the wiper blades based on these environmental parameters.

[0089] Step S120: Call the configured remaining life prediction model to predict the remaining life of the wiper rubber strip based on the state data and environmental parameter characteristics. The remaining life prediction model is an end-to-end deep neural network model.

[0090] In this embodiment, a pre-trained end-to-end deep neural network model (remaining life prediction model) outputs the predicted remaining life of the wiper rubber strip based on the input state data and environmental parameter characteristics.

[0091] Among them, the remaining life prediction model is trained with the status data of the sample wiper rubber strip and its environmental parameter characteristics as input samples and the remaining life of the sample wiper rubber strip as the output label.

[0092] The remaining life prediction model can use neural network architectures such as long short-term memory (LSTM) and Transformer architectures. It uses state data and environmental parameter features as input and outputs an end-to-end prediction of the remaining life of the wiper blades.

[0093] The method provided in an embodiment of the present application constructs a vehicle wiper blade remaining life prediction model using an end-to-end deep neural network model. The model takes wiper blade state data and the vehicle's environmental parameter characteristics as input and outputs a predicted remaining life of the wiper blades. The model input includes multimodal state data of the vehicle wiper strips acquired during the current detection cycle. This multimodal state data provides the model with more state parameters related to the wiper strip life, thereby guiding the model to more accurately predict the remaining life of the wiper strips. Furthermore, the model input also includes the vehicle's environmental parameter characteristics from the time the wiper strips were last replaced. It is understood that the aging rate of the wiper strips varies depending on the vehicle's external environment; for example, the aging rate is significantly accelerated in acid rain. Therefore, by using the vehicle's environmental parameter characteristics as model input parameters, the present application can guide the model to dynamically predict the remaining life of the wiper strips based on the vehicle's environmental parameter characteristics, thereby further improving the accuracy of the remaining life prediction of the wiper strips.

[0094] Furthermore, considering that the aging process of wiper blades is extremely complex, the remaining life of the wiper blades is highly nonlinearly related to their status data and environmental parameter characteristics, making it impossible to accurately calculate the remaining life through manually set mathematical formulas. This application uses an end-to-end deep neural network model, which is pre-trained using sample training data. This model can learn the difficult-to-express, highly nonlinear dependencies between input and output on the training data set, thereby providing more accurate and robust remaining life prediction results.

[0095] Furthermore, related technologies attempt to calculate the remaining life of wiper blades using artificially designed mathematical formulas (typically linear or simple nonlinear combinations, such as weighted sums). These formulas struggle to accurately describe these complex, interdependent relationships. The accuracy of these formulas relies heavily on the modeler's deep understanding of physical and chemical processes and on simplified assumptions, which often fail under complex real-world conditions, resulting in inaccurate calculations.

[0096] Neural networks, particularly deep ones, are essentially powerful universal function approximators. Using training data, they automatically learn the complex, nonlinear mapping relationship between input features (environmental parameters, pressure, noise frequency, crack depth, etc.) and the target output (remaining life). Without manually defining the specific function form, the model can capture subtle patterns and interactions that are difficult to express with explicit formulas, significantly improving the accuracy of predicted remaining life of wiper blades.

[0097] In some embodiments, the process of obtaining the environmental parameter characteristics of the vehicle from the last replacement of the wiper rubber strip to the present time in the aforementioned step S110 is described.

[0098] This embodiment introduces two approaches to acquiring environmental parameter characteristics. One approach involves detecting environmental data using onboard environmental sensors. For example, this can be achieved using onboard light sensors, rain sensors, temperature sensors, and the like. Another approach involves acquiring weather data for the vehicle's location via a network, such as through the Internet of Vehicles (IoV), providing weather forecast information for smog, sandstorms, and the like.

[0099] Environmental and weather data can be acquired at a fixed, configured time interval. In one optional implementation, the vehicle's environmental sensors can acquire environmental data at a set interval (e.g., minutes, hours, or other durations). The vehicle's computer can also acquire daily weather data at a fixed time.

[0100] Understandably, environmental and meteorological data, such as UV intensity (0-1000W / m²), rainfall pH, and extreme temperature fluctuations (-30°C to 80°C), significantly impact the lifespan of wiper blades. For example, for every 100W / m² increase in UV intensity, the lifespan of rubber blades decreases by 15%, and acid rain (pH <5) increases the corrosion rate of wiper blades by two times.

[0101] For the environmental data and meteorological data obtained through any one or more channels, the environmental parameter characteristics can be further determined.

[0102] In an optional example, all environmental data and meteorological data acquired since the last replacement of the wiper rubber strips to the present time may be used as environmental parameter features.

[0103] By using the full amount of environmental data and meteorological data obtained as environmental parameter features, more comprehensive reference data can be provided for the remaining life prediction model, thereby improving the accuracy of the model prediction results.

[0104] In another alternative example, considering that in certain scenarios (such as when the current time interval is far from the last wiper blade replacement), the amount of complete environmental and meteorological data acquired may be too large, which may exceed the input data requirements of the remaining life prediction model. In addition, excessive input data can also reduce the model's inference speed. To this end, in this embodiment, all environmental and meteorological data acquired since the last wiper blade replacement can be statistically summarized and used as environmental parameter features.

[0105] Among them, when statistically summarizing environmental data and meteorological data, statistics can be performed separately according to different environmental parameter dimensions. The environmental parameters of different dimensions include but are not limited to: temperature, rainfall, ultraviolet rays, rainfall pH value, light, haze, sandstorms, etc.

[0106] Taking temperature parameters as an example, when statistics are collected on all acquired temperature parameters, in an exemplary solution, the duration that the vehicle is in different temperature ranges may be counted.

[0107] Taking rainfall as an example, when statistics are collected on all the acquired rainfall parameters, in an exemplary solution, the sum of all rainfall parameters may be calculated to obtain a total rainfall parameter.

[0108] By statistically summarizing the acquired environmental and meteorological data, we can ensure that important information is not lost while reducing the amount of data on environmental parameter characteristics, thereby meeting the requirements of the remaining life prediction model for the amount of input data, improving the model inference speed, and ensuring the accuracy of the model prediction results.

[0109] In some embodiments of the present application, a remaining life prediction model is introduced.

[0110] The remaining life prediction model utilizes an end-to-end deep neural network architecture. In one possible implementation, the model can employ a traditional deep neural network model, such as an LSTM architecture or an RNN architecture. The training data used in the training process includes the state data and environmental parameter characteristics of sample wiper blades as training samples, and the remaining life of the sample wiper blades as sample labels.

[0111] In another possible implementation, the RUL prediction model can also adopt a large model structure. For example, the RUL prediction model uses a general large model as its base, fine-tunes the base large model using domain data, and further trains the fine-tuned large model using task data, making the large model suitable for the RUL prediction task of wiper blades.

[0112] Among them, domain data may include status data of wiper strips of different modes, descriptive data of the association between the remaining life (aging status) of the wiper strips, descriptive data of the association between different environmental parameters and the remaining life (aging status) of the wiper strips, etc.

[0113] As mentioned above, some examples of domain data include but are not limited to:

[0114] Field data 1: When the wiper rubber strip is in the normal life stage, its pressure value is between 10N and 15N. When the pressure value drops by more than 30%, it can be determined that the wiper rubber strip shows signs of aging. The more the pressure value drops, the more serious the aging of the wiper rubber strip, that is, the shorter the remaining life.

[0115] Field Data 2: Collects audio signals from windshield wipers in operation and calculates the energy percentage within a set abnormal noise frequency band. Typically, the abnormal noise frequency band is between 5kHz and 7kHz. If the energy percentage in this frequency band exceeds 30%, it can be determined that the wiper rubber is showing signs of aging.

[0116] Field data 3: When the depth of the wear lines on the wiper strip surface is less than 1mm, it can be judged as light wear; when the depth of the wear lines on the wiper strip surface is between 1mm-2mm, it can be judged as moderate wear; when the depth of the wear lines on the wiper strip surface is greater than 2mm, it can be judged as heavy wear.

[0117] Field Data 4: For every 100W / m² increase in UV radiation, the life of rubber is shortened by 15%; the corrosion rate increases by 2 times in acid rain (pH < 5).

[0118] By using domain data to fine-tune the base large model, the large model can learn domain knowledge related to the prediction of the remaining life of wiper strips.

[0119] The domain-tuned large model is further trained using task data to adapt it to the task paradigm of predicting the remaining life of wiper blades. The task data can be constructed according to the input-output paradigm of the task. The following example shows a piece of task data:

[0120] Input prompt: Please refer to the provided wiper blade state data for each mode and the vehicle's environmental parameter characteristics to predict the remaining life of the wiper blade.

[0121] [status data: xxx];

[0122] [Environmental parameter characteristics: xxx];

[0123] Output: The remaining usable time of the wiper blades is xx hours.

[0124] Based on the above trained remaining life prediction model, the remaining life of the wiper rubber strip can be predicted in the inference stage.

[0125] In one possible implementation, when the remaining life prediction model is a large model structure, the process of calling the configured remaining life prediction model and predicting the remaining life of the wiper rubber strip based on the state data and environmental parameter characteristics may include:

[0126] Get the prompt template. The prompt template includes task instructions and data slots. The data slots are used to fill in status data and environmental parameter characteristics. The task instructions are used to instruct the model to refer to the data in the data slots to predict the remaining life of the wiper rubber strips.

[0127] The acquired status data and environmental parameter characteristics are filled into the data slot to obtain a first prompt instruction prompt, and the first prompt instruction prompt is sent to the remaining life prediction model to obtain the remaining life of the wiper rubber strip output by the model.

[0128] In this embodiment, by leveraging the powerful knowledge understanding and logical reasoning capabilities of the large model base, the accuracy of the remaining life prediction results of the wiper strips can be improved.

[0129] Furthermore, the embodiments of the present application also provide a fault warning strategy. Specifically:

[0130] Based on the predicted remaining life of the wiper rubber strip and state data of at least one mode, it is determined whether a set warning condition is met.

[0131] When it is determined that the target warning condition is met, the warning action corresponding to the target warning condition is triggered.

[0132] Among them, the warning conditions include but are not limited to:

[0133] The remaining life of the wiper rubber strip is in the first life range, and at least one of the following conditions exists:

[0134] The pressure value of the wiper rubber strip drops to a first pressure value range;

[0135] When the wipers are running, the abnormal sound frequency of the audio is within a first frequency range, where the abnormal sound frequency is the energy proportion of the audio within the set abnormal sound frequency range;

[0136] The surface wear of the wiper rubber strip is in a first wear range.

[0137] The warning thresholds such as the first life interval, the first pressure value range, the first frequency range, the first wear amount range, etc. included in the above-mentioned warning conditions can support flexible setting by users.

[0138] In this embodiment, when setting the warning conditions, it is not simply based on the predicted remaining life of the wiper strips, but also takes into account additional information such as the pressure of the wiper strips, the frequency of abnormal noise during operation, and the amount of surface wear, making the warning strategy more accurate and reliable.

[0139] In some possible implementations, when target warning information is detected based on environmental data and / or meteorological data, the warning threshold in the warning condition can also be dynamically adjusted. For example:

[0140] When a sandstorm warning is detected, the lower limit of the first pressure value range can be increased to prevent particles from scratching the glass.

[0141] This embodiment provides a hierarchical warning condition, which is exemplary:

[0142] Mild wear: Remaining life>30%, and pressure value drop<20%, or abnormal sound frequency<10%, or wear amount<1mm;

[0143] Critical replacement: Remaining life: 10%-30%, and, pressure value drop: 20%-30%, or abnormal sound frequency: 10%-15%, or wear: 1mm-2mm;

[0144] Emergency replacement: remaining life <10%, and, pressure value drops >30%, or abnormal sound frequency >15%, or wear >2mm.

[0145] Of course, the specific values ​​in the above-mentioned graded warning logic are only used as an example, and relevant technical personnel can dynamically adjust the value size according to actual business needs.

[0146] For different warning conditions, corresponding warning actions can be pre-configured. For example, for the "light wear" warning condition, the corresponding warning action can be an output prompt: the wiper strips are currently lightly worn, please pay attention to the status of the wiper strips. For the "critical replacement" warning condition, the corresponding warning action can be an output prompt: the wiper strips are currently critical for replacement, please replace the wiper strips in time. For the "urgent replacement" warning condition, the corresponding warning action can be an output prompt: the wiper strips are severely worn, please replace the wiper strips urgently.

[0147] Of course, the above is only an example of the early warning action of outputting a prompt. In addition, other early warning actions can also be set.

[0148] This embodiment introduces an optional example of a warning action, which includes but is not limited to the following:

[0149] 1. Multi-channel reminders

[0150] The health status of the wiper blades can be indicated through one or more channels. For example:

[0151] 1.1. Display the health status of the wiper blades on the car screen.

[0152] The on-board screen displays the health status of the wiper blades in a visual form, such as the remaining life percentage, heat map of the wear areas (green represents normal, yellow represents light wear, and red represents urgent replacement), etc.

[0153] 1.2. Voice prompts on the health status of the wiper blades.

[0154] The car terminal's voice assistant can provide voice notifications about the health status of the wiper blades. For example, when starting the vehicle in cold temperatures, it will prompt, "The wiper blades are aging. Please avoid driving at high speeds."

[0155] 1.3. Push wiper blade replacement reminder information to the user terminal device bound to the vehicle.

[0156] For example, when the remaining life of the wiper rubber strips is 10%, a notification "The wiper rubber strips need to be replaced this week" is pushed to the terminal APP.

[0157] In addition, it can also recommend nearby maintenance points based on navigation data and support online appointment for replacement services.

[0158] 2. Linkage control

[0159] Extreme weather response:

[0160] In set extreme weather scenarios, when it is determined that the target warning conditions are met, the vehicle speed is controlled to be reduced.

[0161] For example, in heavy rain, if the system detects that the wiper blades have reached a critical or urgent replacement warning condition, the vehicle can be automatically slowed down to a target speed of, for example, 60 km / h. Other control strategies can also be activated, such as the defogger function to prevent blurred vision.

[0162] In the autonomous driving scenario, when it is determined that the target warning conditions are met, the control starts the redundant camera compensation to collect field of view information.

[0163] For example, in an autonomous driving scenario, if the wiper blades are detected to have reached a critical or emergency replacement warning condition, redundant cameras are activated to compensate for field of view information and send a safety alert via the CAN bus. Emergency braking can also be triggered if necessary.

[0164] The vehicle wiper life detection device provided in an embodiment of the present application is described below. The vehicle wiper life detection device described below and the vehicle wiper life detection method described above can be referenced to each other.

[0165] See also Figure 3 , Figure 3 This is a schematic structural diagram of a vehicle wiper life detection device disclosed in an embodiment of the present application.

[0166] like Figure 3 As shown, the device may include:

[0167] A state data acquisition unit 11 is used to acquire state data of two or more modes of the vehicle wiper rubber strip in the current detection period, wherein the state data is a state parameter related to the life of the wiper rubber strip;

[0168] An environmental data acquisition unit 12 is used to acquire environmental parameter characteristics of the vehicle since the wiper rubber strip was last replaced until the present;

[0169] The computing unit 13 is used to call the configured remaining life prediction model to predict the remaining life of the wiper rubber strip based on the state data and the environmental parameter characteristics. The remaining life prediction model is an end-to-end deep neural network model, which is trained with the state data of the sample wiper rubber strip and its environmental parameter characteristics as input samples and the remaining life of the sample wiper rubber strip as the output label.

[0170] In some possible implementations, the status data includes status data of at least two or more of the following modalities:

[0171] Wiper blade pressure information;

[0172] Audio information when the wipers are running;

[0173] Surface wear information of the wiper blades.

[0174] In some possible implementations, the status data further includes:

[0175] Material information of the wiper blade.

[0176] In some possible implementations, the process of the status data acquisition unit acquiring the pressure information of the wiper rubber strip includes:

[0177] The pressure information of the wiper strip is obtained by setting a piezoresistor at the root of the wiper arm or embedded in the contact surface of the strip.

[0178] In some possible implementations, the process of the status data acquisition unit acquiring audio information when the wiper is running includes:

[0179] The microphone collects the audio signal generated by the friction between the wiper and the window glass during operation, and extracts the audio features of the audio signal.

[0180] In some possible implementations, the process of the status data acquisition unit acquiring the surface wear information of the wiper rubber strip includes:

[0181] The optical sensor detects the depth of the surface wear lines of the wiper strip based on the principle of infrared reflection.

[0182] In some possible implementations, the process of the environmental data acquisition unit acquiring the environmental parameter characteristics of the vehicle since the last replacement of the wiper rubber strips to the present time includes:

[0183] Obtaining environmental data detected by the vehicle's onboard environmental sensors since the last replacement of the wiper rubber strips and weather data at the vehicle's location;

[0184] The environmental parameter characteristics are determined based on the environmental data and the meteorological data.

[0185] In some possible implementations, the process of determining the environmental parameter characteristics by the environmental data acquisition unit based on the environmental data and the meteorological data includes:

[0186] All the environmental data and the meteorological data from the last replacement of the wiper rubber strip to the present are used as the environmental parameter features;

[0187] or,

[0188] All the environmental data and the meteorological data from the last replacement of the wiper rubber strip to the present are statistically summarized to obtain the environmental parameter characteristics.

[0189] In some possible implementations, the remaining life prediction model is a large model structure; the calculation unit calls the configured remaining life prediction model to predict the remaining life of the wiper rubber strip based on the state data and the environmental parameter characteristics, including:

[0190] Obtain a prompt template, the prompt template including a task instruction and a data slot, the data slot being used to populate state data and environmental parameter characteristics, the task instruction being used to instruct the model to reference the data in the data slot to predict the remaining life of the wiper blade;

[0191] The acquired state data and the environmental parameter characteristics are filled into the data slot to obtain a first prompt instruction prompt, and the first prompt instruction prompt is sent to the remaining life prediction model to obtain the remaining life of the wiper rubber strip output by the model.

[0192] In some possible implementations, the apparatus of the present application may further include:

[0193] The early warning unit is used to determine whether a set early warning condition is met based on the remaining life of the wiper rubber strip and status data of at least one mode; when it is determined that the target early warning condition is met, trigger the early warning action corresponding to the target early warning condition.

[0194] In some possible implementations, the warning conditions include:

[0195] The remaining life of the wiper rubber strip is in a first life range, and at least one of the following conditions exists:

[0196] The pressure value of the wiper rubber strip drops to a first pressure value range;

[0197] The abnormal sound frequency of the audio when the wiper is running is in a first frequency range, and the abnormal sound frequency is the energy proportion of the audio in the set abnormal sound frequency range;

[0198] The surface wear of the wiper rubber strip is within a first wear range.

[0199] In some possible implementations, the warning action includes at least one of the following:

[0200] Prompt the health status of the wiper blades through one or more channels;

[0201] In set extreme weather scenarios, when it is determined that the target warning conditions are met, the vehicle speed is controlled to be reduced;

[0202] In the autonomous driving scenario, when it is determined that the target warning conditions are met, the control starts the redundant camera compensation to collect field of view information.

[0203] Each unit in the above-mentioned vehicle wiper life detection device can be implemented in whole or in part through software, hardware, or a combination thereof. Each of the above-mentioned units can be embedded in or independent of a processor in a computer device in hardware form, or can be stored in a memory in the computer device in software form, so that the processor can call and execute the corresponding operations of each of the above-mentioned units.

[0204] An embodiment of the present application further provides a vehicle, comprising:

[0205] processor;

[0206] a memory for storing processor-executable instructions;

[0207] The processor is configured to execute each step of the vehicle wiper life detection method described in any embodiment of the present application.

[0208] The present disclosure also provides a computer-readable storage medium having computer program instructions stored thereon. When the program instructions are executed by a processor, the various steps of the vehicle wiper life detection method described in any embodiment of the present application are implemented.

[0209] Figure 4 is a block diagram illustrating a vehicle 600 according to an exemplary embodiment. For example, vehicle 600 may be a hybrid vehicle, a non-hybrid vehicle, an electric vehicle, a fuel cell vehicle, or another type of vehicle. Vehicle 600 may be an autonomous vehicle, a semi-autonomous vehicle, or a non-autonomous vehicle. Vehicle 600 may also be equipped with a brake-by-wire system.

[0210] Reference Figure 4 Vehicle 600 may include various subsystems, such as an infotainment system 610, a perception system 620, a decision control system 630, a drive system 640, and a computing platform 650. Vehicle 600 may also include more or fewer subsystems, and each subsystem may include multiple components. Furthermore, each subsystem and each component of vehicle 600 may be interconnected via wired or wireless means.

[0211] In some embodiments, the infotainment system 610 may include a communication system, an entertainment system, a navigation system, and the like.

[0212] The perception system 620 may include several sensors for sensing information about the environment surrounding the vehicle 600. For example, the perception system 620 may include a global positioning system (which may be a GPS system, a BeiDou system, or other positioning systems), an inertial measurement unit (IMU), a laser radar, a millimeter-wave radar, an ultrasonic radar, and a camera.

[0213] The decision control system 630 may include a computing system, a vehicle controller, a steering system, a throttle, and a braking system.

[0214] The drive system 640 may include components that provide power to the vehicle 600. In one embodiment, the drive system 640 may include an engine, an energy source, a transmission system, and wheels. The engine may be an internal combustion engine, an electric motor, an air compression engine, or a combination thereof. The engine is capable of converting energy provided by the energy source into mechanical energy.

[0215] Some or all functions of the vehicle 600 are controlled by a computing platform 650. The computing platform 650 may include at least one processor 651 and a memory 652. The processor 651 may execute instructions 653 stored in the memory 652.

[0216] The processor 651 can be any conventional processor, such as a commercially available CPU. The processor can also include a graphics processor (GPU), a field programmable gate array (FPGA), a system on chip (SOC), an application specific integrated circuit (ASIC), or a combination thereof.

[0217] The memory 652 may be implemented by any type of volatile or non-volatile memory device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk, or optical disk.

[0218] In addition to instructions 653 , memory 652 may also store data, such as road maps, route information, and vehicle location, direction, speed, etc. The data stored in memory 652 may be used by computing platform 650 .

[0219] In the embodiment of the present application, the processor 651 can execute the instruction 653 to complete all or part of the steps of the above-mentioned vehicle wiper life detection method.

[0220] In another exemplary embodiment, a computer program product is also provided, including computer-readable instructions. When the computer-readable instructions are executed on an electronic device, the electronic device implements each step of any vehicle wiper life detection method provided in the embodiments of the present application.

[0221] It should also be noted that the device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separate, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed across multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the present embodiment. In addition, in the drawings of the device embodiments provided in this application, the connection relationship between the modules indicates that there is a communication connection between them, which can be specifically implemented as one or more communication buses or signal lines.

[0222] Through the description of the above embodiments, those skilled in the art can clearly understand that the present application can be implemented by means of software plus necessary general hardware, and of course can also be implemented by special hardware including application-specific integrated circuits, special CPUs, special memories, special components, etc. In general, all functions performed by computer programs can be easily implemented with corresponding hardware, and the specific hardware structures used to implement the same function can also be diverse, such as analog circuits, digital circuits or special circuits, etc. However, for the present application, software program implementation is a better implementation method in most cases. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a readable storage medium, such as a computer's floppy disk, USB flash drive, mobile hard disk, ROM, RAM, magnetic disk or optical disk, etc., and includes a number of instructions to enable a computer device (which can be a personal computer, training equipment, or network equipment, etc.) to execute the methods described in each embodiment of the present application.

[0223] In the above embodiments, all or part of the embodiments may be implemented by software, hardware, firmware, or any combination thereof. When implemented by software, all or part of the embodiments may be implemented in the form of a computer program product.

[0224] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from a website, a computer, a training device or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website, a computer, a training device or a data center. The computer-readable storage medium can be any available medium that a computer can store or a data storage device such as a training device, a data center, etc. that includes one or more available media integrations. The available medium can be a magnetic medium, (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive (SSD)).

[0225] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The various embodiments can be combined as needed, and the same or similar parts can be referenced to each other.

Claims

1. A vehicle wiper life detection method, characterized in that: include: Acquire state data of two or more modes of the wiper rubber strip of the vehicle in the current detection period, wherein the state data is a state parameter related to the life of the wiper rubber strip; Obtaining environmental parameter characteristics of the vehicle since the wiper rubber strip was last replaced until the present; The configured remaining life prediction model is called to predict the remaining life of the wiper rubber strip based on the status data and the environmental parameter characteristics. The remaining life prediction model is an end-to-end deep neural network model, which is trained with the status data of the sample wiper rubber strip and its environmental parameter characteristics as input samples and the remaining life of the sample wiper rubber strip as the output label.

2. The method according to claim 1, characterized in that The state data includes state data of at least two or more modes as follows: Wiper blade pressure information; Audio information when the wipers are running; Surface wear information of the wiper blades.

3. The method according to claim 2, characterized in that The status data also includes: Material information of the wiper blade.

4. The method according to claim 2, characterized in that The process of obtaining the pressure information of the wiper rubber strip includes: The pressure information of the wiper strip is obtained by setting a piezoresistor at the root of the wiper arm or embedded in the contact surface of the strip.

5. The method according to claim 2, characterized in that The process of obtaining audio information when the wiper is running includes: The microphone collects the audio signal generated by the friction between the wiper and the window glass during operation, and extracts the audio features of the audio signal.

6. The method according to claim 2, characterized in that The process of obtaining the surface wear information of the wiper strip includes: The optical sensor detects the depth of the surface wear lines of the wiper strip based on the principle of infrared reflection.

7. The method according to claim 1, characterized in that The process of obtaining the environmental parameter characteristics of the vehicle since the last replacement of the wiper rubber strips to the present time includes: Obtaining environmental data detected by the vehicle's onboard environmental sensors since the last replacement of the wiper rubber strips and weather data at the vehicle's location; The environmental parameter characteristics are determined based on the environmental data and the meteorological data.

8. The method according to claim 7, characterized in that The process of determining the environmental parameter characteristics based on the environmental data and the meteorological data includes: All the environmental data and the meteorological data from the last replacement of the wiper rubber strip to the present are used as the environmental parameter features; or, All the environmental data and the meteorological data from the last replacement of the wiper rubber strip to the present are statistically summarized to obtain the environmental parameter characteristics.

9. The method according to claim 1, characterized in that The remaining life prediction model is a large model structure; the process of calling the configured remaining life prediction model to predict the remaining life of the wiper rubber strip based on the state data and the environmental parameter characteristics includes: Obtain a prompt template, the prompt template including a task instruction and a data slot, the data slot being used to populate state data and environmental parameter characteristics, the task instruction being used to instruct the model to reference the data in the data slot to predict the remaining life of the wiper blade; The acquired state data and the environmental parameter characteristics are filled into the data slot to obtain a first prompt instruction prompt, and the first prompt instruction prompt is sent to the remaining life prediction model to obtain the remaining life of the wiper rubber strip output by the model.

10. The method according to any one of claims 1 to 9, characterized in that Also includes: Determining whether a set warning condition is met based on the remaining life of the wiper rubber strip and status data of at least one mode; When it is determined that the target warning condition is met, a warning action corresponding to the target warning condition is triggered.

11. The method according to claim 10, characterized in that The warning conditions include: The remaining life of the wiper rubber strip is in a first life range, and at least one of the following conditions exists: The pressure value of the wiper rubber strip drops to a first pressure value range; The abnormal sound frequency of the audio when the wiper is running is in a first frequency range, and the abnormal sound frequency is the energy proportion of the audio in the set abnormal sound frequency range; The surface wear of the wiper rubber strip is within a first wear range.

12. The method according to claim 10, characterized in that The warning action includes at least one of the following: Prompt the health status of the wiper blades through one or more channels; In set extreme weather scenarios, when it is determined that the target warning conditions are met, the vehicle speed is controlled to be reduced; In the autonomous driving scenario, when it is determined that the target warning conditions are met, the control starts the redundant camera compensation to collect field of view information.

13. A vehicle wiper life detection device, characterized in that: include: a state data acquisition unit, configured to acquire state data of two or more modes of the vehicle wiper rubber strip in a current detection period, wherein the state data is a state parameter related to the life of the wiper rubber strip; An environmental data acquisition unit, configured to acquire environmental parameter characteristics of the vehicle since the wiper rubber strip was last replaced; A computing unit is used to call a configured remaining life prediction model to predict the remaining life of the wiper rubber strip based on the status data and the environmental parameter characteristics. The remaining life prediction model is an end-to-end deep neural network model, which is trained with the status data of the sample wiper rubber strip and its environmental parameter characteristics as input samples and the remaining life of the sample wiper rubber strip as the output label.

14. A vehicle, characterized in that: include: memory and processor; processor; a memory for storing processor-executable instructions; The processor is configured to execute each step of the vehicle wiper life detection method according to any one of claims 1 to 12.

15. A readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, each step of the vehicle wiper life detection method according to any one of claims 1 to 12 is implemented.

16. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, each step of the vehicle wiper life detection method according to any one of claims 1 to 12 is implemented.

Images