Anti-fog control method, storage medium, program product, electronic equipment and vehicle

Abstract

The invention provides an anti-fog control method, a storage medium, a program product, electronic equipment and a vehicle, and the method comprises the steps: obtaining multi-source fogging related data after an anti-fog function is started, and the multi-source fogging related data comprises meteorological data, vehicle state data and image data of a camera; inputting the image data into a pre-constructed fogging scene recognition model to obtain scene label information; splicing the meteorological data, the vehicle state data and the scene label information into a high-dimensional feature vector, and inputting the high-dimensional feature vector into a pre-constructed fogging prediction model to output a fogging risk value; and generating a demisting control instruction according to the fogging risk value and executing the demisting control instruction. Compared with the prior art, the scheme does not depend on a special temperature and humidity sensor and extra hardware cost support, multi-source heterogeneous data is used for fusion, and a plurality of prediction models are integrated to improve the accuracy of fogging prediction.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent vehicles, and in particular to a defogging control method, a storage medium, a program product, an electronic device and a vehicle. BACKGROUND

[0002] Fogging of the windshield is one of the factors leading to traffic accidents, especially in adverse weather conditions, fog will seriously affect the driver's vision, increasing the risk of traffic accidents.

[0003] In related technologies, the dew point temperature and the glass temperature difference are mainly monitored by a special temperature and humidity sensor to trigger a preset program to start defogging. However, in actual application, such as some vehicles arrange the temperature and humidity sensor on the surface of the window in a device of the intelligent driving front view camera, which will be affected by the heating of the device, causing a large error between the displayed data and the actual situation, thus leading to inaccurate fogging prediction and unable to truly reflect the fogging situation. SUMMARY

[0004] The present application aims to at least solve one of the technical problems existing in the prior art. To this end, the present application provides a defogging control method, a storage medium, a program product, an electronic device and a vehicle. In the present application, after the defogging function is turned on, multi-source fogging related data is obtained, including meteorological data, vehicle state data and image data of the camera; the image data is input into a pre-constructed fogging scene recognition model to obtain scene label information; and the meteorological data, vehicle state data and scene label information are spliced into a high-dimensional feature vector and input into a pre-constructed fogging prediction model to output a fogging risk value; a defogging control instruction is generated according to the fogging risk value and executed. Compared with related technologies, the present application does not rely on special temperature and humidity sensors and additional hardware cost support, but uses multi-source heterogeneous data (including vehicle sensor data, intelligent driving front view camera image data and meteorological data) for fusion, and integrates multiple prediction models to improve the accuracy of fogging prediction.

[0005] To achieve the above-mentioned purpose, the present application adopts the following technical solutions:

[0006] In a first aspect, the present application provides a defogging control method, comprising: obtaining multi-source fogging related data after the defogging function is turned on, the multi-source fogging related data including meteorological data, vehicle state data and image data of the camera; inputting the image data into a pre-constructed fogging scene recognition model to obtain scene label information; and splicing the meteorological data, the vehicle state data and the scene label information into a high-dimensional feature vector and inputting into a pre-constructed fogging prediction model to output a fogging risk value; generating a defogging control instruction according to the fogging risk value and executing.

[0007] In some embodiments of the present application, the generating the defogging control instruction according to the fogging risk value specifically comprises: generating the defogging control instruction according to the fogging risk value and a fogging threshold value.

[0008] In some embodiments of the present application, the generating the defogging control instruction according to the fogging risk value and the fogging threshold value specifically comprises: generating a first-level defogging instruction if the fogging risk value is greater than a first threshold value and less than or equal to a second threshold value; or generating a second-level defogging instruction if the fogging risk value is greater than the second threshold value and less than or equal to a third threshold value; or generating a third-level defogging instruction if the fogging risk value is greater than the third threshold value; wherein the second threshold value is greater than the first threshold value and less than the third threshold value.

[0009] In some embodiments of the present application, the method further comprises: if the fogging risk value is less than the first threshold value, continuously acquiring the multi-source fogging-related data.

[0010] In some embodiments of the present application, the method further comprises: continuously monitoring a real-time fogging risk value when the defogging operation is performed; and dynamically adjusting the defogging instruction level according to the real-time fogging risk value.

[0011] In some embodiments of the present application, the dynamically adjusting the defogging instruction level according to the real-time fogging risk value comprises: if the fogging risk value remains unchanged within a set time period, maintaining the current-level defogging instruction or switching to a higher-level defogging instruction; or if the fogging risk value continuously rises within the set time period, switching the current-level defogging instruction to a higher-level defogging instruction; or if the fogging risk value decreases to a lower-level threshold value within the set time period, switching the current-level defogging instruction to a lower-level defogging instruction or turning off the defogging function.

[0012] In some embodiments of the present application, the method further comprises: controlling the air conditioner temperature and / or the air speed to be within a safety threshold range during execution of the defogging control instruction.

[0013] In a second aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, causes a computer to implement the method according to the first aspect.

[0014] In a third aspect, the present application provides a computer program product having instructions stored thereon, wherein the instructions, when executed by a computer, cause the computer to implement the method according to the first aspect.

[0015] In a fourth aspect, the present application provides an electronic device comprising: a memory having a computer program stored thereon; and a processor configured to execute the computer program in the memory to implement the method according to the first aspect.

[0016] In a fifth aspect, the application provides a vehicle, comprising: the electronic device according to the fourth aspect; or a processor, wherein the processor is configured to execute the method according to the first aspect.

[0017] The vehicle and the electronic device have the same advantages as the control method, and details are not repeated here.

[0018] Additional aspects and advantages of the application will be described in the following description, will become apparent from the following description, or will be learned by practice of the application. BRIEF DESCRIPTION OF DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the embodiments of the application, the following will briefly introduce the drawings needed to be used in the embodiment description. Obviously, the drawings in the following description only show some embodiments of the application, and for those skilled in the art, other drawings can also be obtained from these drawings without creative labor.

[0020] In order to more completely understand the application and its beneficial effects, the following will be described in conjunction with the drawings, wherein the same reference numerals in the following description represent the same parts.

[0021] Figure 1 is a flowchart of a defogging control method according to an embodiment of the application;

[0022] Figure 2 is another flowchart of a defogging control method according to an embodiment of the application;

[0023] Figure 3 is a structural diagram of a defogging control system according to an embodiment of the application;

[0024] Figure 4 is a specific implementation diagram of a defogging trigger module based on Figure 3 provided according to an embodiment of the application;

[0025] Figure 5 is a specific implementation diagram of a multi-source data acquisition and preprocessing module based on Figure 3 provided according to an embodiment of the application;

[0026] Figure 6 is a specific implementation diagram of a defogging detection module based on Figure 3 provided according to an embodiment of the application;

[0027] Figure 7 is a schematic diagram of a defogging scene recognition model according to an embodiment of the application;

[0028] Figure 8This is a schematic diagram of a fogging detection model provided according to an embodiment of the present invention;

[0029] Figure 9 Based on the embodiments of the present invention Figure 3 A schematic diagram illustrating the specific implementation of the provided defogging control module;

[0030] Figure 10 This is a schematic diagram illustrating the generation of dynamic defogging commands according to an embodiment of the present invention;

[0031] Figure 11 This is a schematic diagram of the structure of an electronic device provided according to an embodiment of the present invention. Detailed Implementation

[0032] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the protection scope of this application.

[0033] In related technologies, the dew point temperature and the temperature difference between the glass are mainly monitored by dedicated temperature and humidity sensors to trigger a preset program to start defogging. However, in practical applications, such as some vehicles that place the temperature and humidity sensor on the surface of the window within a device of the intelligent driving forward-looking camera, the heat generated by the device can cause a large error between the displayed data and the actual situation. Therefore, the fogging prediction is inaccurate and cannot truly reflect the fogging situation.

[0034] To address this, this invention proposes an anti-fog control method, storage medium, program product, electronic device, and vehicle. In this solution, after the anti-fog function is activated, multi-source fogging-related data is acquired, including meteorological data, vehicle status data, and camera image data. The image data is input into a pre-built fogging scene recognition model to obtain scene label information. The meteorological data, vehicle status data, and scene label information are concatenated into a high-dimensional feature vector, which is then input into a pre-built fogging prediction model to output a fogging risk value. Based on the fogging risk value, a defogging control command is generated and executed. Compared to related technologies, this solution does not rely on dedicated temperature and humidity sensors or additional hardware costs. It utilizes multi-source heterogeneous data (including vehicle sensor data, intelligent driving forward-looking camera image data, and meteorological data) for fusion and integrates multiple prediction models to improve the accuracy of fogging prediction.

[0035] The present invention will now be described in further detail with reference to the embodiments.

[0036] like Figure 1 The diagram shown is a flowchart of an anti-fog control method provided in an embodiment of the present invention. The method includes:

[0037] 101、acquiring multi-source fogging related data after the anti-fog function is turned on, the multi-source fogging related data including weather data, vehicle state data and camera image data.

[0038] For example, the weather data includes navigation weather data and vehicle weather data. The step 101 specifically includes the following contents: when detecting that the anti-fog function of the vehicle is turned on, determining whether the navigation of the vehicle is turned on, if the navigation of the vehicle is turned on, acquiring multi-source data such as navigation weather data, vehicle state data and camera image data, and if the navigation of the vehicle is not turned on, acquiring multi-source data such as vehicle weather data, vehicle state data and camera image data. The weather data acquired through the navigation system of the vehicle or the cloud API includes relative humidity, ambient temperature, rainfall and pressure change rate. The vehicle state data can be collected through CAN bus, including vehicle speed, air conditioning set temperature, vehicle interior temperature, vehicle exterior temperature, interior and exterior temperature difference, and vehicle window opening and closing state, etc. The intelligent driving front-view camera is instructed to collect images of the windshield area, and the images cover the upper area of the front windshield to ensure that the initial area of the glass surface condensation fog is included.

[0039] 102、inputting the image data into a pre-constructed fogging scene recognition model to obtain scene label information, and concatenating the weather data, the vehicle state data and the scene label information into a high-dimensional feature vector to input into a pre-constructed fogging prediction model to output a fogging risk value.

[0040] Optionally, before the step 102, the method further includes pre-processing and synchronizing the multi-source data, specifically as follows: removing noise values and abnormal values in the vehicle state data, removing temperature and humidity noise from the image data using filtering, performing distortion correction and grayscale processing to reduce the calculation burden. In addition, the multi-source data is aligned with the time stamp to ensure fusion consistency.

[0041] Further optionally, the pre-processed image data is input into a pre-constructed fogging scene recognition model to obtain scene label information, and the weather data, the pre-processed vehicle state data and the scene label information are input into a pre-constructed fogging prediction model to output a fogging risk value. Through the pre-processing and synchronization of the data, the accuracy of the fogging prediction can be further improved.

[0042] 103、generating a defogging control instruction according to the fogging risk value and executing the defogging control instruction.

[0043] For example, the step 103 specifically includes the following contents: 103a, generating a defogging control instruction according to the fogging risk value and a fogging threshold value.

[0044] Further optionally, the step 103a specifically includes the following: 103a1, generating a first-level defogging instruction if the fogging risk value is greater than the first threshold value and less than or equal to the second threshold value; or 103a2, generating a second-level defogging instruction if the fogging risk value is greater than the second threshold value and less than or equal to the third threshold value; or 103a3, generating a third-level defogging instruction if the fogging risk value is greater than the third threshold value.

[0045] In the above embodiment, by constructing a multi-level defogging mechanism, it is possible to realize non-sensing intervention before the fog affects the field of view, dynamically adjust the air volume, temperature and air outlet mode according to the fogging severity, and intelligently cooperate with multiple devices such as air conditioning systems, internal and external circulation systems, etc. While ensuring the defogging effect, the system energy consumption is significantly reduced.

[0046] In the above embodiment, by constructing a multi-level defogging mechanism, it is possible to realize non-sensing intervention before the fog affects the field of view, dynamically adjust the air volume, temperature and air outlet mode according to the fogging severity, and intelligently cooperate with multiple devices such as air conditioning systems, internal and external circulation systems, etc. While ensuring the defogging effect, the system energy consumption is significantly reduced.

[0047] Further optionally, the method further includes: 103a4, if the fogging risk value is less than the first threshold value, continuously acquiring the multi-source fogging related data.

[0048] In the above embodiment, if it is judged that the fogging risk value is less than the first threshold value, it means that it is in a safe state, and at this time, defogging is not needed, so the flow returns to the step of continuously acquiring the multi-source fogging related data, which can realize real-time monitoring of the fogging data.

[0049] Optionally, the method further includes: 104, continuously monitoring the real-time fogging risk value when the defogging operation is performed; and 105, dynamically adjusting the defogging instruction level according to the real-time fogging risk value.

[0050] In the above embodiment, the system updates the fogging risk value once every a seconds (for example, 5 seconds), realizing closed-loop feedback control. If the current is a second-level instruction, but the risk value decreases to below b (below the second threshold value) in two consecutive samplings, it is automatically downgraded to a first-level instruction; if the risk value rises to above c (for example, 0.8), it is upgraded to a third-level instruction, realizing "on-demand regulation and dynamic response".

[0051] Optionally, the step 105 specifically includes the following: 105a, if the fogging risk value remains unchanged within a set time period, the current level defogging instruction is maintained or switched to a higher level defogging instruction; or 105b, if the fogging risk value continues to rise within a set time period, the current level defogging instruction is switched to a higher level defogging instruction; or 105c, if the fogging risk value decreases to a lower level threshold value within a set time period, the current level defogging instruction is switched to a lower level defogging instruction or the defogging function is turned off.

[0052] Specifically, the embodiment of the present application also needs to monitor the fogging risk value in real time when performing the defogging operation related to step 103 or after performing the defogging operation related to step 103, and dynamically adjusts the defogging instruction level according to the real-time fogging risk value, for example: the default time period can be 60 seconds, and the user can also adjust it to 30 seconds or 90 seconds through the central control screen. The risk value change judgment logic is: if the fluctuation amplitude is less than 5% (such as 0.65→0.68→0.66) within 60 seconds, the current instruction is maintained to avoid false triggering due to instantaneous disturbance; if the risk value increases by ≥0.05 every 10 seconds (such as 0.6→0.65→0.7), it is determined that the situation is deteriorating, and the instruction is immediately upgraded; if the risk value decreases by ≥0.05 every 10 seconds (such as 0.7→0.65→0.6), it is determined that the situation is improving, and the instruction is downgraded; if the risk value is reduced to below 0.2 and lasts for 30 seconds, the defogging function is automatically turned off, and the default air conditioner settings are restored.

[0053] Optionally, the above method further comprises: 106, controlling the air conditioner temperature and / or the wind speed to be within a safety threshold range during the execution of the defogging control instruction.

[0054] In the above embodiment, for example, the system sets the temperature safety threshold to be 16℃-30℃, if the predicted instruction requires the temperature to be lower than 16℃, it is forced to limit to 16℃ to prevent passengers from being uncomfortable due to excessive cold; the wind speed safety threshold is 1-6 gears, if the instruction requires the wind speed to exceed 6 gears, it is limited to 6 gears to avoid excessive wind noise or air conditioner motor overload. If it is detected that the temperature or wind speed exceeds the safety range, the system triggers the vehicle-mounted screen pop-up window to prompt: “air conditioner output is limited, safety is automatically limited to protect safety”, and records the event to the vehicle log.

[0055] In the above embodiment, by integrating multi-source heterogeneous data from the existing sensors of the vehicle and the camera of the intelligent driving system, the dependence on special temperature and humidity sensors in the traditional scheme is avoided, and the hardware cost and system complexity are reduced. At the same time, by continuously capturing images of the windshield area using the front-view camera, combining external weather and vehicle operating conditions, a prediction model with environmental perception capability is constructed, which significantly improves the accuracy and generalization ability of the fogging risk judgment, especially in scenes with severe light changes, sensor drift or network delay.

[0056] The specific implementation process of the embodiment of the present application will be described in detail based on the above content. The present application provides a vehicle intelligent anti-fogging control method and system based on multi-modal data fusion, the multi-modal data including whole vehicle CAN signal data, intelligent driving front-view camera image data, and weather data; the vehicle including a fuel vehicle and a new energy vehicle; the intelligent anti-fogging control method is based on multi-modal data and multi-model to predict vehicle window fogging, and dynamically executes a defogging control instruction according to different fogging prediction results to realize defogging, and the specific implementation steps are as follows:Figure 2 As shown, the method includes the following steps:

[0057] Step 1: When the anti-fogging function is detected to be enabled, determine whether the vehicle navigation is enabled. If the vehicle navigation is enabled, acquire multiple data sources such as navigation weather data, vehicle CAN signal data, and intelligent driving front-view camera image data. If the vehicle navigation is not enabled, acquire multiple data sources such as vehicle weather data, vehicle CAN signal data, and intelligent driving front-view camera image data.

[0058] Step 2: Synchronize and preprocess the acquired data.

[0059] Step 3: Input the processed image data into the fog scene recognition model and output scene labels.

[0060] Step 4: Input the scene label, processed meteorological data, and vehicle CAN signal data into the fog prediction model and output the fog prediction value.

[0061] Step 5: Determine whether the fogging prediction value has reached the fogging threshold. If not, continue the detection and return to Step 2. If it has, generate different control commands based on the fogging prediction value to perform defogging.

[0062] Step 6: Continuously monitor the defogging effect of Step 5. If the fogging value is less than the fogging threshold, return to Step 2 and update the prediction result. If the fogging value is not within the current defogging command threshold range, it does not meet the requirement to return to Step 5 and regenerate the defogging command. If the fogging value is within the current defogging command threshold range, it meets the requirement to maintain the current defogging command.

[0063] The anti-fog control system provided in this embodiment of the invention includes an anti-fogging trigger module, a data acquisition and preprocessing module, a fogging detection module, and a defogging control module. Specifically, as follows... Figure 3 As shown, where:

[0064] Anti-fogging trigger module S01, this module is used to detect whether the anti-fogging function is activated, such as Figure 4 As shown, this module specifically implements: S11, manually triggering the anti-fogging function to activate; users can activate this function through the intelligent anti-fogging setting in the central control air conditioning settings interface; and S12, triggering the anti-fogging function with a voice command; users can activate the anti-fogging function by speaking a preset voice command (such as "activate anti-fogging function"). The system confirms receipt of the command and execution of the operation through a voice prompt, and the system will reply "anti-fogging function activated". Voice commands allow the driver to operate without taking their hands off the steering wheel, improving driving safety and reducing physical operations by the driver, thus enhancing driving comfort.

[0065] A multi-source data collection and preprocessing module S02, which is used to ensure the quality and compatibility of the input data, involves multi-source data synchronization and cleaning, such as Figure 5 As shown in the figure, this module is specifically used to achieve:

[0066] S21, multi-source data collection collects data including vehicle state data, intelligent driving front-view camera image data, and weather data. The vehicle state data is obtained from the vehicle CAN bus, with a sampling frequency of 1 second, including vehicle speed, outside temperature, inside temperature, co-driver set temperature, left sunlight AD value, right sunlight AD value, air conditioning circulation mode, air conditioning air outlet mode, air conditioning air outlet gear, main driver set temperature, left front window position percentage, left rear window position percentage, right front window position percentage, right rear window position percentage, etc. The intelligent driving front-view camera image data is a continuous video frame collected by the vehicle-mounted intelligent driving front-view camera, which is used to identify easy-fogging environmental conditions such as weather, such as rain, snow, fog, etc., and scenes such as tunnels and basements. The weather data includes real-time weather data for navigation with navigation turned on or real-time weather data for the vehicle with navigation turned off, such as temperature, humidity, precipitation probability, and wind speed, etc.

[0067] S22, data preprocessing and synchronization, removing noise and outliers in the vehicle state data, such as sensor failure or external interference, filling or deleting missing or erroneous data points. For image data, use median filtering to remove temperature and humidity noise, perform distortion correction and grayscale using OpenCV to reduce computational burden. Align multi-source data with timestamps to ensure fusion consistency.

[0068] A fog detection module S03, which integrates multi-source data input to detect the real-time fogging risk of the front windshield, specifically including identifying the current environment and determining whether the current environment is a special fogging scene, and then inputting the environment recognition result and multi-source data into a detection model to output a fogging risk value, such as Figure 6 As shown in the figure, this module is specifically used to:

[0069] S31, obtaining the preprocessed data of the current vehicle, including processed vehicle state data, weather data, and image data.

[0070] S32, fog scene recognition, inputting the image data processed in S31 into a pre-trained fog scene recognition model to output the current scene category. The pre-trained fog scene recognition model is specifically as shown in Figure 7As shown, by collecting massive continuous video frame data of vehicle intelligent driving front view camera, setting corresponding labels, preprocessing and data enhancement, a data set is constructed, and then an initial fogging environment recognition model is constructed, and the data set is used for training, when the model converges, the training is stopped, and a fogging scene recognition model is obtained, which is deployed on the vehicle end, provides an external fogging scene recognition API, and introduces a continuous model evaluation mechanism, regularly uses new data to test and verify the model, and continuously iterates and optimizes the model according to the evaluation result.

[0071] S33, multi-source data fusion, One-Hot encoding of the environment category of S32 fogging scene recognition, S31 obtains vehicle state data and weather data after vehicle preprocessing and splicing into a high-dimensional feature vector, and constructs a window data with a window of 30 seconds and a step of 1 second.

[0072] S34, fog detection, inputting the multi-source data fused in S33 into a pre-trained fog detection model to output a fog risk value. The pre-trained fog detection model is specifically as shown in Figure 8 As shown, by collecting massive data of vehicles in different scenes and preprocessing, a data set is constructed, and then an initial fog detection model is constructed, and the data set is used for training, when the model converges, the training is stopped, and a fog detection model is obtained, which is deployed on the vehicle end, provides an external fog detection API, and introduces a continuous model evaluation mechanism, regularly uses new data to test and verify the model, and continuously iterates and optimizes the model according to the evaluation result.

[0073] A defogging control module S04 is used to automatically adjust the air conditioning system in the vehicle according to the output result of the fog detection module to realize intelligent defogging. As shown Figure 9 The module is specifically used for:

[0074] S41, judging the fogging probability, receiving the fogging probability P output by the fogging detection model S33, presetting three defogging threshold values: a basic fogging threshold value P1 (environment slightly deteriorated, with a tendency to fog), a moderate fogging threshold value P2 (visible mist has appeared, starting to affect the field of view), and a serious fogging threshold value P3 (serious fogging, field of view clarity sharply decreases), judging whether the defogging operation needs to be started, if P>P1, it is determined that intervention is needed, and S42 is entered to generate a corresponding level of defogging instruction. If P ≤ P1, it is determined that it is a safe state, and the defogging is not started, and the process returns to S21 (multi-source data acquisition) for continuous monitoring.

[0075] S42, generating a defogging instruction, the defogging instruction is a dynamic closed loop process, the system will continuously monitor the fogging probability P during the execution of the instruction, and decide whether to strengthen, maintain, weaken or exit the defogging. The defogging instruction includes a first-level defogging instruction, a second-level defogging instruction, and a third-level defogging instruction, as shown Figure 10The specific content is as follows:

[0076] The first-stage defogging instruction is slight defogging, and the fogging probability is between the basic fogging threshold and the medium fogging threshold, that is, P1 < P < P2. The specific control mode mainly slightly increases the air conditioner air speed of the front windshield glass to 10%-20% of the maximum air speed. The temperature is automatically adjusted. If the humidity in the vehicle is high, the air outlet temperature is slightly lower than the ambient temperature (slight cooling) to quickly reduce the air dew point and avoid water vapor condensation. If it is cold outside, the air outlet temperature is slightly higher than the ambient temperature (slight heating) to increase the glass temperature. The current air outlet mode is maintained, and forced switching is not performed, but the "blow foot + front" mode can be preferentially selected to achieve non-conscious intervention.

[0077] The second-stage defogging instruction is medium defogging, and the fogging probability is between the medium fogging threshold and the serious fogging threshold, that is, P2 < P < P3. The front windshield glass has obvious water mist, which causes an observable impact on the field of view. The air conditioner air speed is increased to 40%-60% of the maximum air speed, and the temperature is intelligently adjusted according to the ambient temperature: in a normal / low-temperature environment, the air outlet temperature is quickly increased (hot air defogging); in a high-temperature and high-humidity environment (such as in the rainy season), the air outlet temperature is quickly reduced (cold air dehumidification); forced switching is performed to the front windshield glass defogging mode, and all air flows attack the fogging area.

[0078] The third-stage defogging instruction is serious defogging, and the fogging probability exceeds the serious fogging threshold, that is, P > P3. The front windshield glass has serious fog, the field of view is blurred, and the driving safety is directly threatened. The air outlet temperature is adjusted to the highest (preferably hot air, because hot air has a much stronger ability to carry water than cold air, and can evaporate water at the fastest speed), the air outlet mode is forced to switch to the front windshield glass defogging mode, the circulation mode is forced to lock the external circulation, and the AC compressor is forced to start.

[0079] S43, a defogging operation is performed, a defogging instruction of a corresponding level is executed according to S42, and the air conditioner system in the vehicle is automatically adjusted to perform corresponding operations. The defogging effect is monitored in real time, and the instruction level is adjusted as needed to ensure that the fog on the windshield glass is quickly and effectively removed.

[0080] S44, continuously detecting the fogging risk, continuously receiving the real-time fogging probability evaluation result provided by the fogging detection module when the defogging operation is performed. According to the change of the real-time fogging probability, the defogging instruction level is dynamically adjusted. If the fogging probability is reduced to below a lower threshold, the defogging instruction level is reduced or the defogging function is turned off. If the fogging probability continuously stays at a high threshold or further rises, the current defogging instruction level is maintained or the defogging instruction level is increased.

[0081] S45, a safety protection mechanism, sets a safety threshold for temperature and wind speed to prevent passengers from being uncomfortable or vehicle systems from being damaged due to extreme temperature or wind speed. The vehicle state is monitored in real time during the defogging process, and if an abnormal condition (such as system overheating) is detected, the defogging operation is immediately interrupted and the driver is reminded.

[0082] As Figure 11 The electronic device 700 includes a processor 701 having one or more processing cores, a memory 702 having one or more computer readable storage media, and a computer program stored on the memory 702 and executable on the processor. The processor 701 is electrically connected to the memory 702.

[0083] The processor 701 is the control center of the electronic device 700, and connects various parts of the electronic device 700 through various interfaces and lines. By running or loading software programs and / or units stored in the memory 702 and calling data stored in the memory 702, the processor 701 performs various functions and processes data of the electronic device 700, thereby monitoring the entire electronic device 700. The processor 701 can be a processor CPU (Central Processing Unit), a graphics processor GPU (Graphics Processing Unit), a network processor (Network Processor, NP), etc., and can implement or execute the methods, steps and logic block diagrams disclosed in the embodiments of the present application.

[0084] In the embodiments of the present application, the processor 701 in the electronic device 700 loads one or more application processes corresponding to the computer program into the memory 702 according to the method or steps of the above embodiments, and runs the application program stored in the memory 702 by the processor 701, thereby executing the above method.

[0085] According to the electronic device of the embodiments of the present application, the above-mentioned method is executed to obtain multi-source fogging related data after the anti-fog function is turned on, the multi-source fogging related data including weather data, vehicle state data and image data of the camera; the image data is input into a pre-constructed fogging scene recognition model to obtain scene label information; and the weather data, vehicle state data and scene label information are spliced into a high-dimensional feature vector and input into a pre-constructed fogging prediction model to output a fogging risk value; and a defogging control instruction is generated according to the fogging risk value and executed. Compared with related technologies, the present solution does not rely on special temperature and humidity sensors and additional hardware cost support, and utilizes multi-source heterogeneous data (including vehicle sensor data, intelligent driving front-view camera image data and weather data) for fusion, and integrates multiple prediction models to improve the accuracy of fogging prediction.

[0086] Embodiments of the present application also provide a computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the vehicle control method described above. For example, the computer readable storage medium can be the memory described above including program instructions executable by the processor of the electronic device to implement or execute the methods, steps and logic block diagrams disclosed in the embodiments of the present application.

[0087] Embodiments of the present application also provide a computer program product having stored instructions which, when executed by a computer, cause the computer to implement the vehicle control method described above. For example, the instructions, when executed by the computer, implement or execute the methods, steps and logic block diagrams disclosed in the embodiments of the present application.

[0088] Embodiments of the present application also provide a vehicle comprising the system described above, or the electronic device, or the processor configured to execute the method described above. The vehicle can be a fuel automobile, a plug-in hybrid electric vehicle or a new energy vehicle, and the present specification does not make specific limitations thereto.

[0089] According to the vehicle of the embodiments of the present application, the electronic device or the control system or the controller executes the above method to obtain multi-source fogging related data after the anti-fog function is turned on, the multi-source fogging related data including weather data, vehicle state data and image data of the camera; the image data is input into a pre-constructed fogging scene recognition model to obtain scene label information; and the weather data, the vehicle state data and the scene label information are spliced into a high-dimensional feature vector and input into a pre-constructed fogging prediction model to output a fogging risk value; and a defogging control instruction is generated according to the fogging risk value and executed. Compared with the related art, the present solution does not rely on a special temperature and humidity sensor and additional hardware cost support, utilizes multi-source heterogeneous data (including vehicle sensor data, intelligent driving front-view camera image data and weather data) for fusion, and integrates multiple prediction models to improve the accuracy of fogging prediction.

[0090] The above embodiments are only used to illustrate the technical solutions of the above method applied to the vehicle, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that the method can also be used in a motor train, a train and a ship, etc., and it does not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application.

[0091] In one embodiment, the vehicle can be configured in a fully or partially autonomous driving mode. For example, the vehicle can control itself while in the autonomous driving mode and can determine a current state of the vehicle and its surrounding environment, determine a possible behavior of at least one other vehicle in the surrounding environment, and determine a confidence level corresponding to a likelihood that the other vehicle will perform the possible behavior based on the determined information, control the vehicle based on the determined information. While the vehicle is in the autonomous driving mode, the vehicle can be placed to operate without human interaction.

[0092] In the description of the present application, the terms "first", "second", etc. are used only for the purpose of description, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the technical features indicated. Therefore, the features defined with "first", "second" can explicitly or implicitly include one or more features. In the description of the present application, the meaning of "a plurality of" is two or more, unless otherwise specifically limited.

[0093] In the description of the present application, the description of the terms "one embodiment", "some embodiments", "exemplary embodiment", "example", "specific example", "optional example" or "some examples" means that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present application. In the description of the present application, the exemplary description of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0094] The embodiments, implementation manners and related technical features of the present application can be combined, replaced with each other without conflict.

[0095] The above is only the preferred embodiment of the present application, and does not limit the present application in any form. Although the description of each embodiment in the embodiment of the present application has its own emphasis, the part not described in detail in a certain embodiment can be referred to the related embodiment of other embodiments. Any simple modification, equivalent change and modification made according to the technical essence of the present application to the above embodiments are still within the scope of the technical solution of the present application.

Claims

1. A method for controlling fog, characterized in that, include: After the anti-fog function is turned on, multi-source fogging-related data is acquired, including meteorological data, vehicle status data, and camera image data. The image data is input into a pre-built fog scene recognition model to obtain scene label information; and the meteorological data, the vehicle status data and the scene label information are concatenated into a high-dimensional feature vector and input into a pre-built fog prediction model to output a fog risk value. Generate and execute defogging control commands based on the fogging risk value.

2. The method according to claim 1, characterized in that, The step of generating a defogging control command based on the fogging risk value specifically includes: Defogging control commands are generated based on the fogging risk value and fogging threshold.

3. The method according to claim 2, characterized in that, The step of generating a defogging control command based on the fogging risk value and fogging threshold specifically includes: If the fogging risk value is greater than the first threshold and less than or equal to the second threshold, a level one defogging command is generated; or... If the fogging risk value is greater than the second threshold and less than or equal to the third threshold, a level two defogging command is generated. If the fogging risk value is greater than the third threshold, a level three defogging command is generated. Wherein, the second threshold is greater than the first threshold and less than the third threshold.

4. The method according to claim 3, characterized in that, The method further includes: If the fogging risk value is less than the first threshold, the multi-source fogging related data will continue to be acquired.

5. The method according to claim 1, characterized in that, The method further includes: During defogging operations, continuously monitor the real-time fogging risk value; The level of the defogging command is dynamically adjusted based on the real-time fogging risk value.

6. The method according to claim 5, characterized in that, The step of dynamically adjusting the defogging command level based on the real-time fogging risk value includes: If the fog risk value remains unchanged within the set time period, maintain the current level of defogging command or switch to a higher level of defogging command; or, If the fog risk value continues to rise within a set time period, the current level of defogging command will be switched to a higher level of defogging command; or, If the fogging risk value decreases to a lower threshold within a set time period, the current level of defogging command will be switched to a lower level of defogging command or the defogging function will be turned off.

7. The method according to claim 1, characterized in that, The method further includes: During the execution of the defogging control command, the air conditioning temperature and / or fan speed are controlled within a safe threshold range.

8. A computer program product, characterized in that, The computer program product stores instructions that, when executed by a computer, cause the computer to perform the method described in any one of claims 1-7.

9. An electronic device, characterized in that, include: A memory on which computer programs are stored; A processor for executing the computer program in the memory to implement the method of any one of claims 1-7.

10. A vehicle, characterized in that, include: The electronic device according to claim 9; Alternatively, a processor, said processor being configured to perform the method according to any one of claims 1-7.

Images