Vehicle sunshade control method and device, electronic equipment and storage medium

Abstract

This invention provides a vehicle sunshade control method, device, electronic device, and storage medium. Based on the vehicle's driving state, ambient light information, and driver state, it determines the vehicle's driving scenario category, driver workload level, and shading requirement information. Based on the vehicle's driving scenario category, it determines the vehicle's visibility safety priority and makes a preliminary sunshade decision. Based on the vehicle's driving scenario category, driver workload level, shading requirement information, and preliminary sunshade decision, it generates a final sunshade decision and controls the vehicle's sunshade execution unit to execute the final sunshade decision. This invention can resolve the fundamental contradiction between sunshade and obstructed visibility, and can also cope with complex dynamic light sources, meeting the vehicle's sunshade needs in diverse driving situations.

Description

Technical Field

[0001] This invention relates to the field of vehicle intelligent control technology, and in particular to a vehicle sunshade control method, device, electronic device, and storage medium. Background Technology

[0002] Existing vehicle sunshade control technologies generally suffer from a lack of simplistic decision-making logic and a disconnect from overall vehicle safety. Specifically, this manifests in the following ways: they fail to resolve the fundamental contradiction between sunshade and obstructed vision, and cannot cope with complex dynamic light sources (such as oncoming headlights and building reflections). Although they can achieve targeted shading of local areas based on visual tracking, their decision-making objective is singular: "shading occurs upon detecting glare." They do not consider the secondary safety risks that the shading behavior itself may bring in different driving scenarios, making it difficult to adapt to diverse driving situations. Summary of the Invention

[0003] In view of this, the purpose of the present invention is to provide a vehicle sunshade control method, device, electronic device and storage medium to alleviate the above-mentioned problems existing in the related art.

[0004] In a first aspect, embodiments of the present invention provide a vehicle sunshade control method, comprising: determining the vehicle's driving scenario category, driver load level, and shading requirement information based on the vehicle's driving state, ambient light information, and driver state; determining the vehicle's visibility safety priority and preliminary sunshade decision based on the vehicle's driving scenario category; generating a final sunshade decision based on the vehicle's driving scenario category, driver load level, shading requirement information, and preliminary sunshade decision, and controlling the vehicle's sunshade execution unit to execute the final sunshade decision.

[0005] Secondly, embodiments of the present invention also provide a vehicle sunshade control device, comprising: a processing module, configured to determine the vehicle's driving scenario category, driver load level, and shading requirement information based on the vehicle's driving state, ambient light information, and driver state; a determining module, configured to determine the vehicle's visibility safety priority and preliminary sunshade decision based on the vehicle's driving scenario category; and a decision module, configured to generate a final sunshade decision based on the vehicle's driving scenario category, driver load level, shading requirement information, and preliminary sunshade decision, and control the vehicle's sunshade execution unit to execute the final sunshade decision.

[0006] Thirdly, embodiments of the present invention also provide an electronic device, including a processor and a memory, wherein the memory stores computer-executable instructions that can be executed by the processor, and the processor executes the computer-executable instructions to implement the method described in the first aspect above.

[0007] Fourthly, embodiments of the present invention also provide a computer-readable storage medium storing computer-executable instructions, which, when invoked and executed by a processor, cause the processor to implement the method described in the first aspect above.

[0008] This invention provides a vehicle sunshade control method, device, electronic device, and storage medium. Based on the vehicle's driving state, ambient light information, and driver state, it determines the vehicle's driving scenario category, driver workload level, and shading requirement information. Based on the vehicle's driving scenario category, it determines the vehicle's visibility safety priority and makes a preliminary sunshade decision. Based on the vehicle's driving scenario category, driver workload level, shading requirement information, and preliminary sunshade decision, it generates a final sunshade decision and controls the vehicle's sunshade execution unit to execute the final sunshade decision. Using this technology, different vehicle information can be integrated to determine the vehicle's visibility safety priority, thereby obtaining and executing the final sunshade decision. This resolves the fundamental contradiction between sunshade and obstructed visibility, and can also cope with complex dynamic light sources, meeting the vehicle's sunshade needs in diverse driving scenarios.

[0009] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention are realized and obtained in accordance with the structures particularly pointed out in the description, claims and drawings.

[0010] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0011] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0012] Figure 1 This is a flowchart illustrating a vehicle sunshade control method according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the overall architecture of the vehicle sunshade control system in an embodiment of the present invention; Figure 3 This is a flowchart illustrating the workflow of the two-layer decision engine in this embodiment of the invention. Figure 4 This is a schematic diagram of the structure of a vehicle sunshade control device according to an embodiment of the present invention; Figure 5 This is a schematic diagram of the structure of an electronic device according to an embodiment of the present invention. Detailed Implementation

[0013] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0014] Currently, existing vehicle sunshade control technologies generally suffer from a lack of simplistic decision-making logic and a disconnect from the overall vehicle safety situation.

[0015] Based on this, the present invention provides a vehicle sunshade control method, device, electronic device and storage medium, which can alleviate the above-mentioned problems existing in related technologies.

[0016] To facilitate understanding of this embodiment, a vehicle sunshade control method disclosed in this invention will first be described in detail, see [link to relevant documentation]. Figure 1 As shown, the method may include the following steps: Step S102: Based on the vehicle's driving status, ambient light information, and driver status, determine the vehicle's driving scenario category, driver load level, and occlusion requirement information.

[0017] In practical applications, three types of data streams can be acquired in parallel: (1) Vehicle dynamics and environmental data stream (including vehicle driving status, vehicle environment and other information): vehicle status signals (including but not limited to: vehicle speed, lateral / longitudinal acceleration, yaw rate, steering wheel angle and turning rate, turn signal status, gear information, brake pedal status, etc.) are obtained in real time through the vehicle navigation module or high-precision map interface, and the curvature, slope and predetermined path information of the road in front of the vehicle are obtained through the vehicle navigation module or high-precision map interface, as well as the ambient light intensity and rainfall information of the vehicle's surrounding environment are obtained through the environmental sensors.

[0018] (2) Driver status data stream (including driver status): The following indicators are obtained and calculated by using an infrared camera facing the driver and possible biosensors (such as a heart rate sensor integrated on the steering wheel): the precise three-dimensional coordinates of the driver's eyes, the driver's gaze direction, head posture angle, etc., and fatigue indicators (such as PERCLOS, i.e. the proportion of time during which the eyelids cover the pupils for more than 80% of the time) and attention distraction indicators (such as the duration and frequency of the gaze deviating from the road ahead).

[0019] (3) Traditional optical sensing data stream (including information related to light sources in the vehicle's environment): External strong light source information is detected and identified by a forward-facing wide-angle camera and a multi-point spectral light sensor array arranged around the windshield. The identified content may include: light source type (such as the sun, oncoming vehicle high / low beam headlights, streetlights, ground reflections, etc.), the coordinates, azimuth and pitch angles of the light source in the vehicle coordinate system, the absolute light intensity of the light source and its contrast relative to the ambient light, and the moving speed and trajectory of the light source (for vehicle light sources), etc.

[0020] The driver's line of sight can be obtained in the following two ways: The first method is based on the approximation of the gaze direction by head posture: Specifically, a three-dimensional head model can be fitted by facial feature points (nose tip, corner of eye, corner of mouth, etc.) to output head posture angles (yaw, pitch, roll). Especially when facial features are not obvious, the head posture angles can be used as an approximation of the gaze direction. The second approach is based on accurate gaze estimation using an eye model: Specifically, a three-dimensional model of the eye can be established based on the assumption of a spherical cornea, the Purkinje image (the infrared spot reflected by the cornea) can be detected, the relative position of the pupil center and the corneal reflection can be calculated, and the gaze direction can be accurately estimated.

[0021] The head attitude angles (yaw, pitch, and roll) can be obtained using the Perspective-n-Point (PnP) method, as follows: 1) Feature point detection: Extract key facial feature points (usually including the tip of the nose, the left and right corners of the eyes, the left and right corners of the mouth, etc.); 2) 3D model alignment: Match the detected 2D feature points (i.e., key facial feature points) with a standard 3D head model; 3) Attitude solution: By solving the PnP problem, the rotation matrix R and translation vector t are obtained, and then three Euler angles (yaw, pitch, and roll) are obtained as the head attitude angles. The solution for R and t is as follows: Let the 3D feature point in the world coordinate system be Pi, and the 2D projection point in the image coordinate system be pi. Then pi = K[R|t]Pi, where K is the camera intrinsic parameter matrix. R and t can be calculated through multiple feature points.

[0022] The calculation process for PERCLOS is as follows: Eye state recognition: Calculate eye opening (distance between upper and lower eyelids); set an eye opening threshold. If the eye opening is greater than the eye opening threshold or the eyelid covers no more than 80% of the pupil, the eye is determined to be "open". If the eye opening is not greater than the eye opening threshold or the eyelid covers more than 80% of the pupil, the eye is determined to be "closed". PERCLOS calculation: PERCLOS = (number of frames with eyes closed / total number of frames) × 100%; the sampling period is set to 30 seconds (corresponding to 900 frames). In each sampling period, if the number of frames with eyes closed is less than 180 and the PERCLOS value is less than 20%, the driver is determined to be in a conscious state. If the number of frames with eyes closed is not less than 180 and the PERCLOS value is not less than 20%, the driver is determined to be in a fatigued state. If the driver keeps his eyes closed for more than 2 seconds, the driver is determined to be in a state of extreme fatigue.

[0023] The calculation process for the attention distraction index (specifically, the duration and frequency of the driver's gaze deviating from the road ahead) is as follows: The driver's field of vision is pre-divided into multiple areas, including the road ahead area, the rearview mirror area, and the dashboard area. Based on the driver's gaze direction vector and head posture, the driver's current gaze area (i.e., the field of vision area) is determined. If the driver's gaze is not in the "road ahead area," it is determined that the driver is in a state of gaze deviation (i.e., the driver's gaze is deviating from the road ahead). The duration of each instance of gaze deviation is calculated (i.e., the duration of the driver's gaze deviating from the road ahead), and the number of times the driver is in a state of gaze deviation per unit time is calculated (i.e., the frequency of the driver's gaze deviating from the road ahead). A normal value for the attention distraction index is set (e.g., setting the duration of the driver's gaze deviating from the road ahead). The normal values ​​for duration and frequency of driver's gaze deviating from the road ahead are 1 second and 2 times / minute, respectively, and there are attention distraction thresholds (for example, setting attention distraction thresholds for duration and frequency of driver's gaze deviating from the road ahead as 2 seconds and 5 times / minute, respectively). If the duration and frequency of driver's gaze deviating from the road ahead are both less than the corresponding normal values ​​(for example, the duration of driver's gaze deviating from the road ahead is less than 1 second, and the frequency of driver's gaze deviating from the road ahead is less than 2 times / minute), then the driver is determined to be in a normal state (i.e., the driver's attention is not distracted). If the duration or frequency of driver's gaze deviating from the road ahead is greater than the corresponding attention distraction threshold (for example, the duration of driver's gaze deviating from the road ahead is less than 2 seconds, or the frequency of driver's gaze deviating from the road ahead is less than 5 times / minute), then the driver is determined to be in a normal state (i.e., the driver's attention is not distracted).

[0024] Step S104: Based on the vehicle's driving scenario category, determine the vehicle's visibility safety priority and make a preliminary sunshade decision.

[0025] Step S106: Based on the vehicle's driving scenario category, driver load level, shading demand information, and preliminary shading decision, generate the final shading decision, and control the vehicle's shading execution unit to execute the final shading decision.

[0026] This invention provides a vehicle sunshade control method that, based on the vehicle's driving state, ambient light information, and driver state, determines the vehicle's driving scenario category, driver workload level, and shading requirement information. Based on the vehicle's driving scenario category, it determines the vehicle's visibility safety priority and makes a preliminary sunshade decision. Based on the vehicle's driving scenario category, driver workload level, shading requirement information, and preliminary sunshade decision, it generates a final sunshade decision and controls the vehicle's sunshade execution unit to execute the final sunshade decision. Using this technology, different vehicle information can be integrated to determine the vehicle's visibility safety priority, thereby obtaining and executing the final sunshade decision. This resolves the fundamental contradiction between sunshade and obstructed visibility, and can also cope with complex dynamic light sources, meeting the vehicle's sunshade needs in diverse driving scenarios.

[0027] As one possible implementation, step S102 (i.e., determining the vehicle's driving scenario category, driver workload level, and occlusion requirement information based on the vehicle's driving status, ambient light information, and driver status) may include: Step A1: Determine the vehicle's driving scenario category based on driving status and ambient light information.

[0028] Step A2: Determine the driver load level of the vehicle based on the driver's status.

[0029] Step A3: Determine the vehicle's shading requirements based on ambient light information and driver status.

[0030] For example, step A1 above (i.e., determining the driving scenario category of the vehicle based on driving status and ambient light information) may include: classifying the driving scenario of the vehicle based on driving status and ambient light information using a pre-trained neural network model to determine the driving scenario category of the vehicle.

[0031] Continuing from the previous example, based on vehicle dynamics and environmental data streams, a pre-trained lightweight neural network model is used to determine the micro-driving scenario in which the vehicle is located in real time, thereby determining the vehicle's driving scenario category. The driving scenario category can include: straight-line cruise, following other vehicles, emergency braking, rapid acceleration, driving on sharp curves, lane changing (including the intention period and execution period), merging / exiting on ramps, turning at intersections, parking status, etc. The lightweight neural network model outputs a corresponding confidence probability for each scenario category, and the category to which the scenario with the highest confidence probability value belongs is the vehicle's driving scenario category.

[0032] For example, step A2 above (i.e., determining the driver load level of the vehicle based on the driver's state) may include: a) calculating fatigue, distraction, and workload based on the driver's state; b) determining the driver load coefficient of the vehicle based on the calculated fatigue, distraction, and workload; and c) determining the driver load level of the vehicle based on the driver load coefficient.

[0033] Continuing from the previous example, a comprehensive driver load factor is calculated based on the driver status data stream. This driver load factor is a function of the driver's fatigue level, attention distraction, and current operational load (such as whether the central control screen is being operated). The higher the driver load factor value, the weaker the driver's ability to process additional information (including glare), and the more proactive the driver needs to intervene in the vehicle to ensure driving safety. Assuming the driver load factor DLF ∈ [0,1], the formula for calculating the driver load factor DLF can be: DLF = α × F + β × D + γ × O Where F represents the fatigue component, D represents the attention distraction component, O represents the workload component, and α, β, and γ are weights, α + β + γ = 1 (the values ​​of α, β, and γ can be dynamically adjusted, preferably 0.4, 0.4, and 0.2 respectively). The formula for calculating the fatigue component F (based on PERCLOS) can be: F = min (1, (PERCLOS - T low ) / (T high - T low )) Among them, T low Indicates the slight fatigue threshold (T) low The preferred value is 0.15), T high Indicates the severe fatigue threshold (T) high The preferred value is 0.35. When PERCLOS is less than 0.15, F=0; when PERCLOS is greater than 0.35, F=1. The formula for calculating the attention distraction component D can be: D = min (1, (average deviation duration / 1.5 + number of deviations / 12)) The average deviation duration refers to the average duration of each time the driver's gaze deviates from the road ahead within the sampling period, and the number of deviations refers to the number of times the driver's gaze deviates from the road ahead within the sampling period. For example, if the driver's gaze deviates from the road ahead twice within 30 seconds, and each deviation lasts an average of 1.2 seconds, then D = 1.2 / 1.5 + 2 / 12 = 0.8 + 0.17 = 0.97. The formula for calculating the operating load component O can be: O = min (1, t op / 2) Among them, t op This indicates the duration (in seconds) during which the driver is currently performing non-driving tasks such as operating the central control screen; If the DLF value is in the range of [0, 0.4), it indicates normal load; only recording is performed, and no prompts are made. If the DLF value is in the range of [0.4, 0.65), it indicates slight overload; a mild reminder (such as HUD flashing) is given, and non-critical notifications are delayed. If the DLF value is in the range of [0.65, 0.85), it indicates moderate overload; audible and visual alarms are triggered, and the complexity of vehicle-machine interaction is reduced (such as disabling some touch functions). If the DLF value is in the range of [0.85, 1], it indicates severe overload; it is recommended to stop and rest, and advanced driver assistance system takeover is triggered. Assuming a 30-second window: PERCLOS = 0.28, average deviation duration 1.2 seconds, number of deviations 3, and 1.5 seconds of operation on the central control screen, the driver load factor (DLF) is calculated as follows: F = (0.28 0.15) / (0.35 0.15) = 0.65 D = 1.2 / 1.5 + 3 / 12 = 0.8 + 0.25 = 1.0 O = 1.5 / 2.0 = 0.75 α = 0.4, β = 0.4, γ = 0.20 DLF = 0.4 × 0.65 + 0.4 × 1.0 + 0.2 × 0.75 = 0.26 + 0.4 + 0.15 =0.81 A DLF value of 0.81 indicates that the driver is in a state of "moderate overload". The driver should take the initiative to intervene in the vehicle, and it is recommended to suspend non-driving tasks and give a rest reminder.

[0034] In practical applications, the above time window (i.e., the above sampling period) can be adjusted according to requirements, for example, one frame or 100ms; the weight β of D can also be reduced when the vehicle speed is less than a certain value (e.g., the vehicle speed is less than 10km / h).

[0035] For example, the shading requirement information may include the shading area on the vehicle's windshield and its shading parameters; based on this, the above step A3 (i.e., determining the vehicle's shading requirement information based on ambient light information and driver status) may include: using a preset geometric optics model to perform calculations based on ambient light information and driver status to determine the shading area and its shading parameters.

[0036] Continuing from the previous example, based on the driver's eye coordinates and the coordinates of the identified strong light source, the theoretical shading area to be processed on the windshield plane (or virtual projection plane) and the required theoretical shading depth (specifically, the degree of light transmittance reduction) are calculated in real time using a geometric optics model.

[0037] The calculation process for the theoretical shading area and its theoretical shading depth is as follows: Define the following symbol: E l E r The left and right eyes represent their respective three-dimensional coordinates (unit: mm, world coordinate system), S represents the coordinates of a strong light source (such as oncoming headlights, sunlight, front taillights, streetlights, etc.), and W represents the windshield plane (mathematical model: ax + by + cz + d = 0). For a single eye, the reflected light path satisfies: angle of incidence = angle of reflection, and the incident ray, reflected ray, and normal are coplanar; in plane mirror reflection, we can find the mirror image point S′ (virtual light source) of S relative to plane W:

[0038] Then the reflection point P is the intersection of line ES′ and plane W:

[0039] P is the key point where glare occurs on the windshield; P is calculated separately for the left and right eyes. l (Corresponding to the left eye) and P r (Corresponding to the right eye); Let the diameter of a single pupil be d. p ≈ 2 8 mm, the radius of the projected area (circular area) on the windshield is:

[0040] Among them, f e ≈ 17 mm (approximate focal length of the eyeball); Combine the two circular regions calculated by the left and right eyes, take their union, and extend them by a safety margin. δ (e.g., 20mm), to obtain the final shading region:

[0041] Define the incident angle deviation factor k θ :

[0042] Where θ represents the angle between the line of sight and the reflected light (unit: degrees), and θ0 represents the angle threshold (usually taken as 10 degrees); θ greater than θ0 indicates no occlusion; Define the light source intensity weight k I :

[0043] Where I represents the actual illuminance or luminous intensity of the strong light source, and I0 represents the reference threshold (e.g., I0 = 50 lx for oncoming vehicle headlights). Let the original light transmittance of the glass be T0 (e.g., 70%), the target light transmittance T to be achieved can be expressed as follows: T = T0 × [1 – η k θ k I ] Where η is the maximum modulation depth (e.g., 0.8, meaning T can be reduced to a minimum of T0 × 0.2). The theoretical shading depth can be represented by the reduction in light transmittance, ΔT: ΔT = T0 - T = T0 × η k θ k I .

[0044] As one possible implementation, the final sunshade decision may include a sunshade command and shading parameters; based on this, the step S106 above, which controls the vehicle's sunshade execution unit to execute the final sunshade decision, may include: sending the sunshade command and shading parameters to the vehicle's sunshade execution unit so that the sunshade execution unit executes the sunshade command and shades the shading area according to the shading parameters.

[0045] As one possible implementation, the final shading decision may include an interactive instruction; based on this, after generating the final shading decision in step S106 based on the vehicle's driving scenario category, driver load level, shading demand information, and preliminary shading decision, the vehicle shading control method may further include: sending the interactive instruction to the vehicle's target domain controller so that the target domain controller executes the interactive instruction.

[0046] Continuing from the previous example, the final execution strategy (i.e., the final shading decision) can be a multi-dimensional vector, including the following information: Executable shading commands: These represent the theoretical shading area of ​​the vehicle, indicating whether shading is allowed, prohibited, or delayed, and are sent to the vehicle's sunshade execution unit (such as the sun visor actuator) for execution. Shading parameters: If the theoretical shading area allows shading, then define the outline of the final shading area (which may be a subset of the theoretical shading area), the shading depth (such as the percentage reduction in transmittance), the gradient of the transition zone, etc. Interactive commands: Coordination information that needs to be sent to the corresponding domain controller of the vehicle.

[0047] Following the previous example, the sunshade actuator is driven according to the final execution strategy; the sunshade actuator can be a partitioned electrochromic glass (EC), a polymer dispersed liquid crystal (PDLC) dimming film array, or a servo motor that controls the rotation of a traditional mechanical sunshade. The decision engine can broadcast the decision logic and results of the final execution strategy to the relevant domain controllers via in-vehicle Ethernet or CAN FD. For example, it can send corresponding interactive commands to the cockpit domain controller to display simple icons and text (such as "Side window shading is limited in the curve") on the instrument cluster or head-up display, or adjust the entertainment system volume to provide a gentle reminder; it can send corresponding interactive commands containing the driver load factor and vehicle decision status to the ADAS (Advanced Driver Assistance Systems) domain controller. When the vehicle takes strong shading intervention due to high driver load, it can suggest that the ADAS temporarily enhance the sensitivity of lane keeping assist or forward collision warning; it can send commands to the body domain controller to adjust the tint of the windows or the air conditioning fan speed in a coordinated manner to achieve overall comfort adjustment.

[0048] For ease of understanding, the implementation process of the above vehicle sunshade control method is described below using a specific application as an example.

[0049] Design a vehicle sunshade control system to implement the above-mentioned vehicle sunshade control method, such as Figure 2 As shown, the vehicle sunshade control system has a six-layer architecture: sensor layer (various cameras, buses, other sensors, etc.), data preprocessing layer (including data cleaning, format standardization, preliminary filtering, etc.), fusion and feature extraction layer (including multi-source data fusion, feature extraction, feature vector generation, etc.), two-layer decision engine (including rule / fast decision layer and AI / deep decision layer), execution layer (including sunshade driving logic), and collaborative interaction layer (connected to various domain controllers, such as body domain controller, chassis domain controller, cockpit domain controller, powertrain domain controller, etc.). Figure 3 The workflow logic of the two-layer decision engine is demonstrated, and the decision-making process is shown in detail: the input of the two-layer decision engine is "driving scenario", "driver load" and "theoretical occlusion requirements"; the first layer situational safety evaluator (i.e., rule / fast decision layer) outputs "visual safety priority" and primary policy (allow occlusion / prohibit occlusion / delay occlusion) according to the scenario; the second layer optimization decision engine (i.e., AI / deep decision layer) integrates all the inputs of the two-layer decision engine and the primary policy to generate the final execution policy containing specific parameters.

[0050] Combination Figure 2 and Figure 3 The implementation process of the above-mentioned vehicle sunshade control method is as follows: Step S1: Synchronous acquisition and preprocessing of multi-source heterogeneous data.

[0051] The vehicle sunshade control system collects the above three types of data streams in parallel (i.e., vehicle dynamics and environment data stream, driver status data stream, and traditional optical sensing data stream), and preprocesses the collected data streams (such as data cleaning, format standardization, and preliminary filtering).

[0052] Step S2: Data fusion and contextual feature extraction.

[0053] The vehicle sunshade control system sends the preprocessed data obtained in step S1 to the fusion processing unit (belonging to the fusion and feature extraction layer) to perform the following operations: dynamic driving scene classification to determine the vehicle's driving scene category, driver load level assessment to determine the vehicle's driver load level, basic optical shading requirement (i.e., shading requirement information) calculation to determine the vehicle's shading area and its shading parameters, and constructs and outputs feature vectors based on the driving scene category, driver load level, shading area and its shading parameters.

[0054] Step S3: Optimize decisions using a two-layer decision engine.

[0055] The decision engine receives all features output from step S2 (i.e., the feature vector mentioned above, which includes information such as driving scenario category, driver load level, occlusion area and its occlusion parameters), and performs the following two operations: The first layer of operation involves the scenario safety evaluator calculating and outputting a "visual safety priority" index and its corresponding primary strategy based on the dynamic driving scenario classification results. This index is dominated by the dynamic driving scenario classification results. For example, when the system determines with high confidence that the vehicle is in the "lane change execution period" or "ramp merging period," to prevent obstruction of the lateral view, the system assigns the highest "visual safety priority" to such scenarios and directly outputs a primary strategy of "prohibit obstruction" or "only allow very slight hue cues." When the system determines that the vehicle is "driving on a high-curvature curve," it assigns a high priority to the A-pillar area inside the curve, limiting the depth of obstruction of this area. When the system determines that the vehicle is "cruising in a straight line" and the following distance is stable, the system assigns a lower "visual safety priority," allowing obstruction to be performed.

[0056] The second layer of operation: Based on the primary strategy output by the first layer, the decision-maker optimizes the driver load factor and theoretical shielding requirements (including theoretical shielding area and its theoretical shielding parameters) to generate the final execution strategy.

[0057] Step S4: Strategy Execution and Cross-Domain Collaboration.

[0058] The dual-layer decision engine sends the final execution strategy to the sun visor actuator, driving the sun visor actuator to execute the final execution strategy so as to perform sun shading actions that meet the theoretical shading requirements. The dual-layer decision engine broadcasts the final execution strategy to the relevant domain controllers through the vehicle Ethernet or CAN FD so as to execute the corresponding interactive instructions through the relevant domain controllers and realize cross-domain collaborative interaction.

[0059] In practical applications, the aforementioned vehicle sunshade control system mainly includes the following components: 1. Multimodal sensor array: including forward vision module, cabin vision module, vehicle bus interface, ambient light sensor array, etc.; 2. Central Fusion Processing Unit: This is an automotive-grade system-on-a-chip with AI acceleration capabilities, integrating a dynamic driving scenario classification model and a driver state analysis algorithm; 3. Two-layer decision engine: Deployed as a software module in the central processing unit or a standalone security microcontroller, it includes a context security evaluator and an optimization decision engine; 4. Strategy executor: including the sunshade body (zone dimming type or intelligent mechanical type) and its high-precision drive controller; 5. Vehicle Cooperative Gateway: Responsible for standardizing the instructions from the decision engine and distributing them to the cockpit domain, ADAS domain, and body domain, etc.

[0060] For ease of understanding, specific examples are described below.

[0061] Example 1: A composite scene of a highway curve and strong light.

[0062] Scenario 1: A vehicle is traveling on a mountain highway on a sunny afternoon and is about to enter a sharp left turn, while sunlight shines in from the west at a low angle from the front.

[0063] The implementation process of Example 1 is as follows: Step T1, Perception: The navigation and inertial sensors indicate that there is a high-curvature left turn 500 meters ahead, the steering wheel angle begins to increase continuously, and the light sensor detects a high-intensity, low-angle light source (sun) on the right front side.

[0064] Step T2, Fusion and Evaluation: The dynamic scene classification model outputs "driving on a high curvature curve (turning left)" with high confidence; monitors and displays driver status-related information (such as driver attention concentration, driver load coefficient is moderate); the optical calculation module calculates and outputs the theoretical shading area and its theoretical shading depth. The results output by the optical calculation module reflect that the vehicle needs to be deeply shaded in the right front area of ​​the windshield to eliminate direct sunlight glare.

[0065] Step T3, Decision: First-level assessment: The situational safety evaluator identifies the current scenario as a sharp left turn. Based on preset rules, it assigns a very high "visibility safety priority" to the driver's observation of the left front of the curve exit and the left A-pillar area, and generates a primary strategy. At the same time, the visibility safety priority is also relatively high for the right side area that may affect the observation of the right rearview mirror. The second layer of decision-making: The decision-maker arbitrates the decision. Although the sunlight on the right is strong, the decision-maker determines that the absolute transparency of the left field of vision must be guaranteed. Therefore, the final execution strategy is: to perform moderate shading only on the rightmost local area of ​​the windshield (which does not affect the left front and rearview mirror field of vision at all), and to use an extremely smooth gradient transition for the parts of the theoretically shading area that may intrude on the left field of vision to ensure that there are no hard shading edges.

[0066] Step T4, Execution and Coordination: Drive the corresponding controller to control the PDLC film to darken only in the designated rightmost area; Coordinate the gateway to send information to the instrument panel, briefly displaying an icon (a curve icon and a sun icon superimposed) to prompt the driver "Curve sun, shading optimized".

[0067] Example 2: Urban following scenario when the driver is distracted.

[0068] Scenario 2: In the evening, on a congested urban road, vehicles move in and out of traffic; the brake lights of the vehicle in front frequently illuminate, creating a periodic red glare; the monitoring in the cockpit detects that the driver's gaze frequently and briefly deviates from the road (to look at his mobile phone).

[0069] The implementation process of Example 2 is as follows: Step U1, Perception: Vehicle speed is low and frequently drops to zero. The forward-facing camera identifies the taillights of the vehicle ahead as periodically strong light sources. The in-cabin camera detects that the driver's gaze shifts downwards multiple times, indicating an increased level of driver distraction.

[0070] Step U2, Fusion and Evaluation: The dynamic scene classification model outputs "congested following" with high confidence, and the visual safety priority is medium (attention to the front is required, but the speed is slow); the driver load coefficient is evaluated as high due to distraction; the optical solution module calculates and outputs the theoretical occlusion area and its theoretical occlusion depth. The results output by the optical solution module reflect that the vehicle needs to dynamically track and occlude the taillight area of ​​the vehicle in front.

[0071] Step U3, Decision: First-level assessment: The situational safety evaluator identifies the current driving scenario of the vehicle, assigns "visual safety priority" according to preset rules, and generates a primary strategy. The "visual safety priority" assignment result and the primary strategy reflect that the vehicle is allowed to obscure the taillight area of ​​the vehicle in front in the following scenario, but must ensure that the outline of the vehicle in front and the brake light switch status are distinguishable. Second-level decision: Given the high driver load factor, the optimization decision-maker outputs the following enhanced intervention strategy: perform rapid, deep (but not completely black) masking of the taillight area of ​​the vehicle in front to significantly reduce its strong stimulation to the human eye and help the driver concentrate; at the same time, the instructions contained in this enhanced intervention strategy are used to maintain the weak visibility of the outline of the vehicle in front at the edge of the masked area.

[0072] Step U4, Execution and Coordination: Drive the corresponding controller to control the PDLC film to dynamically track the position of the taillights of the vehicle in front and perform deep dimming; the coordination gateway sends a signal to the cabin domain, triggering a gentle prompt sound, and simultaneously sends the "driver attention distracted" status mark to ADAS, based on which ADAS slightly increases the advance amount of the forward collision warning.

[0073] In practical applications, the following alternative solutions can also be adopted: In addition to using cameras to monitor driver status, capacitive steering wheel sensors can be used to monitor hand grip and heart rate, or voice recognition can be used to analyze driver tone of voice to help determine their level of tension or fatigue, serving as supplementary input to the driver load factor calculation model. The sunshade actuator can be replaced by a mechanical micro louver array; each micro louver unit is independently controlled by a micro motor to open and close the angle, and precisely blocks light through physical means, achieving the same regional and gradient shading effect. The rules in the two-layer decision engine can be replaced by a deep reinforcement learning model. This deep reinforcement learning model takes multimodal data as input, "reduced driver discomfort" and "zero safety incidents" as long-term reward objectives, and learns the optimal occlusion strategy autonomously through training with a large amount of simulation and real vehicle data, and can continuously optimize online. The core decision-making function can be integrated as a software module into the vehicle's intelligent cockpit domain controller or central computing unit to directly access data from each domain, thereby simplifying the hardware structure and enabling deeper resource integration and collaboration.

[0074] The beneficial effects of the above-mentioned vehicle sunshade control method can be mainly reflected in the following aspects: (i) Significantly improve system safety: The sunshade function is incorporated into the vehicle's dynamic safety framework to ensure that the highest priority of visibility safety is not sacrificed in any driving situation. The vehicle can actively suppress or adjust the shading behavior at critical moments (such as lane changing and turning) to prevent new blind spots caused by shading. (ii) Achieving personalized and adaptive experience: It not only responds to ambient light, but also to the state of the "person" and the "vehicle". It can provide more active shading protection for fatigued drivers, reduce unnecessary interference for focused drivers, and optimize decision preferences through continuous learning. (III) Empowering intelligent collaboration of the whole vehicle: Through cross-domain collaboration, the sunshade system is transformed from an information island into an intelligent connected node, which can provide valuable driver state context for ADAS and facilitate the construction of a more comprehensive vehicle safety protection network.

[0075] Based on the above-described vehicle sunshade control method, this invention also provides a vehicle sunshade control device, see [link to relevant documentation]. Figure 4 As shown, the device may include the following modules: The processing module 402 is used to determine the vehicle's driving scenario category, driver load level, and occlusion requirement information based on the vehicle's driving status, ambient light information, and driver status.

[0076] The determination module 404 is used to determine the vehicle's visibility safety priority and make preliminary sunshade decisions based on the vehicle's driving scenario category.

[0077] The decision module 406 is used to generate a final sunshade decision based on the vehicle's driving scenario category, driver load level, shading demand information, and preliminary sunshade decision, and to control the vehicle's sunshade execution unit to execute the final sunshade decision.

[0078] By adopting the above-mentioned vehicle sunshade control device, different vehicle information can be integrated to determine the vehicle's visibility safety priority, thereby obtaining and executing the vehicle's final sunshade decision. This can resolve the fundamental contradiction between sunshade and obstructed visibility, and can also cope with complex dynamic light sources, meeting the vehicle sunshade needs in diverse driving scenarios.

[0079] The vehicle sunshade control device provided in this embodiment of the invention has the same implementation principle and technical effect as the aforementioned method embodiment. For the sake of brevity, any parts not mentioned in the device embodiment can be referred to the corresponding content in the aforementioned method embodiment.

[0080] This invention also provides an electronic device, such as... Figure 5 The diagram shows the structure of the electronic device, which includes a processor 51 and a memory 50. The memory 50 stores computer-executable instructions that can be executed by the processor 51. The processor 51 executes the computer-executable instructions to implement the above-mentioned vehicle sunshade control method.

[0081] exist Figure 5 In the illustrated embodiment, the electronic device further includes a bus 52 and a communication interface 53, wherein the processor 51, the communication interface 53, and the memory 50 are connected via the bus 52.

[0082] The memory 50 may include high-speed random access memory (RAM) and may also include non-volatile memory, such as at least one disk storage device. Communication between this system network element and at least one other network element is achieved through at least one communication interface 53 (which can be wired or wireless), such as the Internet, wide area network, local area network, metropolitan area network, etc. The bus 52 may be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus, or an EISA (Extended Industry Standard Architecture) bus, etc. The bus 52 can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 5 The symbol is represented by a single double-headed arrow, but this does not mean that there is only one bus or one type of bus.

[0083] The processor 51 may be an integrated circuit chip with signal processing capabilities. In implementation, each step of the aforementioned vehicle sunshade control method can be completed through the integrated logic circuitry in the processor 51 or through software instructions. The processor 51 can be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), etc.; it can also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the vehicle sunshade control method disclosed in this embodiment can be directly implemented by a hardware decoding processor, or implemented by a combination of hardware and software modules in the decoding processor. The software modules can reside in random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. The storage medium is located in the memory. The processor 51 reads the information in the memory and, in conjunction with its hardware, completes the steps of the vehicle sunshade control method of the aforementioned embodiment.

[0084] This invention also provides a computer-readable storage medium storing computer-executable instructions. When these computer-executable instructions are called and executed by a processor, they cause the processor to implement the aforementioned vehicle sunshade control method. For specific implementation details, please refer to the foregoing method embodiments, which will not be repeated here.

[0085] The computer program products of the vehicle sunshade control method, device and electronic device provided in the embodiments of the present invention include a computer-readable storage medium storing program code. The instructions included in the program code can be used to execute the vehicle sunshade control method described in the preceding method embodiments. For specific implementation, please refer to the method embodiments, which will not be repeated here.

[0086] Unless otherwise specifically stated, the relative steps, numerical expressions, and values ​​of the components and steps described in these embodiments do not limit the scope of the invention.

[0087] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a processor-executable, non-volatile, computer-readable storage medium. Based on this understanding, the technical solution of this invention, essentially, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0088] In the description of this invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0089] Finally, it should be noted that the above-described embodiments are merely specific implementations of the present invention, used to illustrate the technical solutions of the present invention, and not to limit it. The scope of protection of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments within the technical scope disclosed in the present invention, or make equivalent substitutions for some of the technical features; and these modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for controlling vehicle sunshades, characterized in that, include: Based on the vehicle's driving status, ambient light information, and driver status, determine the vehicle's driving scenario category, driver workload level, and occlusion requirements. Based on the vehicle's driving scenario category, determine the vehicle's visibility safety priority and make preliminary sunshade decisions; Based on the vehicle's driving scenario category, driver load level, shading demand information, and preliminary shading decision, a final shading decision is generated, and the vehicle's shading execution unit is controlled to execute the final shading decision.

2. The method according to claim 1, characterized in that, Based on the vehicle's driving status, ambient light information, and driver status, the vehicle's driving scenario category, driver workload level, and occlusion requirements are determined, including: Based on the driving status and the ambient light information, the driving scenario category of the vehicle is determined; Based on the driver's condition, determine the driver's workload level of the vehicle; Based on the ambient light information and the driver's state, the vehicle's shading requirements are determined.

3. The method according to claim 2, characterized in that, Based on the driving status and the ambient light information, the vehicle's driving scenario category is determined, including: Based on the driving status and the ambient light information, a pre-trained neural network model is used to classify the driving scenario in which the vehicle is located in order to determine the driving scenario category of the vehicle.

4. The method according to claim 2, characterized in that, Based on the driver's condition, the driver's workload level of the vehicle is determined, including: Based on the driver's condition, fatigue level, attention distraction, and workload are calculated. Based on the calculated fatigue level, attention distraction, and operational load, the driver load factor of the vehicle is determined; Based on the driver load factor, the driver load level of the vehicle is determined.

5. The method according to claim 2, characterized in that, The shading requirement information includes the shading area on the vehicle's windshield and its shading parameters; based on the ambient light information and the driver's state, the vehicle's shading requirement information is determined, including: Based on the ambient light information and the driver's state, a preset geometric optics model is used to perform calculations to determine the shading area and its shading parameters.

6. The method according to claim 5, characterized in that, The final sunshade decision includes a sunshade command and the shading parameters; the sunshade execution unit controlling the vehicle executes the final sunshade decision, including: The sunshade command and the shading parameters are sent to the sunshade execution unit of the vehicle, so that the sunshade execution unit executes the sunshade command and shades the shading area according to the shading parameters.

7. The method according to claim 1, characterized in that, The final shading decision includes interactive instructions; after generating the final shading decision based on the vehicle's driving scenario category, driver workload level, shading requirement information, and preliminary shading decision, it also includes: The interaction command is sent to the target domain controller of the vehicle so that the target domain controller executes the interaction command.

8. A vehicle sunshade control device, characterized in that, include: The processing module is used to determine the vehicle's driving scenario category, driver load level, and occlusion requirements based on the vehicle's driving status, ambient light information, and driver status. The determination module is used to determine the vehicle's visibility safety priority and make preliminary sunshade decisions based on the vehicle's driving scenario category; The decision-making module is used to generate a final sunshade decision based on the vehicle's driving scenario category, driver load level, shading demand information, and preliminary sunshade decision, and to control the vehicle's sunshade execution unit to execute the final sunshade decision.

9. An electronic device, characterized in that, The method includes a processor and a memory, the memory storing computer-executable instructions executable by the processor, the processor executing the computer-executable instructions to implement the method of any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions that, when invoked and executed by a processor, cause the processor to perform the method according to any one of claims 1 to 7.

Images