High-applicability illumination control method, controller, storage medium and program product

Abstract

The invention discloses a high-applicability illumination control method, a controller, a storage medium and a program product, and relates to the field of automobile illumination, and the method comprises the steps: collecting traffic density data, road condition data and weather condition data in a preset range around a vehicle; generating vehicle scene feature data according to the traffic density data, the road condition data and the weather condition data; inputting the vehicle scene feature data into a preset illumination demand model to obtain vehicle recommended illumination parameters; and sending the vehicle recommended illumination parameters to the programmable LED array to control each LED pixel point in the programmable LED array to adjust the illumination brightness, the illumination direction and the illumination color temperature according to the vehicle recommended illumination parameters. By implementing the method, the accuracy of automatic illumination adjustment is improved.

Description

Technical Field

[0001] The present application relates to the field of automotive lighting, and in particular to a highly applicable lighting control method, controller, storage medium and program product. Background Art

[0002] When driving at night or in bad weather conditions, a good vehicle lighting system is particularly important to ensure driving safety. Therefore, the intelligence, adaptability and efficiency of vehicle lighting systems have become the research focus in the field of automotive lighting.

[0003] At present, the vehicle lighting system mainly uses light sensors to detect the intensity of ambient light and automatically adjust the brightness and switching of the lights to adapt to different driving environments. For example, when the external environment becomes dark, the vehicle lighting system will trigger the lights to turn on automatically to ensure driving safety.

[0004] Although the relevant technology can realize automatic adjustment of lighting systems to a certain extent, there are still some limitations. When the vehicle is driving in rainy and foggy weather, complex road conditions or traffic-dense areas, the vehicle lighting system is difficult to meet the actual lighting needs in different scenarios. For example, in foggy or rainy days, relying solely on light sensors to adjust the lights may not provide sufficient visual clarity, thus affecting driving safety. Summary of the invention

[0005] The present application provides a highly applicable lighting control method, controller, storage medium and program product for improving the accuracy of automatic lighting adjustment.

[0006] In a first aspect, the present application provides a highly applicable lighting control method, which is applied to a controller, and the method includes: collecting traffic density data, road condition data and weather condition data within a preset range around a vehicle, the traffic density data is used to represent the number distribution of vehicles and pedestrians, the road condition data is used to represent the road structure characteristics and road environment status, and the weather condition data is used to represent the atmospheric environment parameters that affect the driving field of view; based on the traffic density data, the road condition data and the weather condition data, generating vehicle scene feature data, the vehicle scene feature data is used to represent the comprehensive lighting demand characteristics of the vehicle in the current driving scene; inputting the vehicle scene feature data into a preset lighting demand model to obtain vehicle recommended lighting parameters, the input layer of the lighting demand model receives the scene feature matrix, and the output layer generates ideal lighting parameters, the vehicle recommended lighting parameters include light brightness value, light direction value and light color temperature value; sending the vehicle recommended lighting parameters to a programmable LED array to control each LED pixel in the programmable LED array to adjust the light brightness, light direction and light color temperature according to the vehicle recommended lighting parameters.

[0007] By adopting the above technical solution, the controller collects traffic density data, road condition data and weather condition data within a preset range around the vehicle in real time, and converts these multi-dimensional data into vehicle scene feature data to fully perceive the complexity of the driving environment. The controller generates accurate vehicle recommended lighting parameters through a preset lighting demand model, and the programmable LED array dynamically adjusts the light intensity, light direction and light color temperature according to the vehicle recommended lighting parameters. This scene-aware adaptive lighting control method significantly improves driving safety at night and in severe weather conditions. Compared with traditional vehicle lighting systems that rely solely on light sensors, it can effectively alleviate the problem that related technologies are difficult to meet the actual lighting needs in different scenarios, so that they can better adapt to complex and changing road environments and provide drivers with optimal viewing conditions.

[0008] In combination with some embodiments of the first aspect, in some embodiments, vehicle scene feature data is generated based on the traffic density data, the road condition data and the weather condition data, specifically including: converting the traffic density data into a traffic feature matrix; converting the road condition data into a road feature matrix; converting the weather condition data into a weather feature matrix; and performing feature fusion on the traffic feature matrix, the road feature matrix and the weather feature matrix to obtain the vehicle scene feature data.

[0009] By adopting the above technical solutions, the traffic feature matrix reflects the dynamic distribution of surrounding vehicles and pedestrians, the road feature matrix reflects the spatial characteristics of road structure and road conditions, and the weather feature matrix reflects the degree of influence of the atmospheric environment on the field of vision. The controller integrates these three types of feature matrices into unified vehicle scene feature data through feature fusion algorithms, which not only retains the key information of each dimension, but also establishes the correlation between each dimension, providing standardized feature input for subsequent lighting parameter optimization, significantly improving the accuracy and intelligence level of lighting control.

[0010] In combination with some embodiments of the first aspect, in some embodiments, before the step of inputting the vehicle scene feature data into a preset lighting requirement model to obtain vehicle recommended lighting parameters, the input layer of the lighting requirement model receives a scene feature matrix, and the output layer generates ideal lighting parameters, and the vehicle recommended lighting parameters include light illuminance values, lighting direction values, and lighting color temperature values, the method also includes: obtaining ideal lighting parameters corresponding to the driving scene and a scene feature matrix corresponding to the driving scene; using the scene feature matrix as input data and the ideal lighting parameters as output data to train the preset model; and obtaining the lighting requirement model when the accuracy of the preset model exceeds a preset accuracy threshold.

[0011] By adopting the above technical solution, the controller collects the ideal lighting parameters corresponding to the driving scene and the scene feature matrix corresponding to the driving scene to establish a data-driven lighting demand model training process. This model training method based on measured data enables the vehicle lighting system to accurately understand the optimal lighting requirements under different driving scenarios and generate corresponding lighting control parameters. Compared with the lighting control parameters generated by preset rules, the lighting demand model has stronger scene adaptability and control accuracy, so it can better meet the dynamic lighting requirements in actual road environments.

[0012] In combination with some embodiments of the first aspect, in some embodiments, the vehicle recommended lighting parameters are sent to a programmable LED array to control each LED pixel in the programmable LED array to adjust the light brightness, lighting direction and lighting color temperature according to the vehicle recommended lighting parameters, specifically including: based on the physical layout structure of the programmable LED array, determining the independent control parameters of each pixel according to the vehicle recommended lighting parameters; according to the independent control parameters, grouping the LED pixels located in the same lighting area in the programmable LED array and determining the priority coefficient of each group of LED pixels; based on the priority coefficient, calculating the collaborative control parameters of each group of LED pixels, the collaborative control parameters are used to achieve synchronous dimming of the LED pixels in the group; according to the independent control parameters and the collaborative control parameters, generating an LED drive timing, the LED drive timing is used to control the on-off time and dimming amplitude of each LED pixel.

[0013] By adopting the above technical solution, first, the controller calculates independent control parameters for each LED pixel based on the physical layout structure of the programmable LED array to ensure basic control accuracy. Then, the controller groups the LED pixels in the same lighting area and assigns priority coefficients to establish a regionalized collaborative control system. Next, the controller calculates the collaborative control parameters for each group of LED pixels based on the priority coefficients to ensure the synchronous dimming of the LEDs in the group and avoid uneven display and flickering problems. Finally, the controller generates the LED drive timing to accurately control the on-off time and dimming amplitude of each pixel, so that the entire programmable LED array can complete complex lighting tasks in a coordinated manner. This multi-level control architecture not only ensures the continuity of the overall lighting effect, but also maintains the ability to accurately control a single LED pixel, significantly improving the controllability and lighting quality of the vehicle lighting system.

[0014] In combination with some embodiments of the first aspect, in some embodiments, after generating the LED driving timing according to the independent control parameter and the collaborative control parameter, the method also includes: monitoring the temperature change and current change of each LED pixel point; calculating the performance attenuation index of each LED pixel point based on the temperature change and the current change; and dynamically adjusting the priority coefficient of each group of LED pixel points based on the performance attenuation index.

[0015] By adopting the above technical solutions, the controller establishes a real-time performance monitoring and dynamic optimization mechanism for the programmable LED array, which not only extends the service life of the programmable LED array, but also ensures that the vehicle lighting system can maintain the ideal lighting effect when the LED performance gradually degrades. At the same time, by dynamically adjusting the priority coefficient, the controller can reasonably distribute the load, avoid local overload, and improve the reliability and durability of the vehicle lighting system.

[0016] In combination with some embodiments of the first aspect, in some embodiments, after the step of sending the vehicle recommended lighting parameters to the programmable LED array to control each LED pixel in the programmable LED array to adjust the light brightness, light direction and light color temperature according to the vehicle recommended lighting parameters, the method also includes: collecting facial image data of the driver within a preset time period; based on the facial image data, extracting the driver's pupil size change rate, gaze point offset angle and blinking frequency; generating visual fatigue according to the pupil size change rate, the gaze point offset angle and the blinking frequency; when the visual fatigue exceeds a preset visual fatigue threshold, adjusting the light brightness and light color temperature of the programmable LED array according to a preset visual comfort curve.

[0017] By adopting the above technical solution, the controller monitors the driver's facial image data in real time, extracts key indicators such as pupil size change rate, gaze point offset angle and blinking frequency, so as to accurately assess the driver's visual fatigue level. When the visual fatigue exceeds the preset visual fatigue threshold, the controller automatically adjusts the light brightness and light color temperature of the programmable LED array according to the preset visual comfort curve to create a more comfortable visual environment for the driver. This lighting control solution based on physiological feedback effectively reduces the visual fatigue caused by long-term driving, improves driving safety, and also enhances the driver's comfort and driving experience.

[0018] In combination with some embodiments of the first aspect, in some embodiments, after the step of sending the vehicle recommended lighting parameters to the programmable LED array to control each LED pixel in the programmable LED array to adjust the light brightness, lighting direction and lighting color temperature according to the vehicle recommended lighting parameters, the method also includes: detecting the position information and speed information of the target object within a preset distance in front of the vehicle through a forward millimeter-wave radar; acquiring the reflection characteristics and surface material of the target object through a visual sensor; calculating the contrast of the target object under the current lighting conditions based on the position information, the speed information, the reflection characteristics and the surface material; when the contrast is lower than a preset contrast threshold, adaptively adjusting the lighting direction to improve the visibility of the target object.

[0019] By adopting the above technical solutions, the controller integrates millimeter-wave radar and visual sensors to build an active target object lighting enhancement mechanism, which significantly improves the ability to identify potentially dangerous objects when driving at night. Especially in complex road environments, it can actively enhance the visibility of target objects and provide drivers with a clearer field of vision, thereby effectively reducing the risk of accidents.

[0020] In a second aspect, an embodiment of the present application provides a controller, comprising: one or more processors and a memory; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code includes computer instructions, and the one or more processors call the computer instructions to enable the controller to execute the method described in the first aspect and any possible implementation method of the first aspect.

[0021] In a third aspect, an embodiment of the present application provides a computer program product comprising instructions, which, when the computer program product runs on a controller, enables the controller to execute the method described in the first aspect and any possible implementation of the first aspect.

[0022] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, comprising instructions, which, when executed on a controller, cause the controller to execute the method described in the first aspect and any possible implementation of the first aspect.

[0023] It is understandable that the controller provided in the second aspect, the computer program product provided in the third aspect, and the computer storage medium provided in the fourth aspect are all used to execute the method provided in the embodiment of the present application. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects in the corresponding method, which will not be repeated here.

[0024] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: 1. By adopting the above technical solution, the controller collects traffic density data, road condition data and weather condition data within a preset range around the vehicle in real time, and converts these multi-dimensional data into vehicle scene feature data to fully perceive the complexity of the driving environment. The controller generates accurate vehicle recommended lighting parameters through a preset lighting demand model, and the programmable LED array dynamically adjusts the light intensity, light direction and light color temperature according to the vehicle recommended lighting parameters. This scene-aware adaptive lighting control method significantly improves driving safety at night and in severe weather conditions. Compared with traditional vehicle lighting systems that rely solely on light sensors, it can effectively alleviate the problem that related technologies are difficult to meet the actual lighting needs in different scenarios, so that they can better adapt to complex and changing road environments and provide drivers with optimal viewing conditions.

[0025] 2. By adopting the above technical solution, the controller monitors the driver's facial image data in real time, extracts key indicators such as pupil size change rate, gaze point offset angle and blinking frequency, so as to accurately assess the driver's visual fatigue. When the visual fatigue exceeds the preset visual fatigue threshold, the controller automatically adjusts the light brightness and light color temperature of the programmable LED array according to the preset visual comfort curve to create a more comfortable visual environment for the driver. This lighting control solution based on physiological feedback effectively reduces the visual fatigue caused by long-term driving, improves driving safety, and also enhances the driver's comfort and driving experience.

[0026] 3. By adopting the above technical solutions, the controller integrates millimeter-wave radar and visual sensors to build an active target object lighting enhancement mechanism, which significantly improves the ability to identify potentially dangerous objects when driving at night. Especially in complex road environments, it can actively enhance the visibility of target objects and provide drivers with a clearer field of vision, thereby effectively reducing the risk of accidents. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 It is a flowchart of a highly applicable lighting control method in an embodiment of the present application; Figure 2 is another flowchart of a highly applicable lighting control method in an embodiment of the present application; Figure 3 It is a schematic diagram of a physical device structure of a controller in an embodiment of the present application. DETAILED DESCRIPTION

[0028] The terms used in the following embodiments of the present application are only for the purpose of describing specific embodiments, and are not intended to be used as limitations to the present application. As used in the specification of the present application, the singular expressions "one", "a kind of", "above", "the" and "this" are intended to also include plural expressions, unless there is a clear indication to the contrary in the context. It should also be understood that the term "and / or" used in the present application refers to any or all possible combinations comprising one or more of the listed items.

[0029] In the following, the terms "first" and "second" are used for descriptive purposes only and are not to be understood as suggesting or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features, and in the description of the embodiments of the present application, unless otherwise specified, "plurality" means two or more.

[0030] The following is a description of the process of the method provided by this implementation. Figure 1 , which is a flow chart of a highly applicable lighting control method in an embodiment of the present application.

[0031] S101, collecting traffic density data, road condition data and weather condition data within a preset range around the vehicle, wherein the traffic density data is used to indicate the number distribution of vehicles and pedestrians, the road condition data is used to indicate road structure characteristics and road environment status, and the weather condition data is used to indicate atmospheric environment parameters that affect driving vision; Among them, the controller is an electronic device used to execute the vehicle lighting control algorithm, including a central processing unit, a memory and a communication interface; the preset range refers to a specific distance range centered on the vehicle, usually a fan-shaped area of ​​100-200 meters in front and 20-30 meters on the side; traffic density data is used to indicate the number and distribution of vehicles and pedestrians per unit area or unit road length; road condition data is a collection of information reflecting the geometric characteristics of the road and the road condition environment, including road slope, curvature, road material and humidity; weather condition data refers to environmental parameters that affect driving vision, including visibility, precipitation, haze concentration and other atmospheric condition indicators.

[0032] Specifically, the controller uses the on-board millimeter-wave radar to obtain the location and speed information of traffic participants within a preset range around the vehicle, and combines the image data from the on-board camera to identify the specific types and numbers of vehicles and pedestrians, thereby generating traffic density data. At the same time, the controller uses lidar scanning to obtain the three-dimensional contour features of the road, and cooperates with optical sensors to detect the reflectivity and roughness of the road surface to generate road condition data. In addition, the controller also uses on-board meteorological sensors to measure parameters such as temperature, humidity, and air pressure in real time, and combines the data from the rain sensor and light sensor to comprehensively generate weather condition data.

[0033] S102, generating vehicle scene feature data according to the traffic density data, the road condition data and the weather condition data, wherein the vehicle scene feature data is used to represent comprehensive lighting demand characteristics of the vehicle in the current driving scene; Among them, vehicle scene feature data refers to the digital description of the current driving scene, which includes feature parameters of multiple dimensions; comprehensive lighting demand characteristics refer to the ideal lighting configuration requirements calculated based on the current driving scene; the current driving scene refers to the immediate road environment status of the vehicle.

[0034] Specifically, first, the controller performs spatial partitioning on the traffic density data, divides the area around the vehicle into multiple sector-shaped areas, calculates the vehicle density value and pedestrian density value in each area, and generates a traffic feature matrix. Then, the controller extracts features from the road condition data, including calculating parameters such as the road curvature coefficient, slope coefficient, and road friction coefficient, and constructs a road feature matrix. At the same time, the controller normalizes the weather condition data to generate a weather feature matrix containing parameters such as visibility index, precipitation intensity index, and atmospheric transmittance. Finally, the controller uses a deep neural network to perform a fusion operation on these three feature matrices and outputs a unified scene feature vector, the dimension of which is usually 128-256, and each dimension corresponds to a specific scene attribute.

[0035] Optionally, under normal circumstances, the generation of vehicle scene feature data based on the traffic density data, the road condition data and the weather condition data can be achieved in the following ways, which are not limited here: converting the traffic density data into a traffic feature matrix; converting the road condition data into a road feature matrix; converting the weather condition data into a weather feature matrix; and performing feature fusion on the traffic feature matrix, the road feature matrix and the weather feature matrix to obtain the vehicle scene feature data.

[0036] S103, inputting the vehicle scene feature data into a preset lighting demand model to obtain vehicle recommended lighting parameters, wherein the input layer of the lighting demand model receives the scene feature matrix, and the output layer generates ideal lighting parameters, wherein the vehicle recommended lighting parameters include light illuminance value, light direction value, and light color temperature value; Among them, the lighting demand model refers to a trained deep neural network model, which is used to map the scene feature matrix into the ideal lighting parameter number; the input layer refers to the data receiving layer of the neural network, which is used to receive the scene feature matrix; the output layer refers to the result output layer of the neural network, which is used to generate the ideal lighting parameters; the scene feature matrix refers to the multidimensional data matrix that describes the current driving scene; the ideal lighting parameters refer to the most suitable lighting configuration under the current driving scene; the light illuminance value is used to indicate the luminous intensity of the LED, usually in nits; the light direction value is used to indicate the projection angle of the LED light beam, including the angle parameters in the horizontal and vertical directions; the light color temperature value is used to indicate the color temperature of the light emitted by the LED, in Kelvin, which is used to adjust the warmth or coldness of the light.

[0037] Specifically, first, the controller reconstructs the 128-dimensional scene feature vector into an 8×16 scene feature matrix as the input data of the lighting demand model. The lighting demand model adopts a multi-layer convolutional neural network structure, including 4 convolutional layers and 2 fully connected layers, each of which is equipped with a ReLU activation function and a Batch Normalization layer. The middle layer of the lighting demand model also incorporates an attention mechanism, which can adaptively focus on important scene features. After forward calculation, the output layer of the lighting demand model generates ideal lighting parameters as the recommended lighting parameters for the vehicle. The recommended lighting parameters for the vehicle include: light illuminance values ​​in 20 different areas (range 0-100%), 16 groups of lighting direction adjustment angles (horizontal ±30°, vertical ±15°) and lighting color temperature values ​​in 5 areas (range 3000K-6500K). The delay of the entire calculation process is controlled within 10 milliseconds to ensure that the recommended lighting parameters for the vehicle can be updated in real time.

[0038] S104, sending the vehicle recommended lighting parameters to a programmable LED array to control each LED pixel in the programmable LED array to adjust light brightness, light direction and light color temperature according to the vehicle recommended lighting parameters.

[0039] Among them, the programmable LED array is a lighting device composed of multiple independently controlled LED pixels; an LED pixel refers to a single controllable light-emitting unit in a programmable LED array, which has independent light brightness, light angle and light color temperature adjustment capabilities.

[0040] Specifically, the controller parses the recommended lighting parameters of the vehicle into specific control instructions for each LED pixel. For the equipped 8×24 matrix programmable LED array (a total of 192 LED pixels), the controller allocates a 16-bit control word to each LED pixel, of which 6 bits are used for light brightness adjustment (64 levels), 6 bits are used for light direction adjustment (32 levels horizontally, 32 levels vertically), and 4 bits are used for light color temperature adjustment (16 levels). The controller sends these control words to the LED driver chip at a rate of 2Mbps via the CAN bus. The LED driver chip uses PWM modulation technology to accurately control each LED pixel at a frequency of 20kHz, with a resolution of 12 bits (4096 levels). At the same time, the controller also implements a gradual control algorithm to ensure smooth transition during parameter adjustment and avoid visual discomfort caused by sudden changes. The response time of the entire control process is controlled within 5 milliseconds, realizing real-time adjustment of the lighting effect.

[0041] By adopting the above technical solution, the controller collects traffic density data, road condition data and weather condition data within a preset range around the vehicle in real time, and converts these multi-dimensional data into vehicle scene feature data to fully perceive the complexity of the driving environment. The controller generates accurate vehicle recommended lighting parameters through a preset lighting demand model, and the programmable LED array dynamically adjusts the light intensity, light direction and light color temperature according to the vehicle recommended lighting parameters. This scene-aware adaptive lighting control method significantly improves driving safety at night and in severe weather conditions. Compared with traditional vehicle lighting systems that rely solely on light sensors, it can effectively alleviate the problem that related technologies are difficult to meet the actual lighting needs in different scenarios, so that they can better adapt to complex and changing road environments and provide drivers with optimal viewing conditions.

[0042] The following is a more detailed description of the process of the method provided by this implementation. Figure 2 , is another flow chart of a highly applicable lighting control method in an embodiment of the present application.

[0043] S201, collecting traffic density data, road condition data and weather condition data within a preset range around the vehicle, wherein the traffic density data is used to indicate the number distribution of vehicles and pedestrians, the road condition data is used to indicate road structure characteristics and road environment status, and the weather condition data is used to indicate atmospheric environment parameters that affect driving vision; For details, please refer to step S101, which will not be described in detail here.

[0044] S202, generating vehicle scene feature data according to the traffic density data, the road condition data and the weather condition data, wherein the vehicle scene feature data is used to represent comprehensive lighting demand characteristics of the vehicle in the current driving scene; For details, please refer to step S102, which will not be described in detail here.

[0045] S203, obtaining an ideal lighting parameter corresponding to a driving scene and a scene feature matrix corresponding to the driving scene; using the scene feature matrix as input data and the ideal lighting parameter as output data to train a preset model; when the accuracy of the preset model exceeds a preset accuracy threshold, obtaining the lighting demand model; The steps to build a deep learning-based lighting demand model are as follows: First, the controller obtains the ideal lighting parameters corresponding to the driving scene and the scene feature matrix corresponding to the driving scene. Driving scenes may include driving on highways, urban roads, rural roads, tunnels, driving in bad weather (rainy and snowy days), etc. The ideal lighting parameters refer to the optimal lighting parameters used by the vehicle in the aforementioned driving scenes. The scene feature matrix corresponding to the driving scene is determined by obtaining traffic density data, road condition data and weather condition data in the driving scene. The specific implementation method can refer to step S102 and will not be repeated here. The controller stores the collected scene feature matrix and ideal lighting parameters in the data set D, and the format of each data is (scene feature matrix, ideal lighting parameters). Among them, the scene feature matrix is ​​the input feature of the model training, and the ideal lighting parameters are the output features of the model training.

[0046] Secondly, the controller performs data preprocessing on the data set D and deletes the missing data and abnormal data in the data set D.

[0047] Then, the controller constructs a LSTM-based recurrent neural network, which includes an input layer, two LSTM hidden layers, a fully connected layer, and an output layer. The input layer inputs the scene feature matrix, the number of nodes in the hidden layer is set to 64, the number of nodes in the fully connected layer is set to 32, and the output layer outputs the ideal lighting parameters.

[0048] Next, the controller uses the Adam optimizer, the learning rate is set to 0.001, and the training batch size is 32. It can also be set according to the actual situation and is not limited here. 80% of the data in the data set is divided into a training set and 20% is divided into a validation set. 100 epochs are trained and the model with the highest accuracy of the validation set is saved. It can also be set according to the actual situation and is not limited here. Epoch is the process in which the training data set passes through the neural network once. In machine learning and deep learning, epoch is a unit used to measure the number of times the entire training set is repeatedly learned. Specifically, when the neural network completes a forward calculation and back propagation process, that is, all data has been processed by the network once, this completes an epoch. The controller uses binary cross entropy as the loss function and Early Stopping to prevent overfitting. When the value of the loss function exceeds the preset function threshold, it is determined that the model training is completed and the lighting demand model is obtained. Early Stopping is a technology in deep learning and machine learning to prevent model overfitting. It determines when to stop training by monitoring the performance of the model on the validation set.

[0049] Finally, the controller inputs the input features in the validation set into the lighting demand model, and then obtains the predicted output of the lighting demand model. The predicted output of the lighting demand model is compared with the actual output features in the validation set, and the performance of the lighting demand model is evaluated using some performance indicators such as accuracy, precision, recall, F1 score, mean square error (MSE), etc. According to the performance of the lighting demand model on the validation set, the parameters of the lighting demand model are adjusted, including adjusting the learning rate, changing the model complexity (such as increasing or decreasing the number of layers or nodes of the neural network), modifying the regularization strength, etc. This process may require multiple iterations, each of which is adjusted based on the previous learning results to optimize the lighting demand model.

[0050] S204, inputting the vehicle scene feature data into a preset lighting demand model to obtain vehicle recommended lighting parameters, wherein the input layer of the lighting demand model receives the scene feature matrix, and the output layer generates ideal lighting parameters, wherein the vehicle recommended lighting parameters include light illuminance value, light direction value, and light color temperature value; For details, please refer to step S103, which will not be described in detail here.

[0051] S205. Based on the physical layout structure of the programmable LED array, determine the independent control parameters of each pixel according to the recommended lighting parameters of the vehicle; according to the independent control parameters, group the LED pixels located in the same lighting area in the programmable LED array and determine the priority coefficient of each group of LED pixels; based on the priority coefficient, calculate the collaborative control parameters of each group of LED pixels, the collaborative control parameters are used to achieve synchronous dimming of the LED pixels in the group; according to the independent control parameters and the collaborative control parameters, generate an LED drive timing sequence, the LED drive timing sequence is used to control the on-off time and dimming amplitude of each LED pixel; Among them, the physical layout structure refers to the spatial arrangement of each LED pixel in the programmable LED array, including the array size, pixel pitch and pixel arrangement; independent control parameters refer to specific control instructions for a single LED pixel, including adjustment values ​​in three dimensions: light brightness, light direction and light color temperature; the same lighting area refers to a spatial range with similar lighting requirements; the priority coefficient refers to the importance of different LED pixel groups in lighting control; the collaborative control parameter refers to the control instruction used to achieve unified adjustment of LED pixels within the group; synchronous dimming refers to the synchronization requirement of brightness changes of LED pixels within the group; LED drive timing is used to represent the time series arrangement of control signals; on-off time refers to the turn-on and turn-off times of LED pixels; dimming amplitude refers to the specific value of light brightness adjustment.

[0052] Specifically, first, the controller reads the physical layout configuration file of the programmable LED array and obtains the detailed parameters of the 8×24 pixel matrix, including the precise spatial coordinates (x, y, z), projection angle range (θh, θv) and optical characteristic parameters (luminous efficiency, color temperature range) of each LED pixel. The controller establishes a spatial mapping model to convert the regional lighting requirements in the vehicle's recommended lighting parameters into pixel-level control instructions, and generates a 32-bit control word for each LED pixel, which includes: light brightness (0-4095, 12 bits), horizontal angle (-30°~+30°, 8 bits), vertical angle (-15°~+15°, 6 bits), and light color temperature (3000K-6500K, 6 bits).

[0053] Then, the controller uses a regional clustering algorithm to automatically group the LED pixels based on their spatial position and control parameter similarity. For example, the controller can divide the programmable LED array into three main functional areas: high beam area (2×24), low beam area (4×24), and turn fill area (2×24). In each functional area, the controller uses the DBSCAN clustering algorithm to divide the LED pixels into multiple pixel subgroups based on the similarity of the expected brightness and projection direction of the LED pixels. For each pixel subgroup, its priority coefficient P is calculated: P=w1×position weight+w2×brightness weight+w3×angle weight, where the position weight is related to the importance of the driving field of view, the brightness weight is related to the lighting intensity, and the angle weight is related to the lighting direction.

[0054] Next, the controller calculates the reference dimming parameters based on the priority coefficient of each pixel subgroup: ΔL=L0×P×K, where L0 is the nominal brightness change and K is the scene correction coefficient. The controller generates a unified dimming timing curve, using an S-shaped gradient function to ensure the smoothness of the light brightness change: L(t)=L1+(L2-L1)×(1-cos(πt / T)) / 2, where L1 and L2 are the starting brightness and target brightness, and T is the transition time (usually 50-200ms). For each LED pixel in the pixel subgroup, while maintaining its independent direction and light color temperature, the dimming timing curve is uniformly applied to achieve a synchronous gradient effect.

[0055] Finally, the controller uses a multi-channel PWM controller to generate independent LED drive timing for each LED pixel. The LED timing generation adopts a hierarchical structure: (1) Based on the independent control parameters of the LED pixels, a basic PWM waveform is generated with a frequency of 20kHz and a duty cycle corresponding to the target light brightness; (2) The dimming timing curve of the coordinated control parameters is superimposed to achieve a gradual effect by dynamically adjusting the PWM duty cycle. The controller also implements a timing compensation algorithm, taking into account the LED response characteristics and the delay of the drive circuit, and pre-distorting the PWM signal to ensure that the actual light output effect is consistent with the expected.

[0056] S206, monitoring the temperature change and current change of each LED pixel point; calculating the performance degradation index of each LED pixel point according to the temperature change and the current change; and dynamically adjusting the priority coefficient of each group of LED pixels based on the performance degradation index; Among them, temperature change refers to the heat accumulation during the operation of the LED pixel, usually in degrees Celsius; current change refers to the fluctuation of the working current flowing through the LED, in milliamperes; the performance attenuation index is a quantitative indicator used to represent the degree of performance degradation of the LED pixel, with a value range of 0-1.

[0057] Specifically, the controller collects temperature data at a frequency of 100Hz through the temperature sensor integrated near each LED pixel to determine the temperature change of the LED pixel. At the same time, the controller monitors the LED working current through a high-precision sampling circuit with a sampling accuracy of 0.1mA and a sampling rate of 1kHz to determine the current change of the LED pixel. Based on the temperature change and current change, the controller calculates the performance attenuation index D of each LED pixel: D=1-(w1×VT+w2×VI), where VT is the temperature influence factor, VI is the current influence factor, and w1 and w2 are weight coefficients. When the average performance attenuation index of a certain pixel subgroup is lower than the preset performance attenuation index threshold, the controller automatically reduces the priority coefficient of the pixel subgroup and correspondingly increases the priority coefficients of other normally working pixel subgroups to ensure the stability of the overall lighting effect.

[0058] S207, collecting facial image data of the driver within a preset time period; based on the facial image data, extracting the driver's pupil size change rate, gaze point offset angle and blinking frequency; generating visual fatigue according to the pupil size change rate, the gaze point offset angle and the blinking frequency; when the visual fatigue exceeds a preset visual fatigue threshold, adjusting the light illuminance and light color temperature of the programmable LED array according to a preset visual comfort curve; Among them, facial image data refers to the driver's facial feature information collected by the in-vehicle camera; pupil size change rate refers to the relative change in pupil size per unit time; gaze point offset angle refers to the deviation angle between the driver's line of sight and the front; blinking frequency refers to the number of times the driver blinks per unit time; visual fatigue is used to quantify the driver's visual fatigue level; the preset visual comfort curve refers to the optimal lighting parameter curve under different fatigue levels.

[0059] Specifically, first, the controller collects the driver's facial image data through the infrared camera in the car, with a resolution of 1920×1080 and a frame rate of 30fps. Then, the controller uses a computer vision algorithm to detect facial feature points in real time, focusing on tracking pupil contours and eyelid movements, and calculates the pupil size change rate ΔP, gaze point offset angle θ, and blinking frequency f in the last 5 minutes. Based on these parameters, the controller uses a weighted calculation formula to derive the visual fatigue F: F=α×ΔP+β×θ+γ×f, where α, β, and γ are empirical weight coefficients. When the visual fatigue exceeds the preset visual fatigue threshold, the controller queries the preset visual comfort curve, selects the appropriate light brightness and light color temperature, and adjusts the output characteristics of the programmable LED array to relieve the driver's visual fatigue.

[0060] S208. Detecting the position information and speed information of a target object within a preset distance in front of the vehicle through a forward millimeter-wave radar; acquiring the reflection characteristics and surface material of the target object through a visual sensor; calculating the contrast of the target object under current lighting conditions based on the position information, the speed information, the reflection characteristics, and the surface material; and adaptively adjusting the lighting direction to improve the visibility of the target object when the contrast is lower than a preset contrast threshold.

[0061] Among them, the forward millimeter-wave radar is an on-board sensor used to detect obstacles ahead; position information refers to the spatial coordinates of the target object relative to the vehicle; speed information refers to the moving speed vector of the target object; reflection characteristics are used to indicate the ability of the surface of the target object to reflect light; surface material refers to the material properties of the target object; contrast refers to the degree of difference in brightness between the target object and the background; adaptive adjustment refers to the process of dynamically optimizing the direction of light according to the characteristics of the target object.

[0062] Specifically, first, the controller uses a 77GHz millimeter-wave radar to scan the target objects within 200 meters in front, and obtains the relative distance, azimuth and relative speed of the target objects, with a positioning accuracy better than 0.1 meters. At the same time, the controller uses a high dynamic range camera to capture images of the target objects, and analyzes the material characteristics and reflectivity of the target objects through a deep learning algorithm. Based on the kinematic and optical characteristics of the target object, the controller calculates the contrast C under the current lighting conditions: C=(Lt-Lb) / (Lt+Lb), where Lt is the brightness of the target object and Lb is the brightness of the background. When the contrast is lower than the preset contrast threshold, the controller immediately adjusts the projection angle and light brightness of the relevant LED pixel group to improve the visibility of the target object while avoiding glare to other road users.

[0063] The controller in the embodiment of the present invention is described below from the perspective of hardware processing. Figure 3 , is a schematic diagram of a physical device structure of a controller in an embodiment of the present application.

[0064] It should be noted that Figure 3 The structure of the controller shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0065] like Figure 3As shown, the controller includes a central processing unit (CPU) 301, which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) 302 or the program loaded from the storage part 308 to the random access memory (RAM) 303, such as executing the method described in the above embodiment. In the RAM 303, various programs and data required for system operation are also stored. The CPU 301, the ROM 302 and the RAM 303 are connected to each other through the bus 304. The input / output (I / O) interface 305 is also connected to the bus 304.

[0066] The following components are connected to the I / O interface 305: an input section 306 including an audio input device, a button switch, etc.; an output section 307 including a liquid crystal display (LCD) and an audio output device, an indicator light, etc.; a storage section 308 including a hard disk, etc.; and a communication section 309 including a network interface card such as a LAN (Local Area Network) card, a modem, etc. The communication section 309 performs communication processing via a network such as the Internet. A drive 310 is also connected to the I / O interface 305 as needed. A removable medium 311, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 310 as needed so that a computer program read therefrom is installed into the storage section 308 as needed.

[0067] In particular, according to an embodiment of the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present invention includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a computer program for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through the communication part 309, and / or installed from a removable medium 311. When the computer program is executed by the central processing unit (CPU) 301, various functions defined in the present invention are performed.

[0068] It should be noted that specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in combination with an instruction execution system, apparatus, or device.

[0069] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present invention. Each box in the flowchart or block diagram may represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for implementing the specified logical functions. It should also be noted that in some alternative implementations, the functions marked in the box may also occur in an order different from that marked in the accompanying drawings.

[0070] Specifically, the controller of this embodiment includes a processor and a memory. The memory stores a computer program. When the computer program is executed by the processor, the highly applicable lighting control method provided by the above embodiment is implemented.

[0071] As another aspect, the present invention further provides a computer-readable storage medium, which may be included in the controller described in the above embodiment; or may exist independently without being assembled into the controller. The above storage medium carries one or more computer programs, and when the above one or more computer programs are executed by a processor of the controller, the controller implements the highly applicable lighting control method provided in the above embodiment.

[0072] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application.

[0073] As used in the above embodiments, the term "when..." may be interpreted to mean "if..." or "after..." or "in response to determining..." or "in response to detecting...", depending on the context. Similarly, the phrases "upon determining..." or "if (the stated condition or event) is detected" may be interpreted to mean "if determining..." or "in response to determining..." or "upon detecting (the stated condition or event)" or "in response to detecting (the stated condition or event)", depending on the context.

[0074] Those skilled in the art can understand that to implement all or part of the processes in the above-mentioned embodiments, the processes can be completed by computer programs to instruct related hardware, and the programs can be stored in computer-readable storage media. When the programs are executed, they can include the processes of the above-mentioned method embodiments. The aforementioned storage media include: ROM or random access memory RAM, magnetic disk or optical disk and other media that can store program codes.

Claims

1. A highly applicable lighting control method, characterized in that: Applied to a controller, the method comprises: Collecting traffic density data, road condition data and weather condition data within a preset range around the vehicle, wherein the traffic density data is used to indicate the number and distribution of vehicles and pedestrians, the road condition data is used to indicate the road structure characteristics and road environment status, and the weather condition data is used to indicate the atmospheric environment parameters that affect the driving field of vision; Generating vehicle scene feature data according to the traffic density data, the road condition data and the weather condition data, wherein the vehicle scene feature data is used to represent comprehensive lighting demand characteristics of the vehicle in the current driving scene; Inputting the vehicle scene feature data into a preset lighting demand model to obtain vehicle recommended lighting parameters, wherein the input layer of the lighting demand model receives the scene feature matrix, and the output layer generates ideal lighting parameters, wherein the vehicle recommended lighting parameters include light illuminance value, light direction value, and light color temperature value; The vehicle recommended lighting parameters are sent to a programmable LED array to control each LED pixel in the programmable LED array to adjust the light brightness, light direction and light color temperature according to the vehicle recommended lighting parameters.

2. The method according to claim 1, characterized in that Generating vehicle scene feature data according to the traffic density data, the road condition data and the weather condition data specifically includes: Converting the traffic density data into a traffic feature matrix; Converting the road surface condition data into a road feature matrix; Converting the weather condition data into a weather feature matrix; The traffic feature matrix, the road feature matrix and the weather feature matrix are subjected to feature fusion to obtain the vehicle scene feature data.

3. The method according to claim 1, characterized in that Before the step of inputting the vehicle scene feature data into a preset lighting demand model to obtain vehicle recommended lighting parameters, wherein the input layer of the lighting demand model receives a scene feature matrix, and the output layer generates ideal lighting parameters, wherein the vehicle recommended lighting parameters include a light illuminance value, a light direction value, and a light color temperature value, the method further includes: Acquire ideal lighting parameters corresponding to a driving scene and a scene feature matrix corresponding to the driving scene; Using the scene feature matrix as input data and the ideal lighting parameters as output data, training a preset model; When the accuracy of the preset model exceeds a preset accuracy threshold, the lighting demand model is obtained.

4. The method according to claim 1, characterized in that: The sending of the vehicle recommended lighting parameters to the programmable LED array to control each LED pixel in the programmable LED array to adjust the light brightness, light direction and light color temperature according to the vehicle recommended lighting parameters specifically includes: Based on the physical layout structure of the programmable LED array, determining the independent control parameters of each pixel according to the recommended lighting parameters of the vehicle; According to the independent control parameters, LED pixels located in the same lighting area in the programmable LED array are grouped and a priority coefficient of each group of LED pixels is determined; Based on the priority coefficient, calculating the collaborative control parameters of each group of LED pixels, wherein the collaborative control parameters are used to achieve synchronous dimming of the LED pixels in the group; An LED driving timing sequence is generated according to the independent control parameters and the coordinated control parameters, and the LED driving timing sequence is used to control the on-off time and dimming amplitude of each LED pixel.

5. The method according to claim 4, characterized in that After generating the LED driving timing according to the independent control parameter and the coordinated control parameter, the method further includes: Monitor the temperature and current changes of each LED pixel; Calculating the performance attenuation index of each LED pixel point according to the temperature change and the current change; Based on the performance decay index, the priority coefficient of each group of LED pixels is dynamically adjusted.

6. The method according to claim 1, characterized in that After the step of sending the vehicle recommended lighting parameters to the programmable LED array to control each LED pixel in the programmable LED array to adjust the light brightness, light direction and light color temperature according to the vehicle recommended lighting parameters, the method further includes: Collecting facial image data of the driver within a preset time period; Based on the facial image data, extracting the driver's pupil size change rate, gaze point offset angle, and blinking frequency; generating a visual fatigue degree according to the pupil size change rate, the gaze point offset angle and the blinking frequency; When the visual fatigue exceeds a preset visual fatigue threshold, the light brightness and light color temperature of the programmable LED array are adjusted according to a preset visual comfort curve.

7. The method according to claim 1, characterized in that After the step of sending the vehicle recommended lighting parameters to the programmable LED array to control each LED pixel in the programmable LED array to adjust the light brightness, light direction and light color temperature according to the vehicle recommended lighting parameters, the method further includes: Detecting position information and speed information of a target object within a preset distance in front of the vehicle by a forward millimeter-wave radar; Acquire the reflection characteristics and surface material of the target object through a visual sensor; Calculating the contrast of the target object under current lighting conditions according to the position information, the speed information, the reflection characteristics and the surface material; When the contrast is lower than a preset contrast threshold, the illumination direction is adaptively adjusted to improve the visibility of the target object.

8. A controller, characterized in that: The controller includes: one or more processors and a memory; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code includes computer instructions, and the one or more processors call the computer instructions to enable the controller to execute the method described in any one of claims 1-7.

9. A computer-readable storage medium comprising instructions, characterized in that: When the instructions are executed on a controller, the controller is caused to execute the method according to any one of claims 1 to 7.

10. A computer program product, characterized in that When the computer program product runs on a controller, the controller is caused to execute the method according to any one of claims 1 to 7.

Images