Defogging processing method and device based on fusion model and model training method and device

Abstract

The invention relates to a defogging processing method and device based on a fusion model and a model training method and device, and relates to the technical field of vehicles. The illumination radiation power of a first vehicle window is determined according to illumination intensity and vehicle window structure information of the first vehicle window. And according to a preset first model, the in-vehicle temperature, the out-vehicle temperature and the illumination radiation power of the first vehicle window are processed, and the glass temperature of the first vehicle window is obtained. According to a preset second model, the environment information, the vehicle window information, the glass temperature of the first vehicle window and the pre-collected glass temperature of the second vehicle window are processed, and defogging recommendation information is obtained. In the process, a fusion model of the first model used for predicting the temperature of the vehicle window glass and the second model used for recommending the demisting recommendation information is combined, the demisting recommendation information suitable for different vehicle windows is generated, a processing mode of simply continuing to use a front windshield is replaced, and the demisting effect of each vehicle window in the vehicle is greatly optimized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of vehicles, in particular to a defogging processing method and device based on a fusion model and a model training method. BACKGROUND

[0002] In modern automobile design, the clarity of the vehicle window is crucial for driving safety and passenger comfort. Fogging of the vehicle window is a common problem, especially in environments with high humidity or large temperature differences.

[0003] In related technologies, defogging processing is mainly performed on the front windshield. A temperature sensor is usually arranged at the front windshield to collect the temperature of the front windshield, and then the temperature is combined to predict whether the front windshield is fogging and to perform defogging operation. For other vehicle windows (such as side windows or rear windows), temperature sensors are usually not arranged, and the defogging processing method follows the processing method of the front windshield to perform unified defogging processing.

[0004] However, since the temperature and structure information of these vehicle windows are usually different from those of the front windshield, directly and uniformly following the defogging strategy of the front windshield will result in poor defogging effect of other vehicle windows. SUMMARY

[0005] The purpose of the present application is to provide a defogging processing method and device based on a fusion model to optimize the defogging effect of the vehicle window.

[0006] To achieve the above purpose, the technical solution adopted by the present application is as follows:

[0007] According to a first aspect of the present application, a defogging processing method based on a fusion model is provided, comprising:

[0008] obtaining environment information and vehicle window information of a vehicle; wherein the environment information includes indoor temperature, outdoor temperature and light intensity, the vehicle window information includes glass humidity of the vehicle window and / or vehicle window structure information, and the vehicle window includes a first vehicle window and a second vehicle window;

[0009] determining the light radiation power of the first vehicle window according to the light intensity and the vehicle window structure information of the first vehicle window;

[0010] processing the indoor temperature, the outdoor temperature and the light radiation power of the first vehicle window according to a preset first model to obtain the glass temperature of the first vehicle window, wherein the first model is a model trained based on historical glass temperature data of the second vehicle window and used to predict the glass temperature of the first vehicle window;

[0011] According to the preset second model, the environment information, the window information, the glass temperature of the first window and the pre-acquired glass temperature of the second window are processed to obtain defogging recommendation information; wherein the second model is a model trained based on historical defogging data and used for generating defogging recommendation information.

[0012] In an embodiment, the first window comprises a side window of the vehicle, and the second window comprises a front windshield of the vehicle.

[0013] In an embodiment, the window structure information comprises an installation angle and a surface area of the window glass; and the determining of the light radiation power of the first window according to the light intensity and the window structure information of the first window comprises:

[0014] determining a relative light intensity of the first window according to the installation angle of the first window and the light intensity;

[0015] determining the light radiation power of the first window according to the relative light intensity and the surface area.

[0016] In an embodiment, the method further comprises:

[0017] determining an environmental heat radiation power of the first window according to the temperature inside the vehicle, the temperature outside the vehicle and the surface area;

[0018] the determining of the light radiation power of the first window according to the relative light intensity and the surface area comprises:

[0019] determining an initial light radiation power of the first window according to the relative light intensity and the surface area;

[0020] obtaining the light radiation power according to the environmental heat radiation power and the initial light radiation power.

[0021] In an embodiment, the second model comprises a preprocessing module, a risk prediction module and a recommendation module; and the processing of the environment information, the window information, the glass temperature of the first window and the pre-acquired glass temperature of the second window according to the preset second model to obtain defogging recommendation information comprises:

[0022] processing the glass temperature and the glass humidity of the first window and the glass temperature and the glass humidity of the second window respectively through the preprocessing module to obtain the temperature and humidity changes of the first window and the second window;

[0023] The risk prediction module processes at least one of the glass temperature, the glass humidity, and the temperature and humidity change information of the first window, and at least one of the glass temperature, the glass humidity, and the temperature and humidity change information of the second window, to determine a corresponding fogging level of the first window and the second window respectively.

[0024] The recommendation module generates defogging recommendation information corresponding to the first window and the second window according to the fogging level, the environment information, and the window information.

[0025] In an embodiment, the processing of the glass temperature and the glass humidity of the first window and the glass temperature and the glass humidity of the second window respectively includes:

[0026] The first-order derivative of the glass temperature and the glass humidity of the first window is obtained to obtain the temperature and humidity change information of the first window.

[0027] The first-order derivative of the glass temperature and the glass humidity of the first window is obtained to obtain the temperature and humidity change information of the second window.

[0028] The processing of at least one of the glass temperature, the glass humidity, and the temperature and humidity change information of the first window, and at least one of the glass temperature, the glass humidity, and the temperature and humidity change information of the second window includes:

[0029] According to at least one of the glass temperature, the glass humidity, and the temperature and humidity change information of the first window, the fogging level of the first window is determined; and according to at least one of the glass temperature, the glass humidity, and the temperature and humidity change information of the second window, the fogging level of the second window is determined.

[0030] In an embodiment, the determination of the fogging level of the first window according to the glass temperature and the temperature change information thereof, and the glass humidity and the humidity change information thereof includes:

[0031] According to the temperature and humidity change information of the first window, a corresponding fogging probability change coefficient of the temperature and humidity change information is searched from a preset mapping table to obtain the fogging probability change coefficient of the first window; wherein the mapping table includes fogging probability change coefficients corresponding to different temperature and humidity change information.

[0032] According to at least one of the glass temperature, the glass humidity of the first window, and the fogging probability change coefficient of the first window, the fogging level of the first window is determined.

[0033] In an embodiment, the determining the fogging level of the second window according to the temperature change information and the humidity change information of the second window comprises:

[0034] According to the temperature and humidity change information of the second window, searching a preset mapping table to obtain a fogging probability change coefficient corresponding to the temperature and humidity change information of the second window, and obtaining the fogging probability change coefficient of the second window.

[0035] According to at least one of the glass temperature, the glass humidity of the second window, and the fogging probability change coefficient of the second window, determining the fogging level of the second window.

[0036] In an embodiment, the method further comprises:

[0037] According to the defogging recommendation information, controlling a target actuator to perform a corresponding defogging operation; wherein the defogging recommendation information comprises one of the following recommendation information: opening the first window and / or the second window, adjusting a ventilation setting, and adjusting an air conditioning setting; and the target actuator comprises at least one of the following: a window control system, a ventilation system, and an air conditioning system.

[0038] In an embodiment, the controlling the target actuator to perform the corresponding defogging operation according to the defogging recommendation information comprises:

[0039] Displaying the defogging recommendation information through a center control screen, and / or prompting the defogging recommendation information through an audio device;

[0040] In response to a touch operation on the center control screen, and / or a voice input operation on the audio device, determining defogging setting information, wherein the defogging setting information is the same as or different from the defogging recommendation information.

[0041] According to the defogging setting information, controlling the target actuator to perform a corresponding defogging operation.

[0042] In an embodiment, the method further comprises:

[0043] Monitoring trend information of humidity change and / or trend information of temperature change of the window in real time;

[0044] In response to monitoring that the trend information of humidity change and / or the trend information of temperature change is lower than a preset threshold, controlling the target actuator to stop the current defogging operation.

[0045] In an embodiment, the environmental information further comprises vehicle driving information and / or rainfall information, and the window information further comprises a window position.

[0046] According to a second aspect of the present application, a model training method is provided, comprising:

[0047] collecting first historical sample data, the first historical sample data comprising historical indoor temperature, historical outdoor temperature, historical light intensity and historical glass temperature of the second window in the historical preset time period;

[0048] determining light radiation power of the second window according to the historical light intensity and window structure information of the second window;

[0049] training a first model by taking the historical indoor temperature, the historical outdoor temperature, the historical light radiation power of the second window as input and the historical glass temperature of the second window as output.

[0050] According to a third aspect of the present application, a model training method is provided, comprising:

[0051] collecting second historical sample data, the second historical sample data comprising historical environment information, historical glass temperature of the second window collected in advance and corresponding historical defogging data, the historical defogging data comprising expected defogging setting information of a user under corresponding historical conditions;

[0052] determining light radiation power of the first window according to the historical light intensity and window structure information of the first window, and predicting the historical glass temperature of the first window by the first model according to the light radiation power of the first window;

[0053] training a second model by taking the historical environment information, window information, the historical glass temperature of the first window and the historical glass temperature of the second window as input and the expected defogging setting information as output.

[0054] According to a fourth aspect of the present application, a defogging processing device based on a fusion model is provided, comprising:

[0055] an acquisition module configured to acquire environment information and window information of a vehicle, wherein the environment information comprises indoor temperature, outdoor temperature and light intensity, and the window information comprises glass humidity of a window, the window comprising a first window and a second window;

[0056] a first determination module configured to determine light radiation power of the first window according to the light intensity and window structure information of the first window;

[0057] a first model processing module configured to process the indoor temperature, the outdoor temperature and the light radiation power of the first window according to a preset first model to obtain glass temperature of the first window, wherein the first model is a model trained based on historical glass temperature data of the second window and used for predicting the glass temperature of the first window.

[0058] a second model processing module, configured to process the environment information, the window information, the glass temperature of the first window and the pre-acquired glass temperature of the second window according to a preset second model to obtain defogging recommendation information, wherein the second model is a model trained based on historical defogging data and used to generate defogging recommendation information.

[0059] According to a fifth aspect of the present application, a model training device is provided, comprising:

[0060] a first acquisition module, configured to acquire first historical sample data in a historical preset time period, the first historical sample data comprising historical indoor temperature, historical outdoor temperature, historical light intensity and historical glass temperature of a second window in the historical preset time period;

[0061] a second determination module, configured to determine light radiation power of the second window according to the historical light intensity and window structure information of the second window;

[0062] a first training module, configured to take the historical indoor temperature, the historical outdoor temperature and the historical light radiation power of the second window as input and take the historical glass temperature of the second window as output to train a first model.

[0063] According to a sixth aspect of the present application, a model training device is provided, comprising:

[0064] a second acquisition module, configured to acquire second historical sample data, the second historical sample data comprising historical environment information, pre-acquired historical glass temperature of a second window and corresponding historical defogging data, the historical defogging data comprising expected defogging setting information of a user under corresponding historical conditions;

[0065] a third determination module, configured to determine light radiation power of a first window according to the historical light intensity and window structure information of the first window and predict historical glass temperature of the first window through the first model according to the light radiation power of the first window;

[0066] a second training module, configured to take the historical environment information, window information, historical glass temperature of the first window and historical glass temperature of the second window as input and take the expected defogging setting information as output to train a second model.

[0067] According to a seventh aspect of the present application, a defogging processing system based on a fusion model is provided, comprising a cloud, a central controller in communication connection with the cloud, a vehicle controller in communication connection with the central controller and at least one sensor in communication connection with the vehicle controller, wherein

[0068] The cloud is configured to store historical data of the vehicle, the historical data including historical glass temperature data and / or historical defogging data of the second window.

[0069] The at least one sensor is configured to collect environmental information and window information of the vehicle.

[0070] The central controller is configured to train a first model according to the historical glass temperature data of the second window, and train a second model according to the historical defogging data.

[0071] The vehicle controller is configured to perform the defogging processing method based on the fusion model according to any one of the first aspect.

[0072] In an embodiment, the vehicle further comprises a central control screen and / or an audio device in communication with the vehicle controller, wherein the central control screen is configured to display the defogging recommendation information, and / or the audio device is configured to prompt the defogging recommendation information.

[0073] According to an eighth aspect of the present application, a vehicle is provided, comprising a memory, a processor and a communication interface.

[0074] The memory stores computer execution instructions.

[0075] The communication interface receives historical data transmitted by the cloud, the historical data including historical glass temperature data and / or historical defogging data of the second window.

[0076] The processor executes the computer execution instructions stored in the memory, so that the processor performs the defogging processing method based on the fusion model according to any one of the first aspect, or the model training method according to the second aspect or the third aspect.

[0077] The defogging processing method based on the fusion model, the model training method and the device provided by the present application combine the fusion model of the first model for predicting the glass temperature of the window and the second model for recommending defogging recommendation information to perform defogging processing. The first model can accurately predict the glass temperature of the window according to the information such as the temperature inside and outside the vehicle and the light radiation power. The second model generates defogging recommendation information suitable for different windows by processing the predicted temperature and other related information, instead of simply following the processing mode of the front windshield, which greatly optimizes the defogging effect of each window in the vehicle. In addition, the defogging processing mode based on the above fusion model can quickly respond to environmental changes and timely adjust the defogging strategy, thereby improving the defogging efficiency. BRIEF DESCRIPTION OF DRAWINGS

[0078] Figure 1 One of the possible scene schematic diagrams provided by the embodiments of the present application;

[0079] Figure 2 Fig. 2 is a schematic diagram of another possible scenario for providing an embodiment of the present application;

[0080] Figure 3 Fig. 3 is a schematic diagram of another possible scenario for providing an embodiment of the present application;

[0081] Figure 4 Fig. 4 is a schematic diagram of a flow of a defogging processing method based on a fusion model for providing an embodiment of the present application;

[0082] Figure 5 Fig. 5 is a schematic diagram of a flow of a prediction of a glass temperature of a vehicle window for providing an embodiment of the present application;

[0083] Figure 6 Fig. 6 is a schematic diagram of an exemplary structure of a second model for providing an embodiment of the present application;

[0084] Figure 7 Fig. 7 is a schematic diagram of a flow of a model training and processing at a vehicle end for providing an embodiment of the present application;

[0085] Figure 8 Fig. 8 is a schematic diagram of a flow of a defogging processing method based on a fusion model for providing an exemplary embodiment of the present application;

[0086] Figure 9 Fig. 9 is a schematic diagram of a flow of a model training method for providing an embodiment of the present application;

[0087] Figure 10 Fig. 10 is a schematic diagram of a flow of another model training method for providing an embodiment of the present application;

[0088] Figure 11 Fig. 11 is a schematic diagram of a structure of a defogging processing method based on a fusion model for providing an embodiment of the present application;

[0089] Figure 12 Fig. 12 is a schematic diagram of a structure of a model training apparatus for providing an embodiment of the present application;

[0090] Figure 13 Fig. 13 is a schematic diagram of a structure of another model training apparatus for providing an embodiment of the present application;

[0091] Figure 14 Fig. 14 is a schematic diagram of a structure of a defogging processing system based on a fusion model for providing an embodiment of the present application;

[0092] Figure 15 Fig. 15 is a schematic diagram of a structure of a vehicle for providing an embodiment of the present application. DETAILED DESCRIPTION

[0093] Other advantages and effects of the present application can be easily understood by those skilled in the art from the description of the present application. The present application can also be implemented or applied by other different specific embodiments, and various modifications or changes can be made to the details of the description based on different views and applications without departing from the spirit of the present application. It should be understood that the preferred embodiments are only for illustrating the present application, but not for limiting the protection scope of the present application.

[0094] It should be noted that the drawings provided in the following embodiments only schematically illustrate the basic concept of the present application, and only the components related to the present application are shown in the drawings, but not drawn according to the number, shape and size of the components in actual implementation. The shape, number and proportion of each component in actual implementation can be arbitrarily changed, and the layout pattern of the components can be more complex.

[0095] In the description of the present application, it should be understood that the terms "first", "second", "third" and the like are only used to distinguish similar objects, and do not have to be used to describe a specific order or sequence, nor can they be understood as indicating or implying relative importance. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances. In addition, in the description of the present application, "a plurality of" means two or more, unless otherwise specified. "And / or", which describes the association between objects, means that there can be three relationships, for example, A and / or B can mean that A exists alone, A and B exist together, and B exists alone. The character " / " generally represents an "or" relationship between the associated objects before and after it.

[0096] It should be noted that due to the limitation of the length of the specification, all optional embodiments are not enumerated in the present application, and those skilled in the art should be able to think of any combination of technical features as long as the technical features do not contradict each other, which can constitute optional embodiments. The embodiments are described in detail below.

[0097] In order to facilitate the understanding of the embodiments of the present application, first, the embodiments of the present application are explained in combination with application scenarios. The dehazing processing method based on the fusion model provided by the embodiments of the present application can be applied to the application scenario of intelligent driving, more specifically, it can be applied to the automatic driving application scenario based on vehicle cloud computing. For example, the execution subject of the method provided by the embodiments of the present application can be a vehicle or a server such as a cloud server (i.e. cloud). The following describes the execution subject of the method provided by the embodiments of the present application in combination with possible application scenarios:

[0098] Figure 1 is a possible application scenario provided by the embodiments of the present application, for example, Figure 1As shown, the application scenario includes a cloud server 110 and a vehicle 120, the cloud server 110 and the vehicle 120 are communicatively connected, the cloud server 110 can be used to store historical data of the vehicle (corresponding to the first historical sample data and the second historical sample data in the following), the historical data includes but is not limited to historical glass temperature data of the vehicle window, historical outdoor temperature, historical indoor temperature, historical light intensity, historical glass temperature, historical glass humidity, historical vehicle driving information (such as navigation information), historical user information, historical network environment data, historical defogging data (such as defogging setting information expected by the user under the corresponding historical conditions), and the like, the vehicle 120 can be used to obtain the related historical data from the cloud server 110 to train a first model for predicting the temperature of the vehicle window glass and a second model for generating defogging recommendation information, and combine the first model and the second model, and then control the execution of the corresponding defogging operation according to the defogging recommendation information output by the model, and can upload the processing data (such as all data from input to output) to the cloud server 110. Exemplarily, in combination with Figure 2 As shown, the vehicle 120 can include a central controller 121, a vehicle controller 122, at least one sensor 123 (such as a front windshield temperature sensor, a humidity sensor, a sunlight intensity sensor, an indoor temperature sensor, an outdoor temperature sensor, and the like), a central control screen 124, an audio device 125, and a plurality of actuators 126 (such as a vehicle window control system, a ventilation system, an air conditioning system, and the like). The vehicle 120 can be communicatively connected to the cloud server 110 through the central controller 121, the central controller 121 is used to train the first model and the second model according to the related historical data of the cloud server 110, the vehicle controller 122 can use the first model to predict the glass temperature of the vehicle window (for example, the side window without installing the glass temperature sensor), and use the glass temperature of the vehicle window and the related sensor data as the input of the second model to predict the defogging recommendation information of each vehicle window. After predicting the defogging recommendation information of the vehicle window, the central control screen 124 and / or the audio device 125 can be used to interact with the user to determine the final defogging setting (for example, it can be the defogging setting of the defogging recommendation information, such as opening the side window), and determine the target actuator corresponding to the defogging setting from the plurality of actuators 126, and the vehicle controller 122 controls the one or more target actuators to perform the defogging operation. In this application scenario, the model training process is performed by the vehicle end, and the trained model is used for defogging processing.

[0099] Figure 3 Another possible application scenario provided by the embodiment of the present application is as follows: Figure 3As shown, the application scenario can also include a cloud server 110 and a vehicle 120 connected in communication with the cloud server 110. The cloud server 110 is configured to store historical data of the vehicle, and can train a first model for predicting a temperature of a window glass of the vehicle 120 and a second model for generating defogging recommendation information for the vehicle 120 by using the relevant historical data, and transmit the first model and the second model to the vehicle 120. The vehicle 120 implements defogging operation by using the first model and the second model. In this application scenario, the model training process is performed by the cloud, and the model is transmitted to the vehicle end, so that the vehicle end can perform defogging processing by using the trained model. Alternatively, the cloud can simultaneously train the first model and the second model corresponding to each vehicle for multiple vehicles.

[0100] Alternatively, the cloud server 110 can be a cloud server that provides cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms, and the like. In some embodiments, the cloud server 110 can also be replaced by a standalone physical server, or a server cluster or distributed system composed of multiple physical servers, and the present embodiment does not particularly limit this.

[0101] The embodiment of the present application provides a defogging processing scheme based on a fusion model, environment information and window information of a vehicle are acquired, the environment information comprises an indoor temperature, an outdoor temperature and an illumination intensity, the window information comprises glass humidity of a window and / or window structure information, and the window comprises a first window and a second window. According to the illumination intensity and the window structure information of the first window, illumination radiation power of the first window is determined. According to a preset first model, the indoor temperature, the outdoor temperature and the illumination radiation power of the first window are processed to obtain a glass temperature of the first window, the first model is a model trained based on historical glass temperature data of the second window and used for predicting the glass temperature of the first window. And according to a preset second model, the environment information, the window information, the glass temperature of the first window and the pre-acquired glass temperature of the second window are processed to obtain defogging recommendation information, the second model is a model trained based on historical defogging data and used for generating the defogging recommendation information. In this process, the fusion model of the first model used for predicting the glass temperature of the window and the second model used for recommending the defogging recommendation information is combined, the first model can accurately predict the glass temperature of the window according to the indoor and outdoor temperature and the illumination radiation power and other information, and the second model generates the defogging recommendation information suitable for different windows by processing the predicted temperature and other related information, instead of simply using the processing mode of the front windshield, so that the defogging effect of each window in the vehicle is greatly optimized. In addition, the defogging processing mode based on the fusion model can quickly respond to environmental changes, timely adjust the defogging strategy, and further improve the defogging efficiency.

[0102] The application scenarios of the embodiments of the present application are briefly introduced above, and the application scenarios are taken as examples, the execution subject is a vehicle 120, and a defogging processing method based on a fusion model proposed by the embodiment is introduced in detail as shown in the following Figure 1 application scenarios are taken as examples, the execution subject is a vehicle 120, and a defogging processing method based on a fusion model proposed by the embodiment is introduced in detail as shown in the following Figure 4 application scenarios are taken as examples, the execution subject is a vehicle 120, and a defogging processing method based on a fusion model proposed by the embodiment is introduced in detail as shown in the following

[0103] Step S401, environment information and window information of a vehicle are acquired; wherein the environment information comprises an indoor temperature, an outdoor temperature and an illumination intensity, the window information comprises glass humidity of a window and / or window structure information, and the window comprises a first window and a second window.

[0104] Exemplarily, the environment information and the window information can be collected through various sensors installed on the vehicle, for example, a front windshield temperature sensor, a humidity sensor, a sunlight intensity sensor, an indoor temperature sensor, an outdoor temperature sensor and the like. In some embodiments, the vehicle can also acquire relevant data by communicating with other devices, for example, the vehicle can exchange data with a smart phone or other electronic devices placed in the vehicle and having a sensor function through a wireless communication module, so as to acquire the environment information and the window information of the vehicle.

[0105] Optionally, the first vehicle window can be a side window of the vehicle, and the second vehicle window can be a front windshield of the vehicle. In some embodiments, the first vehicle window can also be another vehicle window (e.g., a rear window) without a temperature sensor or with inaccurate temperature measurement (e.g., a faulty temperature sensor), and the second vehicle window can also be another vehicle window (e.g., a left side window) with a temperature sensor. The embodiments of the present application do not particularly limit the types of the first vehicle window and the second vehicle window.

[0106] In step S402, the solar radiation power of the first vehicle window is determined according to the light intensity and the vehicle window structure information of the first vehicle window.

[0107] For example, the vehicle window structure information can include the installation angle, surface area, and glass thickness of the vehicle window glass. Optionally, the material properties of the glass can also be included, which can affect the light absorption of the glass and in turn affect the solar radiation power of the glass. Specifically, the corresponding absorption coefficient can be obtained in combination with the material properties of the glass.

[0108] In the related art, the influence of light intensity on the temperature of the vehicle window is not considered during the defogging process of the vehicle window, especially the different solar radiation powers caused by different vehicle window structure information, resulting in inaccurate temperature of the vehicle window. In the embodiments, the light intensity and the vehicle window structure information are used to calculate the solar radiation power of the vehicle window, so as to correct the glass temperature of the vehicle window in the subsequent steps, and in turn provide a more accurate glass temperature prediction value for the first vehicle window.

[0109] In one way, the vehicle window structure information includes the installation angle and surface area of the vehicle window glass. The determination of the solar radiation power of the first vehicle window according to the light intensity and the vehicle window structure information of the first vehicle window can be performed in the following way: determining the relative light intensity of the first vehicle window according to the installation angle and the light intensity of the first vehicle window; and determining the solar radiation power of the first vehicle window according to the relative light intensity and the surface area.

[0110] For example, the installation angle can include the horizontal angle and the circumferential angle (north) of the glass. The installation angle of the glass affects the angle of light incident on the surface of the vehicle window, thereby changing the effective intensity of the light. The embodiments consider the incident angle of the light, and can more accurately evaluate the actual influence of the light on the vehicle window.

[0111] Optionally, the calculation method of the relative light intensity hsolar of the first vehicle window can be as follows:

[0112] h solar = h * cos θ

[0113] cos θ = sin γ * cos β + cos γ * sin β * cos δ

[0114] In the formula, h represents the light intensity, cos θ represents the cosine value of the incident angle of the sunlight relative to the glass surface, γ represents the solar horizontal angle, α represents the solar orientation angle, β represents the horizontal angle of the glass, ξ represents the circumferential angle of the glass, and δ represents the angle difference between the incident direction of the sunlight and the facing direction of the glass, where δ = α - ξ. The solar horizontal angle and the solar orientation angle can be calculated by time and latitude and longitude, and are known quantities.

[0115] In the calculation of the relative light intensity h solar , the surface area of the glass is combined, and the light radiation power of the vehicle window can be calculated. Alternatively, in order to further improve the accuracy of the light radiation power, that is, the solar radiation power absorbed by the glass, the embodiment considers the environmental thermal radiation power, that is, the air thermal radiation power, to calculate the light radiation power. Specifically, the method can further include the following steps: determining the environmental thermal radiation power of the first vehicle window according to the temperature inside the vehicle, the temperature outside the vehicle, and the surface area.

[0116] Alternatively, the calculation method of the environmental thermal radiation power q rad may be: q rad = 0.0000000567 * 0.95 * area * (T2^4 - T1^4).

[0117] In the formula, T1 is the temperature outside the vehicle, T2 is the temperature inside the vehicle (as an approximation of the glass surface temperature), and area is the surface area of the glass.

[0118] The above determination of the light radiation power of the first vehicle window according to the relative light intensity and the surface area can be performed in the following manner: determining the initial light radiation power of the first vehicle window according to the relative light intensity and the surface area; and obtaining the light radiation power according to the environmental thermal radiation power and the initial light radiation power.

[0119] For example, the calculation method of the initial light radiation power q solar may be: q solar = h solar * area. The calculation method of the solar radiation power P solar absorbed by the glass may be: P sola = abscoeff * q solar + q rad . In the formula, abscoeff represents the absorption coefficient corresponding to the material properties of the glass.

[0120] Through the above technical solution, the solar radiation power absorbed by the glass can be accurately calculated, and the accuracy of the vehicle window glass temperature prediction in the subsequent model processing process can be improved.

[0121] In step S403, the in-vehicle temperature, the outside temperature, and the light radiation power of the first window are processed according to a preset first model to obtain the glass temperature of the first window, wherein the first model is a model trained based on historical glass temperature data of the second window and used for predicting the glass temperature of the first window.

[0122] For example, the first window is a side window, and the second window is a front windshield. In vehicle design, temperature sensors (or other temperature measuring instruments) are usually not installed at each window position, and more attention is paid to defogging of the front windshield for defogging of the window. Therefore, temperature sensors (or other temperature measuring instruments) are only installed at the front windshield to accurately measure the temperature of the front windshield. However, a clear side window view is also crucial for lane changing, turning, and observing blind spots. If the side window is not effectively defogged, it will also affect the driver's judgment of the surrounding environment and increase the risk of accidents.

[0123] Instead of the original unified defogging method based on the front windshield, or directly using the temperature of the front windshield to predict the temperature of the side window (or other windows), the side window is not accurately predicted or the defogging effect is poor. The first model is used in this embodiment to predict the glass temperature of the side window in combination with the in-vehicle temperature, the outside temperature, and the light radiation power of the side window. Specifically, based on the principles of heat transfer and energy balance (i.e., equivalent heat exchange), it can be known from the equivalent heat exchange principle (similar to the front windshield equivalent heat exchange principle and the side window glass equivalent heat exchange principle) that the change of the glass temperature is mainly affected by the heat exchange process of the surrounding environment. The in-vehicle temperature and the outside temperature represent the environmental conditions on both sides of the glass, which affect the heat absorption and dissipation of the glass through convection and conduction. Specifically, the in-vehicle temperature can affect the inner surface temperature of the glass through convective heat exchange between the in-vehicle air and the glass, while the outside temperature can affect the outer surface temperature of the glass through convective heat exchange between the external air and the glass. The light radiation power is another important factor because solar radiation directly affects the energy absorption of the glass, and the solar energy absorbed by the glass will cause its temperature to rise, which is particularly significant in strong sunlight. This embodiment considers these factors to establish a model to predict the temperature of the glass. The model is based on the basic principles of heat transfer and combines environmental temperature and radiation energy to effectively simulate the dynamic change of the glass temperature under different conditions. The model is trained in combination with the measured value of the glass temperature of the front windshield, and the trained model can accurately predict the glass temperature.

[0124] For example, the front windshield and side window follow the same basic physical principles in heat transfer. The temperature of both is affected by ambient temperature, interior temperature, and solar radiation. The equivalent heat exchange model describes the heat exchange process of the glass through convection, conduction, and radiation, which are consistent on different types of vehicle window glass. Therefore, in this embodiment, to solve the problem of lack of accurate temperature data due to the absence of temperature sensors for side window glass (or other vehicle windows), the actual temperature data of the front windshield (or other vehicle windows with temperature sensors installed) is used to train the model, and the model is used to predict the temperature of the side window glass. In the model training process, the input parameters include ambient temperature, interior temperature, and solar radiation power. Among them, since the interior and exterior temperatures tend to be balanced, the interior temperature of the front windshield and the interior temperature of the side window in the same vehicle are approximately the same, and the exterior temperature of the front windshield and the exterior temperature of the side window are approximately the same. When calculating the solar radiation power of the vehicle window, this embodiment considers the different structural information of the vehicle window, so that the relationship between the solar radiation power and the glass temperature accurately reflects the characteristics of different vehicle windows. In this way, by using the temperature data of the front windshield for model training, the temperature of the side window glass can be more accurately predicted without a direct temperature sensor. It can be understood that the equivalent heat exchange model of the glass is dT / dt = Ptotal / cp*mass, where dT / dt represents the temperature change rate of the glass, cp*mass is a constant, cp represents the specific heat capacity of the glass, and mass represents the mass. P 总 = P solar + P ambient + P cabin , P 总 total represents the total heat exchange power of the glass, P solar solar represents the solar radiation power, P ambient ambient represents the heat exchange power between the glass and the external environment, and P cabin interior represents the heat exchange power between the glass and the interior environment. The specific heat capacity of the glass (such as the front windshield and the side window) of the same vehicle model is consistent. The side window glass and the front windshield may differ in mass or area. To further improve the accuracy of predicting the temperature of the side window glass, since the mass and area of the front windshield and the side window glass are known parameters, the model parameters of the front windshield model can also be adjusted (such as adjusting the weights in the trained first model in proportion to the mass ratio and / or area ratio between the front windshield and the side window glass) through the difference in mass or area between the front windshield and the side window glass, so as to accurately predict the temperature of the side window glass without a temperature sensor.

[0125] For the convenience of understanding the embodiments of the present application, the embodiments further illustrate this. The embodiments are based on the equivalent heat exchange principle:

[0126]

[0127] where dT / dt is the rate of change of the glass temperature, cp, mass are the specific heat capacity and mass of the glass, respectively, and are known constants; is the heat transfer coefficient of the glass inside the vehicle; is the heat transfer coefficient of the glass outside the vehicle, and the side window glass and the front windshield in the vehicle are approximately equal to the above heat transfer coefficients. Among them,

[0128]

[0129]

[0130] where T 车内 represents the temperature inside the vehicle, T 玻璃 represents the temperature of the glass, and T 车外 represents the ambient temperature. By arranging the above formula, we have:

[0131]

[0132] By converting the above formula, we have:

[0133]

[0134] Based on the above formula, the left side can be regarded as a function of the glass temperature, which can correspond to the model output, and the right side can be regarded as the model training input parameters: solar radiation power (i.e. ), the temperature inside the vehicle, the temperature outside the vehicle. The trained model represents the correlation between the glass temperature (such as the temperature of the front windshield) and , , .

[0135] Therefore, even if the A (area), mass (mass) and cp (specific heat capacity) of the side window glass and the front windshield are different, for these known parameters, the same can be adjusted by equal proportion or other adjustment methods, and the training parameters obtained by training the historical glass temperature of the front windshield, the historical ambient temperature, the historical temperature inside the vehicle, the historical solar radiation power can be obtained by fine-tuning the above known parameters to obtain more accurate training parameters, thereby improving the prediction accuracy of the temperature of the side window glass.

[0136] Exemplarily, a first model for predicting the temperature of the side window glass is trained with the temperature data of the front windshield of the same vehicle model. The model is trained by collecting historical data of the front windshield, which can be data of the front windshield under different working conditions (e.g., historical data collected under different air blower gear conditions, air blowing modes, vehicle speeds, etc.). Based on the above formula, the function for representing the model can be as shown in the following formula, and it is exemplified that w1, w2, and w3 are multiplied by the intensity of solar radiation, the temperature inside the vehicle, and the temperature outside the vehicle.

[0137]

[0138] wherein the weight w1 corresponds to the above formula / (A1*mass2*A2*mass1). In the case where the front windshield and the side window glass have different masses and areas, the equal ratio adjustment of the weight coefficient can be w1’ = w1 / A1 / mass2*A2*mass1, given the area A1 of the front windshield, the mass mass1, the area A2 of the side window glass, and the mass mass2.

[0139] Similarly, the weight w2 corresponds to the above formula / (A1*mass2*A2*mass1). In the case where the front windshield and the side window glass have different masses and areas, the equal ratio adjustment of the weight coefficient can be w2’ = w2 / A1 / mass2*A2*mass1.

[0140] Similarly, the weight w3 corresponds to the above formula In the case where the front windshield and the side window glass have different masses and areas, the equal ratio adjustment of the weight coefficient can be w3’ = w3 / A1 / mass2*A2*mass1.

[0141] It should be understood that the above model can be applied to the prediction of the temperature of the side window glass in different vehicle models. In the model training process, for each vehicle model, the above model training method can be used in combination with the historical data of the vehicle model to train a side window temperature prediction model corresponding to the vehicle model, thereby realizing the prediction of the side window temperature of different vehicle models.

[0142] As can be seen, since the front windshield and the side window glass follow the same basic physical principles in heat transfer, and the specific heat capacity and mass of the vehicle window are known parameters, even if they are not exactly the same, the model can be fine-tuned based on these known parameters to accurately predict the temperature of the side window glass without a temperature sensor. In this way, even if the side window lacks direct temperature data, the temperature of the side window glass can still be accurately predicted based on the historical input data of the front windshield using the model.

[0143] Exemplarily, the training process of the first model can be collecting first historical sample data, the first historical sample data including historical in-vehicle temperature, historical outside-vehicle temperature, historical light intensity and historical glass temperature of the second window in a historical preset time period. And according to the historical light intensity and the window structure information of the second window, the light radiation power of the second window is determined. The historical in-vehicle temperature, the historical outside-vehicle temperature, the historical light radiation power of the second window are taken as inputs, and the historical glass temperature of the second window is taken as output, and the first model is trained.

[0144] Optionally, the first model can adopt an artificial intelligence (AI) model, such as a machine learning model.

[0145] Optionally, in the data collection stage, collecting the first historical sample data can include in-vehicle temperature data recorded in a preset time period, historical outside-vehicle temperature data recorded in the same time period, historical light intensity data recorded in the same time period, and second window glass temperature data recorded in the same time period. Before model training, the above data can be preprocessed, such as handling missing values and outliers, to improve data accuracy, and aligning these data at the same timestamp to facilitate model training. Then, according to the historical light intensity and the window structure information of the second window, the light radiation power of the second window is calculated, and the calculation process is similar to the calculation method of the light radiation power of the first window, and the relevant description will not be repeated. By taking the historical in-vehicle temperature, the historical outside-vehicle temperature, and the historical light radiation power of the second window as input features of the model, and taking the historical glass temperature of the second window as the output target of the initial model (such as an AI model), the model is trained using the training data set, the model parameters are adjusted to minimize the error between the predicted output and the actual output (such as gradient descent optimization algorithm), or when the maximum number of iterations is reached, the first model is trained.

[0146] In some embodiments, the first model can also be used for glass temperature prediction of the second window (such as the front windshield), for example, when the temperature sensor of the front windshield is not turned on, etc. The glass temperature prediction of the front windshield is similar to the glass temperature prediction of the first window, and the relevant description will not be repeated here. Optionally, the glass temperature prediction of the front windshield and the glass temperature prediction of the side window can be as shown in Figure 5 In this way, the glass temperature of the corresponding window can be more accurately obtained when the temperature measurement function of the window is missing or not turned on, providing data support for subsequent defogging processing.

[0147] In step S404, the environment information, the window information, the glass temperature of the first window, and the pre-acquired glass temperature of the second window are processed according to a preset second model to obtain defogging recommendation information. The second model is trained based on historical defogging data and is used to generate the defogging recommendation information.

[0148] In this embodiment, the second model is trained based on historical defogging data and can identify defogging requirements and corresponding optimal defogging strategies (such as user-expected defogging setting information) under different conditions. By inputting the environment information (such as the indoor temperature and the outdoor temperature), the window information (such as the glass humidity of the window), the glass temperature of the first window, and the glass temperature of the second window into the second model, the model can output defogging recommendation information for the first window and the second window. Optionally, the window recommendation information can be for different windows respectively or for a specific window, for example, the suggestion of opening the side window, starting the wiper, adjusting the heating of the window (the first window and / or the second window), starting the defogger, opening or adjusting the defogging mode of the air conditioning system, and the like. In addition, other operation suggestions can also be provided, such as adjusting the air circulation mode in the vehicle or opening the air vent, and the like. In some embodiments, feedback information of the actual defogging effect can also be collected to evaluate the effectiveness of the recommendation information. Based on these feedbacks, the second model is further optimized to improve its prediction accuracy and recommendation effect under different environmental conditions.

[0149] In some embodiments, the environment information can further include one or more of the following: vehicle driving information (navigation information, vehicle speed, etc.), rainfall information, current air conditioning setting state, network environment data, user information, and the like. The window information can further include the window position, and the like.

[0150] It can be understood that the pre-acquired glass temperature of the second window can be acquired by the second window with temperature measurement function or predicted by the first model.

[0151] Exemplarily, the training manner of the second model can be that second historical sample data is collected, the second historical sample data including historical environment information, a pre-collected historical glass temperature of the second window, and corresponding historical defogging data, the historical defogging data including expected defogging setting information of a user under a corresponding historical condition. According to the historical light intensity and the window structure information of the first window, the light radiation power of the first window is determined, and the historical glass temperature of the first window is predicted according to the light radiation power of the first window by the first model. The historical environment information, the window information, the historical glass temperature of the first window, and the historical glass temperature of the second window are taken as inputs, and the expected defogging setting information is taken as output, so as to train the second model. Optionally, the second model can adopt the same initial model as the first model or a different initial model for training, which is not particularly limited in the embodiment. Correspondingly, the training process of the second model can also be similar to that of the first model, which can be referred to the training process of the first model.

[0152] In some embodiments, in order to further optimize the defogging efficiency, the embodiment combines the temperature and humidity changes of the window in the defogging recommendation process to predict fogging, thereby providing more optimal fogging recommendation information. As shown in Figure 6 The second model can include an input module 601, a preprocessing module 602, a risk prediction module 603, a recommendation module 604, and an output module 605. In the above steps, according to the second model, the environment information, the window information, the glass temperature of the first window, and the pre-collected glass temperature of the second window are processed to obtain defogging recommendation information, specifically: the input environment information, the window information, the glass temperature of the first window, and the pre-collected glass temperature of the second window are received by the input module 601. The glass temperature and humidity of the first window and the glass temperature and humidity of the second window are processed by the preprocessing module 602 to obtain the temperature and humidity changes of the first window and the temperature and humidity changes of the second window. At least one of the glass temperature, the glass humidity, and the temperature and humidity change information of the first window, and at least one of the glass temperature, the glass humidity, and the temperature and humidity change information of the second window are processed by the risk prediction module 603 to determine the corresponding fogging grades of the first window and the second window. According to the fogging grades, the environment information, and the window information, the defogging recommendation information corresponding to the first window and the second window is generated by the recommendation module 604. The defogging recommendation information is output by the output module 605.

[0153] Exemplarily, the pre-processing module can calculate the glass temperature and the glass humidity to obtain the temperature and humidity change. The pre-processing module can directly calculate the temperature and humidity change by using a first-order derivative algorithm, or use a time series analysis method to extract features to reflect the change trend of the temperature and humidity. Specifically, the temperature and humidity change is calculated by first-order derivation as follows: the first-order derivation is performed on the glass temperature and the glass humidity of the first vehicle window to obtain the temperature and humidity change information of the first vehicle window. The first-order derivation is performed on the glass temperature and the glass humidity of the first vehicle window to obtain the temperature and humidity change information of the second vehicle window.

[0154] The glass humidity can be the environmental humidity in the vehicle, which can be obtained by using a humidity sensor installed in the vehicle, or obtained by using other prior art. The specific acquisition method of the glass humidity is not particularly limited in the embodiment.

[0155] Exemplarily, the risk prediction module can predict the fogging level according to the temperature and humidity change obtained by the pre-processing module. A classification model (such as a decision tree) can be used to divide the fogging level into multiple categories (such as no fog and fogging).

[0156] In the related art, in order to improve the prediction accuracy of the fogging caused by the water vapor content and the dew point temperature in the air, the number of people in the vehicle can be determined by using a seat or a vehicle camera. However, for some vehicle models, there is no rear seat weight sensor or vehicle camera, so it is impossible to determine the number of people in the vehicle. In addition, it is difficult to accurately determine the physical signs of people, such as height, weight, metabolism, water vapor generated by people in the vehicle, and changes in the environment, which has low precision and a complex implementation process. It is found through research that the temperature and humidity change (including temperature change and humidity change) will affect the water vapor content and the dew point temperature in the air, and then affect the formation or dissipation of the fogging. The embodiment replaces the above technical solution by combining the temperature and humidity change in the model to predict the fogging. Specifically, the humidity change in the vehicle is affected by the number of passengers in the vehicle and the physical signs of the passengers. The glass temperature change can also represent external sudden weather changes, such as sudden rainfall. The embodiment can solve the problem of inaccurate automatic defogging function caused by the number of people in the vehicle, different physical signs, and environmental humidity changes. The vehicle camera or seat gravity sensing device can be omitted, and the prediction accuracy of the fogging is effectively improved.

[0157] Exemplarily, the recommendation module generates defogging recommendation information according to the predicted fogging level and environmental information, vehicle window information, etc. The recommendation module can use a rule engine or a generation model to provide specific operation suggestions (such as adjusting air conditioning settings, turning on a defogger, etc.). Specifically, the fogging level of the first vehicle window can be determined according to at least one of the glass temperature, the glass humidity, and the temperature and humidity change information of the first vehicle window. In addition, the fogging level of the second vehicle window can be determined according to at least one of the glass temperature, the glass humidity, and the temperature and humidity change information of the second vehicle window.

[0158] Through the above steps, the temperature and humidity change of the vehicle window is combined in the defogging recommendation process to predict the fogging, further optimizing the prediction and recommendation effect of the vehicle window defogging.

[0159] Further, according to the glass temperature and the temperature change information thereof, and the glass humidity and the humidity change information thereof, the fogging level of the first vehicle window is determined, which can be performed in the following manner: according to the temperature and humidity change information of the first vehicle window, a fogging probability change coefficient corresponding to the temperature and humidity change information is searched from a preset mapping table to obtain the fogging probability change coefficient of the first vehicle window; wherein the mapping table comprises fogging probability change coefficients corresponding to different temperature and humidity change information; and according to at least one of the glass temperature, the glass humidity of the first vehicle window and the fogging probability change coefficient of the first vehicle window, the fogging level of the first vehicle window is determined. In this embodiment, the first-order derivative and the second-order derivative of the glass temperature and the glass humidity of the vehicle window can be obtained respectively, four new variables are obtained: h1 (derivative of humidity) represents the humidity change, t1 (derivative of glass temperature) represents the change of glass temperature, h2 (second-order derivative of humidity) represents the speed of humidity change, and t2 (second-order derivative of temperature) represents the speed of temperature change, wherein the second-order derivative can be used as the exit condition of defogging, which will be described in detail in the following embodiment. In the calculation process, the relative humidity can be converted into the moisture content to avoid the influence of temperature and thus to separate the two parameters. In the related art, the glass temperature and humidity are used to calculate the dew point temperature, and when the glass temperature is lower than the dew point temperature, the fogging can be determined. However, due to the influence of the sensor arrangement position and other factors, the measuring point only covers one area. In this embodiment, the fogging probability change coefficient is introduced, a two-dimensional mapping table is established, the first-order derivative of the newly introduced humidity and temperature is taken as the input, and the fogging probability change coefficient is output (wherein the two-dimensional mapping table can be determined according to a large amount of experimental data or prior data). The faster the humidity increases, the larger the coefficient is; the faster the temperature decreases, the larger the coefficient is, for example, if the threshold is set to 0.3 (or 30%), if the fogging probability change coefficient is lower than 0.3, it is considered that the vehicle window is not fogged. Or if the fogging probability change coefficient is higher than or equal to 0.3, it is considered that the vehicle window is fogged. Alternatively, according to the fogging probability coefficient corresponding to the temperature and humidity, for example, the fogging probability coefficient corresponding to the difference between the glass temperature and the dew point temperature is obtained (the determination of the fogging probability coefficient can be performed in the existing manner), the final fogging probability coefficient is obtained by weighted average calculation of the fogging probability change coefficient and the original fogging probability change coefficient, and the like.

[0160] In some embodiments, the fogging level of the first vehicle window can also be determined by the glass temperature, for example, by comparing the glass temperature with the dew point temperature to determine the fogging level. For example, when the glass temperature is lower than the dew point temperature, such as the glass temperature is higher than the dew point temperature by more than 2°C, it is determined that the fogging level is no fog, or such as the glass temperature is lower than or equal to the dew point temperature by more than 2°C, it is determined that the fogging level is fog. In some embodiments, the fogging level of the first vehicle window can also be determined by the glass humidity, for example, when the glass humidity is lower than 30%, it is determined that the fogging level is no fog, or such as the glass humidity is higher than or equal to 30%, it is considered to be fog.

[0161] In some embodiments, the fogging level of the first vehicle window can also be determined by the glass temperature, for example, by comparing the glass temperature with the dew point temperature to determine the foging level. For example, when the glass temperature is lower than the dew point temperature, such as the glass temperature is higher than the dew point temperature by more than 2°C, it is determined that the fogging level is no fog, or such as the glass temperature is lower than or equal to the dew point temperature by more than 2°C, it is determined that the fogging level is fog. In some embodiments, the fogging level of the first vehicle window can also be determined by the glass humidity, for example, when the glass humidity is lower than 30%, it is determined that the fogging level is no fog, or such as the glass humidity is higher than or equal to 30%, it is considered to be fog. 3， In the formula, Y is the fogging level index, X1 is the glass temperature of the first vehicle window, X2 is the glass humidity of the first vehicle window, and X3 is the fogging probability change coefficient of the first vehicle window. A, B, and C represent the weight coefficients corresponding to the glass temperature, the glass humidity, and the fogging probability change coefficient, respectively (the weight coefficients can be the same or different, and can be adaptively determined by a person skilled in the art in combination with actual application or experience values). For example, when the fogging level index reaches a preset index threshold, it is considered that the fogging level is fog, and when it is lower than the preset index threshold, it is determined that the fogging level is no fog.

[0162] It should be noted that a person skilled in the art can pre-establish the mapping table in combination with actual application or a large number of experience values. The mapping table is part of the model, so that the model can efficiently determine and adjust the fogging probability, thereby improving the prediction accuracy and real-time response ability.

[0163] Correspondingly, the determination of the fogging level of the second vehicle window according to the temperature change information and the humidity change information of the second vehicle window can be performed in the following manner: according to the temperature and humidity change information of the second vehicle window, the fogging probability change coefficient corresponding to the temperature and humidity change information is searched from the pre-established mapping table to obtain the fogging probability change coefficient of the second vehicle window; and at least one of the glass temperature of the second vehicle window, the glass humidity of the second vehicle window, and the fogging probability change coefficient of the second vehicle window is used to determine the fogging level of the second vehicle window.

[0164] It should be noted that the determination of the fogging level of the second vehicle window is similar to the determination of the fogging level of the first vehicle window, and the related description will not be repeated here.

[0165] In some embodiments, after generating the defogging recommendation information, the target actuators can be directly controlled to perform corresponding defogging operations using the defogging recommendation information, so as to improve the defogging efficiency. Specifically, the method can further include the following steps: controlling the target actuators to perform corresponding defogging operations according to the defogging recommendation information; wherein the defogging recommendation information includes one of the following recommendation information: opening the first window and / or the second window, adjusting the ventilation setting, and adjusting the air conditioning setting; and the target actuators include at least one of the following: a window control system, a ventilation system, and an air conditioning system.

[0166] In this embodiment, the target actuators, which are the actuators corresponding to the defogging recommendation information, can include other actuators in addition to the above-mentioned actuators, such as a Heating, Ventilation, and Air Conditioning (HVAC) system, a damper system, a curtain system, a sunroof system, etc., which can be used to perform operations corresponding to the defogging recommendation information, such as opening / closing of the window, ventilation, etc., according to the control.

[0167] For example, when the model prediction result is fogging, the defogging recommendation information can be generated in combination with the predicted fogging level (e.g., both the first window and the second window are predicted to be fogging), which suggests opening the window to promote air circulation, adjusting the ventilation setting to increase air flow, and adjusting the air conditioning to reduce humidity. The corresponding defogging operations can be performed by automatically controlling multiple target actuators. For example, the window control system partially opens the first window to speed up air circulation, the wiper performs rain wiping on the second window, etc.

[0168] In further examples of this embodiment, the defogging process can also be implemented through user interaction to optimize the user experience. Specifically, the above-mentioned controlling the target actuators to perform corresponding defogging operations according to the defogging recommendation information can be implemented in the following ways.

[0169] Method one: displaying the defogging recommendation information on the center control screen, determining defogging setting information in response to a touch operation on the center control screen, the defogging setting information being the same as or different from the defogging recommendation information, and controlling the target actuators to perform corresponding defogging operations.

[0170] In this method, by displaying defogging recommendations on the central control screen, the driver or passengers can view system-generated defogging recommendations, such as suggesting opening windows, adjusting ventilation settings, or adjusting the air conditioning. Users can respond to these recommendations via touchscreen operation on the central control screen. Based on the user's touch input, the system determines the final defogging settings, which can be the same as the recommendations (if the user accepts them) or different (if the user makes custom adjustments). The vehicle then further controls the target actuators to perform the corresponding defogging operations based on these settings, such as opening windows, adjusting ventilation, or adjusting air conditioning settings.

[0171] This approach allows users to adjust settings to their individual needs, enabling personalized defogging control and enhancing user experience and driving safety.

[0172] Method 2: Defogging recommendation information is prompted through an audio device. In response to voice input operation on the audio device, defogging setting information is determined. If the defogging setting information is the same as or different from the defogging recommendation information, the target actuator is controlled to perform the corresponding defogging operation.

[0173] It should be noted that the interaction method through the audio device is similar to that of the central control screen mentioned above. For relevant explanations and effects, please refer to Method 1 above. It will not be elaborated further here.

[0174] Method 3: It can also be combined with the central control screen and audio device to achieve specific defogging operation. The relevant principle is similar to that of Method 1 and Method 2 above, and will not be elaborated here.

[0175] The following explanation uses the vehicle-side model training and processing (omitting the temperature prediction process of the first model) and voice interaction as an example to illustrate the defogging process. Figure 7 As shown, this includes training data input, model training, model deployment, (current) data input, model processing, and defogging recommendation information output. Among these,

[0176] The model training and deployment phase begins with inputting training data, including historical data such as outside temperature, inside temperature, light intensity, glass temperature, glass humidity, current air conditioning settings, network environment data, navigation information, and user information, as well as corresponding historical defogging data (defogging operation results and user's desired defogging settings), which corresponds to the second historical sample data discussed later. This training data can be stored in the cloud and retrieved from the cloud for model training when needed on the vehicle, or it can be directly input into the vehicle for training. After the vehicle completes model training, model deployment can proceed. For example, the model can be converted into a form that can run on the vehicle's local system (such as the in-vehicle controller), making it compatible with the vehicle's hardware and other software modules.

[0177] In the model processing stage, firstly, the (current) data input is performed, such as inputting the defogging related data into the second model for processing to output the defogging recommendation information. Optionally, the input data for model processing can be transmitted to the cloud for storage as historical data to facilitate model optimization. In one possible manner, the model can combine a machine learning algorithm (such as a decision tree, a particle swarm algorithm, etc.). Taking the decision tree algorithm as an example, the defogging level of each window can be determined according to various real-time input data (such as the temperature outside the vehicle, the temperature inside the vehicle, the light intensity, the glass temperature, the glass humidity, the current air conditioning setting state, the network environment data, the navigation information, the user information, etc.), and then the corresponding defogging recommendation information can be generated in combination with other factors such as temperature, such as opening / closing the window, air conditioning setting (such as air conditioning switch, temperature setting, automatic (Automatic switch, AUTO) switch, air conditioner compressor (Air Conditioner switch, AC) switch, mode setting), ventilation setting, inside / outside circulation setting, defogging through window / skylight position (such as slightly opening the window to make the air inside and outside the vehicle flow, reducing the humidity difference), window heating to defog. The particle swarm algorithm can be used to optimize the parameters in the defogging process, such as finding the combination of air conditioning air volume, temperature, etc. to further optimize the defogging effect. It should be noted that this process is only one possible example of the present embodiment, and is not a limitation of the present embodiment.

[0178] Optionally, the model output can also include a voice interaction stage. Specifically, according to the defogging recommendation information output by the model, the user is prompted through the audio device, for example, the voice informs “it is recommended to start the defogging mode, and the first window starts the wiper”, and the feedback information of the user for the defogging recommendation information can be received to realize the final defogging setting, thereby optimizing the user experience and improving the driving safety and comfort.

[0179] In some embodiments, considering that the defogging process in the related art lacks a reasonable exit mechanism, it can cause frequent exit or entry of the defogging process. To further optimize the user driving experience, the trend information of the humidity change and / or the trend information of the temperature change of the window can also be monitored in real time. In response to monitoring that the trend information of the humidity change and / or the trend information of the temperature change is lower than a preset threshold, the control target executor stops the current defogging operation.

[0180] In this embodiment, the trend information of the humidity change can be the derivative of the humidity change, i.e. the second derivative of the glass humidity mentioned above, which can represent the speed of the change of the humidity inside the vehicle. The trend information of the temperature change is the same.

[0181] It should be noted that the preset threshold can be adaptively set by a person skilled in the art according to actual application or experience value, and the specific value of the preset threshold is not particularly limited in the embodiment of the application.

[0182] By introducing the above exit mechanism, the problem that the time interval between the exit and re-entry of the current automatic defogging function is unreasonable and may lead to frequent start-up exit operations, affecting the comfort of the passenger compartment, is considered, and the user experience is further optimized.

[0183] For the convenience of understanding the embodiments of the application, further introduction will be made in combination with an exemplary embodiment. As shown in Figure 8 The flow includes the following:

[0184] The input parameters (including environmental information, vehicle window information, etc.) are obtained. For example, the environmental information can include the temperature outside the vehicle, the temperature inside the vehicle, the light intensity, the current air conditioning setting state, network environment data, navigation information, user information, etc. The vehicle window information can include the glass temperature of the vehicle window, the glass humidity, etc.

[0185] The related parameters in the input parameters (including direct input or processed input, and the related parameters and specific processing process can be understood in combination with the above embodiments) are input into the trained first model for processing to predict the glass temperature of the vehicle window (lacking the glass temperature measurement result).

[0186] The glass temperature of the vehicle window predicted by the first model and the related parameters in the input parameters are input into the trained second model for processing to predict the defogging recommendation information. Specifically, the second model can include a fogging level prediction module and a recommendation module. The prediction module can analyze the model input information to determine the fogging level of each vehicle window. The recommendation module can output the defogging recommendation information in combination with the fogging level, the environmental information and the vehicle window information.

[0187] After outputting the defogging recommendation information, a safety rule boundary can be introduced to determine the specific content in the defogging recommendation information, for example, the defogging recommendation information recommends air conditioning defogging, and the recommended air conditioning defogging temperature is outside the reasonable temperature range that can be adjusted by the vehicle air conditioning system (for example, the lower limit of the vehicle air conditioning refrigeration temperature is about 16℃, and the upper limit of the heating temperature is about 32℃). If the recommended temperature is lower than the lower limit or higher than the upper limit, it is determined that the safety rule boundary is not met. If the safety rule boundary is not met, the process can be ended, and the user is recommended the defogging recommendation information determined based on the existing technology, such as recommending the defogging recommendation information obtained based on the rule algorithm to the user. For the defogging recommendation information that meets the safety rule boundary, the audio device can be called to broadcast the related defogging recommendation, or the large screen can be prompted to set according to the intelligent recommendation, such as the user selecting to confirm the use, and then the related defogging operation is performed according to the defogging recommendation information, otherwise the original state is maintained unchanged. During the whole defogging process, the cloud can record the data of the whole process, so as to facilitate the optimization of the subsequent model and the analysis of the user preference, etc.

[0188] Figure 9 is a flowchart of a model training method provided by an embodiment of the present application, as shown in Figure 9 , the method comprises:

[0189] Step S901, collect first historical sample data, the first historical sample data comprising historical indoor temperature, historical outdoor temperature, historical light intensity and historical glass temperature of the second window in a historical preset time period;

[0190] Step S902, determine the light radiation power of the second window according to the historical light intensity and the window structure information of the second window;

[0191] Step S903, take the historical indoor temperature, the historical outdoor temperature and the historical light radiation power of the second window as input, and take the historical glass temperature of the second window as output, to train a first model.

[0192] It should be noted that the model training process has been described in the above model processing embodiment, and can be understood in combination with the above embodiment, and no further description is given.

[0193] Figure 10 is a flowchart of a model training method provided by an embodiment of the present application, as shown in Figure 10 , the method comprises:

[0194] Step S1001, collect second historical sample data, the second historical sample data comprising historical environmental information, pre-collected historical glass temperature of the second window and corresponding historical defogging data, the historical defogging data comprising expected defogging setting information of the user under the corresponding historical conditions;

[0195] Step S1002, determining the light radiation power of the first window according to the historical light intensity and the window structure information of the first window, and predicting the historical glass temperature of the first window according to the light radiation power of the first window through a first model;

[0196] Step S1003, training a second model by taking the historical environment information, the window information, the historical glass temperature of the first window and the historical glass temperature of the second window as inputs, and taking the expected defogging setting information as output.

[0197] It should be noted that the model training process has been described in the above model processing embodiment, which can be understood in combination with the above embodiment, and the related description will not be repeated here.

[0198] Figure 11 A structure diagram of a defogging processing device based on a fusion model provided by an embodiment of the present application is shown in FIG. 11. Figure 11 As shown in FIG. 11, the device 1100 includes:

[0199] The acquisition module 1101 is configured to acquire environment information and window information of a vehicle, wherein the environment information includes an indoor temperature, an outdoor temperature and a light intensity, and the window information includes a glass humidity of a window, and the window includes a first window and a second window.

[0200] The first determination module 1102 is configured to determine the light radiation power of the first window according to the light intensity and the window structure information of the first window.

[0201] The first model processing module 1103 is configured to process the indoor temperature, the outdoor temperature and the light radiation power of the first window according to a preset first model to obtain the glass temperature of the first window, wherein the first model is a model trained based on historical glass temperature data of the second window and used for predicting the glass temperature of the first window.

[0202] The second model processing module 1104 is configured to process the environment information, the window information, the glass temperature of the first window and the pre-acquired glass temperature of the second window according to a preset second model to obtain defogging recommendation information, wherein the second model is a model trained based on historical defogging data and used for generating the defogging recommendation information.

[0203] In an embodiment, the first window includes a side window of the vehicle, and the second window includes a front windshield of the vehicle.

[0204] In an embodiment, the vehicle window structure information comprises an installation angle and a surface area of the vehicle window glass; the first determining module 1102 comprises: a relative light intensity determining unit configured to determine a relative light intensity of the first vehicle window according to the installation angle of the first vehicle window and the light intensity; and a first power determining unit configured to determine a light radiation power of the first vehicle window according to the relative light intensity and the surface area.

[0205] In an embodiment, the second power determining module is further configured to determine an environmental heat radiation power of the first vehicle window according to an indoor temperature, an outdoor temperature and the surface area.

[0206] The first power determining unit is specifically configured to: determine an initial light radiation power of the first vehicle window according to the relative light intensity and the surface area; and obtain the light radiation power according to the environmental heat radiation power and the initial light radiation power.

[0207] In an embodiment, the second model comprises a preprocessing module, a risk prediction module and a recommendation module. The second model processing module 1104 comprises:

[0208] The first processing unit is configured to process the glass temperature and the glass humidity of the first vehicle window and the glass temperature and the glass humidity of the second vehicle window respectively by the preprocessing module to obtain a temperature and humidity change of the first vehicle window and a temperature and humidity change of the second vehicle window.

[0209] The second processing unit is configured to process at least one of the glass temperature, the glass humidity and the temperature and humidity change information of the first vehicle window, and at least one of the glass temperature, the glass humidity and the temperature and humidity change information of the second vehicle window by the risk prediction module to determine a respective fogging grade of the first vehicle window and the second vehicle window.

[0210] The recommendation generating unit is configured to generate defogging recommendation information corresponding to the first vehicle window and the second vehicle window according to the fogging grade, the environmental information and the vehicle window information by the recommendation module.

[0211] In an embodiment, the first processing unit is specifically configured to: perform first-order derivation on the glass temperature and the glass humidity of the first vehicle window respectively to obtain the temperature and humidity change information of the first vehicle window; and perform first-order derivation on the glass temperature and the glass humidity of the first vehicle window respectively to obtain the temperature and humidity change information of the second vehicle window.

[0212] The second processing unit is specifically configured to: determine the fogging grade of the first vehicle window according to at least one of the glass temperature, the glass humidity and the temperature and humidity change information of the first vehicle window; and determine the fogging grade of the second vehicle window according to at least one of the glass temperature, the glass humidity and the temperature and humidity change information of the second vehicle window.

[0213] In an embodiment, the fogging level of the first vehicle window is determined according to the glass temperature and the temperature change information thereof, and the glass humidity and the humidity change information thereof, specifically: according to the temperature and humidity change information of the first vehicle window, a fogging probability change coefficient corresponding to the temperature and humidity change information is found from a preset mapping table, to obtain the fogging probability change coefficient of the first vehicle window; wherein the mapping table comprises fogging probability change coefficients corresponding to different temperature and humidity change information; and the fogging level of the first vehicle window is determined according to at least one of the glass temperature, the glass humidity, and the fogging probability change coefficient of the first vehicle window.

[0214] In an embodiment, the fogging level of the second vehicle window is determined according to the temperature change information and the humidity change information of the second vehicle window, specifically: according to the temperature and humidity change information of the second vehicle window, a fogging probability change coefficient corresponding to the temperature and humidity change information is found from a preset mapping table, to obtain the fogging probability change coefficient of the second vehicle window; and the fogging level of the second vehicle window is determined according to at least one of the glass temperature, the glass humidity, and the fogging probability change coefficient of the second vehicle window.

[0215] In an embodiment, the device further comprises:

[0216] The control module is configured to control the target actuator to perform a corresponding defogging operation according to the defogging recommendation information; wherein the defogging recommendation information comprises one of the following recommendation information: opening the first vehicle window and / or the second vehicle window, adjusting the ventilation setting, and adjusting the air conditioning setting; and the target actuator comprises at least one of the following: a vehicle window control system, a ventilation system, and an air conditioning system.

[0217] In an embodiment, the control module comprises:

[0218] The interaction unit is configured to display the defogging recommendation information through the center control screen, and / or prompt the defogging recommendation information through the audio device;

[0219] The response unit is configured to determine defogging setting information in response to a touch operation on the center control screen, and / or a voice input operation on the audio device; wherein the defogging setting information is the same as or different from the defogging recommendation information.

[0220] The defogging unit is configured to control the target actuator to perform a corresponding defogging operation according to the defogging setting information.

[0221] In an embodiment, the device further comprises:

[0222] The monitoring module is configured to monitor trend information of humidity change and / or trend information of temperature change of the vehicle window in real time.

[0223] The control module is further configured to control the target actuator to stop the current defogging operation in response to the trend information of the humidity change and / or the trend information of the temperature change being lower than a preset threshold.

[0224] In an embodiment, the environmental information further comprises vehicle driving information and / or rainfall information, and the window information further comprises a window position.

[0225] Figure 12 A structural schematic diagram of a model training device provided by an embodiment of the present application is shown in FIG. 12. Figure 12 As shown in FIG. 12, the device 1200 comprises:

[0226] A first collection module 1201 is configured to collect first historical sample data in a historical preset time period, the first historical sample data comprising historical indoor temperature, historical outdoor temperature, historical light intensity and historical glass temperature of a second window in the historical preset time period.

[0227] A second determination module 1202 is configured to determine light radiation power of the second window according to the historical light intensity and window structure information of the second window.

[0228] A first training module 1203 is configured to take the historical indoor temperature, the historical outdoor temperature, the historical light radiation power of the second window as input, and take the historical glass temperature of the second window as output, to train a first model.

[0229] Figure 13 Another model training device provided by an embodiment of the present application is shown in FIG. 13. Figure 13 As shown in FIG. 13, the device 1300 comprises:

[0230] A second collection module 1301 is configured to collect second historical sample data, the second historical sample data comprising historical environmental information, historical glass temperature of a second window pre-collected, and corresponding historical defogging data, the historical defogging data comprising expected defogging setting information of a user under corresponding historical conditions.

[0231] A third determination module 1302 is configured to determine light radiation power of a first window according to the historical light intensity and window structure information of the first window, and to predict historical glass temperature of the first window according to the light radiation power of the first window by using the first model.

[0232] A second training module 1302 is configured to take the historical environmental information, window information, historical glass temperature of the first window and historical glass temperature of the second window as input, and take expected defogging setting information as output, to train a second model.

[0233] Figure 14 A structural schematic diagram of a defogging processing system based on a fusion model provided by an embodiment of the present application is shown in FIG. 14.Figure 14 As shown in the figure, the system 1400 includes a cloud 1401, a central controller 1402 connected in communication with the cloud 1401, a vehicle controller 1403 connected in communication with the central controller 1402, and at least one sensor 1404 connected in communication with the vehicle controller 1403; wherein,

[0234] The cloud 1401 is configured to store historical data of the vehicle, the historical data including historical glass temperature data and / or historical defogging data of the second window.

[0235] The at least one sensor 1404 is configured to collect environmental information and window information of the vehicle.

[0236] The central controller 1402 is configured to train a first model according to the historical glass temperature data of the second window, and train a second model according to the historical defogging data.

[0237] The vehicle controller 1403 is configured to perform the defogging processing method based on the fusion model provided in the above method embodiments.

[0238] Optionally, the system further includes a central control screen and / or an audio device connected in communication with the vehicle controller; wherein the central control screen is configured to display defogging recommendation information, and / or the audio device is configured to prompt the defogging recommendation information.

[0239] Figure 15 A structural schematic diagram of a vehicle provided by an embodiment of the present application is shown in the figure, Figure 15 As shown in the figure, the vehicle 1500 includes a memory 1501, a processor 1502, and a communication interface 1503.

[0240] The memory 1501 stores computer execution instructions.

[0241] The communication interface 1503 receives historical data transmitted by the cloud, the historical data including historical glass temperature data and / or historical defogging data of the second window.

[0242] The processor 1502 executes the computer execution instructions stored in the memory 1501, so that the processor 1502 performs the defogging processing method based on the fusion model provided in the above method embodiments, or the model training method provided in the above method embodiments.

[0243] An embodiment of the present application also provides a computer storage medium, a computer readable storage medium stores computer execution instructions, and the computer execution instructions are executed by a processor to implement the method corresponding to any of the above embodiments.

[0244] The above-mentioned readable storage medium can be realized by any type of volatile or nonvolatile storage devices 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 storage, flash memory, magnetic disk or optical disk. The readable storage medium can be any available medium that can be accessed by a general or special purpose computer.

[0245] An exemplary readable storage medium is coupled to the processor so that the processor can read information from the readable storage medium and write information to the readable storage medium. Of course, the readable storage medium can also be a part of the processor. The processor and the readable storage medium can be located in an application specific integrated circuit (ASIC). Of course, the processor and the readable storage medium can also exist as discrete components in a device.

[0246] The division of units is only a logical functional division, and in actual implementation, there can be another division manner. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units shown or discussed can be indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0247] The units described as separate components can or can not be physically separated, and the components shown as units can or can not be physical units, that is, they can be located in one place or distributed on multiple network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiment.

[0248] In addition, the functional units in each embodiment of the present application can be integrated in one processing unit, or each unit can be physically present separately, or two or more units can be integrated in one unit.

[0249] If the functions are implemented in the form of software function units and sold or used as independent products, they can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application or the parts of the technical solutions that essentially contribute to the prior art or the parts of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the embodiments of the method of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various media that can store program codes.

[0250] It can be understood by those skilled in the art that all or part of the steps of the above-mentioned method embodiments can be completed by program instruction related hardware. The aforementioned program can be stored in a computer readable storage medium. When the program is executed, the steps of the above-mentioned method embodiments are executed. The aforementioned storage medium includes: ROM, RAM, magnetic disk or optical disk, and various media that can store program codes.

[0251] Finally, it should be noted that other embodiments of the present application will be readily apparent to those skilled in the art upon considering the description set forth herein. The present application is intended to cover any variations, uses, or adaptations of the application following, in general, the principles of the application and including such departures from the present disclosure as come within known or customary practice in the art to which the application pertains and as can be applied to the essential features herein set forth and fall within the scope of the application. The scope of the present application is limited only by the claims appended hereto.

[0252] The above embodiments are only preferred embodiments of the present application, and the protection scope of the present application is not limited thereto. Any equivalent replacement or transformation of the present application based on the present application is within the protection scope of the present application.

Claims

1. A defogging method based on a fusion model, characterized by, The method comprises: obtaining environment information and window information of a vehicle, wherein the environment information comprises an indoor temperature, an outdoor temperature and an illumination intensity, and the window information comprises a glass humidity of a window and / or window structure information, and the window comprises a first window and a second window; determining an illumination radiation power of the first window according to the illumination intensity and the window structure information of the first window; processing the indoor temperature, the outdoor temperature and the illumination radiation power of the first window according to a preset first model to obtain a glass temperature of the first window, wherein the first model is a model trained based on historical glass temperature data of the second window and used for predicting the glass temperature of the first window; processing the environment information, the window information, the glass temperature of the first window and a pre-acquired glass temperature of the second window according to a preset second model to obtain defogging recommendation information, wherein the second model is a model trained based on historical defogging data and used for generating defogging recommendation information.

2. The method of claim 1, wherein, The first window comprises a side window of the vehicle, and the second window comprises a front windshield of the vehicle.

3. The method according to claim 1 or 2, characterized in that, The window structure information comprises an installation angle and a surface area of the window glass, and the determination of the illumination radiation power of the first window according to the illumination intensity and the window structure information of the first window comprises: determining a relative illumination intensity of the first window according to the installation angle of the first window and the illumination intensity; and determining the illumination radiation power of the first window according to the relative illumination intensity and the surface area.

4. The method of claim 3, wherein, The method further comprises: determining an environmental thermal radiation power of the first window according to the indoor temperature, the outdoor temperature and the surface area; and the determination of the illumination radiation power of the first window according to the relative illumination intensity and the surface area comprises: determining an initial illumination radiation power of the first window according to the relative illumination intensity and the surface area; and obtaining the illumination radiation power according to the environmental thermal radiation power and the initial illumination radiation power.

5. The method according to claim 1 or 2, characterized in that, The second model comprises a preprocessing module, a risk prediction module and a recommendation module, and the processing of the environment information, the window information, the glass temperature of the first window and the pre-acquired glass temperature of the second window according to the preset second model to obtain the defogging recommendation information comprises: processing the glass temperature and the glass humidity of the first window and the glass temperature and the glass humidity of the second window respectively through the preprocessing module to obtain a temperature and humidity change of the first window and a temperature and humidity change of the second window; processing at least one of the glass temperature, the glass humidity and the temperature and humidity change information of the first window and at least one of the glass temperature, the glass humidity and the temperature and humidity change information of the second window through the risk prediction module to determine a respective corresponding fogging grade of the first window and the second window; and generating defogging recommendation information corresponding to the first window and the second window according to the fogging grade, the environment information and the window information through the recommendation module.

6. The method of claim 5, wherein, The processing of the glass temperature and the glass humidity of the first vehicle window and the glass temperature and the glass humidity of the second vehicle window respectively comprises: The first-order derivative of the glass temperature and the glass humidity of the first vehicle window is performed to obtain the temperature and humidity change information of the first vehicle window; The first-order derivative of the glass temperature and the glass humidity of the first vehicle window is performed to obtain the temperature and humidity change information of the second vehicle window; The processing of at least one of the glass temperature, the glass humidity and the temperature and humidity change information of the first vehicle window, and at least one of the glass temperature, the glass humidity and the temperature and humidity change information of the second vehicle window comprises: According to at least one of the glass temperature, the glass humidity and the temperature and humidity change information of the first vehicle window, the fogging grade of the first vehicle window is determined; and according to at least one of the glass temperature, the glass humidity and the temperature and humidity change information of the second vehicle window, the fogging grade of the second vehicle window is determined.

7. The method of claim 6, wherein, The determination of the fogging grade of the first vehicle window according to the glass temperature and the temperature change information thereof, and the glass humidity and the humidity change information thereof comprises: According to the temperature and humidity change information of the first vehicle window, a fogging probability change coefficient corresponding to the temperature and humidity change information is searched from a preset mapping table to obtain the fogging probability change coefficient of the first vehicle window; wherein the mapping table comprises fogging probability change coefficients corresponding to different temperature and humidity change information. The determination of the fogging grade of the first vehicle window according to at least one of the glass temperature, the glass humidity and the fogging probability change coefficient of the first vehicle window.

8. The method of claim 6, wherein, The determination of the fogging grade of the second vehicle window according to the temperature change information and the humidity change information thereof comprises: According to the temperature and humidity change information of the second vehicle window, a fogging probability change coefficient corresponding to the temperature and humidity change information is searched from a preset mapping table to obtain the fogging probability change coefficient of the second vehicle window. The determination of the fogging grade of the second vehicle window according to at least one of the glass temperature, the glass humidity and the fogging probability change coefficient of the second vehicle window.

9. The method of any one of claims 1, 2, 4, 6-8, wherein, Further comprising: According to the defogging recommendation information, a target actuator is controlled to perform a corresponding defogging operation; wherein the defogging recommendation information comprises one of the following recommendation information: opening the first vehicle window and / or the second vehicle window, adjusting the ventilation setting and adjusting the air conditioning setting; the target actuator comprises at least one of the following: a vehicle window control system, a ventilation system and an air conditioning system.

10. The method of claim 9, wherein, The control of the target actuator to perform the corresponding defogging operation according to the defogging recommendation information comprises: The defogging recommendation information is displayed through a center control screen, and / or the defogging recommendation information is prompted through an audio device; In response to a touch operation on the center control screen, and / or a voice input operation on the audio device, defogging setting information is determined, the defogging setting information being the same as or different from the defogging recommendation information; According to the defogging setting information, a target actuator is controlled to perform a corresponding defogging operation.

11. The method of claim 9, wherein, Further comprising: monitor trend information of humidity change and / or trend information of temperature change of the vehicle window in real time; in response to monitoring that the trend information of humidity change and / or the trend information of temperature change is lower than a preset threshold, control the target actuator to stop the current defogging operation.

12. The method according to any one of claims 1, 2, 4, 6-8, 10, 11, characterized in that, The environmental information further comprises vehicle driving information and / or rainfall information, and the vehicle window information further comprises vehicle window position.

13. A model training method, comprising: comprising: collecting first historical sample data, the first historical sample data comprising historical vehicle interior temperature, historical vehicle exterior temperature, historical light intensity and historical glass temperature of a second vehicle window in a historical preset time period; determining light radiation power of the second vehicle window according to the historical light intensity and vehicle window structure information of the second vehicle window; training a first model by taking the historical vehicle interior temperature, the historical vehicle exterior temperature, the historical light radiation power of the second vehicle window as input and the historical glass temperature of the second vehicle window as output.

14. A model training method, comprising: comprising: collecting second historical sample data, the second historical sample data comprising historical environmental information, historical glass temperature of a second vehicle window pre-collected and corresponding historical defogging data, the historical defogging data comprising expected defogging setting information of a user under corresponding historical conditions; determining light radiation power of the first vehicle window according to the historical light intensity and vehicle window structure information of the first vehicle window, and predicting historical glass temperature of the first vehicle window by the first model according to the light radiation power of the first vehicle window; training a second model by taking the historical environmental information, vehicle window information, the historical glass temperature of the first vehicle window and the historical glass temperature of the second vehicle window as input and the expected defogging setting information as output.

15. A defogging treatment device based on a fusion model, characterized by, comprising: an acquisition module configured to acquire environmental information and vehicle window information of a vehicle, wherein the environmental information comprises vehicle interior temperature, vehicle exterior temperature and light intensity, and the vehicle window information comprises glass humidity of a vehicle window, the vehicle window comprising a first vehicle window and a second vehicle window; a first determination module configured to determine light radiation power of the first vehicle window according to the light intensity and vehicle window structure information of the first vehicle window; a first model processing module configured to process the vehicle interior temperature, the vehicle exterior temperature and the light radiation power of the first vehicle window according to a preset first model to obtain glass temperature of the first vehicle window, wherein the first model is a model trained based on historical glass temperature data of the second vehicle window and used to predict the glass temperature of the first vehicle window; a second model processing module configured to process the environmental information, the vehicle window information, the glass temperature of the first vehicle window and pre-collected glass temperature of the second vehicle window according to a preset second model to obtain defogging recommendation information, wherein the second model is a model trained based on historical defogging data and used to generate defogging recommendation information.

16. A model training apparatus, comprising: comprising: a first acquisition module configured to acquire first historical sample data in a historical preset time period, the first historical sample data comprising historical vehicle interior temperature, historical vehicle exterior temperature, historical light intensity and historical glass temperature of a second vehicle window in the historical preset time period; a second determining module, configured to determine the light radiation power of the second window according to the historical light intensity and the window structure information of the second window; a first training module, configured to train a first model by taking the historical indoor temperature, the historical outdoor temperature, and the historical light radiation power of the second window as inputs and taking the historical glass temperature of the second window as output.

17. A model training apparatus, comprising: comprising: a second collecting module, configured to collect second historical sample data, the second historical sample data comprising historical environment information, historical glass temperature of the second window collected in advance, and corresponding historical defogging data, the historical defogging data comprising expected defogging setting information of a user under corresponding historical conditions; a third determining module, configured to determine the light radiation power of the first window according to the historical light intensity and the window structure information of the first window, and predict the historical glass temperature of the first window according to the light radiation power of the first window by using the first model; a second training module, configured to train a second model by taking the historical environment information, the window information, the historical glass temperature of the first window, and the historical glass temperature of the second window as inputs and taking the expected defogging setting information as output.

18. A defogging processing system based on a fusion model, characterized by, comprising a cloud, a central controller in communication connection with the cloud, a vehicle controller in communication connection with the central controller, and at least one sensor in communication connection with the vehicle controller; wherein the cloud is configured to store historical data of a vehicle, the historical data comprising historical glass temperature data of the second window and / or historical defogging data; the at least one sensor is configured to collect environment information and window information of the vehicle; the central controller is configured to train a first model according to the historical glass temperature data of the second window and train a second model according to the historical defogging data; the vehicle controller is configured to perform the defogging processing method based on the fusion model according to any one of claims 1-12.

19. The system of claim 18, wherein, further comprising a central control screen and / or an audio device in communication connection with the vehicle controller; wherein the central control screen is configured to display the defogging recommendation information, and / or the audio device is configured to prompt the defogging recommendation information.

20. A vehicle characterized by comprising a memory, a processor, and a communication interface; the memory stores computer execution instructions; the communication interface receives historical data transmitted by the cloud, the historical data comprising historical glass temperature data of the second window and / or historical defogging data; the processor executes the computer execution instructions stored in the memory, so that the processor performs the defogging processing method based on the fusion model according to any one of claims 1-12 or the model training method according to claim 13 or 14.

Images