Cleaning method and cleaning device for vehicle windshield glass, vehicle and medium

Abstract

The invention provides a vehicle windshield cleaning method and device, a vehicle and a medium. The vehicle windshield cleaning method comprises the steps that current state data, corresponding to a windshield of the vehicle, collected by a sensor installed at the designated position of the vehicle is obtained; pollutant data corresponding to the windshield glass are determined based on the current state data; calculating a cleaning difficulty score corresponding to the windshield by using the pollutant data, the current environment temperature, at least one historical cleaning failure rate and a cleaning difficulty evaluation rule; and based on the cleaning difficulty score and the pollutant coverage rate corresponding to each preset area, determining a target cleaning strategy of the windshield by using a cleaning strategy determination rule, and controlling a corresponding cleaning execution module to execute a corresponding cleaning action according to the target cleaning strategy. According to the method and the device, the recognition precision and the removal capability of the high-adhesion stubborn pollutants are remarkably improved.

Description

A method, device, vehicle, and medium for cleaning a vehicle windshield. Technical Field

[0001] This application relates to the field of vehicle cleaning technology, and in particular to a method, device, vehicle, and medium for cleaning a vehicle windshield. Background Technology

[0002] With the continuous development of the automotive industry and the increasing demand for driving comfort, windshield washer systems have become a standard feature in modern cars. Traditional windshield washer systems combine washer fluid spraying with mechanical swiping to provide drivers with clear visibility. This system is primarily designed for lightly soiled conditions encountered while driving: when the driver activates the washer switch, the washer pump immediately starts and sprays washer fluid onto the glass surface. Subsequently, the wiper motor drives the rubber blades in a reciprocating motion, simultaneously removing the washer fluid and dirt. Its cleaning efficiency mainly relies on the flushing action of the washer fluid and the physical swiping function of the wiper blades.

[0003] However, when vehicles are parked for extended periods, various dried and solidified contaminants often accumulate on the windshield surface. These contaminants have a high bonding strength with the glass surface, making it difficult for traditional washing systems, which rely on immediate rinsing and mechanical scraping, to effectively remove them, resulting in limited cleaning effectiveness. Summary of the Invention

[0004] In view of this, the purpose of this application is to provide a method, apparatus, vehicle, and medium for cleaning a vehicle windshield. By acquiring current state data collaboratively collected by optical and image sensors, the method accurately identifies the type of pollutant and the pollution coverage of each preset area, dynamically calculates a cleaning difficulty score, and then determines a cleaning strategy accordingly. This eliminates reliance on the driver's subjective judgment or a fixed preset procedure, instead closely matching the actual pollution state, significantly improving the accuracy of identifying and removing highly adhesive and stubborn pollutants.

[0005] In a first aspect, embodiments of this application provide a method for cleaning a vehicle windshield. The cleaning method includes: acquiring current state data of the windshield of the vehicle collected by a sensor installed at a designated location on the vehicle; wherein the sensor includes an optical sensor and an image sensor, and the current state data includes optical data and image data; determining contaminant data corresponding to the windshield based on the current state data; wherein the contaminant data includes at least one contaminant type corresponding to contaminants on the windshield, and a contaminant coverage rate corresponding to each preset area of ​​the windshield; acquiring the current ambient temperature and at least one historical cleaning failure rate corresponding to at least one contaminant type, and calculating a cleaning difficulty score corresponding to the windshield using the contaminant data, the current ambient temperature, the at least one historical cleaning failure rate, and a cleaning difficulty assessment rule; determining a target cleaning strategy for the windshield based on the cleaning difficulty score and the contaminant coverage rate corresponding to each preset area using a cleaning strategy determination rule, and controlling a corresponding cleaning execution module to perform a corresponding cleaning action according to the target cleaning strategy.

[0006] Furthermore, determining the pollutant data corresponding to the windshield based on the current state data includes: inputting the image data into a pre-trained pollutant type detection model to determine at least one pollutant type; for each preset area, statistically analyzing the effective photosensitive area covered by all sensing points marked as polluted units within the preset area based on the optical data, and determining the pollution coverage rate of the preset area based on the effective photosensitive area and the area corresponding to the preset area.

[0007] Furthermore, the step of calculating the cleaning difficulty score corresponding to the windshield using the pollutant data, the current ambient temperature, the at least one historical cleaning failure rate, and the cleaning difficulty assessment rules includes: determining a target stubbornness index from the stubbornness index corresponding to at least one pollutant type, and determining a pollutant stubbornness score based on the target stubbornness index and a first weight; determining a pollutant coverage index based on the pollution coverage rate of each preset area and the area weight corresponding to each preset area, and determining a pollutant coverage score based on the pollutant coverage index and a second weight; determining a temperature index based on the current ambient temperature, and determining a temperature score based on the temperature index and a third weight; determining a target failure rate from the at least one historical cleaning failure rate, and determining a failure rate score based on the target failure rate and a fourth weight; and determining the cleaning difficulty score as the sum of the pollutant stubbornness score, the pollutant coverage score, the temperature score, and the failure rate score.

[0008] Furthermore, the target cleaning strategy includes a target spray pattern for the target cleaning fluid, an immersion time for the target cleaning fluid, and a target wiper pattern. The step of determining the target cleaning strategy for the windshield based on the cleaning difficulty score and the contaminant coverage rate corresponding to each preset area, using cleaning strategy determination rules, includes: determining the total coverage rate based on the contaminant coverage rate corresponding to each preset area; when the total coverage rate is greater than or equal to a coverage rate threshold, using a global uniform spray pattern as the target spray pattern; when the total coverage rate is less than the coverage rate threshold, using a local precision spray pattern as the target spray pattern; calculating the immersion time based on the base immersion time, amplification factor, and the cleaning difficulty score; when the cleaning difficulty score is greater than or equal to a score threshold, using a powerful wiper pattern as the target wiper pattern; when the cleaning difficulty score is less than the score threshold, using a standard wiper pattern as the target wiper pattern; and calculating the number of wiper strokes for the target wiper pattern using the formula for calculating the number of strokes corresponding to the target wiper pattern and the cleaning difficulty score.

[0009] Furthermore, the target cleaning fluid is determined through the following steps: determining a target stubbornness index from the stubbornness index corresponding to at least one type of contaminant; calculating a score corresponding to each preset cleaning fluid based on the target stubbornness index value; determining the selection probability corresponding to each preset cleaning fluid using the score corresponding to each preset cleaning fluid; and selecting the preset cleaning fluid with the highest probability from among multiple preset cleaning fluids as the target cleaning fluid.

[0010] Furthermore, before the corresponding cleaning action is executed by the corresponding cleaning execution module according to the target cleaning strategy, the cleaning method further includes: when the current ambient temperature is detected to be lower than the temperature threshold and the target stubbornness index is greater than the preset index threshold, generating a heating command and sending the heating command to the cleaning fluid heater to control the cleaning fluid heater to heat the target cleaning fluid to the target temperature.

[0011] Furthermore, after the cleaning execution module completes its execution, the cleaning method further includes: acquiring the post-cleaning state data of the windshield collected by the sensor; when it is determined based on the post-cleaning state data that there are still residual pollutants on the windshield, updating the target cleaning strategy, and controlling the cleaning execution module to perform the corresponding cleaning action according to the updated target cleaning strategy until there are no residual pollutants on the windshield; and dynamically updating the cleaning difficulty assessment rule and the cleaning strategy determination rule using an incremental learning algorithm based on the current state data and the updated target cleaning strategy.

[0012] Secondly, embodiments of this application also provide a vehicle windshield cleaning device, the cleaning device comprising: a data acquisition module, configured to acquire current state data of the windshield of the vehicle collected by sensors installed at a designated location on the vehicle; wherein the sensors include optical sensors and image sensors, and the current state data includes optical data and image data; a contaminant data determination module, configured to determine contaminant data corresponding to the windshield based on the current state data; wherein the contaminant data includes at least one contaminant type corresponding to contaminants on the windshield, and a contaminant coverage rate corresponding to each preset area of ​​the windshield; a cleaning difficulty score calculation module, configured to acquire the current ambient temperature and at least one historical cleaning failure rate corresponding to at least one contaminant type, and calculate the cleaning difficulty score corresponding to the windshield using the contaminant data, the current ambient temperature, the at least one historical cleaning failure rate, and a cleaning difficulty assessment rule; and a cleaning strategy determination module, configured to determine a target cleaning strategy for the windshield based on the cleaning difficulty score and the contaminant coverage rate corresponding to each preset area, using a cleaning strategy determination rule, and control a corresponding cleaning execution module to perform a corresponding cleaning action according to the target cleaning strategy.

[0013] Thirdly, embodiments of this application also provide a vehicle, including: a processor, a memory, and a bus. The memory stores machine-readable instructions executable by the processor. When the vehicle is running, the processor communicates with the memory via the bus. When the machine-readable instructions are executed by the processor, the steps of the vehicle windshield cleaning method described above are performed.

[0014] Fourthly, embodiments of this application also provide a computer-readable storage medium storing a computer program that, when executed by a processor, performs the steps of the vehicle windshield cleaning method described above.

[0015] This application provides a method, apparatus, vehicle, and medium for cleaning a vehicle windshield. First, current state data of the windshield is acquired from sensors installed at designated locations on the vehicle. These sensors include optical sensors and image sensors, and the current state data includes both optical and image data. Then, contaminant data corresponding to the windshield is determined based on the current state data. This contaminant data includes at least one contaminant type on the windshield and the contaminant coverage rate for each preset area of ​​the windshield. A cleaning difficulty score for the windshield is calculated using the contaminant data, the current ambient temperature, at least one historical cleaning failure rate, and a cleaning difficulty assessment rule. Finally, based on the cleaning difficulty score and the contaminant coverage rate for each preset area, a target cleaning strategy for the windshield is determined using a cleaning strategy determination rule, and the corresponding cleaning execution module is controlled to perform the corresponding cleaning action according to the target cleaning strategy.

[0016] This application accurately identifies pollutant types and pollution coverage rates in preset areas by acquiring current state data through the collaborative collection of optical and image sensors, and dynamically calculates a cleaning difficulty score to determine the cleaning strategy accordingly. This eliminates reliance on driver judgment or fixed preset procedures for cleaning actions, instead closely matching the actual pollution state, significantly improving the accuracy of identifying and removing highly adhesive and stubborn pollutants.

[0017] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0018] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 is a flowchart of a vehicle windshield cleaning method provided in an embodiment of this application; Figure 2 is a schematic diagram of a windshield preset area division effect provided in an embodiment of this application; Figure 3 is a structural schematic diagram of a vehicle windshield cleaning device provided in an embodiment of this application; Figure 4 is a structural schematic diagram of a vehicle provided in an embodiment of this application. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely represents selected embodiments of this application. Based on the embodiments of this application, every other embodiment obtained by those skilled in the art without inventive effort falls within the scope of protection of this application.

[0021] First, the applicable scenarios for this application will be introduced. This application can be applied to the field of vehicle cleaning technology.

[0022] With the continuous development of the automotive industry and the increasing demand for driving comfort, windshield washer systems have become a standard feature in modern cars. Traditional windshield washer systems combine washer fluid spraying with mechanical swiping to provide drivers with clear visibility. This system is primarily designed for lightly soiled conditions encountered while driving: when the driver activates the washer switch, the washer pump immediately starts and sprays washer fluid onto the glass surface. Subsequently, the wiper motor drives the rubber blades in a reciprocating motion, simultaneously removing the washer fluid and dirt. Its cleaning efficiency mainly relies on the flushing action of the washer fluid and the physical swiping function of the wiper blades.

[0023] However, research has found that when vehicles are parked for extended periods, various dried and solidified contaminants tend to accumulate on the windshield surface. These contaminants have a high bonding strength with the glass surface, making it difficult for traditional washing systems, which rely on immediate rinsing and mechanical scraping, to effectively remove them, resulting in limited cleaning effectiveness.

[0024] Based on this, the present application provides a method for cleaning a vehicle windshield, which makes the cleaning action no longer dependent on the driver's subjective judgment or a fixed preset process, but closely matches the actual pollution state, significantly improving the identification accuracy and removal ability of highly adhesive and stubborn pollutants.

[0025] Please refer to Figure 1, which is a flowchart of a vehicle windshield cleaning method provided in an embodiment of this application. As shown in Figure 1, the cleaning method provided in this embodiment includes: S101, acquiring current state data of the vehicle's windshield collected by a sensor installed at a designated location on the vehicle.

[0026] Here, according to the embodiments provided in this application, the sensor includes an optical sensor and an image sensor. The current state data includes optical data collected by the optical sensor and image data collected by the image sensor. Specifically, the optical sensor and the image sensor can be installed on the back of the rearview mirror inside the vehicle, covering the main viewing area of ​​the windshield, and can also be arranged on both sides of the A-pillar of the vehicle, covering the edge area of ​​the glass, to ensure that there are no blind spots in the perception of the entire windshield.

[0027] In specific implementation of step S101, optical data and image data of the windshield are acquired by optical sensors and image sensors installed at designated locations on the vehicle.

[0028] S102, Based on the current state data, determine the pollutant data corresponding to the windshield.

[0029] Here, the pollutant data includes at least one pollutant type corresponding to the pollutants on the windshield, and the pollutant coverage rate corresponding to each preset area of ​​the windshield.

[0030] Regarding step S102 above, in specific implementation, the pollutant data corresponding to the windshield is determined based on the current state data obtained in step S102.

[0031] Please refer to Figure 2, which is a schematic diagram of the windshield preset area division effect provided in an embodiment of this application. As shown in Figure 2, the windshield is regarded as a two-dimensional plane, and a coordinate system is established with the projection point v of the driver's eye on the windshield as the origin, with the x-axis horizontal to the right and the y-axis vertically upward. The eye position varies depending on the vehicle design and driving position (such as left-hand drive or right-hand drive). According to the importance of the driver's field of vision, the windshield is divided into three preset areas, including the main cleaning focus area, the secondary cleaning focus area, and the peripheral cleaning focus area. The main cleaning focus area includes a circle with v as the center and a radius of... The circular area and the triangular area of ​​column A (i.e., the red area in Figure 2) are the areas of concern for secondary cleaning. The secondary cleaning concern area is a circle with V as the center and an inner radius of V. Outer radius The circular area is defined as the area in question (yellow in Figure 2), and the surrounding cleaning focus area is the remaining portion of the windshield (blue in Figure 2). A weighting factor is assigned to each area, with the weight of the main cleaning focus area being [value missing]. The weight of the secondary cleaner's focus area is The weight of the surrounding cleanliness focus area is , + + = 1, > > .

[0032] As an optional embodiment, according to the above step S102, determining the pollutant data corresponding to the windshield based on the current state data includes: Step 1021, inputting the image data into a pre-trained pollutant type detection model to determine at least one pollutant type.

[0033] Here, the pollutant type detection model is a model trained using sample pollutant images and pollutant type labels, and is used to detect the types of pollutants present in the image data.

[0034] For the above step 1021, in specific implementation, the image data is input into a pre-trained pollutant type detection model, and image processing algorithms are used to generate fingerprint information based on the optical characteristics (such as color, texture, shape, etc.) of different pollutants under natural light. For example, bird droppings are usually irregular, protruding, white or milky white lumps; tree sap is a spattered, semi-transparent viscous spot; oil film is a uniformly covered, rainbow-like interference fringe; dust is a uniform, dull covering layer; mud spots are irregular solidified spots with a granular roughness to determine at least one pollutant type.

[0035] Step 1022, for each preset area, based on the optical data, statistically calculate the effective photosensitive area covered by all sensing points marked as contaminated units within the preset area, and determine the pollution coverage rate of the preset area based on the effective photosensitive area and the area of the corresponding preset area.

[0036] For the above step 1022, in specific implementation, an independent two-dimensional coordinate mapping relationship is established for each preset area to ensure that the optical data (such as light transmittance, reflectance, brightness value, light intensity attenuation coefficient, etc.) collected by the optical sensor can be accurately associated with the spatial position of the corresponding pixel or sensing unit. Using calibration parameters (including sensor installation pose, field of view angle, distortion model, and glass surface geometry compensation parameters), the original optical measurement values (such as the relative light transmittance T∈[0,1] of each sensing point) are mapped to the corresponding physical area on the windshield surface, and threshold discrimination is performed on the optical data of all effective sensing points within each preset area: set a pollution determination optical threshold T0 (for example, a light transmittance lower than 0.75 is considered to have occlusive pollution), and mark the sensing points that satisfy T<T0 as contaminated units. Then, statistically calculate the effective photosensitive area covered by all sensing points marked as contaminated units within the preset area. The effective photosensitive area is calculated by weighted summation, and the actual glass surface area corresponding to each sensing point is determined by its spatial position, normal angle, and sensor resolution, and is corrected by surface unfolding. Finally, calculate the pollution coverage rate of the preset area, obtain the area of the preset area, and determine the pollution coverage rate of the preset area by the ratio between the effective photosensitive area and the area.

[0037] S103, obtain the current ambient temperature and at least one historical cleaning failure rate corresponding to at least one type of pollutant, and calculate the cleaning difficulty score corresponding to the windshield using the pollutant data, the current ambient temperature, the at least one historical cleaning failure rate and the cleaning difficulty assessment rules.

[0038] Regarding step S103 above, in practical implementation, after determining at least one type of pollutant and the pollution coverage rate of each preset area, it is also necessary to obtain the current ambient temperature and the historical cleaning failure rate corresponding to each type of pollutant. Here, as an example, the historical cleaning failure rate can be statistically analyzed using real vehicle data. Based on vehicles where this cleaning method has been deployed, cleaning task execution data and cleaning effect feedback data are continuously collected in real usage scenarios, and the percentage of cleaning that did not achieve the expected cleaning effect for each type of pollutant is calculated as the historical cleaning failure rate. Then, based on the pollutant data, the current ambient temperature, and at least one historical cleaning failure rate, the cleaning difficulty score corresponding to the windshield is calculated using pre-built cleaning difficulty assessment rules.

[0039] As an optional embodiment, for step S103 above, the step of calculating the cleaning difficulty score corresponding to the windshield using the pollutant data, the current ambient temperature, the at least one historical cleaning failure rate and the cleaning difficulty assessment rule includes: step 1031, determining a target stubbornness index from the stubbornness index corresponding to at least one pollutant type, and determining the pollutant stubbornness score based on the target stubbornness index and the first weight.

[0040] For each different type of pollutant There is a corresponding stubbornness index. For example, when the type of pollutant When it is bird droppings, the corresponding stubbornness index It is 0.9, when the pollutant type When it is a resin, the corresponding stubbornness index It is 0.8, when the pollutant type When it is an oil film, the corresponding stubbornness index It is 0.6, when the pollutant type When the substance is dust or mud, the corresponding stubbornness index is... It is 0.3.

[0041] Regarding step 1031 above, in specific implementation, the persistence index corresponding to the pollutant type is first determined. If only one pollutant type exists, the persistence index corresponding to that pollutant type is determined as the target persistence index; if multiple pollutant types exist, the largest persistence index among the multiple persistence indices is taken as the target persistence index. Then, based on the target persistence index and a first weight, a pollutant persistence score is determined. The pollutant persistence score represents the cleaning difficulty caused by the chemical and physical properties of the pollutant itself. First weight The value can be preset to 0.4; this application does not specifically limit this setting. The contaminant persistence score is... , The larger the value, the more stubborn the pollutant.

[0042] Step 1032: Determine the pollutant coverage index based on the pollution coverage rate of each preset area and the area weight corresponding to each preset area, and determine the pollutant coverage score based on the pollutant coverage index and the second weight.

[0043] Regarding step 1032 above, in specific implementation, firstly, the pollutant coverage rate index is determined based on the pollution coverage rate of each preset area and the corresponding regional weight of each preset area. Here, continuing the example in Figure 2, when there are three preset areas, Pollution coverage rate and pollutant coverage index for each preset area Calculated using the following formula:

[0044] The pollutant coverage score indicates the degree to which the location of pollutants affects driving safety. (Second weight) The value can be preset to 0.3; this application does not specify a particular value for it. The pollutant coverage score is... .

[0045] Step 1033: Determine the temperature index based on the current ambient temperature, and determine the temperature score based on the temperature index and the third weight.

[0046] Regarding step 1033 above, in specific implementation, the current ambient temperature... Based on the current ambient temperature Determine the temperature index Specifically, when If the temperature is <0℃, then freezing conditions are met, and T=1.5; if 0℃≤ If the temperature is below 5℃, then the low-temperature condition is reached, T=1.2; when If the temperature is ≥5℃, then normal conditions are met, and T=1. Third weight. The value can be preset to 0.2; this application does not specifically limit this setting. The temperature score represents the negative impact of the current ambient temperature on the cleaning effect. The temperature score is... .

[0047] Step 1034: Determine the target failure rate from the at least one historical cleaning failure rate, and determine the failure rate score based on the target failure rate and the fourth weight.

[0048] Regarding step 1034 above, in specific implementation, if only one type of contaminant exists, the target failure rate is determined by the historical cleaning failure rate corresponding to that contaminant type. If multiple types of contaminants exist, the highest historical cleaning failure rate among the multiple historical cleaning failure rates will be used as the target failure rate. Fourth weight The value can be preset to 0.1; this application does not specifically limit this setting. The failure rate score represents the expected cleaning difficulty, and the temperature score is... .

[0049] Step 1035: The sum of the contaminant stubbornness score, the contaminant coverage score, the temperature score, and the failure rate score is determined as the cleaning difficulty score.

[0050] Regarding step 1035 above, in specific implementation, the sum of the contaminant stubbornness score, contaminant coverage score, temperature score, and failure rate score is determined as the cleaning difficulty score. .Right now, .

[0051] S104. Based on the cleaning difficulty score and the pollutant coverage rate corresponding to each preset area, the target cleaning strategy for the windshield is determined using the cleaning strategy determination rules, and the corresponding cleaning execution module is controlled to perform the corresponding cleaning action according to the target cleaning strategy.

[0052] Regarding step S104 above, in specific implementation, based on the cleaning difficulty score determined in the above steps and the pollutant coverage rate corresponding to each preset area, the target cleaning strategy for the windshield is determined using the cleaning strategy determination rules, and the corresponding cleaning execution module is controlled to perform the corresponding cleaning action according to the target cleaning strategy.

[0053] Here, according to the embodiments provided in this application, the target cleaning strategy includes a target spray pattern of the target cleaning fluid, a target immersion time of the target cleaning fluid, and a target wiper pattern. The immersion time of the target cleaning fluid refers to the duration during which the cleaning fluid remains on a designated area of ​​the windshield surface and undergoes physical / chemical action from the time the cleaning fluid completes all spraying actions under the target spray pattern until the wiper module starts wiping.

[0054] Furthermore, the cleaning execution module includes a cleaning fluid spraying module and a wiper module. The cleaning fluid spraying module supports both global uniform spraying and localized precise enhanced spraying of the cleaning fluid, and controls the spray pipeline via a LIN bus to achieve gradient and intermittent spraying; the wiper module supports activating the wipers after the cleaning fluid has soaked the area to remove contaminants.

[0055] Specifically, regarding step S104 above, the step of determining the target cleaning strategy for the windshield based on the cleaning difficulty score and the pollutant coverage rate corresponding to each preset area using cleaning strategy determination rules includes: step 1041, determining the total coverage rate based on the pollutant coverage rate corresponding to each preset area; when the total coverage rate is greater than or equal to the coverage rate threshold, using the global uniform spraying mode as the target spraying mode; when the total coverage rate is less than the coverage threshold, using the local precision spraying mode as the target spraying mode.

[0056] Regarding step 1041 above, in specific implementation, the total coverage rate is determined based on the pollutant coverage rate corresponding to each preset area. When the total coverage rate is greater than or equal to the coverage rate threshold, the global uniform spraying mode is used as the target spraying mode; when the total coverage rate is less than the coverage rate threshold, the local precision spraying mode is used as the target spraying mode. Here, when performing the local precision spraying mode, optical data (such as transmittance attenuation gradient) is first used to identify high-pollution areas with abnormal light transmission on the windshield, and their spatial centroid is calculated as the spraying target point; or, image data is used through a pollutant type detection model to extract the center of its bounding box as the spraying target point. If the distance between the above two types of target points is less than a preset threshold (such as 15 mm), a weighted fusion is performed to generate the final spraying position. Subsequently, when controlling the cleaning fluid spraying module to perform local spraying, the cleaning fluid spraying module only performs fixed-point spraying of cleaning fluid at the final spraying position.

[0057] Step 1042: Calculate the soaking time based on the base soaking time, the amplification factor, and the cleaning difficulty score.

[0058] Regarding step 1042 above, in specific implementation, the immersion time is calculated based on the base immersion time, the amplification factor, and the cleaning difficulty score. Specifically, the immersion time is calculated using the following formula:

[0059] in, Indicates the duration of immersion. Indicates the basic immersion time, 5 seconds ≤ ≤60 seconds Indicates the magnification factor (cleaning difficulty score) The higher the value, the greater the magnification factor.

[0060] Step 1043: When the cleaning difficulty score is greater than or equal to the score threshold, the powerful wiper mode is taken as the target wiper mode; when the cleaning difficulty score is less than the score threshold, the standard wiper mode is taken as the target wiper mode. The number of wiper cycles for the target wiper mode is calculated using the formula for calculating the number of wiper cycles corresponding to the target wiper mode and the cleaning difficulty score.

[0061] Regarding step 1043 above, in specific implementation, the cleaning difficulty score is compared with a scoring threshold. For example, the scoring threshold can be set to 0.6, which is not specifically limited in this application. When the cleaning difficulty score is greater than or equal to the scoring threshold, the powerful wiper mode is used as the target wiper mode; when the cleaning difficulty score is less than the scoring threshold, the standard wiper mode is used as the target wiper mode. Then, the number of wiper cycles for the target wiper mode is calculated using the formula for calculating the number of wiper cycles corresponding to the target wiper mode and the cleaning difficulty score. Specifically, when the target wiper mode is the powerful wiper mode, the number of wiper cycles is... When the target wiper mode is standard wiper mode, the number of wiper cycles is... .

[0062] As an optional embodiment, the target cleaning fluid is determined by the following steps: A: Determine the target stubbornness index from the stubbornness index corresponding to at least one type of contaminant, and calculate the score corresponding to each preset cleaning fluid based on the target stubbornness value.

[0063] Regarding step A above, in practical implementation, the method for determining the target stubbornness index is the same as that in the previous steps, and it achieves the same technical effect, so it will not be repeated here. Then, based on the target stubbornness value, a score is calculated for each preset cleaning fluid. As an example, the preset cleaning fluids include general-purpose, oil film decomposition, and high-solubility types. General-purpose formulation scoring function. = 2 The higher the stubbornness, the less suitable it is for general-purpose formulations; therefore, the coefficient is represented by -2 to indicate a negative correlation. (Score function for oil film decomposition type formulations) = A score of 0.5 indicates a stubbornness greater than 0.5, which is positive and suitable for treating moderately stubborn oil film contaminants. This is a scoring function for high-solubility formulations. = 3 1. The higher the stubbornness, the stronger the dissolving power is required, so the coefficient is 3; to avoid the possibility that even with slight contamination, the high-solubility type will receive a positive score and may be mistakenly selected, so it is -1.

[0064] B: Determine the selection probability of each preset cleaning solution by using the score corresponding to each preset cleaning solution, and select the preset cleaning solution with the highest probability from among the multiple preset cleaning solutions as the target cleaning solution.

[0065] Regarding step B above, in practical implementation, firstly, the selection probability of each preset cleaning solution is determined using the score corresponding to each preset cleaning solution. Then, the preset cleaning solution with the highest probability is selected as the target cleaning solution from among the various preset cleaning solutions. Specifically, the selection probability is calculated using the following formula:

[0066] in, Indicates selecting the first One recipe The probability of.

[0067] As an optional embodiment, before the corresponding cleaning execution module is controlled to perform the corresponding cleaning action according to the target cleaning strategy, the cleaning method provided in this application further includes: when the current ambient temperature is detected to be lower than the temperature threshold and the target stubbornness index is greater than the preset index threshold, generating a heating command and sending the heating command to the cleaning fluid heater to control the cleaning fluid heater to heat the target cleaning fluid to the target temperature.

[0068] Here, the temperature threshold can be set to 10℃, and the preset index threshold can be set to 0.5. This application does not make specific limitations on these.

[0069] Regarding the above steps, in specific implementation, before the corresponding cleaning execution module executes the corresponding cleaning action according to the target cleaning strategy, when the current ambient temperature is detected to be lower than the temperature threshold and the target stubbornness index is detected to be greater than the preset index threshold, a heating command is generated and sent to the cleaning fluid heater to control the cleaning fluid heater to heat the target cleaning fluid to the target temperature. For example, the target temperature can be 40°C, and this application does not specifically limit it.

[0070] As an optional embodiment, after the cleaning execution module completes its execution, the cleaning method provided in this application further includes: I: acquiring the post-cleaning status data of the windshield collected by the sensor.

[0071] Regarding step I above, in specific implementation, after the cleaning execution module completes all actions of the target cleaning strategy (including spraying, wetting, and wiper wiping), a delayed acquisition mechanism is triggered: wait for the wipers to stop and remain still for 1.5–2.0 seconds, and then simultaneously start the optical sensor and image sensor to perform a new round of data acquisition; among them, the optical sensor acquires the real-time light transmittance distribution data of each preset area of ​​the windshield after cleaning, and the image sensor acquires the image data of the windshield after cleaning as the post-cleaning status data.

[0072] II: When it is determined based on the post-cleaning status data that there are still residual pollutants on the windshield, the target cleaning strategy is updated, and the cleaning execution module is controlled to perform the corresponding cleaning action according to the updated target cleaning strategy until there are no residual pollutants on the windshield.

[0073] Regarding step II above, in practice, the post-cleaning status data is used to determine whether residual contaminants still exist on the windshield. Specifically, based on optical data, the post-cleaning contaminant coverage rate of each preset area can be calculated. If the post-cleaning contaminant coverage rate of any area is greater than a first threshold, or the post-cleaning contaminant coverage rate of the main cleaning focus area is greater than a second threshold (the second threshold is less than the first threshold), then it is determined that residues exist. Based on image data, the post-cleaning image is input into a detection model for the same contaminant type. If contaminant instances are still detected, then it is determined that visually visible residues exist.

[0074] A strategy update is triggered when any condition is met. For example, a penalty coefficient γ (γ = 0.3 × number of residual areas) is added to the original cleaning difficulty score, and the cleaning difficulty score is recalculated; or, the spray mode in the original target cleaning strategy is forcibly upgraded to local precision spray, the immersion time is increased by 20%, the wiper mode is forcibly switched to strong mode and an additional brush stroke is added, etc. The updated strategy is immediately issued and implemented until there are no residual pollutants on the windshield.

[0075] III: Based on the current state data and the updated target cleaning strategy, the cleaning difficulty assessment rule and the cleaning strategy determination rule are dynamically updated using an incremental learning algorithm.

[0076] Regarding step III above, in specific implementation, the complete closed-loop data of this cleaning task (including initial state data, original strategy, update strategy, sensor feedback at each stage, and the final cleaning effect label "qualified / unqualified") is used as a training sample and stored in the learning cache. When the number of cached samples reaches a threshold N or the cumulative runtime reaches a preset duration, lightweight online learning can be triggered. As an example, an L2-regularized linear weighted regression model can be used, with the goal of minimizing the prediction error of the cleaning difficulty score, dynamically adjusting the weights (first to fourth weights) in the cleaning difficulty assessment rule; or, a decision tree pruning optimization algorithm can be used to perform small-step gradient corrections on the key thresholds (coverage threshold, score threshold, temperature threshold) in the cleaning strategy determination rule. In this way, after each cleaning, the cleaning effect is fed back by sensors. If the cleaning is insufficient, the strategy is updated and cleaning is performed again. The system automatically learns and optimizes the next cleaning strategy to achieve adaptive cleaning.

[0077] The vehicle windshield cleaning method provided in this application embodiment first acquires the current state data of the vehicle's windshield collected by sensors installed at designated locations on the vehicle; wherein the sensors include optical sensors and image sensors, and the current state data includes optical data and image data; then, based on the current state data, the contaminant data corresponding to the windshield is determined; wherein the contaminant data includes at least one contaminant type corresponding to the contaminants on the windshield, and the contaminant coverage rate corresponding to each preset area of ​​the windshield; the cleaning difficulty score corresponding to the windshield is calculated using the contaminant data, the current ambient temperature, at least one historical cleaning failure rate, and cleaning difficulty assessment rules; finally, based on the cleaning difficulty score and the contaminant coverage rate corresponding to each preset area, a target cleaning strategy for the windshield is determined using cleaning strategy determination rules, and the corresponding cleaning execution module is controlled to perform the corresponding cleaning action according to the target cleaning strategy.

[0078] This application accurately identifies pollutant types and pollution coverage rates in preset areas by acquiring current state data through the collaborative collection of optical and image sensors, and dynamically calculates a cleaning difficulty score to determine the cleaning strategy accordingly. This eliminates reliance on driver judgment or fixed preset procedures for cleaning actions, instead closely matching the actual pollution state, significantly improving the accuracy of identifying and removing highly adhesive and stubborn pollutants.

[0079] Please refer to Figure 3, which is a structural schematic diagram of a vehicle windshield cleaning device provided in an embodiment of this application. As shown in Figure 3, the cleaning device 300 includes: a data acquisition module 301, used to acquire current state data of the windshield of the vehicle collected by sensors installed at a designated location on the vehicle; wherein the sensors include optical sensors and image sensors, and the current state data includes optical data and image data; a contaminant data determination module 302, used to determine contaminant data corresponding to the windshield based on the current state data; wherein the contaminant data includes at least one contaminant type corresponding to the contaminants on the windshield, and the contaminant coverage rate corresponding to each preset area of ​​the windshield; a cleaning difficulty score calculation module 303, used to acquire the current ambient temperature and at least one historical cleaning failure rate corresponding to at least one contaminant type, and calculate the cleaning difficulty score corresponding to the windshield using the contaminant data, the current ambient temperature, the at least one historical cleaning failure rate, and cleaning difficulty assessment rules; and a cleaning strategy determination module 304, used to determine the target cleaning strategy for the windshield based on the cleaning difficulty score and the contaminant coverage rate corresponding to each preset area, using cleaning strategy determination rules, and control the corresponding cleaning execution module to perform the corresponding cleaning action according to the target cleaning strategy.

[0080] Furthermore, when the pollutant data determination module 302 determines the pollutant data corresponding to the windshield based on the current state data, the pollutant data determination module 302 is also used to: input the image data into a pre-trained pollutant type detection model to determine at least one pollutant type; for each preset area, based on the optical data, calculate the effective photosensitive area covered by all sensing points marked as polluted units within the preset area, and determine the pollution coverage rate of the preset area based on the effective photosensitive area and the area corresponding to the preset area.

[0081] Furthermore, when the cleaning difficulty score calculation module 303 calculates the cleaning difficulty score corresponding to the windshield using the pollutant data, the current ambient temperature, the at least one historical cleaning failure rate, and the cleaning difficulty assessment rules, the cleaning difficulty score calculation module 303 is also used to: determine a target stubbornness index from the stubbornness index corresponding to at least one pollutant type, and determine a pollutant stubbornness score based on the target stubbornness index and a first weight; determine a pollutant coverage index based on the pollution coverage rate of each preset area and the area weight corresponding to each preset area, and determine a pollutant coverage score based on the pollutant coverage index and a second weight; determine a temperature index based on the current ambient temperature, and determine a temperature score based on the temperature index and a third weight; determine a target failure rate from the at least one historical cleaning failure rate, and determine a failure rate score based on the target failure rate and a fourth weight; and determine the cleaning difficulty score as the sum of the pollutant stubbornness score, the pollutant coverage score, the temperature score, and the failure rate score.

[0082] Furthermore, the target cleaning strategy includes a target spray pattern for the target cleaning fluid, an immersion time for the target cleaning fluid, and a target wiper pattern. When the cleaning strategy determination module 304 determines the target cleaning strategy for the windshield based on the cleaning difficulty score and the contaminant coverage rate corresponding to each preset area using cleaning strategy determination rules, the cleaning strategy determination module 304 is also used to: determine the total coverage rate based on the contaminant coverage rate corresponding to each preset area; when the total coverage rate is greater than or equal to a coverage rate threshold, use the globally uniform spray pattern as the target cleaning strategy. In the standard spray mode, when the total coverage is less than the coverage threshold, the local precision spray mode is used as the target spray mode; the immersion time is calculated based on the basic immersion time, the magnification factor, and the cleaning difficulty score; when the cleaning difficulty score is greater than or equal to the score threshold, the powerful wiper mode is used as the target wiper mode; when the cleaning difficulty score is less than the score threshold, the standard wiper mode is used as the target wiper mode, and the number of wiper strokes for the target wiper mode is calculated using the formula for calculating the number of wiper strokes for the target wiper mode and the cleaning difficulty score.

[0083] Furthermore, the cleaning strategy determination module 304 is also used to determine the target cleaning fluid through the following steps: determining a target stubbornness index from the stubbornness index corresponding to at least one type of contaminant, calculating a score corresponding to each preset cleaning fluid based on the target stubbornness value; determining the selection probability corresponding to each preset cleaning fluid using the score corresponding to each preset cleaning fluid, and selecting the preset cleaning fluid with the highest probability from among multiple preset cleaning fluids as the target cleaning fluid.

[0084] Furthermore, the cleaning device 300 also includes a heating module. Before the corresponding cleaning execution module is controlled to perform the corresponding cleaning action according to the target cleaning strategy, the heating module is used to: generate a heating command and send the heating command to the cleaning fluid heater when the current ambient temperature is detected to be lower than the temperature threshold and the target stubbornness index is greater than the preset index threshold, so as to control the cleaning fluid heater to heat the target cleaning fluid to the target temperature.

[0085] Furthermore, the cleaning device 300 also includes a strategy update module. After the cleaning execution module completes its execution, the strategy update module is used to: acquire the post-cleaning state data of the windshield collected by the sensor; when it is determined based on the post-cleaning state data that there are still residual pollutants on the windshield, update the target cleaning strategy, and control the cleaning execution module to perform the corresponding cleaning action according to the updated target cleaning strategy until there are no residual pollutants on the windshield; and dynamically update the cleaning difficulty assessment rule and the cleaning strategy determination rule using an incremental learning algorithm based on the current state data and the updated target cleaning strategy.

[0086] Please refer to Figure 4, which is a schematic diagram of the structure of a vehicle provided in an embodiment of this application. As shown in Figure 4, the vehicle 400 includes a processor 410, a memory 420, and a bus 430.

[0087] The memory 420 stores machine-readable instructions that can be executed by the processor 410. When the vehicle 400 is running, the processor 410 and the memory 420 communicate via the bus 430. When the machine-readable instructions are executed by the processor 410, the steps of the vehicle windshield cleaning method in the method embodiment shown in Figure 1 above can be performed. For specific implementation details, please refer to the method embodiment, which will not be repeated here.

[0088] This application also provides a computer-readable storage medium storing a computer program. When the computer program is run by a processor, it can execute the steps of the vehicle windshield cleaning method in the method embodiment shown in FIG1 above. For specific implementation details, please refer to the method embodiment, which will not be repeated here.

[0089] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0090] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the shown or discussed mutual couplings, direct couplings, or communication connections may be through some communication interfaces; indirect couplings or communication connections between devices or units may be electrical, mechanical, or other forms.

[0091] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0092] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0093] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a processor-executable, non-volatile, computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0094] Finally, it should be noted that the above-described embodiments are merely specific implementations of this application, used to illustrate the technical solutions of this application, and not to limit them. The scope of protection of this application is not limited thereto. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments, or make equivalent substitutions for some of the technical features, within the scope of the technology disclosed in this application. Such modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be covered within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for cleaning a vehicle windshield, characterized in that, The cleaning method includes: acquiring current state data of the windshield of the vehicle collected by sensors installed at designated locations on the vehicle; wherein the sensors include optical sensors and image sensors, and the current state data includes optical data and image data; determining contaminant data corresponding to the windshield based on the current state data; wherein the contaminant data includes at least one contaminant type corresponding to the contaminants on the windshield, and the contaminant coverage rate corresponding to each preset area of ​​the windshield; acquiring the current ambient temperature and at least one historical cleaning failure rate corresponding to at least one contaminant type, and calculating the cleaning difficulty score corresponding to the windshield using the contaminant data, the current ambient temperature, the at least one historical cleaning failure rate, and cleaning difficulty assessment rules; determining the target cleaning strategy for the windshield based on the cleaning difficulty score and the contaminant coverage rate corresponding to each preset area using cleaning strategy determination rules, and controlling the corresponding cleaning execution module to perform the corresponding cleaning action according to the target cleaning strategy.

2. The cleaning method according to claim 1, characterized in that, The step of determining the pollutant data corresponding to the windshield based on the current state data includes: inputting the image data into a pre-trained pollutant type detection model to determine at least one pollutant type; for each preset area, statistically analyzing the effective photosensitive area covered by all sensing points marked as polluted units within the preset area based on the optical data, and determining the pollution coverage rate of the preset area based on the effective photosensitive area and the area corresponding to the preset area.

3. The cleaning method according to claim 1, characterized in that, The step of calculating the cleaning difficulty score for the windshield using the pollutant data, the current ambient temperature, the at least one historical cleaning failure rate, and the cleaning difficulty assessment rules includes: determining a target stubbornness index from the stubbornness index corresponding to at least one pollutant type, and determining a pollutant stubbornness score based on the target stubbornness index and a first weight; determining a pollutant coverage index based on the pollution coverage rate of each preset area and the area weight corresponding to each preset area, and determining a pollutant coverage score based on the pollutant coverage index and a second weight; determining a temperature index based on the current ambient temperature, and determining a temperature score based on the temperature index and a third weight; determining a target failure rate from the at least one historical cleaning failure rate, and determining a failure rate score based on the target failure rate and a fourth weight; and determining the cleaning difficulty score as the sum of the pollutant stubbornness score, the pollutant coverage rate score, the temperature score, and the failure rate score.

4. The cleaning method according to claim 1, characterized in that, The target cleaning strategy includes a target spray pattern for the target cleaning fluid, an immersion time for the target cleaning fluid, and a target wiper pattern. The step of determining the target cleaning strategy for the windshield based on the cleaning difficulty score and the contaminant coverage rate corresponding to each preset area, using cleaning strategy determination rules, includes: determining the total coverage rate based on the contaminant coverage rate corresponding to each preset area; when the total coverage rate is greater than or equal to a coverage rate threshold, using a global uniform spray pattern as the target spray pattern; when the total coverage rate is less than the coverage rate threshold, using a local precision spray pattern as the target spray pattern; calculating the immersion time based on the base immersion time, amplification factor, and the cleaning difficulty score; when the cleaning difficulty score is greater than or equal to a score threshold, using a powerful wiper pattern as the target wiper pattern; when the cleaning difficulty score is less than the score threshold, using a standard wiper pattern as the target wiper pattern; and calculating the number of wiper strokes for the target wiper pattern using the formula for calculating the number of strokes corresponding to the target wiper pattern and the cleaning difficulty score.

5. The cleaning method according to claim 4, characterized in that, The target cleaning fluid is determined by the following steps: determining the target stubbornness index from the stubbornness index corresponding to at least one type of contaminant, and calculating the score corresponding to each preset cleaning fluid based on the target stubbornness value; The selection probability of each preset cleaning solution is determined by the score corresponding to each preset cleaning solution, and the preset cleaning solution with the highest probability is selected as the target cleaning solution from among the multiple preset cleaning solutions.

6. The cleaning method according to claim 5, characterized in that, Before the corresponding cleaning action is executed by the corresponding cleaning execution module according to the target cleaning strategy, the cleaning method further includes: when the current ambient temperature is detected to be lower than the temperature threshold and the target stubbornness index is greater than the preset index threshold, generating a heating command and sending the heating command to the cleaning fluid heater to control the cleaning fluid heater to heat the target cleaning fluid to the target temperature.

7. The cleaning method according to claim 1, characterized in that, After the cleaning execution module completes its work, the cleaning method further includes: acquiring the post-cleaning state data of the windshield collected by the sensor; when it is determined based on the post-cleaning state data that there are still residual pollutants on the windshield, updating the target cleaning strategy, and controlling the cleaning execution module to perform the corresponding cleaning action according to the updated target cleaning strategy, until there are no residual pollutants on the windshield; and dynamically updating the cleaning difficulty assessment rule and the cleaning strategy determination rule using an incremental learning algorithm based on the current state data and the updated target cleaning strategy.

8. A vehicle windshield cleaning device, characterized in that, The cleaning device includes: a data acquisition module, used to acquire current state data of the windshield of the vehicle collected by sensors installed at a designated location on the vehicle; wherein the sensors include optical sensors and image sensors, and the current state data includes optical data and image data; a contaminant data determination module, used to determine contaminant data corresponding to the windshield based on the current state data; wherein the contaminant data includes at least one contaminant type corresponding to the contaminants on the windshield, and the contaminant coverage rate corresponding to each preset area of ​​the windshield; a cleaning difficulty score calculation module, used to acquire the current ambient temperature and at least one historical cleaning failure rate corresponding to at least one contaminant type, and calculate the cleaning difficulty score corresponding to the windshield using the contaminant data, the current ambient temperature, the at least one historical cleaning failure rate, and cleaning difficulty assessment rules; and a cleaning strategy determination module, used to determine the target cleaning strategy for the windshield based on the cleaning difficulty score and the contaminant coverage rate corresponding to each preset area, using cleaning strategy determination rules, and control the corresponding cleaning execution module to perform the corresponding cleaning action according to the target cleaning strategy.

9. A vehicle, characterized in that, include: The system includes a processor, a memory, and a bus. The memory stores machine-readable instructions executable by the processor. When the vehicle is in operation, the processor communicates with the memory via the bus. The machine-readable instructions are executed by the processor to perform the steps of the vehicle windshield cleaning method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, performs the steps of the vehicle windshield cleaning method as described in any one of claims 1 to 7.

Images