A motor vehicle headlamp

Abstract

The application discloses a car headlamp and relates to the field of vehicle lamps, which comprises a lamp shell body, the inside of the lamp shell body is provided with a low beam, a high beam, a position lamp and a turn signal lamp, the surface of the lamp shell body is provided with a temperature and humidity sensing module and a driving mechanism for periodically changing the spatial orientation of the temperature and humidity sensing module, the driving mechanism is used for collecting temperature and humidity data, the surface of the lamp shell body is provided with a hot air mechanism, the hot air mechanism comprises at least one fan, an electric heating wire and a cover plate arranged correspondingly for controlling the flow direction of the hot air, the directional hot air intervention is started after the risk point is verified by predicting the fogging trend, so that the problem is suppressed in the budding state, the local precise heating is more energy-saving than the overall heating, the uniform damage of thermal stress to the lampshade material is avoided, the visual blind area which is difficult to be directly detected by a traditional sensor is identified, and a light angle adjusting instruction is directly generated.

Description

Technical Field

[0001] This invention relates to the field of vehicle lighting technology, specifically to an automotive headlight. Background Technology

[0002] As a core component of vehicle active safety, automotive headlights are undergoing a profound transformation from passive lighting to intelligent environmental perception and active control. With the evolution of automotive electronic architecture towards intelligence, modern vehicles are equipped with rich sensor networks and powerful on-board computing units, providing a hardware foundation for headlight systems to achieve data fusion and intelligent decision-making.

[0003] Traditional automotive headlight systems have poor environmental adaptability. Fogging inside the headlight housing generally relies on passive ventilation or simple heating films, which cannot monitor temperature and humidity gradients in different areas in real time and perform precise active dehumidification and anti-fogging, affecting light transmittance and driving safety. Although adaptive headlight systems have emerged, their judgment of blind spots such as curves and slopes relies heavily on indirect signals such as steering wheel angle and vehicle speed, lacking real-time direct analysis of actual road images. This results in limited prediction accuracy and response speed, making it difficult to cope with sudden changes in road conditions. Summary of the Invention

[0004] (a) Technical problems to be solved In view of the above-mentioned shortcomings of the prior art, the present invention provides an automotive headlight that can effectively solve the problems in the prior art.

[0005] (II) Technical Solution To achieve the above objectives, the present invention provides the following technical solution: This invention discloses an automotive headlight, comprising: The lamp housing body has a low beam lamp, a high beam lamp, a position lamp and a turn signal inside. A temperature and humidity sensing module is installed on the surface of the lamp housing body, and a driving mechanism is used to drive the temperature and humidity sensing module to periodically change its spatial orientation. The driving mechanism is used to collect temperature and humidity data. The surface of the lamp housing body is provided with a hot air mechanism, which includes at least one fan, a heating wire, and a cover plate correspondingly provided for controlling the direction of hot air flow. An analysis unit is provided on the surface of the lamp housing body, and the analysis unit includes: The trend analysis module is used to receive temperature and humidity data collected by the drive mechanism from different directions, perform time-series analysis on the temperature and humidity data through a pre-trained trend recognition model, and construct and output trend data reflecting the temperature and humidity change trends in each direction. The verification module is used to make judgments based on the trend data output by the position lights and a preset temperature and humidity fluctuation threshold, and generate control commands for the drive mechanism. An active adjustment module is used to generate adjustment instructions for the hot air mechanism when it is determined that the trend data in a specific direction meets the preset fluctuation threshold conditions. The adjustment instructions include the opening and closing sequence of the specified cover plate, the start and stop sequence of the corresponding fan, and the working temperature of the heating wire. The headlight adjustment unit is used to preset a direction adjustment model through a deep learning algorithm. The direction adjustment model takes image data provided by the vehicle driving recorder as input and outputs adjustment commands for the low beam or high beam illumination angle.

[0006] Furthermore, the drive mechanism is equipped with sub-components, including a slide groove, a lead screw, a square lead screw nut, and a servo motor. The slide groove is located in the middle of the surface of the lamp housing body. The servo motor is installed on the left side inside the slide groove. The left end of the lead screw is fixedly connected to the output shaft of the servo motor, and the right end of the lead screw is rotatably connected to the inner wall of the right side of the slide groove. The square lead screw nut is rotatably connected to the surface of the slide groove via a thread, and the square lead screw nut is slidably connected to the slide groove. The temperature and humidity sensing module is installed on the surface of the square lead screw nut. The control commands received by the drive mechanism include the target rotation direction of the servo motor, the target rotation angle, and the target dwell time at a specific point, as well as the speed control parameters for the servo motor to drive the lead screw to accelerate to the target point or decelerate and stop at the target point.

[0007] Furthermore, the low beam headlight is installed at the top of the lamp housing body surface, the high beam headlight is installed at the bottom of the lamp housing body surface, the turn signal is installed between the low beam headlight and the high beam headlight, and there are two position lights, which are respectively installed on the upper and lower sides of the lamp housing body surface.

[0008] Furthermore, the front of the lamp housing body is evenly provided with mounting grooves, the fan is installed inside the mounting groove, the edge of the heating wire is fixedly connected to the inner wall of the mounting groove, a second servo motor is installed inside the mounting groove, and the output shaft of the second servo motor is fixedly connected to the bottom end of the cover plate.

[0009] Furthermore, the construction process of the trend recognition model in the trend analysis module is as follows: Receives time-series temperature and humidity data sequences collected from different spatial locations according to a preset sampling period; Using a deep learning framework, feature encoding is performed on the input time series data sequence to extract local dependencies and periodic patterns in the time dimension. The contribution weight of different time points in historical data to the prediction of current and future trends is dynamically calculated through an attention mechanism, with a focus on the abrupt change points and steady-state stages of temperature and humidity data. The model takes historical temperature and humidity data within a previous preset time window as input and outputs a predicted sequence of temperature and humidity for each monitoring point within the next time window. It then compares and fuses this predicted sequence with real-time collected data to generate multidimensional trend data containing predicted values, slope of change, and confidence intervals.

[0010] Furthermore, during the operation phase of the verification module, temperature and humidity fluctuation thresholds related to fog risk are pre-stored. The trend data is compared and analyzed with the fluctuation thresholds. When a fog risk trend exceeding the fluctuation threshold is identified in a specific location, a control command is generated for that risk location, and the drive mechanism is driven to operate according to the operating parameters in the control command, so that the temperature and humidity sensing module is positioned in the risk location for key monitoring. After the temperature and humidity sensing module completes monitoring at the risk point, it obtains the secondary monitoring data fed back by the module and transmits it to the trend analysis module for secondary trend analysis to generate secondary trend data. The verification module verifies the secondary trend data. If it is confirmed that the risk of fogging exceeds the threshold, the hot air mechanism control command is triggered.

[0011] Furthermore, the operating logic of the lighting adjustment unit includes the following steps: Collect image sequence data and corresponding vehicle sensor data containing different driving scenarios, including at least straight roads, curves with different curvatures, undulating road surfaces, following vehicles, and meeting vehicles; annotate the image data, including lane curvature, bounding boxes of vehicles or pedestrians ahead, uneven areas of the road surface, curb outlines, and coordinates of the ideal headlight illumination center point; synchronize the timestamps of the sensor data, including vehicle speed, steering angle, yaw rate, and vehicle attitude data. A feature extraction sub-model framework based on convolutional neural networks is constructed to acquire image data provided by the current vehicle driving recorder and extract key features related to the driving environment through image processing technology. Key features include feature information for identifying blind spots in curves or blind spots on uneven road surfaces. Key feature vectors are fused with vehicle sensor data at the same time stamp to form environmental features. Recurrent neural networks are then used to perform sequence modeling of environmental features at continuous time steps to capture the dynamic temporal change patterns of the driving environment. A regression network is constructed, taking comprehensive environmental features as input and calibrated ideal lighting adjustment parameters as training targets. The adjustment parameters include the vertical deflection angle, horizontal deflection angle, and dimming intensity of the low beam or high beam. Through end-to-end training, the regression network learns the mapping relationship from complex environmental states to optimal lighting adjustment commands. The trained feature extraction sub-model, fusion network, and regression network are integrated to form the orientation adjustment model.

[0012] Furthermore, the formula for calculating the adjustment parameter is as follows: ; In the formula, This represents the vertical adjustment angle, measured in degrees. A positive value indicates that the light beam is tilted upwards, while a negative value indicates that it is tilted downwards. This represents the horizontal adjustment angle, measured in degrees. Using the vehicle's direction of travel as a reference, a positive value indicates the headlight beam will deflect to the right, while a negative value indicates it will deflect to the left. This represents the dimming intensity coefficient, a dimensionless scalar between 0 and 1, used to linearly adjust the headlight output power in specific scenarios, such as when following another vehicle. 1 represents full power, and 0 represents minimum power. The overall mapping function representing the orientation adjustment model. Represents feature extraction and fusion functions. Represents time The raw image data obtained internally, Represents time The synchronously acquired vehicle state data vector includes vehicle speed, steering wheel angle, yaw rate, and vehicle pitch angle. Represents time Time Historical time sequence feature window, in which This is the preset time step.

[0013] Furthermore, the lighting adjustment unit is wirelessly connected to a user interaction module. The user interaction module is also wirelessly connected to the active adjustment module. The user interaction module receives adjustment commands generated by the active adjustment module and the lighting adjustment unit, and provides a confirmation interface to the user. After receiving user confirmation, it triggers the final hot air adjustment or lighting adjustment action. The confirmation interface graphically simulates the change in illumination range before and after lighting adjustment, and allows the user to manually fine-tune the lighting based on the automatically generated adjustment commands according to a preset scale. The lighting adjustment action is then submitted to the adjustment actuators of the low beam and high beam.

[0014] Furthermore, the verification module is interconnected with the trend analysis module and the active adjustment module via a wireless network, and the active adjustment module is interconnected with the lighting adjustment unit via a wireless network.

[0015] (III) Beneficial Effects Compared with the known prior art, the technical solution provided by this invention has the following beneficial effects: 1. By using a fixed-point driven temperature and humidity sensing module to perform spatial scanning and combining it with deep learning algorithms to predict fogging trends, and by initiating directional hot air intervention after verifying risk points, problems can be contained in their infancy. Localized precise heating is more energy-efficient than overall heating and avoids uniform damage to the lampshade material caused by thermal stress, thereby improving the reliability and service life of the system.

[0016] 2. By directly analyzing the image of the road ahead, it identifies blind spots that are difficult for traditional sensors to detect directly and generates direct angle adjustment commands, which illuminates dangerous areas more accurately and in a timely manner, giving drivers valuable reaction time and improving the safety performance of active lighting in complex road conditions, especially improving driving safety at night on complex roads such as mountainous areas and rural areas.

[0017] 3. By making unified decisions and scheduling based on multi-source data, the two major functions of internal environmental control and external lighting adaptation are no longer isolated functions, but rather a whole that can work together according to environmental priorities. The adjustment instructions generated by automatic analysis are given to the user for final confirmation, which not only ensures the advanced nature of automated decision-making, but also respects the driver's main control rights, and improves the trust and acceptance of human-machine interaction. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0019] Figure 1 This is a front view structural diagram of the present invention; Figure 2 This is a side view of the structure of the present invention; Figure 3 This is a schematic diagram of the rear view structure of the present invention; Figure 4 This is a bottom-view cross-sectional structural diagram of the hot air mechanism in this invention; Figure 5 This is a front cross-sectional view of the cover plate and mounting groove in this invention. Figure 6 This is a front view cross-sectional structural diagram of the drive mechanism in this invention; Figure 7 This is a schematic diagram of the analysis unit in this invention.

[0020] The labels in the diagram represent: 1. Main body of the lamp housing; 2. Low beam headlight; 3. High beam headlight; 4. Position lights; 5. Turn signals; 6. Drive mechanism; 61. Slide groove; 62. Lead screw; 63. Square lead screw nut; 64. Servo motor one; 7. Hot air mechanism; 71. Mounting slot; 72. Cover plate; 73. Fan; 74. Heating wire; 75. Servo motor two; 8. Analysis unit; 81. Trend analysis module; 82. Verification module; 83. Active adjustment module; 84. Lighting adjustment unit; 85. User interaction module; 9. Temperature and humidity sensing module. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0022] The present invention will be further described below with reference to embodiments.

[0023] This embodiment provides a type of automotive headlight, such as... Figures 1-7 As shown, it includes: The lamp housing body 1 has a low beam lamp 2, a high beam lamp 3, position lights 4 and turn signals 5 installed inside it. The low beam lamp 2 is installed at the top of the surface of the lamp housing body 1, the high beam lamp 3 is installed at the bottom of the surface of the lamp housing body 1, the turn signals 5 are installed between the low beam lamp 2 and the high beam lamp 3, and there are two position lights 4, which are installed on the upper and lower sides of the surface of the lamp housing body 1 respectively. A temperature and humidity sensing module 9 is mounted on the surface of the lamp housing body 1, along with a drive mechanism 6 that periodically changes the spatial orientation of the temperature and humidity sensing module 9. The drive mechanism 6 is used to collect temperature and humidity data. Sub-components are deployed below the drive mechanism 6, including a slide 61, a lead screw 62, a square nut 63, and a servo motor 64. The slide 61 is located in the middle of the surface of the lamp housing body 1. The servo motor 64 is installed on the left side inside the slide 61. The left end of the lead screw 62 is fixedly connected to the output shaft of the servo motor 64, and the right end of the lead screw 62 is rotatably connected to the inner wall of the right side of the slide 61. The square nut 63 rotates via a thread. The square nut 63 is slidably connected to the surface of the slide groove 61, and the temperature and humidity sensing module 9 is installed on the surface of the square nut 63. The control commands received by the drive mechanism 6 include the target rotation direction, target rotation angle and target dwell time at a specific point of the servo motor 64, as well as the speed control parameters for the servo motor 64 to drive the lead screw 62 to accelerate to the target point or decelerate and stop at the target point. The axis of the lead screw 62 is parallel to the vehicle driving direction or arranged at a preset angle, and the displacement path of the square nut 63 covers the key area inside the lamp housing body 1 where temperature and humidity changes need to be monitored.

[0024] A hot air mechanism 7 is provided on the surface of the lamp housing body 1. The hot air mechanism 7 includes at least one electric fan 73, an electric heating wire 74, and a cover plate 72 for controlling the direction of hot air flow. The front of the lamp housing body 1 is evenly provided with mounting grooves 71. The electric fan 73 is installed inside the mounting groove 71. The edge of the electric heating wire 74 is fixedly connected to the inner wall of the mounting groove 71. A servo motor 75 is installed inside the mounting groove 71. The output shaft of the servo motor 75 is fixedly connected to the bottom end of the cover plate 72. There are three cover plates 72, which correspond to the upper, middle and lower sections inside the lamp housing body 1, respectively. The verification module 82 selectively controls one or more cover plates 72 to open according to the specific location of the temperature and humidity abnormality, and controls the electric fan 73 and electric heating wire 74 of the corresponding section to work in conjunction. The working temperature of the electric heating wire 74 is set to multiple adjustable levels. The active adjustment module 83 matches and selects the corresponding working temperature level according to the specific humidity value or condensation risk level detected by the temperature and humidity sensing module 9.

[0025] An analysis unit 8 is provided on the surface of the lamp housing body 1. The analysis unit 8 includes: The trend analysis module 81 is used to receive temperature and humidity data collected by the drive mechanism 6 from different directions, perform time-series analysis on the temperature and humidity data through a pre-trained trend recognition model, and construct and output trend data reflecting the changing trends of temperature and humidity in each direction; the construction process of the trend recognition model is as follows: Receives time-series temperature and humidity data sequences collected from different spatial locations according to a preset sampling period; Using a deep learning framework, feature encoding is performed on the input time series data sequence to extract local dependencies and periodic patterns in the time dimension. The contribution weight of different time points in historical data to the prediction of current and future trends is dynamically calculated through an attention mechanism, with a focus on the abrupt change points and steady-state stages of temperature and humidity data. The model takes historical temperature and humidity data within a previous preset time window as input and outputs a predicted temperature and humidity sequence for each monitoring point within the next future time window. It then compares and fuses this predicted sequence with real-time collected data to generate multidimensional trend data containing predicted values, slope of change, and confidence intervals. The trend data specifically represents the rate of temperature increase or decrease, the risk level of humidity saturation or condensation, and the probability of change for each location in the future.

[0026] The verification module 82 is used to judge based on the trend data output by the position light 4 and the preset temperature and humidity fluctuation threshold, and generate control commands for the drive mechanism 6. During the operation of the verification module 82, the temperature and humidity fluctuation thresholds related to the risk of fog are pre-stored. The trend data is compared and analyzed with the fluctuation thresholds. When a fog risk trend exceeding the fluctuation threshold is identified in a specific location, a control command for the risk location is generated, and the drive mechanism 6 is driven to run according to the operating parameters in the control command, so that the temperature and humidity sensing module 9 is positioned in the risk location for key monitoring. After the temperature and humidity sensing module 9 completes monitoring at the risk point, it obtains the secondary monitoring data fed back by the sensor and transmits it to the trend analysis module 81 for secondary trend analysis to generate secondary trend data. The verification module 82 verifies the secondary trend data. If it is confirmed that the risk of fogging exceeds the threshold, the hot air mechanism 7 is triggered to issue a control command.

[0027] The active adjustment module 83 is used to generate an adjustment command for the hot air mechanism 7 when it is determined that the trend data of a specific direction meets the preset fluctuation threshold condition. The adjustment command includes specifying the opening and closing sequence of the cover plate 72 and the start and stop sequence of the corresponding electric fan 73 and the working temperature of the heating wire 74. The headlight adjustment unit 84 is used to preset a direction adjustment model through a deep learning algorithm. The direction adjustment model takes image data provided by the vehicle driving recorder as input and outputs adjustment commands for the illumination angle of the low beam headlight 2 or the high beam headlight 3. The headlight adjustment unit 84 is wirelessly connected to a user interaction module 85. The user interaction module 85 is wirelessly connected to the active adjustment module 83. The user interaction module 85 is used to receive the adjustment commands generated by the active adjustment module 83 and the headlight adjustment unit 84 and provide a confirmation interface to the user. After obtaining user confirmation, the final hot air adjustment action or headlight adjustment action is triggered. The confirmation interface graphically simulates and displays the change in illumination range before and after headlight adjustment, and allows the user to manually fine-tune according to a preset scale based on the automatically generated adjustment commands. The headlight adjustment action is submitted to the adjustment actuators of the low beam headlight 2 and the high beam headlight 3.

[0028] The verification module 82 is interconnected with the trend analysis module 81 and the active adjustment module 83 via a wireless network, and the active adjustment module 83 is interconnected with the lighting adjustment unit 84 via a wireless network.

[0029] Compared with existing technologies, by combining a movable temperature and humidity sensing module with a deep learning algorithm, it can proactively predict the risk trend of fogging in specific areas inside the lamp cover, and link the drive mechanism 6 for key verification. Finally, the hot air mechanism 7 achieves localized preventive dehumidification and defogging, fundamentally avoiding the impact of fogging on the uniformity and intensity of illumination. Utilizing the vehicle's existing image data, it intelligently identifies blind spots caused by curves, bumpy roads, etc., and automatically calculates and provides headlight adjustment schemes, which are executed after user confirmation, improving the safety of active lighting and the driver experience in complex road conditions.

[0030] At other levels, in this embodiment, the operating logic of the lighting adjustment unit 84 includes the following steps: Collect image sequence data and corresponding vehicle sensor data containing different driving scenarios. Driving scenarios include at least straight roads, curves with different curvatures, undulating roads, following vehicles, and meeting vehicles. Annotate the image data, including lane curvature, bounding boxes of vehicles or pedestrians ahead, uneven areas of the road surface, curb outlines, and coordinates of the ideal headlight illumination center point. Synchronize the timestamps of the sensor data, including vehicle speed, steering angle, yaw rate, and vehicle attitude data. A feature extraction sub-model framework based on convolutional neural networks is constructed to acquire image data provided by the current vehicle driving recorder and extract key features related to the driving environment through image processing technology. Key features include feature information for identifying blind spots in curves or blind spots on uneven road surfaces. Key feature vectors are fused with vehicle sensor data at the same time stamp to form environmental features. Recurrent neural networks are then used to perform sequence modeling of environmental features at continuous time steps to capture the dynamic temporal change patterns of the driving environment. A regression network is constructed, taking comprehensive environmental features as input and calibrated ideal lighting adjustment parameters as training targets. The adjustment parameters include the vertical deflection angle, horizontal deflection angle, and dimming intensity of the low beam 2 or high beam 3. Through end-to-end training, the regression network learns the mapping relationship from complex environmental states to the optimal lighting adjustment command. The formula for calculating the adjustment parameter is: ; In the formula, This represents the vertical adjustment angle, measured in degrees. A positive value indicates that the light beam is tilted upwards, while a negative value indicates that it is tilted downwards. This represents the horizontal adjustment angle, measured in degrees. Using the vehicle's direction of travel as a reference, a positive value indicates the headlight beam will deflect to the right, while a negative value indicates it will deflect to the left. This represents the dimming intensity coefficient, a dimensionless scalar between 0 and 1, used to linearly adjust the headlight output power in specific scenarios, such as when following another vehicle. 1 represents full power, and 0 represents minimum power. The overall mapping function representing the orientation adjustment model. Represents feature extraction and fusion functions. Represents time The raw image data obtained internally, Represents time The synchronously acquired vehicle state data vector includes vehicle speed, steering wheel angle, yaw rate, and vehicle pitch angle. Represents time Time Historical time sequence feature window, in which For a preset time step, the window data is input into the sequence modeling part of the orientation adjustment model to capture the dynamic changing trends of the environment; The trained feature extraction sub-model, fusion network, and regression network are integrated to form a direction adjustment model. The integrated model is further optimized in a simulated environment using a reinforcement learning framework. The reward functions are maximized lighting effect, minimized glare, and minimized system response delay. The model parameters are fine-tuned to improve its generalization and decision-making ability in unknown scenarios.

[0031] Working principle: In specific implementation, the temperature and humidity sensing module 9 collects temperature and humidity data and submits it to the analysis unit 8 for analysis. If there is an abnormality, the production control command is sent to the servo motor 64. The output shaft of the servo motor 64 drives the lead screw 62 to rotate. Under the action of the thread and the limitation of the movement trajectory of the square lead screw 63 by the slide groove 61, the square lead screw 63 drives the temperature and humidity sensing module 9 to move to the specified position. The temperature and humidity sensing module 9 collects the temperature and humidity data of the specified position and analyzes it through the analysis unit 8. Based on the analysis results of the analysis unit 8, control commands for the hot air mechanism 7 are generated, which causes the servo motor 75 in the specified position to start, and the output shaft of the servo motor 75 drives the cover plate 72 to rotate, so that the cover plate 72 is hidden. The electric fan 73 in the corresponding position is started, and the heating wire 74 is controlled to the preset temperature, so that the hot air acts on the specified position inside the lamp housing body 1. By utilizing image recognition and deep learning technologies, the analysis unit 8 can analyze the road environment in real time, automatically identify blind spots caused by curves, bumpy roads, etc., and intelligently adjust the beam direction of the low beam headlights 2 and high beam headlights 3 to illuminate dangerous areas in advance, thereby improving the active safety of nighttime driving and transforming the vehicle lights from passive lighting devices into intelligent safety nodes that integrate environmental perception, risk warning, and active adaptation.

[0032] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A type of automotive headlight, characterized in that, include: The lamp housing body (1) is provided with a low beam lamp (2), a high beam lamp (3), a position lamp (4) and a turn signal (5) inside the lamp housing body (1). A temperature and humidity sensing module (9) is installed on the surface of the lamp housing body (1), and a driving mechanism (6) is used to drive the temperature and humidity sensing module (9) to periodically change its spatial orientation. The driving mechanism (6) is used to collect temperature and humidity data. The surface of the lamp housing body (1) is provided with a hot air mechanism (7), which includes at least one electric fan (73), an electric heating wire (74) and a cover plate (72) correspondingly provided for controlling the hot air flow direction. An analysis unit (8) is provided on the surface of the lamp housing body (1), and the analysis unit (8) includes: The trend analysis module (81) is used to receive temperature and humidity data collected by the drive mechanism (6) in different directions, perform time series analysis on the temperature and humidity data through a pre-trained trend recognition model, and construct and output trend data reflecting the temperature and humidity change trends in each direction. The verification module (82) is used to make a judgment based on the trend data output by the position light (4) and the preset temperature and humidity fluctuation threshold, and generate control commands for the drive mechanism (6). The active adjustment module (83) is used to generate an adjustment command for the hot air mechanism (7) when it is determined that the trend data of a specific direction meets the preset fluctuation threshold condition. The adjustment command includes specifying the opening and closing sequence of the cover plate (72) and the start and stop sequence of the corresponding electric fan (73) and the working temperature of the heating wire (74). The headlight adjustment unit (84) is used to preset the direction adjustment model through a deep learning algorithm. The direction adjustment model takes the image data provided by the vehicle driving record device as input and outputs the adjustment command of the low beam (2) or high beam (3) illumination angle.

2. The automotive headlight according to claim 1, characterized in that, The drive mechanism (6) has sub-components, including a slide groove (61), a lead screw (62), a square lead screw nut (63), and a servo motor (64). The slide groove (61) is located in the middle of the surface of the lamp housing body (1). The servo motor (64) is installed on the left side inside the slide groove (61). The left end of the lead screw (62) is fixedly connected to the output shaft of the servo motor (64). The right end of the lead screw (62) is rotatably connected to the inner wall on the right side of the slide groove (61). The square lead screw (63) is... The square nut (63) is slidably connected to the surface of the slide groove (61) by a threaded rotation. The temperature and humidity sensing module (9) is installed on the surface of the square nut (63). The control commands received by the drive mechanism (6) include the target rotation direction, target rotation angle and target dwell time at a specific point of the servo motor (64), and the speed control parameters for the servo motor (64) to drive the lead screw (62) to accelerate to the target point or decelerate and stop at the target point.

3. The automotive headlight according to claim 1, characterized in that, The low beam lamp (2) is installed at the top of the surface of the lamp housing body (1), the high beam lamp (3) is installed at the bottom of the surface of the lamp housing body (1), the turn signal (5) is installed between the low beam lamp (2) and the high beam lamp (3), and there are two position lights (4), which are installed on the upper and lower sides of the surface of the lamp housing body (1) respectively.

4. A car headlight according to claim 1, characterized in that, The lamp housing body (1) has a uniformly provided mounting groove (71) on the front side. The electric fan (73) is installed inside the mounting groove (71). The edge of the heating wire (74) is fixedly connected to the inner wall of the mounting groove (71). A servo motor (75) is installed inside the mounting groove (71). The output shaft of the servo motor (75) is fixedly connected to the bottom end of the cover plate (72).

5. A car headlight according to claim 1, characterized in that, The construction process of the trend identification model in the trend analysis module (81) is as follows: Receives time-series temperature and humidity data sequences collected from different spatial locations according to a preset sampling period; A deep learning framework is used to encode the features of the input time series data sequence, extract local dependencies and periodic patterns in the time dimension, and dynamically calculate the contribution weight of different time points in historical data to the prediction of current and future trends through an attention mechanism. The model takes historical temperature and humidity data within a previous preset time window as input and outputs a predicted sequence of temperature and humidity for each monitoring point within the next time window. It then compares and fuses this predicted sequence with real-time collected data to generate multidimensional trend data containing predicted values, slope of change, and confidence intervals.

6. A car headlight according to claim 1, characterized in that, During the operation phase of the verification module (82), a temperature and humidity fluctuation threshold related to fog risk is pre-stored. The trend data is compared and analyzed with the fluctuation threshold. When a fog risk trend exceeding the fluctuation threshold is identified in a specific location, a control command is generated for that risk location, and the drive mechanism (6) is driven to run according to the operating parameters in the control command, so that the temperature and humidity sensing module (9) is positioned in the risk location for key monitoring. After the temperature and humidity sensing module (9) completes monitoring at the risk point, it obtains the secondary monitoring data fed back by the module and transmits it to the trend analysis module (81) for secondary trend analysis to generate secondary trend data. The verification module (82) verifies the secondary trend data. If it is confirmed that the fogging risk exceeds the threshold, the hot air mechanism (7) is triggered to issue a control command.

7. A car headlight according to claim 1, characterized in that, The operating logic of the lighting adjustment unit (84) includes the following steps: Collect image sequence data containing different driving scenarios and corresponding vehicle sensor data, and annotate the image data; A feature extraction sub-model framework based on convolutional neural networks is constructed to acquire image data provided by the current vehicle driving recorder and extract key features related to the driving environment through image processing technology. Key features include feature information for identifying blind spots in curves or blind spots on uneven road surfaces. Key feature vectors are fused with vehicle sensor data at the same time stamp to form environmental features, and recurrent neural networks are used to perform sequence modeling of environmental features at continuous time steps. A regression network is constructed with comprehensive environmental features as input and calibrated ideal lighting adjustment parameters as training targets. The adjustment parameters include the vertical deflection angle, horizontal deflection angle and dimming intensity of the low beam (2) or high beam (3). The trained feature extraction sub-model, fusion network, and regression network are integrated to form the orientation adjustment model.

8. A car headlight according to claim 7, characterized in that, The formula for calculating the adjustment parameter is: ； In the formula, This represents adjusting the angle in the vertical direction. This represents adjusting the angle in the horizontal direction. Represents the dimming intensity coefficient. The overall mapping function representing the orientation adjustment model. Represents feature extraction and fusion functions. Represents time The raw image data obtained internally, Represents time The vehicle status data vector acquired synchronously, Represents time Time Historical time sequence feature window, in which This is the preset time step.

9. A car headlight according to claim 1, characterized in that, The lighting adjustment unit (84) is connected to the user interaction module (85) via a wireless network. The user interaction module (85) is connected to the active adjustment module (83) via a wireless network. The user interaction module (85) is used to receive the adjustment instructions generated by the active adjustment module (83) and the lighting adjustment unit (84) and provide a confirmation interface to the user. After obtaining the user's confirmation, the final hot air adjustment action or lighting adjustment action is triggered. The confirmation interface graphically simulates the change in illumination range before and after lighting adjustment and allows the user to manually fine-tune according to a preset scale based on the automatically generated adjustment instructions. The lighting adjustment action is handed over to the adjustment actuators of the low beam (2) and high beam (3).

10. A car headlight according to claim 1, characterized in that, The verification module (82) is interconnected with the trend analysis module (81) and the active adjustment module (83) via a wireless network, and the active adjustment module (83) is interconnected with the lighting adjustment unit (84) via a wireless network.

Images