Vehicle light decoding control system and method based on lin / can dual bus situational awareness

Abstract

The application discloses a kind of vehicle light decoding control systems and methods based on Lin / CAN double bus situational awareness.The system includes power supply unit, situational awareness device, bus protocol decoding control unit, main control unit and multi-channel power drive module;Power supply unit accesses vehicle battery voltage to provide power supply for system, situational awareness device obtains face image, traffic condition information, environmental illumination intensity and precipitation data, bus protocol decoding control unit receives and analyzes vehicle Lin / CAN bus signal to obtain vehicle state information, main control unit fuses vehicle state information and situational awareness information to generate light control instruction, and multi-channel power drive module is driven according to instruction Corresponding car light is cooperated.The application realizes the active identification and dynamic regulation and control of vehicle light by combining multi-dimensional situational awareness with double bus signal analysis, solves the problem that existing light system needs manual operation, cannot adapt to complex driving environment and driver state, effectively improves driving safety and driving experience.

Description

Technical Field

[0001] This invention relates to the field of vehicle technology, and in particular to a vehicle lighting decoding control system and method based on Lin / CAN dual-bus context awareness. Background Technology

[0002] With the rapid development of automotive electronics technology, vehicle lights are no longer limited to mere illumination; they have gradually evolved into an important medium for interaction between the vehicle and the driver, and between the vehicle and the external environment. Currently, traditional vehicle lighting control systems typically employ a passive response mode. For example, they receive commands from the vehicle control unit (MCU) via CAN or LIN bus, decode them, and then directly drive the corresponding lighting actuators (such as low beam headlights, high beam headlights, and turn signals) to perform simple on / off operations. Although some high-end models have begun to introduce basic ambient lighting modes such as strobe, these modes mainly rely on preset, fixed program triggers.

[0003] However, existing vehicle lighting control systems lack the ability to perceive the vehicle's internal and external environment. Controlling functions such as high beams, low beams, fog lights, and emergency illumination typically still requires manual operation by the driver based entirely on environmental conditions. They lack proactive recognition and response mechanisms and cannot dynamically provide lighting control solutions tailored to complex driving environments (such as rain, snow, or traffic congestion). If a driver forgets to operate the lights in time due to carelessness, it poses a safety hazard. Secondly, existing systems cannot personalize immersive ambient lighting effects based on driver fatigue or mood, failing to provide emotional comfort or psychological alertness to the driver.

[0004] Therefore, existing vehicle lighting control technology cannot automatically adapt to complex and ever-changing driving environments and driver states, making it difficult to effectively improve driving safety and the driver's driving experience. Summary of the Invention

[0005] This invention provides a vehicle lighting decoding control system and method based on Lin / CAN dual-bus context awareness, which can automatically adapt to complex and ever-changing driving environments and driver driving states, thereby effectively improving driving safety and driver experience.

[0006] One embodiment of the present invention provides a vehicle lighting decoding control system based on Lin / CAN dual-bus context awareness, comprising:

[0007] A power supply unit is used to connect to the vehicle battery voltage and provide power output to the control system.

[0008] A context-aware device is used to acquire context-aware information, which includes facial images, traffic condition information, light intensity and precipitation data of the vehicle's environment.

[0009] The bus protocol decoding control unit is used to receive and parse the vehicle's Lin / CAN bus signals to obtain vehicle status information;

[0010] The main control unit, connected to the bus protocol decoding control unit and the context perception device, is used to generate lighting control commands based on the vehicle status information and the context perception information.

[0011] A multi-channel power drive module, connected to the main control unit, is used to collaboratively drive the corresponding vehicle lights according to the lighting control commands.

[0012] According to the second aspect, another embodiment of the present invention provides a vehicle lighting decoding control method based on Lin / CAN dual-bus context awareness, applied to the vehicle lighting decoding control system based on Lin / CAN dual-bus context awareness as described in any of the above schemes, comprising the following steps:

[0013] The bus protocol decoding unit receives and parses bus signals from the vehicle to obtain context-aware information; the context-aware information includes facial images, vehicle status information, traffic condition information, and light intensity and precipitation data of the vehicle's environment.

[0014] Generate lighting control commands based on the context-aware information;

[0015] The corresponding vehicle lights are driven in coordination with the light control commands through the multi-channel power drive module.

[0016] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0017] This technology directly acquires comprehensive information including facial images, traffic conditions, light intensity, and precipitation data through a context-aware device. Simultaneously, a bus protocol decoding control unit parses Lin / CAN signals to obtain vehicle status information. The main control unit integrates these two types of information to generate lighting control commands, which are then collaboratively driven by a multi-channel power drive module. This technology, employing a parallel information acquisition architecture involving both context-aware devices and bus decoding, allows the main control unit to directly grasp the real-time multi-dimensional status of people, vehicles, and the environment, rather than passively receiving fixed program commands. This endows the system with the ability to proactively identify complex driving scenarios and driver states. Combined with multi-channel collaborative driving, it achieves dynamic, scenario-based lighting output that goes beyond simple on / off operations. Addressing the problem that existing vehicle lighting control technologies cannot automatically adapt to complex and changing driving environments and driver states, this invention actively generates lighting control commands and collaboratively drives corresponding vehicle lights by combining vehicle status information and context-aware information. This enables automatic adaptation to complex and changing driving environments and driver states, effectively improving driving safety and the driver's experience. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the structure of a vehicle lighting decoding control system based on Lin / CAN dual-bus context awareness, provided in an embodiment of the present invention.

[0019] Figure 2 This is a schematic diagram of the system architecture of a vehicle lighting decoding control system based on Lin / CAN dual-bus context awareness, provided in another embodiment of the present invention;

[0020] Figure 3 This is a schematic diagram of the system operation of a vehicle lighting decoding control system based on Lin / CAN dual-bus context awareness, provided in an embodiment of the present invention.

[0021] Figure 4 This is a schematic diagram of the working circuit of a low beam headlight provided in an embodiment of the present invention;

[0022] Figure 5 This is a schematic diagram of the working circuit of a high beam headlight provided in an embodiment of the present invention;

[0023] Figure 6 This is a schematic diagram of the working circuit of a turn signal provided in an embodiment of the present invention;

[0024] Figure 7 This is a schematic diagram of a vehicle light driving circuit according to an embodiment of the present invention;

[0025] Figure 8 This is a schematic flowchart of a vehicle lighting decoding and control method based on Lin / CAN dual-bus context awareness, provided by an embodiment of the present invention. Detailed Implementation

[0026] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0027] See Figures 1 to 3 An embodiment of the invention provides a vehicle lighting decoding control system based on Lin / CAN dual-bus context awareness. The Lin / CAN dual-bus context awareness vehicle lighting decoding control system includes: a power supply unit 8, a context awareness device 9, a bus protocol decoding control unit 11, a main control unit 10, and a multi-channel power drive module 15; the context awareness device 9 includes an environmental perception module 12, a vehicle network communication module 13, and an in-vehicle camera module 14; the power supply unit 8 is used to connect to the vehicle battery voltage and provide power output to the control system; the bus protocol decoding control unit 11, connected to the main control unit 10, is used to parse the vehicle's Lin / CAN bus signals to obtain the vehicle's... Status information; an environmental perception module 12, connected to the main control unit, for collecting ambient light intensity (e.g., detected by the light sensor of the environmental perception module) and precipitation data (e.g., detected by the rain sensor of the environmental perception module); a vehicle-to-everything (V2X) communication module 13, connected to the main control unit, for receiving traffic information from the V2X; an in-vehicle camera module 14, connected to the main control unit, for collecting facial images of occupants; and a multi-channel power drive module 15, connected to the main control unit, for driving multiple headlights. The main control unit is used for:

[0028] Feature extraction is performed on the face image, the vehicle status information, and the traffic condition information to obtain the driver's emotion features, vehicle driving status features, and road condition features respectively.

[0029] The spectral attenuation compensation coefficient of the vehicle headlights is calculated based on the ambient light intensity and the precipitation data.

[0030] The driver's emotional characteristics, the vehicle's driving state characteristics, and the road condition characteristics are fused together to construct driving context features;

[0031] Based on driving context characteristics, and with driving safety and driving emotion matching degree as multiple objectives, the target headlight to be controlled is determined and the control parameters of the target headlight are generated.

[0032] Based on the spectral attenuation compensation coefficient, the control parameters of the target vehicle lamp are adaptively corrected to generate optimized lighting control parameters for the target vehicle lamp.

[0033] Based on the optimized lighting control parameters of the target vehicle light, the multi-channel power drive module is controlled to drive the target vehicle light.

[0034] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0035] This technology directly acquires comprehensive information including facial images, traffic conditions, light intensity, and precipitation data through a context-aware device. Simultaneously, a bus protocol decoding control unit parses Lin / CAN signals to obtain vehicle status information. The main control unit integrates these two types of information to generate lighting control commands, which are then collaboratively driven by a multi-channel power drive module. This technology, employing a parallel information acquisition architecture involving both context-aware devices and bus decoding, allows the main control unit to directly grasp the real-time multi-dimensional status of people, vehicles, and the environment, rather than passively receiving fixed program commands. This endows the system with the ability to proactively identify complex driving scenarios and driver states. Combined with multi-channel collaborative driving, it achieves dynamic, scenario-based lighting output that goes beyond simple on / off operations. Addressing the problem that existing vehicle lighting control technologies cannot automatically adapt to complex and changing driving environments and driver states, this invention actively generates lighting control commands and collaboratively drives corresponding vehicle lights by combining vehicle status information and context-aware information. This enables automatic adaptation to complex and changing driving environments and driver states, effectively improving driving safety and the driver's experience.

[0036] For example, see Figures 4 to 7The multi-channel power drive module involved in this system has its hardware circuit designed with independent drive channels according to the type of vehicle lights (such as low beam, high beam, and turn signals). Its core working principle is to receive digital control signals from the main control unit and convert them into electrical actions that can directly drive the high-current vehicle light load through a power switching circuit. Specifically, the main control unit outputs specific high and low levels or PWM waveforms from its GPIO ports (such as PB13, PB12, and PB14 as shown in the diagram) based on optimized lighting control parameters (such as switching commands and PWM dimming duty cycle). Taking the low beam drive channel as an example, when the low beam needs to be lit, the main control unit controls the corresponding GPIO (such as PB13 or PB14) to output a high level. This high-level signal turns on the connected low-power transistor (such as Q11, Q10, or the 8050 transistor not fully marked in the schematic diagram), thereby controlling the gate potential of the subsequent high-power MOSFET (Q2 marked "30A 60V" in the diagram), making it fully conductive. After the MOSFET is turned on, a low-resistance path is formed between its drain and source, allowing the vehicle battery voltage (VBAT) to drive the low beam headlight assembly (L1+) through this path. For brightness adjustment, the main control unit can output a PWM signal with a specific duty cycle to this path, controlling the average current of the headlight assembly by rapidly switching the MOSFET, thereby achieving stepless dimming. The working principle of the high beam drive channel (involving Q3, Q4, etc.) is similar. Specifically, the OUT_low beam is connected to the low beam headlight board enable terminal and is externally pulled up by default. When PB13 outputs a high level, Q1 is turned on, and the OUT_low beam is pulled down to ground. The OUT_high beam is connected to the high beam headlight board enable terminal and is externally pulled up by default. When PB12 outputs a high level, Q3 is turned on, and the OUT_high beam is pulled down to ground.

[0037] For the turn signals, the drive channel integrates a dedicated flashing control chip (OR 3523 in the figure). The main control unit only needs to output a simple enable signal, and the flashing chip will automatically generate periodic on / off signals that comply with regulations, controlling the operation of the subsequent power transistors (such as Q5 and Q8), thereby causing the turn signals to flash at a fixed frequency. All drive channels are designed with necessary protection components, such as base current limiting resistors (R30, R15, etc. in the figure) and gate pull-down resistors (R12 in the figure), to ensure operational stability and safety. In summary, this multi-channel power drive module, as the system's actuator, accurately, reliably, and in real-time converts the abstract control parameters issued by the main control unit, derived from complex situation perception and intelligent decision-making, into specific switching, dimming, and flashing actions for various vehicle lights.

[0038] It is understandable that the hardware composition, design, and working principle of the relevant modules, units, and other system architectures can be referenced from existing technologies, and will not be elaborated here.

[0039] As an example, the main control unit first processes facial images from the in-vehicle camera module. It calls a pre-trained face detection and emotion recognition model, typically based on a convolutional neural network architecture. The model automatically detects and aligns the input facial image, then extracts depth visual features from key facial regions (such as the eyes and mouth), and finally maps them to a multi-dimensional emotion state space, outputting a structured driver emotion feature vector. This vector can contain quantitative scores for dimensions such as fatigue, attention span, tension, and pleasure. Next, the main control unit processes vehicle status information obtained from the bus protocol decoding control unit. This information is a structured data stream from the vehicle's CAN bus and other buses. The control unit parses parameters directly related to driving dynamics, such as current speed, longitudinal acceleration, lateral acceleration, steering wheel angle, and brake pedal status, and standardizes and formats these parameters, organizing them into an ordered vehicle driving state feature vector or matrix to characterize the vehicle's real-time kinematics and handling state. Simultaneously, the main control unit processes traffic condition information received through the vehicle-to-everything (V2X) communication module. This information describes the congestion level, accident warnings, construction zones, and special weather alerts ahead. The control unit parses the protocol fields and content, transforming this discrete semantic information into numerical features with preset dimensions. For example, it assigns specific values ​​to different congestion levels or encodes the presence of accident signs using binary encoding, thereby constructing a road condition feature vector that summarizes the current external driving environment. Finally, the main control unit performs these three feature extraction tasks in parallel, providing structured driver emotion features, vehicle driving state features, and road condition features for subsequent fusion and decision-making processes.

[0040] As an example of the above embodiments, dynamic weighting coefficients of the driver's emotional characteristics, the vehicle driving state characteristics, and the road condition characteristics are calculated based on the current vehicle speed.

[0041] The driver's emotional features, the vehicle's driving state features, and the road condition features are multiplied by their respective dynamic weight coefficients to obtain a weighted feature vector.

[0042] The weighted feature vectors are concatenated to generate the driving context features.

[0043] In this embodiment, the main control unit first calculates dynamic weight coefficients for driver emotion characteristics, vehicle driving state characteristics, and road condition characteristics based on the current vehicle speed, enabling dynamic adjustment of the contribution of each feature according to the vehicle speed. Next, each feature is multiplied by its corresponding dynamic weight coefficient to obtain a weighted feature vector, thus differentiating the importance of features. Then, the weighted feature vectors are concatenated to generate driving context features, improving the accuracy of the context features in representing the real-time driving state. Finally, subsequent control decisions are made based on these driving context features, ultimately enhancing the adaptability of headlight control to complex driving environments. Therefore, this embodiment achieves accurate construction of driving context features through dynamic weight adjustment and feature fusion, thereby enhancing the targeting and effectiveness of headlight control.

[0044] The main control unit first reads the vehicle's current speed from the vehicle driving status information parsed by the bus protocol decoding control unit, and records it as... Current vehicle speed This serves as the core basis for dynamically adjusting the importance of different features. Subsequently, the main control unit adjusts the settings based on the current vehicle speed. The dynamic weighting coefficients for driver emotional characteristics, vehicle driving state characteristics, and road condition characteristics are calculated separately. An exponential function is used to represent the non-linear relationship between the weights and vehicle speed. Specifically, for driver emotional characteristics, the dynamic weighting coefficients are... The calculation formula is: In this formula, This represents the vehicle speed value obtained in real time from the vehicle bus, and its unit is kilometers per hour. It is a preset decay coefficient greater than zero, which is set empirically to control the rate at which the weight of emotional features decreases as the vehicle speed increases; It is a natural constant. This formula shows that as the vehicle speed... The increase in the weight of emotional characteristics The exponential decay indicates that at high speeds, the system's decisions prioritize driving safety-related characteristics over the driver's subjective emotional state, thus prioritizing safety. For vehicle driving state characteristics, the dynamic weighting coefficient... The calculation formula is: In this formula, The same applies to the current vehicle speed; This is another preset coefficient greater than zero, used to control the saturation rate of vehicle state feature weights as vehicle speed increases. This makes the weights of the vehicle driving state features... With vehicle speed The dynamic weighting coefficients of road condition characteristics increase monotonically and approach 1, ensuring that under high-speed conditions, the vehicle's own operating state (such as speed and acceleration) plays a dominant role in situational judgment. This is achieved by calculating the residual weights to ensure that the sum of the three weight coefficients is 1. Through the calculation using the above formula, the system can determine the vehicle speed based on real-time speed. The contribution of each of the three features in the current context is dynamically and continuously quantified, rather than using fixed weight values. This enables the feature fusion process to adapt to changes in the environment and respond more accurately to driving needs at different speeds.

[0045] After calculating the dynamic weight coefficient , and Next, the main control unit performs feature weighting processing. Specifically, the main control unit modifies the driver emotion feature vector, vehicle driving state feature vector, and road condition feature vector extracted in the previous steps with the calculated corresponding dynamic weight coefficients. , and Perform scalar multiplication. This involves multiplying all dimensional components of each feature vector by the same scalar weight. For example, if a driver's emotion feature is a multi-dimensional vector, its form can be represented as... ,in If we consider the dimension of emotional features, then the weighted emotional feature vector is: Similarly, the vehicle driving state feature vector Weighted Road condition feature vector Weighted Through this multiplication operation, each original feature vector is transformed into a weighted feature vector. The weighted feature vector not only contains the original feature information, but also significantly enhances the magnitude of feature components considered more important in the current vehicle speed context by adjusting the weight coefficients, while weakening the magnitude of relatively minor feature components. This achieves the differentiation and emphasis of feature importance at the data level.

[0046] Finally, the main control unit performs vector concatenation to generate driving context features. The specific process is as follows: the main control unit concatenates all dimensions of these vectors in a predetermined order, such as the weighted driver emotion feature vector, the weighted vehicle driving state feature vector, and the weighted road condition feature vector, to form a longer new vector, thus achieving vector concatenation or connection. Assuming the weighted emotion feature vector... Dimensions Vehicle state feature vector Dimensions Road condition feature vector Dimensions Then the new vector generated after concatenation The total dimension is This new vector This is the final constructed driving situation feature. It integrates information from the driver, the vehicle itself, and the external traffic environment, with each information component weighted according to its importance based on real-time vehicle speed. Because the stitching operation preserves all weighted feature information, this driving situation feature can comprehensively represent the overall driving situation at the current moment.

[0047] As an example, the calculation of the spectral attenuation compensation coefficient of the vehicle headlights based on the ambient light intensity and the precipitation data specifically includes:

[0048] The ambient light intensity is divided into discrete illuminance levels, and the current precipitation level is determined based on the precipitation data;

[0049] Based on the illuminance level and the precipitation level, the basic spectral attenuation factor is obtained by indexing a pre-stored multidimensional lookup table.

[0050] The basic spectral attenuation factor is dynamically corrected based on the current vehicle speed. When the vehicle speed is higher than a first threshold, the basic spectral attenuation factor is decreased; when the vehicle speed is lower than a second threshold, the basic spectral attenuation factor is increased.

[0051] Based on the aforementioned basic spectral attenuation factor, the spectral attenuation compensation coefficient of the vehicle lamp is calculated.

[0052] In this embodiment, ambient light intensity and precipitation data collected by the environmental perception module are first used to classify illuminance and precipitation levels, quantifying the impact of environmental factors on light transmission. Then, a basic spectral attenuation factor is obtained by indexing a pre-stored multidimensional lookup table based on the illuminance and precipitation levels, thus quickly acquiring an environmental attenuation benchmark. Next, the basic spectral attenuation factor is dynamically corrected according to the current vehicle speed; the factor is decreased at high speeds and increased at low speeds, improving the adaptability of the compensation coefficient to different speed conditions. Finally, the spectral attenuation compensation coefficient is calculated based on the corrected basic spectral attenuation factor, ultimately improving the lighting effect of the headlights in rain, snow, or dim environments. Therefore, this embodiment achieves adaptive calculation of the spectral attenuation compensation coefficient through environmental level classification and dynamic vehicle speed correction, thereby optimizing the lighting performance of the headlights under different environments and vehicle speeds.

[0053] Specifically, in this embodiment, the main control unit first discretizes and classifies the raw environmental data collected by the environmental sensing module. The main control unit reads the ambient light intensity values ​​collected in real time by the environmental sensing module, which are typically expressed in lux. To map continuous light intensity to a finite number of discrete states, the main control unit pre-stores an illuminance level classification table. This table defines multiple continuous light intensity intervals, each interval corresponding to a unique illuminance level. subscript This indicates the level sequence number. For example, the system can divide light intensity into multiple levels such as "extremely dark," "dark," "normal," "bright," and "extremely bright." The main control unit compares the currently collected light intensity value with the intervals defined in the table to determine its corresponding interval, thereby obtaining the current discrete illuminance level. Simultaneously, the main control unit reads precipitation data collected by the environmental sensing module, which may include information such as precipitation amount and type (e.g., rain, snow). Similarly, the system has pre-stored precipitation level classification rules, determining the current precipitation level based on precipitation intensity and type. subscript This indicates the precipitation level number. For example, the levels can be divided into "no precipitation", "light rain", "moderate rain", "heavy rain", "snow", etc.

[0054] Next, the main control unit, based on the aforementioned determined illuminance level, and precipitation level The basic spectral attenuation factor is indexed from a pre-stored multidimensional lookup table. This multidimensional lookup table is a data structure pre-calibrated experimentally, with its row indices corresponding to all possible illuminance levels. The column index corresponds to all possible precipitation levels. Each cell stores information in a specific context. The typical attenuation coefficient of light propagating in the atmosphere under various environmental conditions, i.e., the fundamental spectral attenuation factor. The main control unit uses the currently determined... and As a composite index key, it allows for quick retrieval of the corresponding data when performing a matching query in the lookup table. Value. This factor This represents the proportional reduction in spectral intensity of light emitted by vehicle headlights due to physical effects such as scattering and absorption before reaching a target (e.g., the road surface or a vehicle ahead) under current lighting and precipitation conditions. Using a lookup table instead of a complex real-time physical model reduces the system's computational complexity and response time.

[0055] Then, the main control unit inputs the current vehicle speed. For the basic spectral attenuation factor Dynamic adjustments are made to obtain a decay factor that better reflects actual driving dynamics. The main control unit obtains the current vehicle speed from the vehicle's driving status information. The system has a preset first vehicle speed threshold. Second vehicle speed threshold And usually The main control unit executes the following correction logic: when the current vehicle speed is detected... Above the first threshold This indicates that the vehicle is traveling at high speed. At this time, to ensure clear visibility at greater distances and faster visual response, the system needs to relatively reduce the compensation for spectral attenuation, i.e., the compensation for the basic spectral attenuation factor. A reduction correction is applied. The specific correction formula is as follows: ,in It is the corrected attenuation factor. It is a preset speed correction coefficient less than 1. Conversely, when the current speed is detected... Below the second threshold This indicates that the vehicle is at low speed or idling. At this time, the need for long-range lighting decreases, but there may be greater focus on the quality and ambiance of short-range lighting. The system allows for a slight increase in attenuation compensation, i.e., for... To increase the correction, the formula is: ,in It is a preset speed correction factor greater than 1. If the vehicle speed... Between and In between, no correction is made, that is... By introducing vehicle speed as a dynamic variable for correction, the spectral attenuation factor not only reflects the influence of the static environment but also couples the changes in lighting requirements due to the vehicle's dynamic driving state, thus improving the scene adaptability of the compensation coefficient.

[0056] Finally, the main control unit based on the dynamically corrected attenuation factor Calculate the final headlight spectral attenuation compensation coefficient. Spectral attenuation compensation coefficient This is used to directly adjust the control parameters of the vehicle lights (such as brightness and color temperature) proportionally in subsequent stages to compensate for light attenuation caused by the environment. Its calculation formula is defined as follows: The physical meaning of this formula is that if the attenuation factor... The percentage of light intensity retained (e.g., 0.8 represents 80% intensity retention) is then its reciprocal. This refers to the factor by which the output light intensity needs to be amplified to achieve the original desired luminous effect (e.g., 1.25 times). After the main control unit completes this calculation, it obtains the final spectral attenuation compensation coefficient. This coefficient will be passed to the subsequent control parameter adaptive correction processing module for use.

[0057] As an example of the above embodiment, the step of determining the target headlight to be controlled and generating control parameters for the target headlight based on driving context features and with driving safety and driving emotion matching degree as multiple objectives specifically includes:

[0058] The driving scenario features are input into a preset vehicle light control recognition model to determine the target vehicle light to be controlled;

[0059] From the pre-stored sample library of historical driving scenarios and lighting control parameter pairs, retrieve multiple historical scenario samples that have an Euclidean distance of less than a preset threshold with respect to the driving scenario features and match the type of the target vehicle light.

[0060] The lighting control parameters corresponding to the multiple historical scenario samples are obtained from the sample library as reference points, and perturbations that meet the preset vehicle light safety constraints are applied to each reference point to generate multiple candidate vehicle light control parameters for the target vehicle light.

[0061] The vehicle driving state feature matrix and the road condition features are input into the driving risk assessment model to calculate the driving risk probability corresponding to each candidate headlight control parameter, and the driving risk probability is converted into a driving safety reward value through a safety probability conversion function.

[0062] The driver's emotional characteristics and each of the candidate headlight control parameters are vector-projected in a preset joint feature space, and the cosine similarity between the projected vectors is calculated as the emotion matching reward value.

[0063] Based on preset safety weight coefficients and emotion weight coefficients, the driving safety reward value and the emotion matching reward value are weighted and summed to obtain a comprehensive reward value, and the candidate headlight control parameter with the largest comprehensive reward value is selected as the control parameter of the target headlight.

[0064] In this embodiment, the main control unit first inputs driving scenario features into a preset headlight control recognition model to determine the target headlight, automatically identifying the type of headlight that needs to be controlled. Next, it retrieves similar scenario samples from a historical sample database and obtains corresponding headlight control parameters as reference points. A perturbation conforming to safety constraints is applied to the reference points to generate multiple candidate headlight control parameters, thus achieving parameter diversification. Then, a driving risk assessment model calculates the driving risk probability of each candidate parameter and converts it into a safety reward value. Simultaneously, it calculates the emotional matching reward value between emotional features and candidate parameters, improving the safety and personalization of the control parameters. Finally, a comprehensive reward value is obtained by weighted summation of safety weight coefficients and emotional weight coefficients, and the maximum value is selected as the control parameter, ultimately improving the driving safety and driving emotion matching degree of headlight control. Therefore, this embodiment achieves intelligent optimization of headlight control parameters through multi-objective optimization and reward value evaluation, thereby balancing safety requirements and emotional comfort effects.

[0065] Specifically, the workflow of this embodiment is as follows:

[0066] The main control unit first inputs the constructed driving scenario feature vector into a pre-defined headlight control recognition model. This model is a trained classification or sequence model, such as a multi-classifier based on a deep neural network. The model's input is the feature vector representing the overall driving scenario, and the output is the identifier of the headlight type that needs to be prioritized for control or adjustment, such as "low beam," "high beam," "turn signal," "position light," or a specific "ambient lighting group." By learning the mapping relationship between a large amount of historical driving data and headlight operation records, the model can determine the headlight unit that most needs intervention based on the current overall scenario; this unit is the target headlight.

[0067] After determining the target headlight type, the main control unit retrieves similar scenarios from a pre-stored database of historical driving scenario-headlight control parameter pairs. This database stores a large number of previously recorded data pairs, each containing a historical driving scenario feature vector and the headlight control parameters actually used or evaluated as optimal in that scenario. The main control unit calculates the Euclidean distance between the current driving scenario feature vector and the feature vector of each historical scenario in the database. The Euclidean distance is calculated by taking the square root of the sum of the squares of the differences in each dimension of the two vectors, used to quantify the overall similarity between the two scenarios. The system sets a preset similarity threshold, retrieving only historical scenario samples whose Euclidean distance is less than this threshold. Simultaneously, a filtering process is performed, retaining only data from historical samples whose headlight type matches the currently determined target headlight type. Ultimately, the system obtains multiple historical scenario samples that are highly similar to the current scenario and target the same type of headlight.

[0068] Next, the main control unit extracts lighting control parameters corresponding to the multiple historical scenario samples retrieved above from the sample library, using these parameters as "baseline points" for generating new parameters. Each baseline point represents a control scheme validated in similar scenarios. To explore potentially better solutions around the baseline points and avoid getting trapped in local optima, the system applies random perturbations to each baseline point that conform to preset vehicle headlight safety constraints. The safety constraints limit the range of the perturbations; for example, for headlight brightness parameters, the perturbed value must be between the maximum brightness value allowed by regulations and the minimum brightness value to ensure visibility; for color temperature parameters, the perturbation must be within the range of human visual comfort. By applying multiple independent perturbations to each baseline point, the system generates a diverse set of candidate headlight control parameters for the target headlight. This expands discrete historical experience points into a candidate parameter space, providing a rich selection for subsequent multi-objective evaluation.

[0069] Then, the main control unit initiates a parallel dual-objective evaluation process. The first evaluation objective is driving safety. The main control unit inputs the current vehicle driving state characteristics (such as a vector or matrix composed of vehicle speed, acceleration, yaw rate, etc.) and road condition characteristics (such as traffic flow, road type, distance to obstacles ahead, etc.) into a pre-trained driving risk assessment model. This model can be a regression model based on gradient boosting decision trees or neural networks, which can evaluate the probability of driving risks (such as oncoming vehicle accidents caused by glare, and self-collision risk caused by insufficient illumination) that may occur when using certain headlight control parameters under given vehicle state and road conditions. The system performs a risk assessment for each candidate headlight control parameter, obtaining a driving risk probability. This probability is then converted into a driving safety reward value using a safety probability conversion function. This transformation function is typically designed as follows: ,in It is a positive scaling constant used to adjust the dimensions and range of the reward value. This function inverts the risk probability into a safety reward; the lower the probability, the higher the safety reward value, reflecting the safety performance of the candidate parameter.

[0070] The second evaluation objective is the matching degree of driving emotions. The extracted driver emotional features (such as vectors representing fatigue, tension, and pleasure) are fused with each candidate headlight control parameter (such as vectors representing brightness, color temperature, and flashing frequency). The system pre-defines a joint feature space that can simultaneously represent emotional states and headlight parameters. The main control unit projects the driver emotional feature vector and the candidate headlight control parameter vector into this joint feature space, obtaining two projected vectors. The cosine similarity between these two projected vectors is calculated, with a value between -1 and 1. Cosine similarity measures the directional closeness between two vectors; the closer the directions, the better the headlight parameter setting matches the driver's current emotional state psychologically and visually. For example, soothing headlights match fatigue, while exhilarating headlights match pleasure. The cosine similarity value serves as the emotion matching reward value. .

[0071] After obtaining the security reward value for each candidate parameter. Reward value for matching emotions Then, the main control unit executes multi-objective fusion decision-making. The system has preset security weight coefficients. And sentiment weight coefficient The sum of these two coefficients is usually 1, and their specific ratio can be configured according to vehicle model positioning or driver preference. For each candidate headlight control parameter, the main control unit calculates its comprehensive reward value. The calculation formula is: This formula unifies the evaluations of the two dimensions into a single scalar using linear weighting. After calculating the comprehensive reward value for all candidate parameters, the main control unit compares them and selects the comprehensive reward value. The largest candidate headlight control parameter is determined as the final control parameter used to control the target headlight.

[0072] In summary, this embodiment uses a vehicle headlight control recognition model to lock onto the control target, retrieves empirical benchmarks through similar historical context retrieval, generates candidate solutions through safety constraint perturbations, quantifies the safety and personalized utility of each solution through a risk assessment model and emotion matching degree calculation, and finally performs multi-objective optimization decision-making through weighted summation of preset weights. This achieves personalized headlight control under the fundamental premise of ensuring driving safety, enabling the generated headlight control parameters to dynamically adapt to complex vehicle states and external road conditions, while also conforming to the driver's internal emotional state, thereby improving safety while optimizing the driving experience.

[0073] As an example of the above embodiment, the step of adaptively correcting the control parameters of the target vehicle headlight based on the spectral attenuation compensation coefficient to generate optimized headlight control parameters for the target vehicle headlight specifically includes:

[0074] The theoretical compensation gain is calculated based on the spectral attenuation compensation coefficient, and the safety redundancy coefficient is determined based on the vehicle's current speed.

[0075] Based on the theoretical compensation gain and the safety redundancy coefficient, the actual compensation gain is calculated.

[0076] Based on the actual compensation gain, the control parameters of the target vehicle headlight are corrected to generate optimized headlight control parameters for the target vehicle headlight.

[0077] In this embodiment, the main control unit first calculates the theoretical compensation gain based on the spectral attenuation compensation coefficient, which quantifies the amount of light compensation required for environmental attenuation. Next, a safety redundancy coefficient is determined based on the current vehicle speed, considering the impact of speed changes on lighting safety. Then, the actual compensation gain is calculated based on the theoretical compensation gain and the safety redundancy coefficient, improving the reliability and safety of the compensation. Finally, the control parameters of the target headlight are corrected based on the actual compensation gain to generate optimized lighting control parameters, ultimately improving the lighting stability of the headlight under environmental attenuation and vehicle speed fluctuations. Therefore, this embodiment achieves adaptive correction of control parameters through gain calculation and redundancy adjustment, thereby ensuring the accuracy and safety of the headlight lighting effect.

[0078] Specifically, the main control unit first uses the calculated spectral attenuation compensation coefficient... To calculate the theoretical compensation gain Spectral attenuation compensation coefficient This reflects the theoretical amplification factor required to compensate for spectral attenuation caused by environmental factors. Theoretical compensation gain. It can be directly set to be equal to this coefficient, that is This converts the compensation coefficient into a gain that can be directly used for parameter adjustment. Simultaneously, the main control unit adjusts the gain based on the vehicle's current speed. Determine a safety redundancy factor The purpose of introducing a safety redundancy coefficient is to dynamically adjust the aggressiveness of compensation based on vehicle speed, ensuring that at any vehicle speed, the enhanced headlights, after compensation, will not create new safety hazards due to excessive brightness (such as glare to drivers ahead or oncoming drivers). The main control unit has a pre-stored mapping relationship between vehicle speed and the safety redundancy coefficient, which was obtained through experimental calibration. Specifically, the safety redundancy coefficient... The calculation is achieved through a piecewise function: when the vehicle speed... Exceeding a set high speed threshold When the speed is high, it indicates that the vehicle is traveling at high speed. To avoid the risks posed by strong light under high-speed relative motion, a more conservative compensation strategy is needed; therefore, the safety redundancy coefficient is [not specified]. Take a value less than 1, for example When the vehicle speed Below a set low speed threshold This indicates that the vehicle is at low speed or stationary, and the environment is relatively static, allowing for more adequate compensation to optimize lighting or ambient effects; therefore, the safety redundancy factor is higher. It can take the value 1 or a value slightly greater than 1, for example When the vehicle speed In and Safety redundancy factor during the period Available and The calculation is performed using linear interpolation, i.e. . In the formula, To obtain the current vehicle speed in real time, and The preset vehicle speed threshold and , and These are constant coefficients determined based on safety calibration. This yields a safety redundancy coefficient negatively correlated with vehicle speed, providing a safety adjustment dimension for the final determination of subsequent gains.

[0079] Next, the main control unit compensates for the gain based on the calculated theoretical gain. With safety redundancy coefficient The actual compensation gain used for parameter correction is calculated. Actual compensation gain It is the result of the combined effect of theoretical requirements and safety constraints, and its calculation formula is: This formula means that the gain ultimately applied to the control parameters is the theoretically required gain multiplied by a safety discount (or safety bonus) factor determined based on the real-time vehicle speed. At that time, actual gain Less than theoretical gain This achieves the suppression of compensation amount to ensure high-speed safety; when At that time, actual gain Equal to or slightly greater than the theoretical gain This allows for sufficient or even slightly aggressive compensation at safe vehicle speeds. This organically combines environmental compensation requirements with dynamic driving safety requirements, resulting in the calculation of a safe and controllable actual operational gain value.

[0080] Finally, the main control unit calculates the actual compensation gain. The control parameters of the target headlight are modified to generate optimized headlight control parameters. The control parameters of the target headlight are a vector containing one or more lighting attributes; for example, for the main headlight, the control parameters are... May contain brightness values and color temperature value etc., that is The correction process involves performing scalar multiplication on these parameters that are significantly affected by light attenuation. Specifically, for the brightness parameter... Its corrected value Calculated as Regarding color temperature parameters In some implementations, spectral attenuation may affect different wavelengths of light differently, thus influencing the perceived color temperature. Therefore, adaptive adjustments can be made based on a calibration model, for example... ,in It is based on the actual compensation gain The color temperature fine-tuning amount is obtained through table lookup or calculation. The main control unit adjusts the control parameter vector according to the above rules. Each corresponding component in the vector is corrected to generate a new parameter vector. .this This refers to the optimized lighting control parameters for the final target vehicle headlights. It incorporates the basic control intent derived from multi-objective optimization, and integrates spectral attenuation compensation for current ambient light and precipitation conditions. Furthermore, this compensation has been adjusted for safety redundancy based on real-time vehicle speed, thereby ensuring the optimal balance of lighting output while meeting multiple requirements such as environmental adaptability, personalization, and safety.

[0081] In summary, this embodiment converts the spectral attenuation compensation coefficient into a theoretical gain and combines it with a safety redundancy coefficient calculated based on vehicle speed dynamics to adjust the theoretical gain under safety constraints, thereby obtaining the actual compensation gain. Finally, this gain is used to perform scalar correction on the original control parameters, making the lighting control not only intelligent but also safe and reliable, ensuring that the vehicle can provide effective and safe lighting under any environment and speed.

[0082] As an example of the above embodiment, controlling the multi-channel power drive module to drive the target vehicle light based on the optimized lighting control parameters of the target vehicle light specifically includes:

[0083] Based on the type identifier of the target vehicle light, determine the target drive channel in the multi-channel power drive module that corresponds to the target vehicle light;

[0084] The optimized lighting control parameters are converted into driving signals and output to the multi-channel power drive module, so as to drive the target vehicle light through the target driving channel of the multi-channel power drive module corresponding to the target vehicle light.

[0085] In this embodiment, the main control unit first determines the corresponding target drive channel in the multi-channel power drive module based on the target vehicle light type identifier, enabling precise positioning of the drive path. Next, the optimized lighting control parameters are converted into drive signals, achieving the conversion from digital control commands to electrical signals. Then, the drive signal is output to the multi-channel power drive module, improving drive efficiency and response accuracy. Finally, the target vehicle light is driven through the target drive channel, ultimately enhancing the accuracy and real-time performance of the vehicle light control. Therefore, this embodiment achieves efficient control of the vehicle light drive through channel identification and signal conversion, ensuring that the lighting effect is accurately achieved as needed. The specific workflow of this embodiment is as follows:

[0086] The main control unit first determines the target drive channel in the multi-channel power drive module corresponding to the target headlight based on the headlight type identifier. The headlight type identifier, determined by the headlight control recognition model in the aforementioned multi-target optimization step, is stored in the main control unit's memory as an enumeration or string, such as "low beam_left", "high beam", "turn signal_right", "ambient light_dashboard", etc. The multi-channel power drive module is a hardware circuit unit containing multiple independent drive channels. Each drive channel is physically and electrically connected to one or a group of specific headlight actuators (such as LED light groups, motors, etc.) and has a unique channel number or logical address. The main control unit pre-stores a "headlight type-drive channel mapping table," which establishes a fixed correspondence between each headlight type identifier and a specific drive channel number in the multi-channel power drive module. Once the main control unit obtains the target headlight type identifier, it queries this mapping table and retrieves the unique drive channel number bound to it by precisely matching the type identifier. For example, when the target headlight is "low beam_left", the mapping table may indicate that its corresponding target drive channel is "channel 3".

[0087] Next, the main control unit converts the optimized lighting control parameters into specific drive signals. The optimized lighting control parameters are a data structure or vector containing one or more lighting attributes; for example, for a conventional light fixture, the parameters... It can be represented as For dynamic effect lights, flashing frequency may also be included. Gradient mode These parameters are digital quantities and cannot directly drive power devices. Therefore, they need to be converted into drive signals that a multi-channel power drive module can recognize and execute. The conversion process is completed by the signal generation submodule within the main control unit. This submodule uses an appropriate conversion algorithm based on the type of parameter and the characteristics of the driven headlight. For brightness control, pulse width modulation (PWM) technology is typically used. The signal generation submodule determines the conversion algorithm based on the brightness value. (Usually normalized to a value between 0 and 1), calculate a corresponding PWM duty cycle. The relationship is generally linear or gamma-corrected. A PWM digital waveform sequence representing this duty cycle is then generated. For color temperature control, if the headlights are achieved by mixing multi-color temperature LEDs, the color temperature value... This will be converted into independent brightness ratios for two or more LED groups, thereby generating multiple PWM signals. For other complex parameters, such as angle adjustment commands (if optimized parameters have been incorporated), they may be converted into pulse sequences and direction signals to drive the stepper motor. All these underlying electronic control signals required to drive the target vehicle light are collectively packaged or synchronously prepared to form a complete drive signal package for the target drive channel.

[0088] Then, the main control unit outputs this drive signal to the multi-channel power drive module. The main control unit connects to the multi-channel power drive module through a specific communication interface (such as CAN FD, Ethernet, or a dedicated parallel bus). The main control unit takes the encapsulated drive signal packet, along with the target drive channel number determined in the first step, and assembles it into one or more data frames according to a predetermined communication protocol, and sends them to the multi-channel power drive module. The multi-channel power drive module receives and parses these messages, extracting the target drive channel number and the corresponding drive signal content.

[0089] Finally, the target headlight is driven by the target drive channel corresponding to the target headlight in the multi-channel power drive module. The microcontroller or logic circuit inside the multi-channel power drive module activates the physical drive circuit corresponding to the received target drive channel number. Simultaneously, the received drive signal content (such as PWM duty cycle data, pulse sequence, etc.) is applied to the activated drive circuit. The drive circuit typically includes power switching devices (such as MOSFETs) that precisely control the current flowing through the target headlight (such as an LED chip), the on / off timing, or the motor movement based on these signals, thereby achieving precise changes in headlight brightness, color temperature, angle, or dynamic effects. For example, when a high duty cycle PWM signal is applied to the low beam headlight drive channel, the low beam LED will emit light at high brightness; when a specific pulse sequence is applied to the low beam headlight channel with an adjustment motor, the low beam illumination angle will change accordingly.

[0090] As an example of the above embodiments, the main control unit is further configured to:

[0091] The vehicle network communication module receives emergency broadcast information about accidents and extracts the current location of the accident.

[0092] Obtain the vehicle's current driving trajectory;

[0093] Based on the vehicle's driving trajectory and the current location of the accident, predict the vehicle's arrival time at the accident scene;

[0094] If the arrival time is less than a preset safety threshold, then light enhancement parameters and vehicle warning light control parameters are generated.

[0095] Based on the control parameters of the warning light, the vehicle's warning light is controlled to emit warning light, and the light enhancement parameters are integrated into the optimized light control parameters to control the multi-channel power drive module to drive the target vehicle light to enhance the light and achieve visual warning.

[0096] In this embodiment, the main control unit first analyzes the emergency broadcast information received by the vehicle network communication module to extract the current location of the accident, enabling real-time acquisition of road hazard information. Then, based on the vehicle's trajectory and the accident location, the arrival time is predicted to determine the potential risk level. If the arrival time is less than a preset safety threshold, headlight enhancement parameters and warning light control parameters are generated to improve the proactive warning capability. Finally, the warning lights are controlled to emit warning light, and the headlight enhancement parameters are integrated into the optimized control parameters to drive the target vehicle's headlights to enhance their illumination, ultimately improving the visual warning effect when the vehicle approaches the accident scene. Therefore, this embodiment achieves proactive safety warnings through accident information processing and headlight enhancement control, thereby reducing the risk of secondary accidents.

[0097] The main control unit first continuously monitors and receives emergency accident broadcasts from the vehicle-to-everything (V2X) cloud platform or roadside units via the vehicle-to-everything (V2X) communication module. This information typically uses a standardized data format (such as BSM, SPAT / TPEG), which encapsulates detailed accident attributes. The main control unit then calls its internal information parsing subroutine to decode the received data packets and extract the crucial "current location of the accident" field. This location information is usually represented in latitude and longitude coordinates, such as a value containing latitude values. and longitude value coordinate pairs The parsing process includes verifying data integrity, parsing the protocol layer, and finally converting the geographic location information into a unified coordinate format used internally by the system for subsequent calculations.

[0098] Next, the main control unit acquires the vehicle's current driving trajectory. This is achieved by integrating real-time data from the vehicle bus (such as CAN), including vehicle driving status information continuously obtained by the main control unit from the bus protocol decoding control unit, such as the current vehicle speed. Heading angle Information such as yaw rate, etc. The main control unit uses this information, combined with timestamps, and fuses it with positioning information provided by dead reckoning or high-precision positioning modules (such as GPS / IMU), to estimate the vehicle's predicted driving trajectory over a short period in the future. This trajectory can be represented as a sequence of consecutive future locations, for example... ,in The current vehicle position coordinates Subsequent points It is the position at a future moment obtained by extrapolating from the current state of motion.

[0099] Then, the main control unit uses the acquired vehicle driving trajectory as a basis. The current location of the accident obtained from the analysis Predict the arrival time of the vehicle at the accident scene. This embodiment employs a prediction algorithm that combines path distance and dynamic vehicle speed. First, it calculates the distance from the vehicle's current position... to the accident location Planar straight-line distance Alternatively, if a detailed map path is available, calculate the distance along the current road network. Assuming a fast estimate is made using straight-line distance, the distance... Calculated using formulas for spherical or planar distances. Then, the arrival time is predicted. Through formula Calculation. Among them, This is the estimated average speed. An improved estimation method is used here: It's not simply the current instantaneous speed. Instead, it considers the possibility that the vehicle may slow down due to approaching an accident, and the calculation formula is: . In the formula, It represents the current real-time vehicle speed, and λ is a preset attenuation coefficient that is greater than zero. It is the calculated distance. It is a natural constant. The physical meaning of this formula is that as the distance between the vehicle and the accident point increases... The decrease in the estimated average speed The decay will be exponential, which better aligns with the expected behavior of drivers slowing down before seeing an accident scene. The formula used to calculate... It is more conservative and safer than simply using the current vehicle speed for calculation.

[0100] The main control unit will calculate the predicted arrival time With a preset safety time threshold Compare this to the security threshold. This represents the minimum time margin at which the system deems proactive warnings necessary. If... If the system determines that the vehicle is about to enter an accident risk area, it will trigger a warning response. At this time, the main control unit generates two types of control parameters: one type is the light enhancement parameters for the regular target vehicle lights. Another type is the control parameters specifically for vehicle warning lights. Light enhancement parameters This could be a scalar gain coefficient (e.g., 1.5), indicating that the brightness needs to be further increased or the flashing mode changed to enhance visual salience, based on the optimized lighting control parameters. Warning light control parameters This defines the specific flashing frequency, synchronization mode, etc. of hazard warning lights (such as hazard lights).

[0101] Finally, the main control unit executes dual-path lighting control. In the first path, the main control unit immediately applies the generated warning light control parameters. Through the drive channel corresponding to the hazard warning lights in the multi-channel power drive module, the vehicle's warning lights are driven to emit high-frequency or specific warning light patterns to immediately alert the driver and surrounding vehicles to potential risks. In the second path, the main control unit generates light enhancement parameters... This is integrated into the ongoing main lighting control process. Specifically, it involves generating optimized lighting control parameters for the target headlights. In the final stage, the system will It is combined with it as an additional correction factor. For example, if Including brightness The final output brightness is then corrected to Subsequently, the main control unit integrates the final parameters of the enhancement intent and controls the multi-channel power drive module to drive the target vehicle lights (such as headlights and fog lights) to enhance the light, for example, by illuminating the front with higher brightness or a specific scanning pattern, thereby illuminating the accident scene area at a greater distance or forming a more conspicuous visual marker, thus playing a role in early visual warning.

[0102] In summary, this embodiment achieves a leap from passively receiving information to proactively providing preventative lighting warnings by real-time analysis of vehicle-to-everything (V2X) accident broadcasts, dynamic prediction of the spatiotemporal relationship between the vehicle and the accident site, and proactively triggering a dual response of warning lights and main headlight enhancement when a risk approaches. This process deeply integrates V2X information into the lighting control decision loop, significantly improving the vehicle's proactive safety when approaching sudden dangerous scenarios such as accidents, and helping to prevent secondary accidents.

[0103] As an example of the above embodiment, the system further includes an external vehicle camera module for acquiring road surface images; the main control unit is connected to the external vehicle camera module, and the main control unit is further configured to:

[0104] When the target vehicle headlights are low beam, the slippery areas of the road surface are identified through the road image.

[0105] The road surface reflectivity index is calculated based on the area ratio of the slippery road surface and the precipitation data.

[0106] When the road surface reflectivity index exceeds the preset reflectivity threshold, an angle adjustment command for the vehicle's low beam headlights corresponding to the road surface reflectivity index is generated.

[0107] The angle adjustment command is incorporated into the optimized lighting control parameters.

[0108] In this embodiment, an external camera module first acquires road surface images and identifies slippery areas, detecting road surface conditions. Then, based on the proportion of slippery areas and precipitation data, a road surface reflectivity index is calculated to quantify the degree of road reflectivity. When the reflectivity index exceeds a preset reflectivity threshold, a corresponding low-beam headlight angle adjustment command is generated, reducing glare from the road surface from interfering with the driver's vision. Finally, the angle adjustment command is integrated into optimized lighting control parameters, ultimately improving the lighting safety and driving comfort of the low beam headlights on slippery roads. Therefore, this embodiment achieves adaptive adjustment of the low beam headlights through road reflectivity recognition and angle adjustment, thereby improving driving visibility at night or in adverse weather conditions.

[0109] For example, when the road surface reflectivity index exceeds a preset reflectivity threshold, generating an angle adjustment command for the vehicle's low beam headlights corresponding to the road surface reflectivity index includes:

[0110] When the road surface reflectivity index exceeds the preset reflectivity threshold, the basic angle adjustment amount is obtained by querying the pre-stored mapping table between reflectivity and angle adjustment amount based on the road surface reflectivity index.

[0111] Based on the difference between the basic angle adjustment amount and the current illumination angle of the low beam, the target illumination angle is calculated, and an angle adjustment command is generated to adjust the low beam to the target illumination angle.

[0112] This embodiment first uses the main control unit to retrieve the baseline angle adjustment amount by querying a pre-stored mapping table of reflectivity and angle adjustment amount when the road surface reflectivity index exceeds a preset reflectivity threshold. This allows for rapid determination of the adjustment benchmark. Next, the target illumination angle is calculated based on the difference between the baseline angle adjustment amount and the current low beam illumination angle, achieving precise angle positioning. Then, an angle adjustment command is generated to adjust the low beam to the target illumination angle, improving adjustment accuracy. Finally, the command is integrated into the control parameters, ultimately improving the efficiency of low beam angle adjustment and optimizing the lighting range. Therefore, this embodiment achieves precise control of the low beam angle through mapping lookup and difference calculation, thereby optimizing lighting distribution and reducing glare. For easier understanding of the above embodiment, the following detailed explanation is provided:

[0113] First, when the target headlight to be controlled is determined to be the low beam headlight through the aforementioned multi-objective optimization process, the main control unit activates the external camera module and acquires its real-time road surface image. The external camera module is typically deployed at the front of the vehicle, such as inside the windshield or at the exterior rearview mirror, and its field of view covers the road area in front of the vehicle. The main control unit receives this road surface image data and calls its integrated image processing and computer vision algorithms to identify slippery areas in the image. This identification process is mainly achieved by analyzing the visual characteristics of the road surface area. For example, the system can use a threshold segmentation method based on color space (such as HSV), because wet or waterlogged road surfaces often exhibit different color saturation and brightness characteristics than dry road surfaces, and have higher specular reflection properties. The main control unit performs pixel-level analysis of the image, identifying continuous pixel areas that meet the visual characteristics of slipperiness (such as low saturation, high brightness, and blurred texture) as potential slippery areas. To further improve accuracy, edge detection algorithms can also be used to observe the deformation or reflection of road markings in suspected slippery areas. Finally, the system outputs a binary image or a list of region contours, clearly identifying all slippery road surfaces in the current image frame.

[0114] Next, the main control unit calculates the area proportion of the identified slippery road surface and, in conjunction with the precipitation data provided by the environmental perception module, calculates a road surface reflectivity index. The calculation process consists of two parts. The first part calculates the area ratio of the slippery region. The main control unit counts the total number of pixels belonging to the slippery area in the binarized image and divides it by the total number of pixels in the image that are predefined as representing the effective road surface area (e.g., obtained through ROI delineation or semantic segmentation) to obtain the result. This value, ranging from 0 to 1, quantifies the extent of the slippery road surface. The second part involves the main control unit reading real-time precipitation data provided by the environmental perception module. This data may include precipitation type (rain, snow) and precipitation intensity level (e.g., none, light, moderate, heavy). The system pre-assigns different reflection influence coefficients for different precipitation types and intensities. For example, the coefficient for heavy rain is higher than that for light rain, and the coefficient for snow may differ from that for rain. Ultimately, this relates to the road surface reflectivity index. It is calculated using a weighted or product formula, for example .in, and These are the preset weights for area proportion and precipitation impact coefficient, used to balance their contributions to overall reflectivity. This index... It is a dimensionless value. The larger the value, the higher the risk of specular reflection and glare of low beam headlights due to water and smooth surfaces under the current road conditions.

[0115] Then, the main control unit calculates the road surface reflectivity index. With a preset reflectivity threshold Compare. The threshold. It is a safety boundary value calibrated through experiments, representing the lowest risk level that the system considers likely to cause significant glare interference or affect visibility to the driver. If the system determines that the current road surface reflectivity exceeds a safe and comfortable range, intervention to adjust the low beam headlight angle is necessary. At this point, the main control unit initiates the process of generating an angle adjustment command.

[0116] The process of generating angle adjustment commands first involves determining the excessive reflectivity index. The system queries a pre-stored mapping table between reflectivity and angle adjustment. This table is a data table calibrated through extensive real-vehicle testing and optical simulation; its input is the reflectivity index. (or the level of discretization), the output is the corresponding basic angle adjustment amount. Basic angle adjustment amount This is typically a negative value (in degrees), indicating the angle at which the low beam headlights should be adjusted downwards in the vertical direction to reduce road glare. The query process uses either exact matching or range matching to find the current... corresponding .

[0117] Subsequently, the main control unit obtains the current illumination angle of the low beam headlights. This angle information can be read from the sensors of the headlight leveling system, or recorded internally by the system based on the state of the most recent adjustment command. Next, the target illumination angle is calculated. The target illumination angle is obtained by adding the base angle adjustment amount to the current illumination angle, i.e. .because A negative value means that the optical axis is adjusted downwards, lowering the cutoff line of the low beam headlights. This allows more light to be projected onto the closer road surface, reducing the amount of strong light projected onto the distant, slippery road surface, thus suppressing glare reflected back to the driver's eyes.

[0118] Finally, the main control unit generates the parameters for adjusting the low beam headlights to illuminate the target. Angle adjustment command. This command contains the target angle value. or relative adjustment amount The main control unit incorporates this angle adjustment command as an additional control requirement into the optimized lighting control parameters of the low beam headlights, which are being generated or have already been generated. This integration can be achieved by adding the command as an independent control dimension to the control parameter vector, for example, expanding the parameter vector from [brightness, color temperature] to [brightness, color temperature, target angle]; or, during the final drive signal conversion stage, converting this angle command along with parameters such as brightness and color temperature into control signals for driving the stepper motor or servo motor. Thus, when the multi-channel power drive module executes the optimized lighting control parameters, the low beam headlight drive channel will not only control the brightness and color temperature of the light but also adjust the illumination angle of the lamp head to the calculated value through the corresponding actuator. This allows the system to proactively adapt to slippery road conditions, ensuring necessary lighting while reducing glare from road surface reflections and improving driving safety at night or in inclement weather.

[0119] See Figure 8 This is a flowchart illustrating a vehicle lighting decoding and control method based on Lin / CAN dual-bus context awareness, provided in an embodiment of the present invention. The vehicle lighting decoding and control method based on Lin / CAN dual-bus context awareness, applied to the vehicle lighting decoding and control system based on Lin / CAN dual-bus context awareness as described in any of the above schemes, includes the following steps:

[0120] The system acquires context-aware information and receives and parses bus signals from the vehicle's MCU through a bus protocol decoding unit to obtain vehicle status information. The context-aware information includes facial images, traffic conditions, light intensity, and precipitation data of the vehicle's environment.

[0121] Generate lighting control commands based on the vehicle status information and the context awareness information;

[0122] The corresponding vehicle lights are driven in coordination with the light control commands through the multi-channel power drive module.

[0123] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0124] This technology directly acquires comprehensive information including facial images, traffic conditions, light intensity, and precipitation data through a context-aware device. Simultaneously, a bus protocol decoding control unit parses Lin / CAN signals to obtain vehicle status information. The main control unit integrates these two types of information to generate lighting control commands, which are then collaboratively driven by a multi-channel power drive module. This technology, employing a parallel information acquisition architecture involving both context-aware devices and bus decoding, allows the main control unit to directly grasp the real-time multi-dimensional status of people, vehicles, and the environment, rather than passively receiving fixed program commands. This endows the system with the ability to proactively identify complex driving scenarios and driver states. Combined with multi-channel collaborative driving, it achieves dynamic, scenario-based lighting output that goes beyond simple on / off operations. Addressing the problem that existing vehicle lighting control technologies cannot automatically adapt to complex and changing driving environments and driver states, this invention actively generates lighting control commands and collaboratively drives corresponding vehicle lights by combining vehicle status information and context-aware information. This enables automatic adaptation to complex and changing driving environments and driver states, effectively improving driving safety and the driver's experience.

[0125] The method further includes:

[0126] The vehicle controller sends lighting control commands to the MCU via the CAN bus. These commands include starting the vehicle, turning off the vehicle, turning on the left turn signal, turning on the right turn signal, turning on the high beam, and turning on the low beam.

[0127] The lighting control command is generated by the MCU outputting the lighting controller and then processing it.

[0128] The lighting control commands are transmitted to the vehicle lighting signal receiver via the Lin bus;

[0129] The vehicle light signal receiver transmits the signal to the Lin bus protocol module;

[0130] The Lin bus protocol module performs protocol parsing on the received signals and then sends them to the MCU;

[0131] The MCU generates drive instructions based on the protocol parsing results and outputs them to the power output module;

[0132] The power output module drives the lamp head drive module, which in turn controls the dual-color display module to execute any one or more of the following lighting modes: welcome, home, left turn, right turn, fog light high beam, and fog light low beam.

[0133] In this embodiment, as an alternative lighting control path, the vehicle control unit (VCU) sends basic lighting control commands to the vehicle's microcontroller unit (MCU) via the CAN bus. These commands include discrete on / off commands such as turning on the vehicle, turning off the vehicle, turning on the left turn signal, turning on the right turn signal, turning on the high beam, and turning on the low beam. Upon receiving these commands, the MCU first performs protocol parsing and validity verification, then outputs them to the lighting controller (e.g., the master control unit or an independent lighting control slave node as described in this system). The lighting controller then generates the corresponding underlying driver logic based on the command type. Next, the lighting control commands are transmitted to each headlight signal receiver via the Lin bus. This Lin bus uses a master-slave communication architecture, with the lighting controller as the master node and each headlight signal receiver as a slave node. The headlight signal receivers transmit the received Lin bus signals to their internal Lin bus protocol module. This module performs frame parsing, verification, and command extraction on the signals, and sends the parsing results back to the MCU or directly to the local power output module. The MCU generates specific drive instructions (such as PWM duty cycle, illumination duration, and flashing frequency) based on the protocol parsing results and outputs them to the power output module. The power output module integrates power switching devices (such as MOSFETs or relays) to control the on / off state or current magnitude of the lamp head drive module according to the drive instructions, thereby driving the dual-color display module to execute any one or more of the following lighting modes: welcome, home, left turn, right turn, fog light high beam, and fog light low beam. This process complements the aforementioned context-aware active control path: when the system does not use the context-aware automatic decision-making mode, this path can respond to the vehicle controller's regular instructions to achieve basic lighting functions; while in context-aware mode, the optimized lighting control parameters generated by the main control unit can also be executed through the same Lin bus and power drive hardware link, thus achieving intelligent lighting output while ensuring compatibility with the original manual / semi-automatic control.

[0134] It is understood that the specific implementation of the above-described vehicle lighting decoding control method based on Lin / CAN dual-bus context awareness can be referred to the relevant content of the above-described vehicle lighting decoding control device based on Lin / CAN dual-bus context awareness, and will not be repeated here.

[0135] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A vehicle lighting decoding control system based on Lin / CAN dual-bus context awareness, characterized in that, include: A power supply unit is used to connect to the vehicle battery voltage and provide power output to the control system. A context-aware device is used to acquire context-aware information, which includes facial images, traffic condition information, light intensity and precipitation data of the vehicle's environment. The bus protocol decoding control unit is used to receive and parse the vehicle's Lin / CAN bus signals to obtain vehicle status information; The main control unit, connected to the bus protocol decoding control unit and the context perception device, is used to generate lighting control commands based on the vehicle status information and the context perception information. A multi-channel power drive module, connected to the main control unit, is used to collaboratively drive the corresponding vehicle lights according to the lighting control commands. The context-aware device includes: An environmental perception module is used to collect data on light intensity and precipitation in the environment where the vehicle is located, and connects to the main control unit through the bus protocol decoding control unit. The vehicle-to-everything (V2X) communication module is used to receive traffic information from the V2X network and connect to the main control unit through the bus protocol decoding control unit. The in-vehicle camera module is used to capture facial images of people inside the vehicle and connects to the main control unit through the bus protocol decoding control unit. The main control unit is also used for: Feature extraction is performed on the face image, the vehicle status information, and the traffic condition information to obtain the driver's emotion features, vehicle driving status features, and road condition features respectively. The spectral attenuation compensation coefficient of the vehicle headlights is calculated based on the ambient light intensity and precipitation data. The driver's emotional characteristics, the vehicle's driving state characteristics, and the road condition characteristics are fused together to construct driving context features; Based on driving context characteristics, and with driving safety and driving emotion matching degree as multiple objectives, the target headlight to be controlled is determined and the control parameters of the target headlight are generated. Based on the spectral attenuation compensation coefficient, the control parameters of the target vehicle lamp are adaptively corrected to generate optimized lighting control parameters for the target vehicle lamp. Based on the optimized lighting control parameters of the target vehicle light, the multi-channel power drive module is controlled to drive the target vehicle light; The process, based on driving context features and with driving safety and driving emotion matching as multiple objectives, optimizes the process to determine the target headlight to be controlled and generates control parameters for the target headlight. Specifically, this includes: The driving scenario features are input into a preset vehicle light control recognition model to determine the target vehicle light to be controlled; From the pre-stored sample library of historical driving scenarios and lighting control parameter pairs, retrieve multiple historical scenario samples that have an Euclidean distance of less than a preset threshold with respect to the driving scenario features and match the type of the target vehicle light. The lighting control parameters corresponding to the multiple historical scenario samples are obtained from the sample library as reference points, and perturbations that meet the preset vehicle light safety constraints are applied to each reference point to generate multiple candidate vehicle light control parameters for the target vehicle light. The vehicle driving state characteristics and road condition characteristics are input into the driving risk assessment model to calculate the driving risk probability corresponding to each candidate headlight control parameter. The driving risk probability is then converted into a driving safety reward value using a safety probability conversion function. The conversion function is... ,in It is a positive scaling constant. For the probability of driving risks, This is a driving safety bonus value; The driver's emotional characteristics and each of the candidate headlight control parameters are vector-projected in a preset joint feature space, and the cosine similarity between the projected vectors is calculated as the emotion matching reward value. Based on preset safety weight coefficients and emotion weight coefficients, the driving safety reward value and the emotion matching reward value are weighted and summed to obtain a comprehensive reward value, and the candidate headlight control parameter with the largest comprehensive reward value is selected as the control parameter of the target headlight.

2. The control system according to claim 1, characterized in that, The process of fusing driver emotional characteristics, vehicle driving state characteristics, and road condition characteristics to construct driving context features specifically includes: Based on the current vehicle speed, calculate the dynamic weighting coefficients of the driver's emotional characteristics, the vehicle's driving state characteristics, and the road condition characteristics; The driver's emotional features, the vehicle's driving state features, and the road condition features are multiplied by their respective dynamic weight coefficients to obtain a weighted feature vector. The weighted feature vectors are concatenated to generate the driving context features.

3. The control system according to claim 2, characterized in that, The calculation of the spectral attenuation compensation coefficient for the vehicle headlights based on the ambient light intensity and precipitation data specifically includes: The ambient light intensity is divided into discrete illuminance levels, and the current precipitation level is determined based on the precipitation data. Based on the illuminance level and the precipitation level, the basic spectral attenuation factor is obtained by indexing a pre-stored multidimensional lookup table. The basic spectral attenuation factor is dynamically corrected based on the current vehicle speed. When the vehicle speed is higher than a first threshold, the basic spectral attenuation factor is decreased; when the vehicle speed is lower than a second threshold, the basic spectral attenuation factor is increased. Based on the aforementioned basic spectral attenuation factor, the spectral attenuation compensation coefficient of the vehicle lamp is calculated.

4. The control system according to claim 3, characterized in that, The process of adaptively correcting the control parameters of the target vehicle headlight based on the spectral attenuation compensation coefficient to generate optimized headlight control parameters specifically includes: The theoretical compensation gain is calculated based on the spectral attenuation compensation coefficient, and the safety redundancy coefficient is determined based on the vehicle's current speed. Based on the theoretical compensation gain and the safety redundancy coefficient, the actual compensation gain is calculated. Based on the actual compensation gain, the control parameters of the target vehicle headlight are corrected to generate optimized headlight control parameters for the target vehicle headlight.

5. The control system according to claim 4, characterized in that, The step of controlling the multi-channel power drive module to drive the target vehicle light based on the optimized lighting control parameters of the target vehicle light specifically includes: Based on the type identifier of the target vehicle light, determine the target drive channel in the multi-channel power drive module that corresponds to the target vehicle light; The optimized lighting control parameters are converted into driving signals and output to the multi-channel power drive module, so as to drive the target vehicle light through the target driving channel of the multi-channel power drive module corresponding to the target vehicle light.

6. The control system according to claim 2, characterized in that, It also includes an external camera module for acquiring road surface images; the main control unit is connected to the external camera module, and the main control unit is further used for: When the target vehicle's headlights are low beam, the slippery areas of the road surface are identified using the road image. The road surface reflectivity index is calculated based on the area ratio of the slippery road surface and the precipitation data. When the road surface reflectivity index exceeds the preset reflectivity threshold, an angle adjustment command for the vehicle's low beam headlights corresponding to the road surface reflectivity index is generated. The angle adjustment command is incorporated into the optimized lighting control parameters.

7. A vehicle lighting decoding control method based on Lin / CAN dual-bus context awareness, applied to the vehicle lighting decoding control system based on Lin / CAN dual-bus context awareness as described in any one of claims 1-6, characterized in that, Includes the following steps: The system acquires context-aware information and receives and parses bus signals from the vehicle via a bus protocol decoding control unit to obtain vehicle status information. The context-aware information includes facial images, traffic conditions, light intensity and precipitation data of the vehicle's environment. Generate lighting control commands based on the vehicle status information and the context awareness information; The corresponding vehicle lights are driven in coordination with the light control commands through the multi-channel power drive module.

Images