Light barrier control method, vehicle and storage medium

Abstract

This application provides a light shield control method, a vehicle, and a storage medium. The method, applied in the field of intelligent vehicle technology, includes: acquiring facial image data of the driver; obtaining facial illumination features of the driver based on the facial image data, and determining the light intensity and direction based on the facial illumination features; generating light shield control commands and / or light shield pre-adjustment commands based on the light intensity, light direction, and preset personalized configuration information, and sending the light shield control commands and / or light shield pre-adjustment commands to the light shield to adjust its position. Thus, by sensing the driver's facial illumination features in real time and generating light shield control commands in conjunction with user-personalized configuration information, the method solves the problems in related technologies where light shielding systems cannot accurately sense the driver's actual light exposure and lack personalized adjustment capabilities. It achieves precise light shielding control based on real-time facial illumination, improving driving comfort and safety.

Description

Technical Field

[0001] This application relates to the field of intelligent vehicle technology, and more specifically, to a light-blocking control method, vehicle, and storage medium in the field of intelligent vehicle technology. Background Technology

[0002] With the rapid development of intelligent vehicle technology, users' demand for intelligent vehicle experiences is increasing. Especially in driving scenarios, direct sunlight (such as sunlight or oncoming headlights) can easily cause glare and visual fatigue for drivers, and in severe cases, lead to traffic accidents. Against this backdrop, intelligent shading systems have emerged and become an important development direction in the field of intelligent cockpits.

[0003] Among the related technologies, it is proposed to collect information on the intensity of light outside the vehicle by using a transparent photosensitive unit, and then use a camera to identify the position of the human eye to control the electrochromic module to adjust the light transmittance; or to determine the head position by using seat data, and then use navigation data and information from the external light sensor to adjust the position of the sun visor.

[0004] However, the shading systems in the above methods cannot accurately sense the driver's actual light exposure and lack personalized adjustment capabilities, which urgently need to be addressed. Summary of the Invention

[0005] This application provides a light shield control method, a vehicle, and a storage medium. The method senses the light characteristics of the driver's face in real time and generates light shield control commands by combining the user's personalized configuration information. This solves the problems in related technologies where the light shielding system cannot accurately sense the driver's actual light exposure and lacks personalized adjustment capabilities. It achieves precise light shielding control based on real-time facial light, thereby improving driving comfort and safety.

[0006] In a first aspect, a method for controlling a light shield is provided. The method includes: acquiring facial image data of a driver; obtaining facial illumination features of the driver based on the facial image data, and determining illumination intensity and illumination direction based on the facial illumination features; generating a light shield control command and / or a light shield pre-adjustment command according to the illumination intensity, the illumination direction, and preset personalized configuration information, and sending the light shield control command and / or the light shield pre-adjustment command to the light shield to adjust the position of the light shield.

[0007] By acquiring the driver's facial image data and extracting facial lighting features through the above technical solution, the actual lighting conditions on the driver's face can be directly perceived, thereby accurately determining the light intensity and direction experienced by the driver. This avoids the shortcomings of related technologies that rely on ambient light sensors and cannot reflect the true lighting conditions. Furthermore, based on real-time light intensity, light direction, and preset personalized configuration information, a light-blocking control command is generated, allowing the adjustment of the light-blocking baffle to fully consider the driver's differences in light sensitivity and shading preferences, achieving a personalized shading strategy. Finally, by sending the command to the light-blocking baffle to adjust its position, a timely response can be made when the light affects the driver, improving driving comfort and safety. Thus, a dimensional upgrade from environmental perception to facial perception is achieved, and a strategy optimization from fixed thresholds to personalized adaptation is implemented, effectively solving the technical problems of inaccurate shading control and lack of personalized adjustment in related technologies.

[0008] In conjunction with the first aspect, in some possible implementations, generating a light-blocking control command and / or a light-blocking pre-adjustment command based on the light intensity, the light direction, and preset personalized configuration information includes: acquiring the vehicle's real-time geographical location information and current time; calculating the real-time azimuth and altitude angles of the sun in the celestial coordinate system based on the real-time geographical location information and the current time; acquiring the vehicle's driving direction and heading, and calculating the trend of the change in the incident angle of sunlight relative to the driver's perspective within a preset future time period based on the azimuth, altitude, driving direction, and heading; comparing the trend of the change in the incident angle with the light direction, and generating the light-blocking pre-adjustment command based on the light intensity and the preset personalized configuration information if the comparison result meets preset matching conditions.

[0009] The above technical solution first acquires the vehicle's real-time geographical location and current time, and then calculates the sun's real-time azimuth and altitude angles using a solar position algorithm, accurately determining the sun's spatial position relative to the vehicle. Next, by integrating the vehicle's driving direction and heading, the trend of the incident angle of sunlight relative to the driver's viewpoint over a future time period is calculated, achieving advanced prediction of lighting changes. Based on this, the predicted incident angle trend is compared and verified with the real-time lighting direction extracted from facial images. A pre-adjustment command for the sun shade is generated only when the two match, ensuring the accuracy of the prediction and avoiding false triggering. Finally, a pre-adjustment command is generated by combining light intensity and personalized configuration information, enabling the sun shade to respond in advance before the actual lighting conditions arrive. This achieves an upgrade from passive response to active prediction, effectively solving the technical problems of delayed response and inability to predict lighting changes in related technologies, thus improving driving comfort and safety.

[0010] In combination with the first aspect and the above implementation methods, in some possible implementation methods, the step of comparing the incident angle change trend with the illumination direction includes: converting the incident angle change trend into a predicted light source direction vector and converting the illumination direction into a measured light source direction vector; calculating the spatial angle between the predicted light source direction vector and the measured light source direction vector; and determining that the comparison result satisfies the preset matching condition if the spatial angle is less than a preset angle.

[0011] The above technical solution converts the predicted trend of incident angle change into a predicted light source direction vector, and simultaneously converts the real-time illumination direction extracted from the facial image into a measured light source direction vector. This unifies illumination information from different sources and dimensions into a single mathematical expression space. Furthermore, the spatial angle between the two vectors is calculated to quantitatively and accurately measure the deviation between the predicted and actual perceived results, avoiding the ambiguity of qualitative judgments. Finally, a preset angle threshold is set as a matching criterion; the comparison is only considered successful when the spatial angle is less than this threshold, ensuring a clear numerical standard for verifying the consistency between prediction and measurement. This achieves an improvement in accuracy from qualitative comparison to quantitative verification, effectively solving the technical problem of effectively fusing and verifying predicted and perceived information, and providing accurate decision-making basis for subsequently generating reliable light-blocking pre-adjustment commands.

[0012] In combination with the first aspect and the above implementation methods, in some possible implementation methods, the step of generating the light-blocking panel pre-adjustment command based on the light intensity and the preset personalized configuration information includes: obtaining the current shading trigger threshold based on the preset personalized configuration information; in response to the light intensity being greater than the current shading trigger threshold, determining the predicted position adjustment target value of the light-blocking panel according to the shading priority mode or field of view priority mode in the preset personalized configuration information; and generating the light-blocking panel pre-adjustment command based on the predicted position adjustment target value of the light-blocking panel.

[0013] The above technical solution obtains the current sunshade trigger threshold from the preset personalized configuration information, serving as a quantitative basis for determining whether sunshade action needs to be initiated. This ensures that sunshade intervention is triggered only when the light intensity reaches the user's acceptable upper limit, avoiding unnecessary frequent adjustments. Furthermore, the subsequent decision-making process only proceeds when the light intensity exceeds this threshold, achieving energy-saving control triggered on demand. Based on this, according to the sunshade priority mode or vision priority mode in the personalized configuration information, the same light conditions are mapped to different predicted position adjustment target values. That is, in the sunshade priority mode, the adjustment range of the sunshade is larger to maximize light blocking, while in the vision priority mode, the adjustment range is smaller to preserve the driver's field of vision, achieving a differentiated response to user preferences. Finally, a pre-adjustment command for the sunshade is generated based on this target value, ensuring that the sunshade action precisely matches the user's personalized needs. Thus, an intelligent upgrade from uniform threshold triggering to personalized strategy adaptation is achieved, effectively solving the technical problems of rigid sunshade logic and inability to meet the differentiated needs of different users in related technologies, improving user experience and system friendliness.

[0014] In combination with the first aspect and the above implementation methods, in some possible implementation methods, sending the light-blocking plate control command and / or the light-blocking plate pre-adjustment command to the light-blocking plate to adjust the position of the light-blocking plate includes: determining the pre-adjustment execution time based on the trend of the incident angle change, wherein the pre-adjustment execution time is before the incident angle of sunlight reaches a first preset threshold; sending the light-blocking plate pre-adjustment command to the light-blocking plate at the pre-adjustment execution time to drive the light-blocking plate to perform coarse positioning adjustment, so that the light-blocking plate reaches the predicted position in advance; based on the facial image data, in response to the light intensity reaching a second preset threshold, sending the light-blocking plate control command to the light-blocking plate to drive the light-blocking plate to perform calibration adjustment from the predicted position, so that the light-blocking plate reaches the target position; wherein the adjustment range of the calibration adjustment is smaller than the adjustment range of the coarse positioning adjustment.

[0015] The above technical solution determines the pre-adjustment execution time based on the trend of the incident angle change, and sets this time before the incident angle of sunlight reaches the first preset threshold. This ensures that the light-blocking plate can act in advance before the actual lighting conditions affect the driver, overcoming the lag problem caused by system response delay. Then, at this time, the pre-adjustment command is sent to the light-blocking plate, driving it to perform coarse positioning adjustment, so that the light-blocking plate reaches the predicted position in advance, achieving rapid response to upcoming lighting changes and large-scale initial shading. Based on this, using real-time facial image data, when the light intensity is detected to reach the second preset threshold, a control command is sent to the light-blocking plate, driving it to perform calibration adjustment from the predicted position, so that the light-blocking plate accurately reaches the target position, achieving precise correction of prediction errors and actual lighting changes. Finally, by limiting the calibration adjustment amplitude to be smaller than the coarse positioning adjustment amplitude, a hierarchical control strategy of "large-step coarse adjustment and small-step fine adjustment" is formed, ensuring both response speed and positioning accuracy. This enables a two-stage collaborative control system that moves from a single response to predictive pre-adjustment and real-time calibration, effectively solving the technical problem of balancing response lag and positioning accuracy in related technologies and improving the overall performance of the shading system.

[0016] In conjunction with the first aspect and the above implementation methods, in some possible implementation methods, the above-mentioned light-blocking plate control method further includes: acquiring the real-time feedback signal of the light-blocking plate position sensor; in response to the fact that the real-time feedback signal of the light-blocking plate position sensor does not change within a preset time period, or the first deviation value between the real-time feedback signal of the light-blocking plate position sensor and the predicted position is greater than a first preset deviation threshold, or the second deviation value between the real-time feedback signal of the light-blocking plate position sensor and the target position is greater than a second preset deviation threshold, determining that the light-blocking plate motor is stalled or the light-blocking plate position sensor is faulty, and stopping the automatic adjustment of the light-blocking plate, and outputting fault prompt information.

[0017] By employing the aforementioned technical solution, feedback signals from the light-blocking plate position sensor are acquired in real time, establishing a continuous monitoring mechanism for the light-blocking plate's execution status and providing a data foundation for fault diagnosis. Furthermore, by monitoring the feedback signal for any change within a preset time period, actuator malfunctions such as light-blocking plate jamming or motor stalling can be detected promptly. Simultaneously, by comparing the feedback signal with the predicted position in the pre-adjustment stage and the target position in the calibration stage, the accuracy of the light-blocking plate reaching the expected position can be quantitatively assessed, allowing for timely detection of positioning anomalies. In the event of any of the above anomalies, it is determined to be a motor stall or sensor malfunction, and automatic adjustment is immediately stopped, effectively preventing potential mechanical damage or safety hazards that might result from continuing to execute commands under fault conditions. Finally, by outputting fault warning information, users or the system are promptly notified for maintenance, improving the system's maintainability and safety. Thus, an intelligent upgrade from blind execution to self-awareness of status, self-diagnosis of faults, and self-protection of safety is achieved, effectively solving the technical problem of the lack of status monitoring and fault protection mechanisms in related technologies for light-blocking systems, and improving the system's reliability and safety.

[0018] In combination with the first aspect and the above implementation methods, in some possible implementation methods, after adjusting the position of the light-blocking plate, the method further includes: obtaining the current brightness adjustment state of the head-up display system; and if the current brightness adjustment state satisfies a preset maximum brightness adjustment threshold and the light intensity is greater than the current shading trigger threshold, reducing the current shading trigger threshold and generating a new shading trigger threshold.

[0019] Through the above technical solution, after the light-blocking plate completes its position adjustment, the current brightness adjustment status of the head-up display (HUD) system is obtained, establishing a collaborative mechanism for cross-system state awareness. Furthermore, by determining whether the HUD system is already at the preset maximum brightness adjustment threshold, the specific scenario of "the HUD (Head-Up Display) is trying its best but still cannot meet display requirements" can be accurately identified. Based on this, combined with the condition that the current light intensity is still greater than the shading trigger threshold, the root cause is confirmed to be that the existing shading strategy is insufficient to eliminate the impact of glare on the HUD display. Finally, by actively lowering the current shading trigger threshold and generating a new threshold, the light-blocking plate can intervene earlier in similar lighting conditions, reducing the display burden on the HUD from the source. Thus, an intelligent upgrade from independent control to cross-system collaborative optimization is achieved. By dynamically optimizing the shading strategy through HUD status feedback, the technical problem of isolated operation of various cockpit functions and inability to collaboratively respond to user needs in related technologies is effectively solved, improving the overall user experience of the intelligent cockpit.

[0020] In combination with the first aspect and the above-described implementation methods, in some possible implementation methods, the light-blocking plate control command and / or the light-blocking plate pre-adjustment command are generated by the intelligent cockpit system-on-a-chip and sent to the electronic control unit of the light-blocking plate via a CAN (Controller Area Network) bus.

[0021] Through the above technical solution, the smart cockpit system-on-a-chip generates control commands and / or pre-adjustment commands for the light-shielding panel, achieving hardware reuse of high-performance computing resources in the cockpit domain. This eliminates the need for a dedicated processor for the light-shielding panel system, reducing system hardware costs and vehicle wiring complexity. Furthermore, the high computing power of the system-on-a-chip processes facial image data and executes complex control algorithms, ensuring the real-time performance and accuracy of computational tasks such as illumination feature extraction, direction comparison, and threshold judgment. Based on this, commands are sent to the electronic control unit of the light-shielding panel via a CAN bus, employing a mature and reliable standard communication protocol in the automotive field. This ensures seamless compatibility of commands with the vehicle's existing electronic and electrical architecture, guaranteeing real-time transmission, stability, and anti-interference capabilities. Thus, this achieves architectural optimization from functional silos to deep integration within the smart cockpit, effectively solving the technical problems of high cost, poor scalability, and difficulty in integrating into the cockpit ecosystem caused by independent deployment of existing light-shielding systems, thereby improving system integration and economy.

[0022] Secondly, a light-blocking plate control device is provided, the device comprising: The first acquisition module is used to acquire the driver's facial image data; The determination module is used to obtain the facial illumination features of the driver based on the facial image data, and to determine the illumination intensity and illumination direction based on the facial illumination features; The adjustment module is used to generate a light-blocking plate control command and / or a light-blocking plate pre-adjustment command based on the light intensity, the light direction and preset personalized configuration information, and send the light-blocking plate control command and / or the light-blocking plate pre-adjustment command to the light-blocking plate to adjust the position of the light-blocking plate.

[0023] In conjunction with the second aspect, in some possible implementations, the adjustment module includes: The acquisition unit is used to acquire the vehicle's real-time geographical location information and current time; The first calculation unit is used to calculate the real-time azimuth and altitude angles of the sun in the celestial coordinate system based on the real-time geographical location information and the current time. The second calculation unit is used to obtain the vehicle's driving direction and front orientation, and calculate the trend of the change of the incident angle of sunlight relative to the driver's viewpoint within a preset future time period based on the azimuth angle, the elevation angle, the driving direction, and the front orientation. The generation unit is used to compare the trend of the incident angle change with the illumination direction, and if the comparison result meets the preset matching conditions, generate the light-blocking plate pre-adjustment command according to the illumination intensity and the preset personalized configuration information.

[0024] In combination with the second aspect and the above implementation methods, in some possible implementations, the generating unit is specifically used for: The trend of the incident angle change is converted into a predicted light source direction vector, and the illumination direction is converted into a measured light source direction vector; Calculate the spatial angle between the predicted light source direction vector and the measured light source direction vector; If the spatial angle is less than a preset angle, the comparison result is determined to meet the preset matching condition.

[0025] In combination with the second aspect and the above implementation methods, in some possible implementations, the generating unit is specifically used for: Based on the preset personalized configuration information, obtain the current shading trigger threshold; In response to the light intensity being greater than the current shading trigger threshold, the target value for adjusting the predicted position of the light-blocking panel is determined according to the shading priority mode or field of view priority mode in the preset personalized configuration information. Based on the predicted position adjustment target value of the light-blocking plate, a pre-adjustment command for the light-blocking plate is generated.

[0026] In combination with the second aspect and the above implementation methods, in some possible implementations, the adjustment module is specifically used for: The pre-adjustment execution time is determined based on the trend of the incident angle change, wherein the pre-adjustment execution time is before the incident angle of sunlight reaches a first preset threshold. At the pre-adjustment execution time, the pre-adjustment command of the light-blocking plate is sent to the light-blocking plate, driving the light-blocking plate to perform coarse positioning adjustment, so that the light-blocking plate reaches the predicted position in advance; Based on the facial image data, in response to the light intensity reaching a second preset threshold, a control command for the light-blocking plate is sent to the light-blocking plate to drive the light-blocking plate to perform calibration adjustment from the predicted position, so that the light-blocking plate reaches the target position; The adjustment range of the calibration adjustment is smaller than that of the coarse positioning adjustment.

[0027] In conjunction with the second aspect and the above-described implementation, in some possible implementations, the aforementioned light-blocking plate control device further includes: The second acquisition module is used to acquire the real-time feedback signal from the light-blocking plate position sensor; The processing module is configured to respond to the following: if the real-time feedback signal of the light-blocking plate position sensor does not change within a preset time period, or if the first deviation value between the real-time feedback signal of the light-blocking plate position sensor and the predicted position is greater than a first preset deviation threshold, or if the second deviation value between the real-time feedback signal of the light-blocking plate position sensor and the target position is greater than a second preset deviation threshold, determine that the light-blocking plate motor is stalled or the light-blocking plate position sensor is faulty, and stop the automatic adjustment of the light-blocking plate, and output a fault prompt message.

[0028] In conjunction with the second aspect and the above implementation methods, in some possible implementations, after adjusting the position of the light-blocking plate, the adjustment module is further configured to: The current brightness adjustment status of the head-up display system is calculated based on the current air quality parameters and target air quality parameters at the location of each occupant. If the current brightness adjustment state meets the preset maximum brightness adjustment threshold and the light intensity is greater than the current shading trigger threshold, the current shading trigger threshold is reduced to generate a new shading trigger threshold.

[0029] In combination with the second aspect and the above implementation methods, in some possible implementation methods, the light-blocking plate control command and / or the light-blocking plate pre-adjustment command are generated by the smart cockpit system-on-a-chip and sent to the electronic control unit of the light-blocking plate via the CAN bus.

[0030] Thirdly, a vehicle is provided, including a controller, the controller including a memory, a processor, and a computer program stored in the memory and executable on the processor, the processor executing the program to implement the allergy protection method described in the above embodiments.

[0031] Fourthly, a computer program product is provided, comprising: computer program code, which, when executed on a computer, causes the computer to perform the light-blocking control method of the first aspect or any possible implementation thereof.

[0032] Fifthly, a computer-readable storage medium is provided that stores computer program code, which, when executed on a computer, causes the computer to perform the light-blocking control method of the first aspect or any possible implementation thereof. Attached Figure Description

[0033] Figure 1 A schematic flowchart illustrating the light-blocking plate control method provided in an embodiment of this application; Figure 2 This is a schematic diagram of the structure of the light-blocking plate control system provided in the embodiments of this application; Figure 3 A block diagram of the light-blocking plate control device provided in the embodiments of this application; Figure 4 This is a schematic diagram of the vehicle structure according to an embodiment of this application. Detailed Implementation

[0034] The technical solutions in this application will be clearly and thoroughly described below with reference to the accompanying drawings. In the description of the embodiments of this application, unless otherwise stated, " / " means "or," for example, A / B can mean A or B. "And / or" in the text is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Furthermore, in the description of the embodiments of this application, "multiple" refers to two or more than two.

[0035] Hereinafter, the terms "first" and "second" are used for descriptive purposes only and should not be construed as implying or suggesting relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature.

[0036] Figure 1 This is a schematic flowchart of a light-blocking plate control method provided in an embodiment of this application.

[0037] Before introducing the light-blocking plate control method proposed in the embodiments of this application, a brief introduction will be given to the light-blocking plate control system involved in this method. For example... Figure 2As shown, the sun visor control system mainly consists of a camera module, an intelligent sun visor driver module, and an electronically controlled sun visor (sunshade) module. The camera module includes a Driver Monitor System (DMS) camera, a driver controller, and a communication transceiver (such as CAN FD or Ethernet). The DMS camera, installed at the top center of the central control screen, captures facial image data of the driver, with its field of view directly facing the driver's face. The driver domain controller encodes the raw image data and transmits it to the intelligent sun visor driver module via communication protocols such as CAN FD or Ethernet. The intelligent sun visor driver module includes a communication transceiver (such as CAN FD, Ethernet, or CAN) and a System on a Chip (SOC). The communication transceiver uses communication protocols such as CAN FD or Ethernet to receive image data sent by the camera module. The smart cockpit SOC can decode, recognize, and analyze the received image data, extract the driver's facial lighting features, determine the light intensity and direction, and generate control commands for the light-blocking baffle based on the user's personalized configuration information. These commands are then sent to the electronically controlled light-blocking baffle module via the CAN bus of the communication transceiver. The electronically controlled light-blocking baffle module includes a communication transceiver and an electronic control unit (ECU). The communication transceiver receives control commands from the smart light-blocking baffle driver module. The ECU parses these commands and drives the motor via signals such as SPI (Serial Peripheral Interface) and PWM (Pulse Width Modulation) to adjust the position or angle of the light-blocking baffle, achieving precise sun shading.

[0038] To facilitate understanding, the light-blocking plate control method proposed in the embodiments of this application will be described in detail below.

[0039] For example, such as Figure 1 As shown, the light-blocking plate control method includes the following steps: In step S101, facial image data of the driver is acquired.

[0040] It is understood that, in this embodiment of the application, facial image data refers to the raw image information containing the driver's facial area captured by the visible light camera of the driver monitoring system. This data can serve as the input basis for subsequent image processing and illumination analysis. The visible light camera can be installed at the top center of the central control screen, with its field of view directly facing the driver's face, thereby stably acquiring high-quality facial images.

[0041] In other words, the system can acquire facial image data of the driver through the camera module, serving as the data foundation for subsequent illumination feature extraction and sunshade decisions. Specifically, the DMS camera in the camera module is reused as a facial image acquisition device. Its installation position ensures that the field of view is directly facing the driver's face, enabling stable acquisition of high-quality facial images including key areas such as the forehead, eye sockets, and nose. The driving domain controller in the camera module can encode the raw image data and transmit it in real time to the intelligent sunshade driver module via an in-vehicle communication network (such as CAN FD or Ethernet). This step achieves real-time acquisition and transmission of facial images, providing a reliable data source for subsequent accurate perception of the driver's actual lighting conditions. At the same time, by reusing the DMS camera, additional hardware costs are avoided, demonstrating the characteristics of high system integration and excellent hardware resource utilization.

[0042] In step S102, the driver's facial illumination features are obtained based on the facial image data, and the illumination intensity and direction are determined based on the facial illumination features.

[0043] It is understood that, in this embodiment, facial illumination features refer to visual feature information extracted from facial image data that reflects the illumination distribution, including but not limited to facial brightness gradient, shape and distribution of nose shadow area, depth and boundary of eye socket shadow, and location and area of ​​facial highlight area. These features can be used to infer the spatial position and intensity of the light source. Illumination intensity is the actual intensity of light received by the driver's face, which can be quantified as illuminance value, irradiance value, or normalized brightness level. Illumination direction is the position and direction of the light source relative to the driver's face, which can usually be expressed as the azimuth and pitch angles of the light source in three-dimensional space.

[0044] In other words, the system decodes and performs image recognition processing on the acquired facial image data through the smart cockpit SOC, and extracts the lighting features of the driver's face using computer vision algorithms. Specifically, the system can analyze the brightness distribution in the facial image data, calculate the brightness gradient changes in areas such as the forehead and cheeks, identify the shape and boundaries of shadow areas such as nose shadow and eye sockets, and detect the location distribution of highlight areas. Based on these facial lighting features, the system can further infer the actual light intensity value received by the driver and the spatial direction of the light source relative to the face through deep learning models or lighting estimation algorithms.

[0045] For example, deep learning models can employ illumination direction estimation networks, illumination intensity estimation networks, and 3D face reconstruction and illumination inversion models. The illumination direction estimation network can use a Convolutional Neural Network (CNN) as its backbone, such as ResNet (Residual Network) or MobileNet (Mobile Neural Network, a lightweight convolutional neural network series), with a fully connected regression layer added at the neck. The model's input can be facial image region data (e.g., 112×112 pixels), and the output can be the azimuth angle (e.g., 0°~360°) and pitch angle (e.g., -90°~90°) of the light source, output as regression values. The model's training data can use a dataset of face images containing multiple angles and illumination conditions, with each image labeled with a light source direction (e.g., "left light 30°", "top light 60°", etc.). The loss function can use mean absolute error or mean squared error to optimize the deviation between the predicted angle and the true angle. The lighting intensity estimation network can also use a CNN backbone network, sharing the feature extraction layer of the lighting direction estimation network. It employs a multi-task learning architecture to simultaneously output lighting intensity and direction. The input to this model can be a facial image region, and the output is a normalized lighting intensity value (0-1) or a specific illuminance value (in lux). Training data for this model can be generated by simultaneously recording ambient lighting intensity using a lux meter while acquiring facial images, constructing an image-illuminance paired dataset. The 3D face reconstruction and illumination inversion model can employ a 3D deformation model or a deep learning-based face reconstruction network to reconstruct the 3D face shape and reflectivity from a single 2D image. Based on the reconstructed 3D face geometry and surface reflectivity, the scene's illumination parameters, including light source direction, light source intensity, and ambient light components, can be decoupled through inverse rendering algorithms (i.e., physically based inverse rendering networks such as IRN (Inverse Rendering Network) and SfSNet (Shape from Shading Network)).

[0046] For illumination estimation algorithms, various methods can be used to extract illumination information from facial images. Among them, the illumination direction estimation method based on shadow analysis detects key facial points such as the tip of the nose and corners of the eyes, extracts the boundaries and directions of shadow regions such as nose shadows and eye sockets, and calculates the azimuth and pitch angles of the light source by combining them with a facial geometric model. This method requires no training data and has low computational cost, making it suitable for lightweight deployment scenarios. The illumination intensity estimation method based on brightness gradient converts the facial image into a grayscale image, calculates the average pixel brightness of high-reflectivity areas such as the forehead and cheeks, and converts the brightness values ​​into illumination values ​​according to a pre-calibrated brightness-illuminance mapping curve. This method is simple to implement, has high real-time performance, and is suitable for embedded systems. The global illumination estimation method based on spherical harmonic illumination can extract the direction and intensity of the main light source by segmenting the facial region and estimating the normal map, establishing a linear equation system between image brightness and the spherical harmonic illumination model, and solving for the spherical harmonic coefficients. This method can represent complex lighting environments such as multiple light sources and ambient light, and has high accuracy but relatively high computational cost. The above methods can be flexibly selected according to the system's computing power and accuracy requirements, providing diverse technical paths for facial illumination perception.

[0047] In step S103, based on the light intensity, light direction and preset personalized configuration information, a light-blocking plate control command and / or a light-blocking plate pre-adjustment command are generated, and the light-blocking plate control command and / or light-blocking plate pre-adjustment command are sent to the light-blocking plate to adjust the position of the light-blocking plate.

[0048] It is understood that, in this embodiment, the preset personalized configuration information refers to the user's preset sunshade preference parameters, including but not limited to the sunshade trigger threshold (i.e., the upper limit of light intensity acceptable to the user), sunshade priority mode (aiming to block light to the maximum extent), field-of-view priority mode (aiming to preserve the driver's field of vision), sunshade response sensitivity, and maximum adjustment range of the light shield. This information can be preset and stored through the vehicle's human-machine interface. The light shield control command refers to the immediate execution command generated based on real-time perceived facial illumination information, used to drive the light shield to perform precise adjustment when the actual illumination conditions arrive, so that the light shield reaches the target position. The light shield pre-adjustment command refers to the advance execution command generated based on the predicted illumination change trend, used to drive the light shield to perform coarse positioning adjustment before the actual illumination conditions arrive, so that the light shield reaches the predicted position in advance, overcoming system response delay.

[0049] In other words, the intelligent cockpit SOC can make comprehensive decisions based on light intensity, light direction and preset personalized configuration information to generate corresponding light shield control commands and / or light shield pre-adjustment commands. Subsequently, the system can send the generated control commands and / or pre-adjustment commands to the electronic control unit of the light shield through the vehicle communication network to drive the light shield to perform the corresponding position adjustment action.

[0050] Therefore, by acquiring the driver's facial image data and extracting facial lighting features, the actual lighting conditions on the driver's face can be directly perceived, thus accurately determining the light intensity and direction experienced by the driver. This avoids the shortcomings of related technologies that rely on ambient light sensors and cannot reflect the true lighting conditions. Furthermore, based on real-time light intensity, light direction, and preset personalized configuration information, control commands for the light shield are generated. This allows the adjustment of the light shield to fully consider the driver's differences in light sensitivity and shading preferences, achieving a personalized shading strategy. Finally, by sending commands to the light shield to adjust its position, a timely response can be made when light affects the driver, improving driving comfort and safety. Thus, a dimensional upgrade from environmental perception to facial perception is achieved, and a strategy optimization from fixed thresholds to personalized adaptation is implemented, effectively solving the technical problems of inaccurate shading control and lack of personalized adjustment in related technologies.

[0051] In one possible implementation, in some embodiments, a light-blocking control command and / or a light-blocking pre-adjustment command are generated based on light intensity, light direction, and preset personalized configuration information. This includes: acquiring the vehicle's real-time geographical location information and current time; calculating the real-time azimuth and elevation angles of the sun in the celestial coordinate system based on the real-time geographical location information and current time; acquiring the vehicle's driving direction and heading, and calculating the trend of the incident angle of sunlight relative to the driver's viewpoint within a preset future time period based on the azimuth, elevation angle, driving direction, and heading; comparing the trend of the incident angle with the light direction, and generating a light-blocking pre-adjustment command based on the light intensity and preset personalized configuration information if the comparison result meets preset matching conditions.

[0052] It is understood that a celestial coordinate system refers to an astronomical coordinate system centered on the observer, projecting celestial bodies onto an imaginary celestial sphere, used to describe the position of celestial bodies such as the sun in the sky. In the embodiments of this application, the azimuth and altitude angles of the sun relative to the vehicle's current geographical location can be calculated using this coordinate system. The azimuth angle is the horizontal angle measured clockwise from due north to the sun's projection point on the celestial sphere, used to describe the sun's horizontal direction. The altitude angle is the angle between the sunlight and the ground plane, used to describe the sun's vertical height. The incident angle change trend refers to the predicted trajectory of the incident angle of sunlight relative to the driver's viewpoint over a preset future time period, including the rate of change of the azimuth angle and the rate of change of the altitude angle. The preset matching condition is a quantitative standard used to determine whether the predicted incident angle change trend is consistent with the real-time perceived direction of illumination.

[0053] Specifically, the system first obtains the vehicle's real-time geographical location information and current time, and calculates the sun's real-time azimuth and elevation angles at the current moment based on astronomical algorithms to determine the sun's absolute spatial position relative to the vehicle. Then, combining the vehicle's driving direction and heading, and comprehensively considering the impact of the vehicle's motion state on the angle of incidence of light, the system can calculate the trend of the change in the angle of incidence of sunlight relative to the driver's perspective within a preset time period, achieving advanced prediction of changes in illumination. Based on this, the system can compare and verify the predicted trend of the change in the angle of incidence with the illumination direction perceived in real time from the facial image in step S102. By calculating whether the deviation between the two meets the preset matching conditions, the system determines whether the prediction result is consistent with the actual perception result. Only when the comparison verification is successful can the system generate a pre-adjustment command for the light shield based on the real-time illumination intensity and the user's preset personalized configuration information.

[0054] Therefore, by acquiring the vehicle's real-time geographical location information and current time, and combining this with a solar position algorithm to calculate the sun's real-time azimuth and altitude angles, the spatial position of the sun relative to the vehicle can be accurately determined. Furthermore, by integrating the vehicle's driving direction and heading, the trend of the incident angle of sunlight relative to the driver's viewpoint over a future time period can be calculated, achieving advanced prediction of lighting changes. Based on this, the predicted trend of incident angle changes is compared and verified with the real-time lighting direction extracted from facial images. A pre-adjustment command for the sun shade is generated only when the two match, ensuring the accuracy of the prediction results and avoiding false triggering. Finally, a pre-adjustment command is generated by combining light intensity and personalized configuration information, enabling the sun shade to respond in advance before the actual lighting conditions arrive. This represents an upgrade from passive response to active prediction, effectively solving the technical problems of delayed response and inability to predict lighting changes in related technologies, thus improving driving comfort and safety.

[0055] As one possible approach, in some embodiments, the trend of incident angle change is compared with the illumination direction, including: converting the trend of incident angle change into a predicted light source direction vector and converting the illumination direction into a measured light source direction vector; calculating the spatial angle between the predicted light source direction vector and the measured light source direction vector; and determining that the comparison result meets the preset matching conditions if the spatial angle is less than a preset angle.

[0056] As can be understood, the predicted light source direction vector refers to converting the predicted trend of incident angle changes (including azimuth and elevation angle changes) into a unit direction vector in three-dimensional space, used to mathematically represent the spatial direction of the predicted light source. The measured light source direction vector refers to converting the illumination direction extracted from the facial image (including azimuth and elevation angles) into a unit direction vector in three-dimensional space, used to mathematically represent the spatial direction of the actual light source. The spatial angle refers to the minimum angle between the predicted and measured light source direction vectors in three-dimensional space, used to quantify the degree of deviation between the two directions, and its value ranges from 0° to 180°. The preset angle is a pre-set spatial angle threshold used to determine whether the predicted direction and the measured direction are sufficiently close. This value can be set according to the system accuracy requirements and the actual application scenario, for example, 5°, 10°, or 15°.

[0057] Specifically, the system first converts the illumination direction extracted from facial image data (e.g., detecting that the current light comes from the driver's left front at a 30° azimuth angle and a 45° elevation angle) into a measured light source direction vector in three-dimensional space (denoted as vector A). Simultaneously, it converts the predicted trend of the incident angle change (e.g., predicting that sunlight will come from the left front at a 28° azimuth angle and a 43° elevation angle in 5 minutes) into a predicted light source direction vector of the same dimension (denoted as vector B), unifying the illumination information from two sources and two dimensions into the same mathematical expression space. Then, the system calculates the spatial angle between the two vectors to accurately measure the deviation between the predicted result and the actual perceived result in a numerical manner. Using the aforementioned data as an example, the calculated spatial angle between the two vectors is approximately 3.2°. Finally, the system compares the calculated spatial angle with a preset angle threshold (e.g., 10°). When this angle is less than the preset angle, it determines that the predicted direction and the measured direction are highly consistent, and the comparison result meets the preset matching conditions, thereby confirming the effectiveness and accuracy of the prediction.

[0058] Therefore, by converting the predicted trend of incident angle change into a predicted light source direction vector, and simultaneously converting the real-time illumination direction extracted from the facial image into a measured light source direction vector, illumination information from different sources and dimensions is unified into a single mathematical expression space. Furthermore, the spatial angle between the two vectors is calculated to quantitatively and accurately measure the degree of deviation between the predicted and actual perceived results, avoiding the ambiguity of qualitative judgment. Finally, by setting a preset angle threshold as a matching criterion, the comparison is considered successful only when the spatial angle is less than this threshold, ensuring that the consistency verification between prediction and measurement has a clear numerical standard. This achieves an improvement in accuracy from qualitative comparison to quantitative verification, effectively solving the technical problem of the difficulty in effectively fusing and verifying predicted and perceived information, and providing an accurate decision-making basis for subsequently generating reliable light-blocking pre-adjustment commands.

[0059] As one possible implementation, in some embodiments, a pre-adjustment command for the light-blocking panel is generated based on the light intensity and preset personalized configuration information, including: obtaining the current shading trigger threshold based on the preset personalized configuration information; in response to the light intensity being greater than the current shading trigger threshold, determining the predicted position adjustment target value of the light-blocking panel according to the shading priority mode or field of view priority mode in the preset personalized configuration information; and generating the pre-adjustment command for the light-blocking panel based on the predicted position adjustment target value of the light-blocking panel.

[0060] It is understandable that the sunshade trigger threshold refers to a user-preset critical value for light intensity. When the actual light intensity received by the driver's face exceeds this threshold, the system can determine that sunshade intervention needs to be initiated. This threshold can be personalized according to the user's light sensitivity; for example, sensitive users can set it to 500 lux, and tolerant users can set it to 1000 lux. The sunshade priority mode refers to an adjustment strategy that prioritizes maximizing light blocking. In this mode, the system tends to adjust the sunshade to a position closer to the driver's line of sight to block light as much as possible, even if it may slightly affect visibility. The vision priority mode refers to an adjustment strategy that prioritizes preserving the driver's field of vision. In this mode, the system tends to adjust the sunshade to a position that only blocks the necessary area, preserving the driver's field of vision as much as possible while providing sun protection. The predicted position adjustment target value refers to the expected position parameters of the sunshade determined based on lighting conditions and user preferences. This can be expressed as specific values ​​such as sliding displacement, rotation angle, or extension / retraction stroke, used to guide the execution of pre-adjustment commands.

[0061] Specifically, the system can first obtain the current sunshade trigger threshold corresponding to the current user from the preset personalized configuration information. For example, if user A is a light-sensitive driver, the sunshade trigger threshold can be set to 500 lux in the personalized configuration; if user B is a light-tolerant driver, the sunshade trigger threshold can be set to 1000 lux. When the light intensity determined in step S102 is 800 lux, for user A, the light intensity is greater than 500 lux, which meets the trigger condition, and the system continues to execute subsequent steps; for user B, the light intensity is less than 1000 lux, which does not meet the trigger condition, and the system will not generate a pre-adjustment command, thereby avoiding unnecessary sunshade intervention.

[0062] Under the condition that the triggering conditions are met, the system can further determine the target value for adjusting the predicted position of the light shield based on the preset personalized configuration information, which specifies either a sunshade priority mode or a vision priority mode. For example, assuming the light source is 30° to the left front: if user A selects the sunshade priority mode, the system tends to adjust the light shield to a position closer to the driver's line of sight, for example, rotating the light shield to a 25° azimuth angle to the left front to maximize light blocking, even if it may obstruct the left front view to some extent; if user A selects the vision priority mode, the system tends to block only the necessary area, for example, rotating the light shield to a 35° azimuth angle to the left front to block the main light while preserving the left front view as much as possible. Finally, the system can generate a pre-adjustment command for the light shield based on the determined predicted position adjustment target value. For example, in sunshade priority mode, the generated command would be "rotate to a 25° azimuth angle position," and this command would be sent to the light shield ECU for execution via the CAN bus.

[0063] Therefore, by obtaining the current sunshade trigger threshold from the preset personalized configuration information as a quantitative basis for determining whether to activate sunshade action, it ensures that sunshade intervention is triggered only when the light intensity reaches the user's acceptable upper limit, avoiding unnecessary frequent adjustments. Furthermore, the subsequent decision-making process is only initiated when the light intensity exceeds this threshold, achieving energy-saving control triggered on demand. Based on this, according to the sunshade priority mode or vision priority mode in the personalized configuration information, the same light conditions are mapped to different predicted position adjustment target values. That is, in the sunshade priority mode, the light shield adjusts more significantly to maximize light blocking, while in the vision priority mode, the adjustment is smaller to preserve the driver's field of vision, achieving a differentiated response to user preferences. Finally, a light shield pre-adjustment command is generated based on this target value, so that the light shield action is precisely matched with the user's personalized needs. Thus, an intelligent upgrade from uniform threshold triggering to personalized strategy adaptation is achieved, effectively solving the technical problems of rigid sunshade logic and inability to meet the differentiated needs of different users in related technologies, improving user experience and system friendliness.

[0064] As one possible implementation, in some embodiments, sending a light-blocking control command and / or a light-blocking pre-adjustment command to the light-blocking plate to adjust its position includes: determining the pre-adjustment execution time based on the trend of the incident angle change, wherein the pre-adjustment execution time is before the incident angle of sunlight reaches a first preset threshold; sending the light-blocking pre-adjustment command to the light-blocking plate at the pre-adjustment execution time to drive the light-blocking plate to perform coarse positioning adjustment, so that the light-blocking plate reaches the predicted position in advance; and, based on facial image data, in response to the light intensity reaching a second preset threshold, sending a light-blocking control command to the light-blocking plate to drive the light-blocking plate to perform calibration adjustment from the predicted position, so that the light-blocking plate reaches the target position; wherein the adjustment range of the calibration adjustment is smaller than the adjustment range of the coarse positioning adjustment.

[0065] Understandably, the pre-adjustment execution time is the point in time determined by the system based on the predicted trend of the incident angle change, used to send the pre-adjustment command in advance. This time can be set before the incident angle of sunlight reaches a first preset threshold to ensure that the sunshade completes its initial positioning before the light actually affects the driver. The first preset threshold refers to the critical value of the incident angle of sunlight used to trigger the pre-adjustment action. When the predicted incident angle is about to reach this first preset threshold, the system determines that sunshade preparation needs to be initiated in advance. For example, the first preset threshold can be set to 30°. The second preset threshold refers to the critical value of the light intensity used to trigger real-time calibration adjustment. When the real-time light intensity detected from the facial image data reaches this second preset threshold, the system determines that the light has actually reached the driver's face and precise calibration needs to be performed. This threshold can be consistent with the sunshade trigger threshold in the preset personalized configuration information. Coarse positioning adjustment is the initial, large-range position adjustment of the sunshade based on the pre-adjustment command. Its purpose is to quickly move the sunshade to the predicted position before the light arrives. The adjustment range is large, but the accuracy requirement is relatively low. Calibration adjustment refers to the precise, small-amplitude position adjustment of the light-blocking plate based on real-time control commands (i.e., light-blocking plate control commands). The purpose is to make fine adjustments based on the actual lighting conditions on the basis of coarse positioning, so that the light-blocking plate can accurately reach the target position. The adjustment range is small, but the accuracy requirement is high.

[0066] Specifically, the system first determines the pre-adjustment execution time based on the predicted trend of the incident angle change. For example, if the system predicts that the sunlight will change from the current 20° incident angle to 30° (i.e., the first preset threshold) in 10 seconds, then the pre-adjustment execution time can be set to 5 seconds later, ensuring sufficient adjustment time before the sunlight reaches 30°. At the pre-adjustment execution time, the system can send a pre-adjustment command to the light-blocking plate, driving it to perform coarse positioning adjustment. For example, it can command the light-blocking plate to quickly rotate to the predicted position, such as a 25° azimuth angle to the left front. This stage prioritizes rapid response and a large adjustment range (e.g., rotating from 0° to 25°, with an adjustment range of 25°), and even a certain degree of error (e.g., the actual position reached is 24° or 26°) is acceptable.

[0067] Subsequently, the system can continuously monitor changes in light intensity based on real-time acquired facial image data. When the light intensity reaches a second preset threshold (e.g., 500 lux), it is determined that the light has actually reached the driver's face. At this point, the system can generate precise control commands for the light-blocking plate based on the real-time perceived light direction and intensity, driving the light-blocking plate to perform calibration adjustments from the current predicted position (e.g., 24°) to accurately reach the target position (e.g., 24.5°). This stage prioritizes positioning accuracy, with a small adjustment range (e.g., adjusting from 24° to 24.5°, an adjustment range of only 0.5°), far smaller than the 25° adjustment range in the coarse positioning stage.

[0068] In other words, the embodiments of this application can divide the light-blocking plate adjustment process into two stages: the first stage is based on a rapid response to predictive information, roughly moving the light-blocking plate to the expected area; the second stage is based on real-time perception for precise correction, compensating for prediction errors and actual changes in illumination. For example, in a scenario where a vehicle is exiting a tunnel into direct sunlight, the system can perform coarse positioning in advance when the vehicle is about to exit the tunnel, pre-adjusting the light-blocking plate to its approximate position; when the vehicle actually exits the tunnel and sunlight instantly shines on the driver's face, the system immediately performs minor calibration, ensuring the light-blocking plate accurately blocks the sunlight. The entire process is both lag-free and accurately positioned, significantly improving the driving experience.

[0069] Therefore, by determining the pre-adjustment execution time based on the trend of the incident angle change and setting this time before the incident angle of sunlight reaches the first preset threshold, the light-blocking plate can be activated in advance before the actual lighting conditions affect the driver, overcoming the lag problem caused by system response delay. Then, at this time, the pre-adjustment command is sent to the light-blocking plate, driving it to perform coarse positioning adjustment, so that the light-blocking plate reaches the predicted position in advance, achieving rapid response to upcoming lighting changes and large-scale initial shading. Based on this, according to real-time facial image data, when the light intensity is detected to reach the second preset threshold, a control command is sent to the light-blocking plate, driving it to perform calibration adjustment from the predicted position, so that the light-blocking plate accurately reaches the target position, achieving precise correction of prediction errors and actual lighting changes. Finally, by limiting the amplitude of calibration adjustment to be smaller than that of coarse positioning adjustment, a hierarchical control strategy of "large-step coarse adjustment and small-step fine adjustment" is formed, ensuring both response speed and positioning accuracy. This enables a two-stage collaborative control system that moves from a single response to predictive pre-adjustment and real-time calibration, effectively solving the technical problem of balancing response lag and positioning accuracy in related technologies and improving the overall performance of the shading system.

[0070] Optionally, in some embodiments, the above-described light-blocking plate control method further includes: acquiring a real-time feedback signal from a light-blocking plate position sensor; in response to the light-blocking plate position sensor's real-time feedback signal remaining unchanged within a preset time period, or the light-blocking plate position sensor's real-time feedback signal deviating from a first deviation value of the predicted position greater than a first preset deviation threshold, or the light-blocking plate position sensor's real-time feedback signal deviating from a second deviation value of the target position greater than a second preset deviation threshold, determining that the light-blocking plate motor is stalled or the light-blocking plate position sensor is faulty, stopping the automatic adjustment of the light-blocking plate, and outputting a fault prompt message.

[0071] It is understandable that the light-blocking plate position sensor is a detection element installed on the light-blocking plate actuator to monitor the actual position of the light-blocking plate in real time. Its types can include Hall effect sensors, potentiometer-type angle sensors, etc., which can convert the mechanical position of the light-blocking plate into an electrical signal and feed it back to the control system. The real-time feedback signal refers to the actual position data of the light-blocking plate collected and returned by the light-blocking plate position sensor in real time, used to compare with the expected position specified by the command, forming a closed-loop control or fault monitoring. The first preset deviation threshold refers to the allowable error range used to determine whether the coarse positioning adjustment is successful. When the deviation between the actual position of the light-blocking plate (obtained from the real-time feedback signal of the light-blocking plate position sensor) and the predicted position specified by the pre-adjustment command exceeds this threshold, it can be determined that the coarse positioning has failed or that an anomaly exists. The second preset deviation threshold refers to the allowable error range used to determine whether the calibration adjustment is successful. When the deviation between the actual position of the light-blocking plate and the target position specified by the real-time control command exceeds this threshold, it is determined that the precise positioning has failed or that an anomaly exists.

[0072] Specifically, during the process of the light-blocking plate executing adjustment commands, the system can continuously acquire real-time feedback signals from the position sensor to monitor the actual positional changes of the light-blocking plate. The system can make judgments on at least three abnormal situations. The first abnormal situation: no response. For example, the system sends a pre-adjustment command to the light-blocking plate, requiring it to rotate from 0° to 25° within 3 seconds, but the position sensor feedback signal remains unchanged at 0° for a preset time period (e.g., 5 seconds). In this case, the system can determine that the light-blocking plate may be stuck or the motor may be stalled (i.e., the light-blocking plate drive motor cannot rotate normally due to mechanical jamming, overload, etc., while powered on, which may lead to motor damage or the light-blocking plate malfunctioning), and therefore cannot respond to the command. The second abnormal situation: excessive coarse positioning deviation. For example, the pre-adjustment command requires the light-blocking plate to reach the predicted position of 25°, but the position sensor feedback signal shows that it has only reached 20°, with a first deviation value of 5°, while the first preset deviation threshold is set to 3°. Since the first deviation value is greater than the first preset deviation threshold, the system determines that the coarse positioning adjustment has not achieved the expected result, possibly due to mechanical jamming or sensor malfunction. The third abnormal situation: excessive calibration deviation. For example, the real-time control command requires the light-blocking plate to be calibrated from the predicted position to the target position of 24.5°, but the position sensor feedback signal shows that it actually reached 26°, with a second deviation value of 1.5°, while the second preset deviation threshold is set to 1°. Since the second deviation value is greater than the second preset deviation threshold, the system determines that the accurate calibration has failed.

[0073] If any of the above-mentioned abnormal situations occur, the system will immediately determine that the light-blocking motor is stalled or the position sensor is faulty, and will execute safety protection actions: stop the automatic adjustment of the light-blocking to avoid mechanical damage or safety hazards that may result from continuing to execute commands in a faulty state; at the same time, output fault prompt information through the vehicle display screen or voice system to promptly notify the user or maintenance personnel to carry out inspection.

[0074] Therefore, by acquiring feedback signals from the light-blocking plate position sensor in real time, a continuous monitoring mechanism for the light-blocking plate's execution status is established, providing a data foundation for fault diagnosis. Furthermore, by monitoring the feedback signal for no change within a preset time period, actuator faults such as light-blocking plate jamming or motor stalling can be detected in a timely manner. Simultaneously, by comparing the deviation of the feedback signal with the predicted position in the pre-adjustment stage and the target position in the calibration stage, the accuracy of whether the light-blocking plate has reached the expected position can be quantitatively assessed, and positioning anomalies can be detected in a timely manner. When any of the above anomalies occurs, it is determined that the motor is stalled or the sensor is faulty, and automatic adjustment is immediately stopped, effectively avoiding mechanical damage or safety hazards that may result from continuing to execute commands under fault conditions. Finally, by outputting fault prompt information, users or the system are promptly notified to perform maintenance, improving the system's maintainability and safety. Thus, an intelligent upgrade from blind execution to state self-awareness, fault self-diagnosis, and safety self-protection is achieved, effectively solving the technical problem of the lack of state monitoring and fault protection mechanisms in related technologies for light-blocking systems, and improving the system's reliability and safety.

[0075] Optionally, in some embodiments, after adjusting the position of the light-blocking plate, the method further includes: obtaining the current brightness adjustment state of the head-up display system; and, if the current brightness adjustment state meets the preset maximum brightness adjustment threshold and the light intensity is greater than the current shading trigger threshold, reducing the current shading trigger threshold and generating a new shading trigger threshold.

[0076] It is understandable that the current brightness adjustment status refers to the current brightness adjustment level of the head-up display system, which can be expressed as a brightness adjustment level (such as 1-10 levels) or a specific brightness value (such as 500 cd / m²). 2 The preset maximum brightness adjustment threshold refers to the highest brightness that the head-up display system can adjust to; that is, the critical point where the system, even at its brightest setting, still cannot meet the display requirements. For example, it can be set to the highest brightness adjustment level, level 10, or a maximum brightness value of 1000 cd / m². 2 .

[0077] Specifically, after the head-up display (HUD) completes its position adjustment, the system can further obtain the current brightness adjustment status of the HUD and determine whether it has reached the preset maximum brightness adjustment threshold. For example, the HUD has 10 brightness levels, and it is currently adjusted to the highest level (level 10) (i.e., the preset maximum brightness adjustment threshold). Simultaneously, if the light intensity determined in step S102 is 800 lux, and the current sunshade trigger threshold is 500 lux, the condition that the light intensity exceeds the current sunshade trigger threshold is met. In this situation, the system identifies a specific scenario: the HUD has been adjusted to its maximum brightness, but the light intensity still exceeds the sunshade trigger threshold, potentially causing glare interference for the driver and making the HUD content unclear. This indicates that the current sunshade strategy is insufficient to effectively solve the lighting problem, and the HUD needs to intervene earlier and more aggressively to provide sunshade.

[0078] To address this, the system can proactively lower the current shading trigger threshold and generate a new one. For example, it can lower the current shading trigger threshold from 500 lux to 400 lux. This means that under subsequent lighting conditions, when the light intensity reaches 400 lux, the system will trigger the baffle adjustment, allowing the baffle to intervene in shading earlier than before, thus reducing the display burden on the head-up display system from the source. For instance, in a summer afternoon with strong direct sunlight, even if the head-up display system is set to its brightest setting, it may still be difficult to see clearly. Upon detecting this, the system automatically lowers the shading trigger threshold, allowing the baffle to intervene earlier under similar lighting conditions, effectively reducing driver glare and improving the readability of the head-up display.

[0079] Therefore, by obtaining the current brightness adjustment status of the head-up display (HUD) system after the light-blocking plate completes its position adjustment, a collaborative mechanism for cross-system state awareness is established. Furthermore, by determining whether the HUD system is already at a preset maximum brightness adjustment threshold, the specific scenario of "the HUD is doing its best but still cannot meet display requirements" can be accurately identified. Based on this, combined with the condition that the current light intensity is still greater than the shading trigger threshold, it is confirmed that the root cause of the problem lies in the existing shading strategy's inadequacy to eliminate the impact of glare on the HUD display. Finally, by actively lowering the current shading trigger threshold and generating a new threshold, the light-blocking plate can intervene earlier in similar lighting conditions in the future, reducing the display burden on the HUD from the source. This achieves an intelligent upgrade from independent control to cross-system collaborative optimization. By dynamically optimizing the shading strategy through HUD status feedback, it effectively solves the technical problem of isolated operation of various cockpit functions and the inability to collaboratively respond to user needs in related technologies, improving the overall user experience of the intelligent cockpit.

[0080] Optionally, in some embodiments, the light-blocking control command and / or light-blocking pre-adjustment command are generated by the smart cockpit system-on-a-chip and sent to the electronic control unit of the light-blocking via the CAN bus.

[0081] In other words, in this embodiment, the system can achieve efficient integration and reliable transmission of the light-shield control function through hardware reuse of the smart cockpit SOC and standard communication via the CAN bus. Specifically, the light-shield control commands and / or light-shield pre-adjustment commands are not generated by a separate dedicated controller, but are uniformly generated by the smart cockpit SOC. As the central computing unit of the vehicle, the smart cockpit SOC originally undertakes multiple functions such as in-vehicle infotainment, navigation, and driver monitoring. In this embodiment, the SOC further reuses its high-performance computing power to undertake tasks such as facial image decoding, illumination feature extraction, illumination direction and intensity calculation, and personalized strategy decision-making, ultimately generating light-shield control commands and / or light-shield pre-adjustment commands. This hardware reuse method avoids configuring a separate processor for the light-shield system, effectively reducing system hardware costs and vehicle wiring complexity. For example, when processing facial images captured by the DMS camera, the smart cockpit SOC can simultaneously perform driver fatigue monitoring and facial illumination analysis, with both calculations occurring in parallel and without interference. When the light analysis module determines that shading is needed, the SOC immediately generates the corresponding light-blocking control command and / or light-blocking pre-adjustment command.

[0082] The generated commands can then be sent to the light deflector ECU via the CAN bus. As a standard automotive communication protocol, the CAN bus boasts high real-time performance and strong anti-interference capabilities, ensuring reliable command transmission even in complex vehicle electromagnetic environments. Upon receiving the commands, the light deflector ECU parses and verifies them, extracts the target position adjustment value, and drives the motor to execute the specific position adjustment action via interfaces such as SPI and PWM.

[0083] Therefore, by generating control commands and / or pre-adjustment commands for the light-shielding system using the system-on-a-chip (SoC), hardware reuse of high-performance computing resources in the cockpit domain is achieved. This eliminates the need for a dedicated processor for the light-shielding system, reducing system hardware costs and vehicle wiring complexity. Furthermore, the high computing power of the SoC is used to process facial image data and execute complex control algorithms, ensuring the real-time performance and accuracy of computational tasks such as illumination feature extraction, direction comparison, and threshold judgment. Based on this, commands are sent to the electronic control unit of the light-shielding system via the CAN bus, employing a mature and reliable standard communication protocol in the automotive field. This ensures seamless compatibility of commands with the vehicle's existing electronic and electrical architecture, guaranteeing real-time transmission, stability, and anti-interference capabilities. This achieves architectural optimization from functional silos to deep integration within the smart cockpit, effectively solving the technical problems of high cost, poor scalability, and difficulty in integrating into the cockpit ecosystem caused by independent deployment of existing light-shielding systems, thus improving system integration and economy.

[0084] Figure 3 This is a schematic diagram of a light-blocking plate control device provided in an embodiment of this application.

[0085] For example, such as Figure 3 As shown, the light-blocking plate control device 10 may include: a first acquisition module 100, a determination module 200, and an adjustment module 300.

[0086] The first acquisition module 100 is used to acquire facial image data of the driver; The determination module 200 is used to obtain the facial illumination features of the driver based on facial image data, and to determine the illumination intensity and illumination direction based on the facial illumination features; The adjustment module 300 is used to generate a light-blocking plate control command and / or a light-blocking plate pre-adjustment command based on the light intensity, light direction and preset personalized configuration information, and send the light-blocking plate control command and / or light-blocking plate pre-adjustment command to the light-blocking plate to adjust the position of the light-blocking plate.

[0087] Optionally, in one embodiment of this application, the adjustment module 300 includes: The acquisition unit is used to acquire the vehicle's real-time geographical location information and current time; The first calculation unit is used to calculate the real-time azimuth and altitude angles of the sun in the celestial coordinate system based on real-time geographic location information and the current time. The second calculation unit is used to obtain the vehicle's driving direction and heading, and calculate the trend of the change of the incident angle of sunlight relative to the driver's viewpoint within a preset future time period based on the azimuth angle, elevation angle, driving direction and heading. The generation unit is used to compare the trend of the incident angle change with the direction of illumination. If the comparison result meets the preset matching conditions, it generates a pre-adjustment command for the light-blocking plate based on the illumination intensity and preset personalized configuration information.

[0088] Optionally, in one embodiment of this application, the generating unit is specifically used for: The trend of incident angle variation is converted into a predicted light source direction vector, and the illumination direction is converted into a measured light source direction vector; Calculate the spatial angle between the predicted light source direction vector and the measured light source direction vector; If the spatial angle is less than a preset angle, the comparison result is determined to meet the preset matching conditions.

[0089] Optionally, in one embodiment of this application, the generating unit is specifically used for: Based on preset personalized configuration information, obtain the current shading trigger threshold; In response to light intensity exceeding the current shading trigger threshold, the target value for adjusting the predicted position of the light-blocking panel is determined based on the shading priority mode or field of view priority mode in the preset personalized configuration information. Based on the predicted position adjustment target value of the light-blocking plate, a pre-adjustment command for the light-blocking plate is generated.

[0090] Optionally, in one embodiment of this application, the adjustment module 300 is specifically used for: The pre-adjustment execution time is determined based on the trend of the incident angle change, wherein the pre-adjustment execution time is before the incident angle of sunlight reaches the first preset threshold. At the time of pre-adjustment execution, the pre-adjustment command of the light-blocking plate is sent to the light-blocking plate, driving the light-blocking plate to perform coarse positioning adjustment, so that the light-blocking plate reaches the predicted position in advance; Based on facial image data, in response to the light intensity reaching the second preset threshold, a control command for the light-blocking plate is sent to the light-blocking plate, driving the light-blocking plate to perform calibration adjustment from the predicted position, so that the light-blocking plate reaches the target position; Among them, the adjustment range of calibration adjustment is smaller than that of coarse positioning adjustment.

[0091] Optionally, in one embodiment of this application, the light-blocking plate control device 10 further includes: The second acquisition module is used to acquire the real-time feedback signal from the light-blocking plate position sensor; The processing module is used to determine whether the light-blocking plate motor is stalled or the light-blocking plate position sensor is faulty in response to the following: the real-time feedback signal of the light-blocking plate position sensor does not change within a preset time period; or the first deviation value between the real-time feedback signal of the light-blocking plate position sensor and the predicted position is greater than a first preset deviation threshold; or the second deviation value between the real-time feedback signal of the light-blocking plate position sensor and the target position is greater than a second preset deviation threshold. The module then stops the automatic adjustment of the light-blocking plate and outputs a fault prompt message.

[0092] Optionally, in one embodiment of this application, after adjusting the position of the light-blocking plate, the adjustment module 300 is further configured to: The current brightness adjustment status of the head-up display system is calculated based on the current air quality parameters and target air quality parameters at each occupant's location; If the current brightness adjustment state meets the preset maximum brightness adjustment threshold and the light intensity is greater than the current shading trigger threshold, the current shading trigger threshold is reduced and a new shading trigger threshold is generated.

[0093] Optionally, in one embodiment of this application, the light-blocking control command and / or light-blocking pre-adjustment command are generated by a smart cockpit system-on-a-chip and sent to the electronic control unit of the light-blocking via a CAN bus.

[0094] In summary, the light shield control device according to the embodiments of this application generates light shield control commands by sensing the light characteristics of the driver's face in real time and combining them with the user's personalized configuration information. This solves the problem in related technologies that the light shielding system cannot accurately sense the actual light exposure of the driver and lacks personalized adjustment capabilities. It achieves precise light shielding control based on real-time facial light illumination, thereby improving driving comfort and safety.

[0095] Figure 4 This is a schematic diagram of the structure of a vehicle provided in an embodiment of this application.

[0096] It should be understood that the methods described above can be applied to... Figure 4 In the vehicle with the structure shown.

[0097] like Figure 4 As shown, the vehicle includes a controller, which may include a memory 401 and a processor 402. The memory 401 stores executable program code, and the processor 402 is used to call and execute the executable program code to perform the light-blocking control method provided in the embodiments of this application.

[0098] Furthermore, the controller also includes a communication interface 404 for communication between the memory 401 and the processor 402.

[0099] This embodiment can divide the vehicle into functional modules based on the above method example. For example, each module can correspond to a separate function module, or two or more functions can be integrated into one processing module. The integrated module can be implemented in hardware. It should be noted that the module division in this embodiment is illustrative and only represents one logical functional division. In actual implementation, there may be other division methods.

[0100] It should be noted that all relevant content of each step involved in the above method embodiments can be referenced from the functional description of the corresponding functional module, and will not be repeated here.

[0101] It should be understood that the vehicle provided in this embodiment is used to execute the above-described light-blocking control method, and therefore can achieve the same effect as the above-described implementation method.

[0102] This embodiment also provides a computer-readable storage medium storing computer program code. When the computer program code is run on a computer, the computer executes the above-described related method steps to implement the light-blocking plate control method provided in the above embodiment.

[0103] This embodiment also provides a computer program product that, when run on a computer, causes the computer to perform the aforementioned related steps to implement the light-blocking plate control method provided in the above embodiment.

[0104] In this embodiment, the vehicle, computer-readable storage medium, computer program product, or chip are all used to execute the corresponding methods provided above. Therefore, the beneficial effects that can be achieved can be referred to the beneficial effects of the corresponding methods provided above, and will not be repeated here.

[0105] Through the above description of the embodiments, those skilled in the art will understand that, for the sake of convenience and brevity, only the division of the above functional modules is used as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above.

[0106] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0107] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for controlling a light-blocking plate, characterized in that, Includes the following steps: Obtain the driver's facial image data; Based on the facial image data, the facial illumination features of the driver are obtained, and based on the facial illumination features, the illumination intensity and illumination direction are determined. Based on the light intensity, the light direction, and the preset personalized configuration information, a light-blocking plate control command and / or a light-blocking plate pre-adjustment command are generated, and the light-blocking plate control command and / or the light-blocking plate pre-adjustment command are sent to the light-blocking plate to adjust the position of the light-blocking plate.

2. The method according to claim 1, characterized in that, The step of generating a light-blocking control command and / or a light-blocking pre-adjustment command based on the light intensity, the light direction, and preset personalized configuration information includes: Obtain the vehicle's real-time location information and current time; Based on the real-time geographic location information and the current time, calculate the real-time azimuth and altitude angles of the sun in the celestial coordinate system; The vehicle's driving direction and heading are obtained, and the trend of the change in the incident angle of sunlight relative to the driver's viewpoint within a preset future time period is calculated based on the azimuth angle, the elevation angle, the driving direction, and the heading. The trend of the incident angle change is compared with the direction of illumination. If the comparison result meets the preset matching conditions, the light-blocking plate pre-adjustment command is generated according to the illumination intensity and the preset personalized configuration information.

3. The method according to claim 2, characterized in that, The step of comparing the trend of the incident angle change with the direction of illumination includes: The trend of the incident angle change is converted into a predicted light source direction vector, and the illumination direction is converted into a measured light source direction vector; Calculate the spatial angle between the predicted light source direction vector and the measured light source direction vector; If the spatial angle is less than a preset angle, the comparison result is determined to meet the preset matching condition.

4. The method according to claim 3, characterized in that, The step of generating a pre-adjustment command for the light-blocking plate based on the light intensity and the preset personalized configuration information includes: Based on the preset personalized configuration information, obtain the current shading trigger threshold; In response to the light intensity being greater than the current shading trigger threshold, the target value for adjusting the predicted position of the light-blocking panel is determined according to the shading priority mode or field of view priority mode in the preset personalized configuration information. Based on the predicted position adjustment target value of the light-blocking plate, a pre-adjustment command for the light-blocking plate is generated.

5. The method according to claim 2, characterized in that, Sending the light-blocking plate control command and / or the light-blocking plate pre-adjustment command to the light-blocking plate to adjust the position of the light-blocking plate includes: The pre-adjustment execution time is determined based on the trend of the incident angle change, wherein the pre-adjustment execution time is before the incident angle of sunlight reaches a first preset threshold. At the pre-adjustment execution time, the pre-adjustment command of the light-blocking plate is sent to the light-blocking plate, driving the light-blocking plate to perform coarse positioning adjustment, so that the light-blocking plate reaches the predicted position in advance; Based on the facial image data, in response to the light intensity reaching a second preset threshold, a control command for the light-blocking plate is sent to the light-blocking plate to drive the light-blocking plate to perform calibration adjustment from the predicted position, so that the light-blocking plate reaches the target position; The adjustment range of the calibration adjustment is smaller than that of the coarse positioning adjustment.

6. The method according to claim 5, characterized in that, Also includes: Obtain the real-time feedback signal from the light-blocking plate position sensor; If the real-time feedback signal of the light-blocking plate position sensor does not change within a preset time period, or if the first deviation value between the real-time feedback signal of the light-blocking plate position sensor and the predicted position is greater than a first preset deviation threshold, or if the second deviation value between the real-time feedback signal of the light-blocking plate position sensor and the target position is greater than a second preset deviation threshold, it is determined that the light-blocking plate motor is stalled or the light-blocking plate position sensor is faulty, and the automatic adjustment of the light-blocking plate is stopped, and a fault prompt message is output.

7. The method according to claim 4, characterized in that, After adjusting the position of the light-blocking plate, the following is also included: Get the current brightness adjustment status of the head-up display system; If the current brightness adjustment state meets the preset maximum brightness adjustment threshold and the light intensity is greater than the current shading trigger threshold, the current shading trigger threshold is reduced to generate a new shading trigger threshold.

8. The method according to claim 1, characterized in that, The light-blocking control command and / or the light-blocking pre-adjustment command are generated by the intelligent cockpit system-on-a-chip and sent to the electronic control unit of the light-blocking plate via the CAN bus.

9. A vehicle comprising a controller, the controller including a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, The processor executes the program to implement the light-blocking plate control method as described in any one of claims 1-8.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed, implements the light-blocking plate control method as described in any one of claims 1-8.

Images