High beam control method, apparatus, device, and storage medium

Abstract

This application provides a high beam control method, device, equipment, and storage medium, relating to the field of vehicle control technology. The method includes: acquiring multi-dimensional scene information and user switching operations on the high beams to generate user operation data; training a neural network model based on the user operation data to obtain a user behavior memory model; generating a high beam control command matching the user's habits based on the user behavior memory model and the vehicle's current real-time driving scenario; and controlling the high beams to turn on or off according to the high beam control command. This method, by collecting user high beam operation habits in different driving scenarios, constructs a personalized user behavior memory model, enabling automatic execution of high beam control matching user habits when the user re-enters the same or similar scenario, achieving personalized active control of the high beams, thereby improving driving safety and convenience.

Description

Technical Field

[0001] This application relates to the field of vehicle control technology, and in particular to a method, device, equipment and storage medium for controlling high beam headlights. Background Technology

[0002] High beams, as a key lighting device for vehicles driving at night, can significantly improve the driver's field of vision and visibility in low-light environments, and significantly enhance the ability to identify road obstacles, pedestrians, non-motorized vehicles and road edges, playing an irreplaceable role in ensuring nighttime driving safety.

[0003] In related technologies, vehicle high beam control mainly relies on manual operation by the driver, who judges the road conditions and ambient lighting conditions and manually turns the high beams on and off. However, when frequently traveling through the same road sections, the driver needs to manually trigger the high beams repeatedly, which affects driving safety. Summary of the Invention

[0004] This application provides a method, device, equipment, and storage medium for controlling high beams, enabling personalized active control of high beams and improving driving safety and convenience.

[0005] In a first aspect, this application provides a method for controlling high beam headlights, the method comprising:

[0006] The system acquires multi-dimensional scene information when a user operates the high beams and the user's on / off operation of the high beams, and generates user operation data. The multi-dimensional scene information includes geographic coordinates, road type, time period category, environmental parameters, and vehicle status data.

[0007] Based on the user operation data, the neural network model is trained to obtain the user behavior memory model;

[0008] Based on the user behavior memory model and the current real-time driving scenario of the vehicle, generate high beam control commands that match the user's habits;

[0009] According to the high beam control command, the high beam is controlled to perform the operation of turning on or off.

[0010] In one possible implementation, the neural network model is a multilayer perceptron model, and the step of training the neural network model based on the user operation data to obtain a user behavior memory model includes:

[0011] Training samples are constructed based on the user operation data. The training samples include the multi-dimensional scene information and the high beam on or off operation corresponding to the multi-dimensional scene information.

[0012] The multilayer perceptron model is trained using the training samples to obtain the user behavior memory model.

[0013] In one possible implementation, generating high beam control commands that match user habits based on the user behavior memory model and the current real-time driving scenario of the vehicle includes:

[0014] The current real-time driving scenario of the vehicle is determined, including the current geographic coordinates, current road type, current time period category, current weather, and current vehicle speed;

[0015] The real-time driving scenario is processed by the user behavior memory model to generate high beam control commands that match the user's habits.

[0016] In one possible implementation, determining the real-time driving scenario currently in which the vehicle is located includes:

[0017] The vehicle's current geographic coordinates are obtained through the vehicle positioning module, and the current road type is determined based on map data matching.

[0018] The current time is obtained through the vehicle clock, and the current time period category is determined based on the current time;

[0019] The current weather is obtained through the vehicle communication module or environmental perception sensor;

[0020] The current vehicle speed is obtained through the vehicle speed sensor;

[0021] The real-time driving scenario is determined by the current geographic coordinates, the current road type, the current time period category, the current weather, and the current vehicle speed.

[0022] In one possible implementation, the method further includes:

[0023] Obtain the user's correction operation for the high beam control command, the correction operation including the user actively turning on the high beam or the user actively turning off the high beam;

[0024] Based on the correction operation and the real-time driving scenario corresponding to the occurrence of the correction operation, correction training samples are generated;

[0025] The user behavior memory model is incrementally trained using the corrected training samples to update the user behavior memory model.

[0026] In one possible implementation, when the high beams are on, the method further includes:

[0027] Simultaneously acquire sensing data from at least two sensing devices, including navigation devices, image acquisition devices, radar devices, and vehicle-to-infrastructure communication devices;

[0028] The at least two types of sensing data are fused to determine whether the light source in front of the vehicle is a moving light source or a fixed light source.

[0029] If the light source in front of the vehicle is a moving light source, a high beam off command is generated; if the light source in front of the vehicle is a fixed light source, the current state of the high beam is maintained.

[0030] In one possible implementation, the method further includes:

[0031] Obtain the current battery level of the vehicle and determine whether the battery level is less than a power threshold;

[0032] If the battery charge is less than the charge threshold, the peak power of the high beam will be limited to below a preset power upper limit, and the illumination distance of the high beam will be adjusted to a preset distance value corresponding to the current vehicle speed.

[0033] If the battery charge is greater than or equal to the charge threshold, the current operating state of the high beam will be maintained.

[0034] Secondly, this application provides a high beam control device, the device comprising:

[0035] The acquisition module is used to acquire multi-dimensional scene information when the user operates the high beam and the user's on / off operation of the high beam, and generate user operation data. The multi-dimensional scene information includes geographic coordinates, road type, time period category, environmental parameters and vehicle status data.

[0036] The first processing module is used to train the neural network model based on the user operation data to obtain a user behavior memory model.

[0037] The second processing module is used to generate high beam control commands that match user habits based on the user behavior memory model and the real-time driving scenario in which the vehicle is currently located.

[0038] The control module is used to control the high beams to perform on or off operations according to the high beam control command.

[0039] Thirdly, this application provides an electronic device, including: a memory and a processor;

[0040] The memory stores computer-executed instructions;

[0041] The processor executes computer execution instructions stored in the memory to implement the method as described in any of the first aspects.

[0042] Fourthly, this application provides a vehicle, including: a vehicle body, and electronic equipment as described in the third aspect.

[0043] Fifthly, this application provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the method described in any of the first aspects above.

[0044] In a sixth aspect, this application provides a computer program product, including a computer program that, when executed by a processor, implements the method described in any of the first aspects above.

[0045] This application provides a method, device, equipment, and storage medium for controlling high beams. The method includes acquiring multi-dimensional scene information and user on / off operations of the high beams when the user operates them, generating user operation data; training a neural network model based on the user operation data to obtain a user behavior memory model; generating high beam control commands matching the user's habits based on the user behavior memory model and the vehicle's current real-time driving scenario; and controlling the high beams to turn on or off according to the high beam control commands. The multi-dimensional scene information includes geographic coordinates, road type, time period category, environmental parameters, and vehicle status data. This method learns the user's operating habits in different scenarios to construct a personalized user behavior memory model, enabling high beam control to have proactive predictive capabilities. When the user re-enters a similar scenario, it automatically executes operations consistent with the user's habits, reducing manual intervention and improving driving safety and convenience. Attached Figure Description

[0046] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0047] Figure 1 A schematic diagram illustrating the application scenarios provided in the embodiments of this application;

[0048] Figure 2 A flowchart illustrating an embodiment of the high beam control method provided in this application;

[0049] Figure 3 A flowchart illustrating Embodiment 2 of the high beam control method provided in this application;

[0050] Figure 4 A flowchart illustrating Embodiment 3 of the high beam control method provided in this application;

[0051] Figure 5 A flowchart illustrating Embodiment 4 of the high beam control method provided in this application;

[0052] Figure 6 A flowchart illustrating Embodiment 5 of the high beam control method provided in this application;

[0053] Figure 7 This is a schematic diagram of the high beam control device provided in the embodiments of this application;

[0054] Figure 8 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application.

[0055] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0056] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0057] High beams, as the core lighting device for vehicles driving at night, can significantly expand the driver's field of vision in low-light environments, improving the ability to identify road obstacles, pedestrians, non-motorized vehicles, and road edges, playing a crucial role in ensuring nighttime driving safety. Especially in complex lighting scenarios such as tunnels, rain, fog, and rural roads, the timely activation and proper deactivation of high beams are directly related to driving safety and user experience.

[0058] In existing technologies, vehicle high beam control mainly relies on manual operation by the driver, who judges the road conditions and ambient lighting conditions and manually turns the high beams on and off. However, this control method lacks initiative. When users frequently drive on the same road sections, they still need to manually trigger the same operation every time, increasing the driving burden and failing to adapt to the user's personalized operating habits.

[0059] To address the aforementioned issues, the inventors considered establishing a high-beam active control mechanism based on user behavior habits. This aims to learn user operating patterns in different scenarios to achieve proactive prediction and autonomous decision-making regarding high beams. Based on this, after numerous experiments, the inventors discovered that by acquiring multi-dimensional scene information and switch operation data from users operating high beams, a neural network model can be used to train the user operation data (including multi-dimensional scene information and corresponding high beam switch operations) to construct a user behavior memory model that includes geographic coordinates, road type, time period category, environmental parameters, and vehicle status data. On this basis, when the vehicle travels to the same or similar scenario again, the model can automatically generate high beam control commands that match the user's habits, thereby achieving proactive control of high beams on recurring road sections and avoiding repeated manual operation by the user. Based on this, this application proposes a high beam control method that can achieve personalized proactive control of high beams by constructing a user behavior memory model, thereby improving driving safety and convenience.

[0060] Figure 1 This is a schematic diagram illustrating an application scenario provided in an embodiment of this application. Please refer to [link / reference]. Figure 1 It can acquire real-time driving scenarios of the vehicle, input them into a user behavior memory model for processing, generate high beam control commands, and then control the high beams to turn on or off based on these commands. The user behavior memory model is trained based on multi-dimensional scenario information and corresponding on / off operations when the user operates the high beams.

[0061] The technical solution of this application and how it solves the above-mentioned technical problems will be described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will be described below with reference to the accompanying drawings.

[0062] Figure 2 This is a flowchart illustrating an embodiment of the high beam control method provided in this application. Please refer to... Figure 2 The method includes:

[0063] S201. Obtain multi-dimensional scene information when the user operates the high beams and the user's on / off operation of the high beams, and generate user operation data.

[0064] The execution subject in this application embodiment can be an electronic device or a high beam control device within an electronic device. The high beam control device can be implemented through software or a combination of software and hardware. The high beam control device can be a processor within an electronic device. For ease of understanding, the following description will use an electronic device as the execution subject.

[0065] In this step, the electronic device can collect and record multi-dimensional scene information and operational behaviors of the user when operating the high beams while driving the vehicle, generating user operation data. The multi-dimensional scene information includes geographic coordinates, road type, time period category, environmental parameters, and vehicle status data.

[0066] Geographic coordinates refer to the vehicle's current latitude and longitude, which identifies the vehicle's specific location when the user activates the high beams. For example, when a user turns on the high beams at night while driving through an intersection or curve, the electronic device can record the corresponding geographic coordinates.

[0067] Regarding road type, this refers to the category of the road the vehicle is currently traveling on. This information is used to distinguish different road conditions such as highways, urban roads, rural roads, tunnels, or bridges. For example, if a user turns on their high beams while driving on a highway, the electronic device can record the current road type as a highway.

[0068] The time period category refers to the time period in which the user operates the high beams. This information is used to distinguish the driving environment characteristics of different time periods, such as early morning, morning, noon, afternoon, dusk, night, late night, and early morning. By finely classifying continuous time periods, it is possible to more accurately capture the user's high beam usage habits under different lighting conditions and traffic flow. For example, when a user operates the high beams at 7:30 PM, the electronic device can record the current time period category as "dusk" according to preset time period classification rules.

[0069] In an alternative implementation, the aforementioned time period category can be replaced with time, that is, directly collecting the specific time point when the user operates the high beams as one of the multi-dimensional scene information. Compared to time period category, time can retain more accurate time information and is suitable for scenarios that need to distinguish user habit differences at different times within the same time period. In practical applications, time period category or time, or a combination of both, can be selected based on model training requirements and data characteristics.

[0070] Regarding environmental parameters, this can refer to information about the external environment when the user operates the high beams. This information reflects natural factors that affect driving visibility, such as weather conditions, light intensity, temperature, or humidity. For example, if a user turns on the high beams while driving in rainy or foggy weather, the electronic device can record the current weather condition as rainy or foggy.

[0071] Vehicle status data refers to the vehicle's operating status information when the user operates the high beams. This information reflects real-time vehicle parameters such as speed, acceleration, steering angle, and battery charge. For example, if a user turns on the high beams at a relatively high speed, the electronic equipment can record the current speed value.

[0072] Through the above data collection, the electronic device can associate and store multi-dimensional scene information of each user's operation of the high beams with the specific type of the operation (on or off), forming a user operation data record. This user operation data can be represented in the following format: {geographic coordinates, road type, time period category, environmental parameters, vehicle status data, operation type}. For example, a complete user operation data record could be: {118.7 degrees East longitude, 31.9 degrees North latitude, highway, dusk, rain / fog, vehicle speed 85 km / h, high beams on}.

[0073] S202. Based on user operation data, train the neural network model to obtain a user behavior memory model.

[0074] In this step, the collected user operation data can be used to train a pre-built neural network model to obtain a user behavior memory model that can represent the user's high beam usage habits.

[0075] In one specific implementation, training samples can first be constructed based on user operation data. Each training sample includes multi-dimensional scene information as input features and the corresponding high beam on / off operation as a label. The geographic coordinates, road type, time period category, environmental parameters, and vehicle status data in the multi-dimensional scene information are preprocessed and converted into vector form suitable for processing by the neural network model.

[0076] For example, geographic coordinates can be converted into relative coordinates or regional codes; road types and time period categories can be represented using one-hot coding; and environmental parameters and vehicle status data can be normalized to eliminate the influence of dimensions.

[0077] In one alternative implementation, the neural network model can be a multilayer perceptron (MLP) model. A multilayer perceptron model includes an input layer, at least one hidden layer, and an output layer. The input layer receives feature vectors corresponding to multidimensional scene information, the hidden layer performs a non-linear transformation using an activation function (such as the ReLU function), and the output layer outputs a prediction of whether the high beams are on or off (e.g., 0 for off, 1 for on).

[0078] Electronic devices can input pre-constructed training samples into a multilayer perceptron model, calculate prediction results through forward propagation, and then adjust the model's network parameters through a backpropagation algorithm (such as stochastic gradient descent) to minimize the loss function (such as cross-entropy loss function) between the prediction results and the actual operation. This process is repeated until the model converges or reaches a preset number of training epochs, thus obtaining the user behavior memory model.

[0079] S203. Based on the user behavior memory model and the current real-time driving scenario of the vehicle, generate high beam control commands that match the user's habits.

[0080] In this step, the electronic device can acquire the real-time driving scenario currently in which the vehicle is located and input the real-time driving scenario into the trained user behavior memory model. The model then outputs a high beam control command that matches the user's habits. The real-time driving scenario includes the current geographic coordinates, current road type, current time period, current weather, and current vehicle speed.

[0081] In one specific implementation, the real-time driving scenario currently in which the vehicle is located can be obtained in the following way:

[0082] The vehicle's current geographic coordinates are obtained through an onboard positioning module, and the current road type is determined by matching map data. This positioning module can be based on systems such as the Global Positioning System (GPS) or the BeiDou Navigation Satellite System. For example, when the vehicle is located at 118.7 degrees east longitude and 31.9 degrees north latitude, the electronic equipment can determine the current road type as a highway using map data.

[0083] The system obtains the current time via the vehicle's clock and determines the current time period category based on preset time period division rules. For example, if the current time is 19:30, the electronic device can determine the current time period category as "dusk" based on preset time period division rules (such as the dusk period being 19:00 to 20:30).

[0084] The current weather is obtained through an onboard communication module and / or environmental perception sensors. The onboard communication module can be a Vehicle-to-Everything (V2X) module, a 4G communication module, or a 5G communication module, used to obtain real-time weather information from the cloud; the environmental perception sensors can be rain sensors or light sensors, used to collect current weather conditions. For example, when the rain sensor detects rainfall, the electronic device can determine that the current weather is "rainy / foggy."

[0085] The current vehicle speed is obtained through vehicle speed sensors. For example, electronic devices can read the current vehicle speed value through wheel speed sensors or through the controller area network bus.

[0086] Specifically, after determining the real-time driving scenario, the electronic device can use the current geographic coordinates, current road type, current time period, current weather, and current vehicle speed as input data. This data is then processed by a user behavior memory model to generate a high beam control command that matches the user's habits. For example, when the vehicle travels to a location where the user frequently uses high beams, the user behavior memory model can output a "turn on high beams" control command based on the current geographic coordinates, current road type, current time period, current weather, and current vehicle speed.

[0087] S204. Control the high beams to turn on or off according to the high beam control command.

[0088] In this step, the electronic device can send corresponding control signals to the high beam control module according to the generated high beam control command, thereby controlling the high beam to turn on or off, and realizing automatic control of the high beam.

[0089] In one specific implementation, the electronic device can send control signals to the high beam control module via a vehicle bus. The vehicle bus can be a controller area network (Controller Area Network) bus or a vehicle Ethernet bus. When the generated high beam control command is "on," the electronic device sends an "on" signal to the high beam control module, which then activates the high beams. When the generated high beam control command is "off," the electronic device sends an "off" signal to the high beam control module, which then extinguishes the high beams.

[0090] In one optional implementation, if the user behavior memory model focuses on scene features other than vehicle speed, the electronic device can also obtain the vehicle's current speed after generating the high beam control command, and independently adjust the execution timing of the high beam control command based on the current speed. For example, when the current speed is greater than a preset high-speed threshold, the execution time of the high beam control command can be advanced by a preset duration; when the current speed is less than a preset low-speed threshold, the execution time of the high beam control command can be delayed by a preset duration. Through the vehicle speed correction mechanism, the matching degree between the execution timing of the high beam control command and the user's driving habits can be further optimized based on the model's decision.

[0091] In this embodiment, the electronic device can acquire multi-dimensional scene information and on / off operations of the user when operating the high beams, generate user operation data, and train a neural network model based on the user operation data to obtain a user behavior memory model. Based on this, the electronic device can acquire the real-time driving scenario currently in which the vehicle is located and input it into the user behavior memory model to generate high beam control commands that match the user's habits, thereby controlling the high beams to perform on / off operations. In the above process, by collecting multi-dimensional scene information of the user when operating the high beams to train the user behavior memory model, the vehicle can automatically learn the user's high beam usage habits on repetitive road sections, transforming manual operation into automated control, reducing the frequency of manual operation by the driver on repetitive road sections, and improving driving safety.

[0092] exist Figure 2 Based on the illustrated embodiment, the following, in conjunction with Figure 3 The above-mentioned high beam control method will be further explained in detail.

[0093] Figure 3This is a flowchart illustrating a second embodiment of the high beam control method provided in this application. Please refer to... Figure 3 The method may include:

[0094] S301. Obtain multi-dimensional scene information when the user operates the high beam headlights and the user's on / off operation of the high beam headlights, and generate user operation data.

[0095] In one specific implementation, the electronic device can record multi-dimensional scene information each time the user manually turns the high beams on or off, and store this information in association with the corresponding operation type to form user operation data.

[0096] Furthermore, multiple user operation data can be organized and stored according to geographic coordinates to form a historical location database. Specifically, the relationship between user operation data and the historical location database can be understood as follows: user operation data is the basic unit that constitutes the historical location database, and a user operation data is generated each time a user operates the high beams; the historical location database then clusters and stores these user operation data according to geographic coordinates, forming a memory structure with location as the core.

[0097] S302. Construct training samples based on user operation data.

[0098] In this step, training samples can be constructed based on user operation data. These training samples include multi-dimensional scene information and the corresponding high beam on / off operations.

[0099] In one specific implementation, the electronic device can extract multiple user operation data from a historical location database. Each user operation data includes geographic coordinates, road type, time period category, environmental parameters, vehicle status data, and corresponding operation type. The multi-dimensional scene information in each user operation data is used as input features, and the corresponding operation type is used as a label to construct a training sample set.

[0100] S303. Train the multilayer perceptron model using training samples to obtain the user behavior memory model.

[0101] In this step, the electronic device can train the multilayer perceptron model based on the constructed training samples, learning the mapping rules between multidimensional scene information and the user's high beam operation. When the number of iterations reaches a preset threshold or the model loss function converges to a preset range, the training is complete, resulting in a user behavior memory model that can represent the user's personalized usage habits.

[0102] Among them, the multilayer perceptron model is a machine learning model based on neural networks. By constructing multilayer nonlinear mapping relationships, it gradually abstracts the multidimensional scene information into high-level features, ultimately outputting the high beam switching decision. The model continuously adjusts the connection weights between neurons, making the output prediction results gradually approximate the user's actual operation, thereby achieving accurate learning of user habits.

[0103] In one alternative implementation, the neural network model can be a convolutional neural network, a recurrent neural network, a long short-term memory network, or a Transformer model, in addition to a multilayer perceptron model.

[0104] For example, when geographic coordinates and road types in multi-dimensional scene information have strong spatial correlations, convolutional neural networks can be used to extract spatial features; when user high-beam operation behavior has strong temporal dependencies, recurrent neural networks or long short-term memory networks can be used to capture operational patterns over time; when it is necessary to process both spatial and temporal features in multi-dimensional scene information simultaneously, the Transformer model can be used, leveraging its self-attention mechanism to globally model the input features, further improving the expressive power and prediction accuracy of the user behavior memory model. In practical applications, the most suitable network structure can be selected based on data characteristics and training results.

[0105] In one alternative implementation, incremental training can be used to continuously optimize the user behavior memory model. When new user operation data is added to the historical location database, the model can be incrementally updated using the new data, enabling the model to dynamically adapt to changes in user habits (such as seasonal changes, improved driving skills, changes in driving routes, etc.).

[0106] S304. Determine the current real-time driving scenario of the vehicle.

[0107] In this step, the electronic device can acquire the real-time driving scenario in which the vehicle is currently located. The real-time driving scenario includes the current geographic coordinates, current road type, current time period, current weather, and current vehicle speed.

[0108] In one specific implementation, the current geographic coordinates can be obtained through the vehicle positioning module and road type can be matched based on map data; the current time can be obtained through the vehicle clock and the time period category can be determined based on the current time; the current weather can be obtained through the vehicle communication module and / or environmental perception sensor; and the current vehicle speed can be obtained through the vehicle speed sensor; thereby determining the real-time driving scenario. Specific implementation details can be found in the description of step S203, and will not be repeated here.

[0109] S305. Process real-time driving scenarios through a user behavior memory model to generate high beam control commands that match user habits.

[0110] In this step, the electronic device can use the determined real-time driving scenario as input data, process it through the user behavior memory model, and generate high beam control commands that match the user's habits.

[0111] In practice, the current geographic coordinates, current road type, current time period, current weather, and current vehicle speed can be used as input data to the user behavior memory model. After forward calculation, the model outputs a prediction result for whether the high beams should be turned on or off. This prediction result is the high beam control command that matches the user's habits.

[0112] For example, when the vehicle travels to a location where the user frequently turns on the high beams, the user behavior memory model can output a "turn on high beams" control command based on the current geographic coordinates, current road type, current time period, current weather, and current vehicle speed; when the vehicle enters a well-lit urban road, the model outputs a "turn off high beams" control command.

[0113] In one optional implementation, the electronic device can also predict the timing of high beam control commands by combining navigation information. Specifically, the electronic device can obtain the vehicle's planned driving path through the navigation module and determine the road segment type ahead based on the planned driving path. When it detects that the vehicle is less than a preset distance threshold from a specific road segment ahead (such as a road segment without streetlights or a tunnel entrance), it can trigger the user behavior memory model in advance to perform reasoning and generate high beam control commands, thereby achieving predictive control of high beams and avoiding the lag problem of responding only after the vehicle reaches the critical road segment.

[0114] S306. Control the high beams to turn on or off according to the high beam control command.

[0115] For example, when the user behavior memory model generates a control command to "turn on the high beams", the electronic device can send an on signal to the high beam control module through the controller local area network bus to turn on the high beams.

[0116] In one alternative implementation, after controlling the high beams, the electronic device can also provide feedback to the user via the vehicle's instrument panel or central control screen. For example, when the high beams turn on automatically, the instrument panel may display "No streetlights ahead, turn on the high beams according to your preference."

[0117] In this embodiment, the electronic device can acquire multi-dimensional scene information and on / off operations when the user operates the high beams, generate user operation data, and construct training samples based on the user operation data. Then, the training samples are used to train a multilayer perceptron model, thereby obtaining a user behavior memory model that can represent the user's personalized usage habits. Based on this, the current real-time driving scenario of the vehicle can be determined, and the real-time driving scenario can be processed through the user behavior memory model to generate high beam control commands that match the user's habits. Finally, the high beams are controlled to turn on or off according to the control commands. In the above process, by constructing training samples to train the multilayer perceptron model, the model can learn the mapping rules between multi-dimensional scene information and user operations, and thus actively output high beam control commands that conform to the user's habits in the real-time driving scenario, making high beam control more intelligent, more precise, and more tailored to the user's individual needs.

[0118] Figure 4 This is a flowchart illustrating Embodiment 3 of the high beam control method provided in this application. Please refer to... Figure 4 In addition to any of the above embodiments, the method further includes:

[0119] S401, Obtain the user's correction operation for the high beam control command.

[0120] In this step, the electronic device can monitor the user's corrective actions regarding the high beam control commands. These corrective actions include the user actively turning on or off the high beams.

[0121] In one specific implementation, users can complete the correction operation through various interactive methods. For example, users can manually turn the high beams on or off using physical buttons or levers; they can control them using voice commands (such as "turn on high beams" or "turn off high beams"); or they can complete the operation by touching virtual buttons on the central control screen or instrument panel through the Human Machine Interface (HMI).

[0122] When a correction operation is detected, the electronic device can record the type of the correction operation (on or off) and simultaneously obtain the real-time driving scenario at the time of the correction operation, including the current geographic coordinates, current road type, current time period category, current weather, and current vehicle speed.

[0123] For example, after the user behavior memory model automatically generates and executes the high beam control command, if the user feels that the automatic control result does not meet their expectations, they can intervene in the high beam operation through any of the aforementioned interaction methods. Specifically, if the model automatically turns off the high beams but the user wants them to remain on, the user can turn them back on via a physical button, by voice inputting "turn on high beams," or by touching the high beam icon on the central control screen. After the electronic device captures this correction operation, it can record the correction type as "on" and simultaneously record the current driving scenario information as a basis for subsequent model optimization.

[0124] S402. Generate correction training samples based on the correction operation and the real-time driving scenario corresponding to the correction operation.

[0125] In one specific implementation, the electronic device uses the real-time driving scenario at the time of the correction operation as input features and the user's correction operation type (on or off) as a label to construct a correction training sample. This correction training sample reflects the user's true operational intent in a specific scenario and corrects the current judgment result of the user behavior memory model.

[0126] For example, when a user manually turns the high beams back on after the model automatically turns them off, the electronic device uses the real-time driving scenario at that moment (e.g., 118.7 degrees east longitude, 31.9 degrees north latitude, rural road, late at night, clear weather, vehicle speed 60 km / h) as input features and "turn on high beams" as the label to generate a corrected training sample. This sample records the user's true preference in a specific scenario, that is, the user's desire to keep the high beams on in that scenario.

[0127] S403. Use the corrected training samples to incrementally train the user behavior memory model and update the user behavior memory model.

[0128] In this step, the electronic device can use the corrected training samples to incrementally train the user behavior memory model, thereby updating the model parameters to adapt to the user's latest operating habits.

[0129] In one specific implementation, the electronic device can add corrected training samples to the training set to fine-tune the user behavior memory model. Since the corrected training samples reflect the discrepancy between the model's current judgment and the user's actual intention, the model can quickly learn and correct this discrepancy through incremental training. Typically, after one or two correction operations, the model can adapt to the user's new habits, achieving a more and more fitting effect with continued use.

[0130] In one alternative implementation, the electronic device can employ an online learning approach, incrementally updating the model each time a correction operation is detected, enabling the model to respond in real time to changes in user habits. Alternatively, the electronic device can temporarily store corrected training samples and then perform incremental training uniformly after accumulating a certain number, thereby improving training efficiency.

[0131] In another alternative implementation, electronic devices can assign higher weights to corrected training samples, enabling them to play a greater role in incremental training and thereby accelerating the model's learning of new habits.

[0132] Optionally, edge computing chips can be deployed in the electronic devices to run user behavior memory models locally. These chips can independently perform incremental training and real-time inference of the model, with a response time controllable to within 100 milliseconds. Even when the vehicle is in an area without network coverage (such as tunnels or remote mountainous areas), the local chip can ensure the continuous availability of high beam control functions, without relying on cloud services. Furthermore, the electronic devices can anonymously upload anonymized user operation data to the cloud for optimizing general scenario models. Through anonymous iteration of fleet data, all users benefit.

[0133] In this embodiment, the system can acquire user correction operations for high beam control commands, generate correction training samples based on the correction operations and corresponding real-time driving scenarios, and incrementally train the user behavior memory model using these training samples to update model parameters, resulting in a user behavior memory model that is continuously synchronized with user habits. Through this feedback iterative learning mechanism, the user behavior memory model can dynamically adapt to changes in user habits, enabling the high beam control strategy to continuously optimize with increased usage, achieving a deep fit between control effectiveness and user preferences.

[0134] Figure 5 This is a flowchart illustrating Embodiment 4 of the high beam control method provided in this application. Please refer to... Figure 5 Based on any of the above embodiments, when the high beam is in the activated state, the method further includes:

[0135] S501, Simultaneously acquire sensing data from at least two sensing devices.

[0136] In this step, the electronic device can simultaneously acquire sensing data from at least two sensing devices to identify and determine the light sources in front of the vehicle. Specifically, the sensing devices may include at least two of the following: navigation devices, image acquisition devices, radar devices, and vehicle-to-everything (V2X) communication devices.

[0137] Navigation devices are used to obtain the vehicle's current geographical location, planned driving route, and road segment attribute information ahead. For example, navigation devices can provide information such as whether the road ahead is a tunnel entrance, a section without streetlights, or a school zone, helping to determine whether there are fixed lighting facilities or scenarios requiring special handling at that location.

[0138] Image acquisition equipment is used to collect image information in front of a vehicle and identify whether there are light sources ahead and their shape characteristics. For example, an in-vehicle camera can capture light sources such as headlights, streetlights, and traffic lights ahead, and use image recognition algorithms to determine the type, color, position, and motion state of the light source.

[0139] Radar equipment is used to detect the distance, speed, and direction of movement of targets ahead. For example, millimeter-wave radar can detect the presence of moving targets ahead and obtain their relative speed, helping to determine whether the light source is an oncoming vehicle, a vehicle traveling in the same direction, or a stationary fixed facility.

[0140] Vehicle-to-infrastructure (V2X) communication equipment is used to exchange information with roadside units or other vehicles to obtain information on road events ahead and vehicle location information. For example, through a V2X module, electronic devices can know in advance that oncoming vehicles are approaching, or receive real-time road event information such as construction areas and accident sections sent by roadside equipment.

[0141] S502. Perform fusion processing on at least two types of perception data to determine whether the light source in front of the vehicle is a moving light source or a fixed light source.

[0142] In this step, at least two types of sensing data can be fused to comprehensively determine the type of light source in front of the vehicle, identifying it as a moving or fixed light source.

[0143] In one specific implementation, the electronic device can employ data-level fusion or decision-level fusion to comprehensively analyze data acquired by different sensing devices. For example, when an image acquisition device detects a light source ahead, the electronic device can combine radar equipment to determine whether the light source has movement characteristics (such as changes in speed or direction), combine navigation equipment to determine whether the location of the light source is a fixed lighting facility (such as a street light, traffic light, or road sign), and combine vehicle-to-infrastructure communication equipment to obtain information about vehicles ahead or road events sent by roadside equipment.

[0144] Multimodal fusion verification can accurately distinguish between moving and stationary light sources. For example, if the radar detects that a light source is moving towards the vehicle and the vehicle-to-infrastructure communication equipment confirms that there is an oncoming vehicle ahead, it can be identified as a moving light source; if the navigation information shows that the location is a street lamp pole and the image recognition identifies it as a fixed light source, it can be identified as a stationary light source.

[0145] Optionally, when the judgments of different sensing devices conflict, the electronic device can use a confidence-weighted approach to determine the final light source type. For example, if the image acquisition device determines it is a moving light source, but the radar device does not detect a moving target, and the vehicle-to-infrastructure communication device does not receive information about oncoming vehicles, the electronic device can combine the confidence levels of each sensing device to determine that the light source in front of the vehicle is a fixed light source.

[0146] S503. If the light source in front of the vehicle is a moving light source, a high beam off command is generated; if the light source in front of the vehicle is a fixed light source, the current state of the high beam is maintained.

[0147] In this step, the electronic device can perform corresponding high beam control operations based on the fusion processing results. When it determines that the light source in front of the vehicle is a moving light source, the electronic device generates a high beam off command; when it determines that the light source in front of the vehicle is a stationary light source, the electronic device maintains the current state of the high beams.

[0148] For example, when an oncoming vehicle is detected, the electronic device can immediately generate a shutdown command to turn off the high beams, thereby preventing glare for the driver of the oncoming vehicle.

[0149] For example, when a vehicle is driving on a well-lit city road, the electronic equipment identifies the fixed light source as a street lamp and can keep the high beams off; when a vehicle is driving on a rural road without streetlights, even if there is a fixed light source in the distance (such as the lights of a distant village), the electronic equipment can still keep the high beams on to provide sufficient illumination for the road ahead.

[0150] In one alternative implementation, the electronic device can also perform zoned control of the high beams to achieve precise obstacle avoidance. Specifically, the high beams can be matrix LEDs, composed of multiple independently controllable light-emitting units. The electronic device can turn off the light-emitting units in the corresponding area based on the target location detected by the sensing device.

[0151] For example, when a pedestrian or non-motorized vehicle is detected ahead, the electronic device can turn off the local light-emitting unit that may illuminate the target, while maintaining normal lighting in other areas. This avoids glare and ensures that the driver's overall view of the road ahead is not affected.

[0152] In this embodiment, the electronic device can simultaneously acquire sensing data from at least two sensing devices, perform fusion processing on the at least two sensing data, and determine whether the light source in front of the vehicle is a moving light source or a fixed light source. If it is determined to be a moving light source, a high beam off command is generated; if it is determined to be a fixed light source, the current state of the high beam is maintained. In the above process, through the multimodal fusion verification mechanism, moving light sources and fixed light sources can be accurately distinguished, effectively avoiding misjudging streetlights as oncoming vehicles and mistakenly turning off the high beams, or misjudging roadside lighting facilities as vehicles and incorrectly executing the high beam off operation, significantly improving the accuracy and reliability of high beam control.

[0153] Figure 6 This is a flowchart illustrating Embodiment 5 of the high beam control method provided in this application. Please refer to... Figure 6 Based on the above embodiment four, the method further includes:

[0154] S601. Obtain the current battery level of the vehicle and determine whether the battery level is less than the power threshold.

[0155] In this step, the current battery level of the vehicle can be obtained through the Battery Management System (BMS), and it can be determined whether the battery level is less than a preset battery level threshold.

[0156] The battery threshold can be set according to the vehicle's actual range requirements and the power consumption characteristics of the high beams. For example, it can be set to 20% of the total battery capacity.

[0157] S602. If the battery charge is less than the charge threshold, the peak power of the high beam will be limited to below the preset power limit value, and the illumination distance of the high beam will be adjusted to a preset distance value corresponding to the current vehicle speed.

[0158] In this step, when the battery level is determined to be less than the power threshold, energy-saving control of the high beams can be implemented to balance lighting safety and driving range.

[0159] In one specific implementation, the electronic device can limit the peak power of the high beam to below a preset power limit. For example, the maximum power of the high beam can be reduced to 70% or 80% of the rated power, thereby reducing the energy consumption of the high beam.

[0160] At the same time, the electronic device can also dynamically adjust the illumination distance of the high beam according to the current vehicle speed, adjusting the illumination distance to a preset distance value corresponding to the current vehicle speed.

[0161] Optionally, the preset distance value can be set to be proportional to the vehicle speed; that is, the faster the vehicle speed, the farther the illumination distance; and the slower the vehicle speed, the closer the illumination distance. For example, when the vehicle speed is 100 km / h, the illumination distance can be adjusted to 150 meters; when the vehicle speed is 60 km / h, the illumination distance can be adjusted to 100 meters; and when the vehicle speed is 30 km / h, the illumination distance can be adjusted to 50 meters.

[0162] It should be noted that the preset distance value should also meet the minimum safety requirements, ensuring that the illumination distance of the high beams is not less than the minimum distance required for driving safety at any vehicle speed.

[0163] S603. If the battery charge is greater than or equal to the charge threshold, the current working state of the high beam will be maintained.

[0164] In this step, when the battery level is determined to be greater than or equal to the power threshold, the electronic device can maintain the current operating state of the high beams without performing additional energy-saving control.

[0165] For example, when the battery is fully charged, the high beams can work normally according to the control commands generated by the user behavior memory model, maintaining the original peak power and illumination distance to ensure the best lighting effect and driving experience.

[0166] In one alternative implementation, the electronic devices can also be linked with an Advanced Driver Assistance System (ADAS). For example, when the vehicle enters a curve, the electronic devices can control the high beams to follow the steering angle, ensuring that the lighting direction is consistent with the vehicle's direction of travel, thus improving safety when driving on curves.

[0167] In another alternative implementation, the electronic device can also interact with roadside units via vehicle-to-infrastructure (V2I) communication equipment to enable advance intervention of high beams. For example, when information about oncoming vehicles is obtained through V2I communication equipment, the electronic device can generate a high beam off command in advance, avoiding a passive response only after the vehicle reaches the meeting point, thereby improving safety and comfort when meeting oncoming traffic. When information about a construction area or accident section ahead is received, the electronic device can turn on the high beams in advance, enhancing the driver's visual anticipation of sudden road conditions. Through the V2I communication mechanism, high beam control can be upgraded from a perception-response mechanism to a predictive intervention mechanism, further improving driving safety.

[0168] In this embodiment, the electronic device can obtain the vehicle's current battery level and determine whether the battery level is less than a threshold. If the battery level is less than the threshold, the peak power of the high beam is limited to below a preset power upper limit, and the illumination distance of the high beam is adjusted to a preset distance value corresponding to the current vehicle speed. If the battery level is greater than or equal to the threshold, the current operating state of the high beam is maintained. In the above process, through the battery level threshold judgment and energy-saving control mechanism, the energy consumption of the high beam can be automatically reduced when the battery level is low, achieving a balance between lighting safety and driving range, better meeting the core requirements of new energy vehicles. At the same time, through the linkage adjustment of vehicle speed and illumination distance, combined with the guarantee of a minimum safe distance, lighting safety at different vehicle speeds can be ensured.

[0169] Figure 7 This is a schematic diagram of the high beam control device provided in an embodiment of this application. Please refer to... Figure 7 The high beam control device 10 includes:

[0170] The acquisition module 11 is used to acquire multi-dimensional scene information when the user operates the high beam and the user's on / off operation of the high beam, and generate user operation data. The multi-dimensional scene information includes geographic coordinates, road type, time period category, environmental parameters and vehicle status data.

[0171] The first processing module 12 is used to train the neural network model based on user operation data to obtain a user behavior memory model.

[0172] The second processing module 13 is used to generate high beam control commands that match user habits based on the user behavior memory model and the real-time driving scenario in which the vehicle is currently located.

[0173] The control module 14 is used to control the high beams to perform on or off operations according to the high beam control command.

[0174] The high beam control device provided in this application embodiment can execute the technical solution shown in the above method embodiment. Its implementation principle and beneficial effects are similar, and will not be described again here.

[0175] In one possible implementation, the neural network model is a multilayer perceptron model, and the first processing module 12 is specifically used for:

[0176] Training samples are constructed based on user operation data. The training samples include multi-dimensional scene information and the corresponding high beam on or off operations.

[0177] The multilayer perceptron model was trained using training samples to obtain a user behavior memory model.

[0178] In one possible implementation, the second processing module 13 is specifically used for:

[0179] Determine the current real-time driving scenario of the vehicle, which includes the current geographic coordinates, current road type, current time period, current weather, and current vehicle speed;

[0180] By processing real-time driving scenarios using a user behavior memory model, high beam control commands that match user habits are generated.

[0181] In one possible implementation, the second processing module 13 is specifically used for:

[0182] The vehicle's current geographic coordinates are obtained through the vehicle positioning module, and the current road type is determined based on map data.

[0183] The current time is obtained through the vehicle's onboard clock, and the current time period category is determined based on the current time.

[0184] The current weather is obtained through the vehicle communication module and / or environmental perception sensors;

[0185] The current vehicle speed is obtained through the vehicle speed sensor;

[0186] Determining the real-time driving scenario includes current geographic coordinates, current road type, current time period, current weather, and current vehicle speed.

[0187] In one possible implementation, the first processing module 12 is further configured to:

[0188] Obtain correction operations from the user's high beam control commands, including user-initiated high beam activation or user-initiated high beam deactivation.

[0189] Based on the correction operation and the real-time driving scenario corresponding to the correction operation, correction training samples are generated;

[0190] The user behavior memory model is incrementally trained using corrected training samples to update the user behavior memory model.

[0191] In one possible implementation, when the high beams are on, the control module 14 is specifically used for:

[0192] Simultaneously acquire perception data from at least two types of perception devices, including navigation devices, image acquisition devices, radar devices, and vehicle-to-infrastructure communication devices;

[0193] At least two types of perception data are fused to determine whether the light source in front of the vehicle is a moving light source or a fixed light source.

[0194] If the light source in front of the vehicle is a moving light source, a command to turn off the high beams will be generated; if the light source in front of the vehicle is a stationary light source, the current state of the high beams will be maintained.

[0195] In one possible implementation, the control module 14 is further configured to:

[0196] Obtain the vehicle's current battery level and determine if the battery level is below a threshold.

[0197] If the battery charge is less than the charge threshold, the peak power of the high beam will be limited to below the preset power limit, and the illumination distance of the high beam will be adjusted to a preset distance value corresponding to the current vehicle speed.

[0198] If the battery level is greater than or equal to the battery threshold, the high beams will remain in their current operating state.

[0199] The high beam control device provided in this application embodiment can execute the technical solution shown in the above method embodiment. Its implementation principle and beneficial effects are similar, and will not be described again here.

[0200] Figure 8 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Please refer to... Figure 8 The electronic device 20 provided in this embodiment can be an in-vehicle controller, a domain controller, or a smart cockpit host. The electronic device 20 includes at least one processor 21 and a memory 22. Optionally, the device 20 also includes a communication component 24. The processor 21, memory 22, and communication component 24 are connected via a bus 23.

[0201] In the specific implementation process, at least one processor 21 executes computer execution instructions stored in memory 22, causing at least one processor 21 to perform the above-described method.

[0202] The specific implementation process of processor 21 can be found in the above method embodiments, and its implementation principle and technical effect are similar. It will not be repeated here.

[0203] In the above embodiments, it should be understood that the processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules within the processor.

[0204] The memory may include random access memory (RAM) and may also include non-volatile memory (NVM), such as at least one disk storage device.

[0205] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of illustration, the buses shown in the accompanying drawings are not limited to a single bus or a single type of bus.

[0206] This application embodiment also provides a vehicle, including a vehicle body and such as Figure 8 The electronic device shown is used to perform the high beam control method in the above method embodiments.

[0207] A computer program product is provided, comprising a computer program that, when executed by a processor, implements the method described above.

[0208] This application also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the above-described method.

[0209] The aforementioned readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The readable storage medium can be any available medium accessible to a general-purpose or special-purpose computer.

[0210] An exemplary readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can reside in an Application Specific Integrated Circuit (ASIC). Alternatively, the processor and the readable storage medium can exist as discrete components in the device.

[0211] The division of units is merely a logical functional division; in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0212] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0213] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0214] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part 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 of 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.

[0215] Those skilled in the art will understand that all or part of the steps of the above-described method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0216] Finally, it should be noted that other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This invention is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein, and is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the invention is limited only by the appended claims.

Claims

1. A method for controlling high beam headlights, characterized in that, The method includes: The system acquires multi-dimensional scene information when a user operates the high beams and the user's on / off operation of the high beams, and generates user operation data. The multi-dimensional scene information includes geographic coordinates, road type, time period category, environmental parameters, and vehicle status data. Based on the user operation data, the neural network model is trained to obtain the user behavior memory model; Based on the user behavior memory model and the current real-time driving scenario of the vehicle, generate high beam control commands that match the user's habits; According to the high beam control command, the high beam is controlled to perform the operation of turning on or off.

2. The method according to claim 1, characterized in that, The neural network model is a multilayer perceptron model. The step of training the neural network model based on the user operation data to obtain a user behavior memory model includes: Training samples are constructed based on the user operation data. The training samples include the multi-dimensional scene information and the high beam on or off operation corresponding to the multi-dimensional scene information. The multilayer perceptron model is trained using the training samples to obtain the user behavior memory model.

3. The method according to claim 1, characterized in that, The step of generating high beam control commands that match user habits based on the user behavior memory model and the current real-time driving scenario of the vehicle includes: The current real-time driving scenario of the vehicle is determined, including the current geographic coordinates, current road type, current time period category, current weather, and current vehicle speed; The real-time driving scenario is processed by the user behavior memory model to generate high beam control commands that match the user's habits.

4. The method according to claim 3, characterized in that, Determining the current real-time driving scenario of the vehicle includes: The vehicle's current geographic coordinates are obtained through the vehicle positioning module, and the current road type is determined based on map data. The current time is obtained through the vehicle clock, and the current time period category is determined based on the current time; The current weather is obtained through the vehicle communication module and / or environmental perception sensors; The current vehicle speed is obtained through the vehicle speed sensor; The real-time driving scenario is determined by the current geographic coordinates, the current road type, the current time period category, the current weather, and the current vehicle speed.

5. The method according to any one of claims 1-4, characterized in that, The method further includes: Obtain the user's correction operation for the high beam control command, the correction operation including the user actively turning on the high beam or the user actively turning off the high beam; Based on the correction operation and the real-time driving scenario corresponding to the occurrence of the correction operation, correction training samples are generated; The user behavior memory model is incrementally trained using the corrected training samples to update the user behavior memory model.

6. The method according to any one of claims 1-4, characterized in that, When the high beams are on, the method further includes: Simultaneously acquire sensing data from at least two sensing devices, including navigation devices, image acquisition devices, radar devices, and vehicle-to-infrastructure communication devices; The at least two types of sensing data are fused to determine whether the light source in front of the vehicle is a moving light source or a fixed light source. If the light source in front of the vehicle is a moving light source, a high beam off command is generated; if the light source in front of the vehicle is a fixed light source, the current state of the high beam is maintained.

7. The method according to claim 6, characterized in that, The method further includes: Obtain the current battery level of the vehicle and determine whether the battery level is less than a power threshold; If the battery charge is less than the charge threshold, the peak power of the high beam will be limited to below a preset power upper limit, and the illumination distance of the high beam will be adjusted to a preset distance value corresponding to the current vehicle speed. If the battery charge is greater than or equal to the charge threshold, the current operating state of the high beam will be maintained.

8. A high beam control device, characterized in that, The device includes: The acquisition module is used to acquire multi-dimensional scene information when the user operates the high beam and the user's on / off operation of the high beam, and generate user operation data. The multi-dimensional scene information includes geographic coordinates, road type, time period category, environmental parameters and vehicle status data. The first processing module is used to train the neural network model based on the user operation data to obtain a user behavior memory model. The second processing module is used to generate high beam control commands that match user habits based on the user behavior memory model and the real-time driving scenario in which the vehicle is currently located. The control module is used to control the high beams to perform on or off operations according to the high beam control command.

9. An electronic device, characterized in that, include: Memory, processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory, causing the processor to perform the method as described in any one of claims 1-7.

10. A vehicle, characterized in that, include: The vehicle body, and the electronic device as described in claim 9.

11. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of claims 1-7.

Images