Intelligent vehicle lamp assembly

Abstract

The invention discloses an intelligent automobile lamp assembly which is used in the technical field of automobile lamps and comprises a sensor module, an instrument prompting module, a processing module, an intelligent light control module and an executing mechanism. The sensor module is used for collecting key data of a vehicle state and an external environment; the instrument prompting module is used for displaying the state of the vehicle lamp system and information related to driving in real time; the processing module is used for analyzing the vehicle state and the key data of the external environment and generating a vehicle lamp adjusting strategy; the intelligent light control module is used for receiving the vehicle lamp adjusting strategy and converting the vehicle lamp adjusting strategy into a specific control instruction; and the execution mechanism is used for executing the control instruction and adjusting the brightness and direction of the vehicle lamp based on the control instruction. Based on the fuzzy logic reasoning method, complex and fuzzy data input can be processed, the vehicle lamp adjustment strategy is flexibly generated according to different driving conditions, and vehicle lamp adjustment has better scene adaptability.

Description

Technical Field

[0001] The present invention relates to the technical field of automobile lamps, and in particular to an intelligent automobile lamp assembly. Background Art

[0002] The intelligent lighting systems widely installed on modern cars mainly include the following: automatic headlights, follow-up steering headlights and adaptive front lighting systems.

[0003] Automatic headlights: This is a light-sensing control system installed on the headlights by some car manufacturers to improve user convenience. When the external light intensity is higher or lower than the preset value, the headlights will be automatically turned on or off.

[0004] Adaptive steering headlights: Also known as automatic steering headlights, they can dynamically adjust the direction of the headlights according to the vehicle's steering wheel angle, deflection rate and driving speed. The system ensures that the headlights are always consistent with the vehicle's current driving direction, especially in poor light or curved roads. It can significantly expand the driver's field of vision and provide optimal lighting for the road ahead, thereby greatly enhancing driving safety.

[0005] Adaptive Front Lighting System (AFS): AFS is a headlight system that can automatically switch between two or more types of light according to changes in vehicle driving conditions; it plays an important role in improving nighttime driving safety; however, even for AFS, the main function is still to automatically adjust the brightness of the main lighting fixtures according to ambient light (such as switching between high and low beams), and does not include automatic control of other lights such as fog lights or hazard warning flashers; AFS is a complex system composed of a sensor group, transmission path, processor and actuator; due to the need to make comprehensive judgments on multiple driving conditions, AFS is objectively a multi-input and multi-output system. system; the information returned by most sensors (such as vehicle speed, vehicle turning angle, vehicle tilt angle, etc.) can be quantified, but some information (such as road surface flatness, rainfall amount, etc.) can only provide qualitative data and cannot be accurately quantified; this results in the central processor of AFS being able to only make fuzzy judgments; in addition, these information are interrelated, for example, in the case of water accumulation on rainy days, the steering angle of the vehicle will be significantly different from that on sunny days; the central processor of AFS not only needs to make fuzzy judgments on these related information, but also needs to continuously correct system parameters to adapt to environmental changes. Therefore, AFS is essentially an adaptive fuzzy system.

[0006] With the development of intelligent driving technology, the ability of vehicles to "think independently" has made a qualitative leap, and consumers' demand for automotive intelligence is also growing; with the foundation of intelligent driving systems, it has become an inevitable development trend for traditional headlights to become more intelligent.

[0007] In the prior art, the defects of automatic headlights, follow-up steering headlights and adaptive headlight systems are as follows:

[0008] 1. The adaptive headlight system mainly focuses on adjusting the brightness of the main lights according to the ambient light (such as automatically switching between high and low beams), but its control range does not include other lights such as fog lights and hazard warning flashers; among the information received by the system, except for a few data such as vehicle speed, vehicle body angle, and vehicle body tilt angle that can be quantified, most of the information fed back by sensors (such as road conditions, rainfall, etc.) can only be described qualitatively and cannot be accurately quantified; for example, the system cannot accurately judge whether the road surface is flat, the amount of rainfall, and other factors, which leads to ambiguity in the system's judgment.

[0009] 2. There is a certain delay in turning on the automatic headlights. After the light sensing system detects that the light has dimmed, it usually takes a few seconds for the headlights to be automatically turned on. However, according to traffic regulations, vehicles must turn on the headlights in advance when entering specific places such as tunnels. Therefore, in actual driving, the driver is still required to manually operate the headlights in advance.

[0010] 3. The adaptive steering headlights are sensitive to light and environment. When the light is insufficient or blocked, the adaptive steering headlights may not work properly.

[0011] Currently, no effective solution has been proposed for the problems in the related technologies. Summary of the invention

[0012] In order to overcome the above problems, the present invention aims to provide an intelligent vehicle lamp assembly to overcome the above technical problems existing in the existing related technologies.

[0013] To this end, the specific technical solution adopted by the present invention is as follows:

[0014] An intelligent vehicle lamp assembly includes a sensor module, an instrument prompt module, a processing module, an intelligent lighting control module and an actuator;

[0015] Wherein, the sensor module is used to collect key data of vehicle status and external environment;

[0016] Instrument prompt module, used to display the status of the lighting system and driving-related information in real time;

[0017] The processing module is used to analyze key data of vehicle status and external environment and generate a light adjustment strategy;

[0018] An intelligent lighting control module is used to receive the lighting adjustment strategy and convert the lighting adjustment strategy into specific control instructions;

[0019] The actuator is used to execute the control instructions and adjust the brightness and direction of the vehicle lights based on the control instructions.

[0020] Optionally, the sensor module includes a camera image acquisition module and a laser radar scanning module;

[0021] A camera image acquisition module is used to capture image data of the road ahead, traffic signs, lane lines and other vehicles in real time through a pre-configured camera, and detect weather conditions to obtain visual information data;

[0022] The lidar scanning module is used to perform a 360-degree scan of the front environment through a pre-configured lidar, measure the distance to obstacles, the distance and shape of road edges and other surrounding objects, and obtain environmental perception data.

[0023] Optionally, the processing module includes:

[0024] Collect visual information data and environmental perception data, and convert the visual information data and environmental perception data into fuzzy data;

[0025] Preset corresponding fuzzy rules according to different driving scenarios, and store the fuzzy rules in a fuzzy rule library;

[0026] The fuzzified data is matched with the fuzzy rules in the fuzzy rule base to find the best rule that meets the current driving conditions and generate the corresponding headlight adjustment strategy.

[0027] Optionally, collecting visual information data and environmental perception data, and converting the visual information data and environmental perception data into fuzzy data includes:

[0028] Preprocess the collected visual information data and environmental perception data;

[0029] Extract feature data from the preprocessed visual information data and environmental perception data;

[0030] The extracted feature data are fuzzified and fuzzy sets suitable for fuzzy logic reasoning are generated.

[0031] Optionally, preprocessing the collected visual information data and environmental perception data includes:

[0032] Define a state vector containing visual information data and environmental perception data;

[0033] Initialize the state transfer matrix, observation matrix, and covariance matrix of process noise and observation noise;

[0034] Set the initial measurement values ​​of the visual information data and the environmental perception data as the initial estimate of the state vector, and initialize the error covariance matrix;

[0035] Use the state transfer matrix to predict the state value at the next moment based on the previous visual information data and environmental perception data; update the prediction error covariance matrix, consider the influence of process noise, and evaluate the error range between the visual information data and environmental perception data at the next moment;

[0036] Calculate the Kalman gain based on the prediction error covariance matrix and the sensor's observation noise covariance matrix;

[0037] The predicted state estimate is corrected using the current visual information data and environmental perception data, and the updated visual information data and environmental perception data estimates are generated in combination with the Kalman gain, and the error covariance matrix is ​​updated.

[0038] Optionally, the expression for calculating the Kalman gain is:

[0039] K k =P k - H k T (H k P k - H k T +R k ) -1 ;

[0040] In the formula, K k represents the Kalman gain, which determines the relative weight of the predicted value and the observed value, the predicted value and the measured value of the equilibrium state estimation;

[0041] P k - Represents the prediction error of visual information data and environmental perception data;

[0042] H k represents the observation matrix;

[0043] H k T represents the transpose of the observation matrix;

[0044] R k Represents measurement noise in visual information data and environmental perception data.

[0045] Optionally, extracting feature data from the preprocessed visual information data and environmental perception data includes:

[0046] Use edge detection algorithms to extract object edge contours from preprocessed visual information data and identify key visual elements;

[0047] Identify specific objects from key visual elements, analyze the brightness values ​​of specific objects and their surroundings, extract light intensity features, and evaluate the lighting conditions of the environment;

[0048] Use laser radar to calculate the distance between the obstacle in front and the vehicle and extract distance feature information;

[0049] By performing time series analysis on the distance feature information of multi-frame lidar, the speed and acceleration of specific objects are calculated, the dynamic features of specific objects are extracted, and the collision risk is predicted.

[0050] Optionally, the extracted feature data is fuzzified and a fuzzy set suitable for fuzzy logic reasoning is generated, including:

[0051] The extracted feature data are defined as fuzzy variables, and a value range is set for each fuzzy variable;

[0052] Create corresponding fuzzy sets for each fuzzy variable;

[0053] Define membership functions for each fuzzy set;

[0054] The corresponding fuzzy sets are input into the membership function, and the membership in each fuzzy set is calculated.

[0055] Optionally, the fuzzified data is matched with the fuzzy rules in the fuzzy rule base to find the best rule that meets the current driving conditions, and the corresponding headlight adjustment strategy is generated, including:

[0056] According to the membership degree of each fuzzy variable, the matching degree of each fuzzy rule is calculated using fuzzy reasoning method;

[0057] An aggregation method is used to combine the results of multiple partial matching rules to form a comprehensive reasoning output;

[0058] Defuzzification is performed based on the aggregated output results to generate a lighting adjustment strategy including turning on the high beam, switching to the low beam, or adjusting the brightness;

[0059] The generated headlight adjustment strategy is passed to the headlight control system.

[0060] Optionally, according to the membership degree of each fuzzy variable, the matching degree of each fuzzy rule is calculated using a fuzzy reasoning method, including:

[0061] For each fuzzy rule, the membership of each fuzzy variable in the antecedent condition of the fuzzy rule is calculated according to the membership value of the relevant fuzzy variable;

[0062] Apply the minimum operation to the membership value of the antecedent condition of each rule to obtain the overall matching degree of the current fuzzy rule;

[0063] According to the matching degree of each fuzzy rule, the fuzzy rule is applied to the corresponding output set to determine the matching degree of the rule.

[0064] Compared with the prior art, the present application has the following beneficial effects: the present invention combines cameras and lidars to acquire visual information and environmental data in all directions, including weather conditions, light intensity, obstacle locations, dynamic characteristics of objects, etc., and realizes high-precision, multi-dimensional perception of the environment. Even in complex or unfavorable environments (such as rainy and foggy weather, night driving), it can provide reliable environmental information support, thereby improving the robustness and accuracy of perception. Based on the fuzzy logic reasoning method, it can process complex and fuzzy data inputs, and flexibly generate headlight adjustment strategies according to different driving conditions (such as weather, light changes, traffic conditions). The fuzzy rule base can adapt to different environments to form efficient and accurate intelligent decision-making. , making the headlight adjustment more scene-adaptive, receiving the headlight adjustment strategy through the intelligent lighting control module, and responding to the data provided by the sensor in real time; for example, automatically switching to low beam when encountering oncoming vehicles or pedestrians, adjusting the brightness when visibility is reduced, etc., which helps to improve driving safety and comfort when driving at night. The actuator can accurately control the brightness and direction of the headlights, especially to provide better lighting effects in special road conditions such as turns and tunnels. The dynamic follow-up control of the headlights not only improves the driving field of vision, but also reduces the driving risks caused by poor light. The intelligent headlight control can automatically reduce the light intensity when high brightness is not required, avoiding unnecessary energy waste and improving the energy efficiency of the vehicle. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] The above characteristics, features and advantages of the present invention and their implementation methods and methods will become more clearly understood in conjunction with the following description of the embodiments, which are described in detail in conjunction with the accompanying drawings. Herein, a schematic diagram is shown:

[0066] Figure 1 It is a principle block diagram of an intelligent vehicle lamp assembly according to an embodiment of the present invention.

[0067] In the figure:

[0068] 1. Sensor module; 2. Instrument prompt module; 3. Processing module; 4. Intelligent lighting control module; 5. Actuator. DETAILED DESCRIPTION

[0069] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0070] According to an embodiment of the present invention, a smart vehicle lamp assembly is provided.

[0071] The present invention will now be further described with reference to the accompanying drawings and specific embodiments. Figure 1 As shown, according to an embodiment of the present invention, the intelligent vehicle lamp assembly includes a sensor module 1, an instrument prompt module 2, a processing module 3, an intelligent lighting control module 4 and an actuator 5;

[0072] Among them, the sensor module 1 is used to collect key data of the vehicle status and external environment.

[0073] Preferably, the sensor module includes a camera image acquisition module and a laser radar scanning module;

[0074] A camera image acquisition module is used to capture image data of the road ahead, traffic signs, lane lines and other vehicles in real time through a pre-configured camera, and detect weather conditions to obtain visual information data;

[0075] The lidar scanning module is used to perform a 360-degree scan of the front environment through a pre-configured lidar, measure the distance to obstacles, the distance and shape of road edges and other surrounding objects, and obtain environmental perception data.

[0076] It should be explained that the camera, as a visual sensor, captures real-time image data of the road ahead, including traffic signs, lane lines, other vehicles, pedestrians, etc. This information is crucial for vehicle positioning, path planning, and compliance with traffic regulations. The camera can also detect weather conditions in the environment, such as rain, haze, light changes, etc., to provide analysis of ambient light intensity and visibility.

[0077] LiDAR uses laser beams to perform 360-degree scanning to generate accurate environmental perception data. It can measure the distance and shape of surrounding objects (such as obstacles, road edges, pedestrians, etc.) and provide high-precision 3D spatial information. LiDAR's high resolution and long-distance measurement capabilities make it very suitable for detecting obstacles in dynamic environments, especially when the camera may be limited by light, weather and other factors. LiDAR can still provide reliable data support. Through continuous scanning, LiDAR can help the system predict the movement trajectory and speed of obstacles ahead, further enhancing the dynamic response capabilities of the autonomous driving system.

[0078] The instrument prompt module 2 is used to display the status of the vehicle lighting system and driving-related information in real time.

[0079] It should be explained that if it is a rainy and foggy day, when the visibility is lower than the safety threshold, the vehicle instrument will immediately prompt "The current visibility is low, the car light rain and fog mode has been turned on for you" or "The fog lights, tail lights and headlights of the whole vehicle have been turned on for you. Please drive carefully in rainy and foggy weather." If the device fails, it will prompt "The automatic adjustment device of the car light mode is faulty." When the ambient visibility is less than 200 meters, the fog lights and tail lights of the whole vehicle will be automatically turned on and the headlight rain and fog mode will be enabled. When the ambient visibility is less than 100 meters, the hazard warning flashers will also be automatically turned on. If it is normal weather, only the high and low beam lights, width lights and tail lights will be automatically turned on in places with low visibility. When there are pedestrians on the opposite side or when meeting other vehicles, the high beam lights will automatically switch back to low beam.

[0080] When passing through a bend or tunnel, the assembly can send relevant information to the smart driving chip through the laser radar and visual camera, so that the smart driving chip can calculate the relevant information processing method in advance. For example, when passing a bend, the curvature of the bend can be calculated in advance, so that the high and low beam lights can automatically perform corresponding light tracking steering in advance, providing a good lighting field for the bend; before passing through a tunnel, turn on the headlights, taillights, and width lights in advance.

[0081] The processing module 3 is used to analyze the key data of the vehicle status and external environment and generate a light adjustment strategy.

[0082] Preferably, the processing module includes:

[0083] Visual information data and environmental perception data are collected, and the visual information data and environmental perception data are converted into fuzzy data.

[0084] Preferably, collecting visual information data and environmental perception data, and converting the visual information data and environmental perception data into fuzzy data includes:

[0085] Preprocess the collected visual information data and environmental perception data;

[0086] Extract feature data from the preprocessed visual information data and environmental perception data;

[0087] The extracted feature data are fuzzified and fuzzy sets suitable for fuzzy logic reasoning are generated.

[0088] Preferably, preprocessing the collected visual information data and environmental perception data includes:

[0089] Define a state vector containing visual information data and environmental perception data;

[0090] Initialize the state transfer matrix, observation matrix, and covariance matrix of process noise and observation noise;

[0091] Set the initial measurement values ​​of the visual information data and the environmental perception data as the initial estimate of the state vector, and initialize the error covariance matrix;

[0092] Use the state transfer matrix to predict the state value at the next moment based on the previous visual information data and environmental perception data; update the prediction error covariance matrix, consider the influence of process noise, and evaluate the error range between the visual information data and environmental perception data at the next moment;

[0093] Calculate the Kalman gain based on the prediction error covariance matrix and the sensor's observation noise covariance matrix;

[0094] The predicted state estimate is corrected using the current visual information data and environmental perception data, and the updated visual information data and environmental perception data estimates are generated in combination with the Kalman gain, and the error covariance matrix is ​​updated.

[0095] Preferably, the expression for calculating the Kalman gain is:

[0096] K k =P k - H k T (H k P k - H k T +R k ) -1 ;

[0097] In the formula, K k represents the Kalman gain, which determines the relative weight of the predicted value and the observed value, the predicted value and the measured value of the equilibrium state estimation;

[0098] P k - Represents the prediction error of visual information data and environmental perception data;

[0099] H k represents the observation matrix;

[0100] H k T represents the transpose of the observation matrix;

[0101] R k Represents measurement noise in visual information data and environmental perception data.

[0102] Preferably, extracting feature data from the preprocessed visual information data and environmental perception data includes:

[0103] Use edge detection algorithms to extract object edge contours from preprocessed visual information data and identify key visual elements;

[0104] Identify specific objects from key visual elements, analyze the brightness values ​​of specific objects and their surroundings, extract light intensity features, and evaluate the lighting conditions of the environment;

[0105] Use laser radar to calculate the distance between the obstacle in front and the vehicle and extract distance feature information;

[0106] By performing time series analysis on the distance feature information of multi-frame lidar, the speed and acceleration of specific objects are calculated, the dynamic features of specific objects are extracted, and the collision risk is predicted.

[0107] Preferably, the extracted feature data is fuzzified and a fuzzy set suitable for fuzzy logic reasoning is generated, including:

[0108] The extracted feature data are defined as fuzzy variables, and a value range is set for each fuzzy variable;

[0109] Create corresponding fuzzy sets for each fuzzy variable;

[0110] Define membership functions for each fuzzy set;

[0111] The corresponding fuzzy sets are input into the membership function, and the membership in each fuzzy set is calculated.

[0112] It should be explained that visual information and environmental perception data are collected through cameras and lidar. These data are organized into a state vector, and the state transfer matrix and observation matrix are initialized to describe dynamic changes. At the same time, the covariance matrix of process and observation noise is set to cope with the uncertainty in the environment. Through the recursive algorithm of Kalman filtering, the system can predict the changes in visual and perceptual information based on historical data and dynamically update the error covariance matrix. This step preprocesses the data into reliable and traceable input, providing a more accurate basis for subsequent feature extraction.

[0113] For the preprocessed data, the visual module uses edge detection algorithms to extract object contours and identify key elements such as roads and traffic signs. For key visual elements, the system can further identify specific objects (such as vehicles and pedestrians) and evaluate the light intensity, providing a basis for adaptively adjusting the headlights to the ambient brightness. The data scanned by the lidar not only helps measure the distance to obstacles, but also derives dynamic features such as the speed and acceleration of obstacles through time series analysis, thereby predicting potential collision risks. These feature data are key inputs for the autonomous driving system to make environmental decisions.

[0114] The feature data is defined as a fuzzy variable, and a value range and fuzzy set are set for it. The fuzziness of the variable is described by a membership function (such as a triangular function or a trapezoidal function). In this way, the system can handle environmental features that contain uncertainty. For example, light intensity can be classified as "weak", "medium", or "strong". After the feature data is input into the membership function, the corresponding membership value is generated to form a fuzzy set, which lays the foundation for subsequent fuzzy logic reasoning. This fuzzification processing enables the system to match the current environmental conditions through a fuzzy rule base and generate decision outputs, such as switching high and low beam lights at the right time to adapt to changing road conditions.

[0115] Corresponding fuzzy rules are preset according to different driving scenarios, and the fuzzy rules are stored in the fuzzy rule base.

[0116] The fuzzified data is matched with the fuzzy rules in the fuzzy rule base to find the best rule that meets the current driving conditions and generate the corresponding headlight adjustment strategy.

[0117] Preferably, the fuzzified data is matched with the fuzzy rules in the fuzzy rule base to find the best rule that meets the current driving conditions, and the corresponding headlight adjustment strategy is generated, including:

[0118] According to the membership degree of each fuzzy variable, the matching degree of each fuzzy rule is calculated using fuzzy reasoning method;

[0119] An aggregation method is used to combine the results of multiple partial matching rules to form a comprehensive reasoning output;

[0120] Defuzzification is performed based on the aggregated output results to generate a lighting adjustment strategy including turning on the high beam, switching to the low beam, or adjusting the brightness;

[0121] The generated headlight adjustment strategy is passed to the headlight control system.

[0122] Preferably, according to the membership degree of each fuzzy variable, using the fuzzy reasoning method to calculate the matching degree of each fuzzy rule includes:

[0123] For each fuzzy rule, the membership of each fuzzy variable in the antecedent condition of the fuzzy rule is calculated according to the membership value of the relevant fuzzy variable;

[0124] Apply the minimum operation to the membership value of the antecedent condition of each rule to obtain the overall matching degree of the current fuzzy rule;

[0125] According to the matching degree of each fuzzy rule, the fuzzy rule is applied to the corresponding output set to determine the matching degree of the rule.

[0126] It should be explained that according to the membership of each fuzzy variable, the system will calculate the membership value for the antecedent condition of each fuzzy rule. Through the "minimum operation" method, the minimum membership value of each variable in the antecedent condition is selected as the matching degree of the rule. For example, if the membership of "weak light intensity" is 0.6, and the membership of "close distance" is 0.8, then the matching degree of this rule is min(0.6, 0.8) = 0.6.

[0127] When multiple partially matched rules are encountered, the results of these partially matched rules are combined through aggregation methods to form a comprehensive reasoning output. For example, if multiple rules simultaneously recommend adjusting the lights (such as switching high beams or adjusting brightness), the system will use the maximum membership principle or weighted average method to determine the final output result.

[0128] The output result after aggregation is fuzzy data, and the system needs to perform defuzzification to convert the fuzzy output into a specific headlight adjustment strategy. This can be achieved through the centroid method or the maximum membership method to generate clear headlight operation instructions, such as turning on the high beam, switching to the low beam, or adjusting the light brightness.

[0129] The intelligent lighting control module 4 is used to receive the vehicle lighting adjustment strategy and convert the vehicle lighting adjustment strategy into a specific control instruction.

[0130] It should be explained that the intelligent lighting control module 4 plays a key role in the automatic driving system. It is responsible for receiving the headlight adjustment strategies generated by fuzzy logic reasoning and converting these strategies into specific control instructions for the vehicle lighting system. It can make intelligent decisions based on real-time environmental perception data (such as light intensity, obstacle distance, road conditions, etc.), such as automatically turning on the high beam in low light conditions, or automatically switching to the low beam when there is an oncoming vehicle. In addition, the intelligent lighting control module can dynamically adjust the brightness of the headlights according to different driving scenarios to ensure the safety of the driver and surrounding traffic participants. This automated lighting adjustment not only improves the safety of night driving, but also optimizes energy consumption and driving comfort.

[0131] The actuator 5 is used to execute the control instruction and adjust the brightness and direction of the vehicle light based on the control instruction.

[0132] It should be explained that the actuator 5 is responsible for executing specific control instructions in the intelligent lighting control system and accurately adjusting the brightness and direction of the headlights according to these instructions. After receiving instructions from the intelligent lighting control module, the actuator can adjust the brightness level of the headlights by driving the motor or servo mechanism, such as increasing the brightness in a low-light environment, or reducing the brightness when the oncoming vehicle approaches to avoid glare. At the same time, the actuator can also adjust the illumination direction of the headlights to adapt to different road conditions and driving needs, such as automatically adjusting the direction of the lights when turning to ensure the best road lighting effect. The precise operation of the actuator ensures that the vehicle has a good field of view in a complex environment and improves driving safety and comfort.

[0133] To sum up, with the help of the above technical scheme of the present invention, the present invention combines cameras and lidars to obtain visual information and environmental data in all directions, covering weather conditions, light intensity, obstacle positions, dynamic characteristics of objects, etc., and realizes high-precision and multi-dimensional perception of the environment. Even in complex or unfavorable environments (such as rainy and foggy weather, night driving), it can provide reliable environmental information support, thereby improving the robustness and accuracy of perception; based on the fuzzy logic reasoning method, it can process complex and fuzzy data inputs, and flexibly generate headlight adjustment strategies according to different driving conditions (such as weather, light changes, traffic conditions). The fuzzy rule base can adapt to different environments to form efficient and accurate intelligent decisions, making the headlight adjustment more scene-adaptive, receiving the headlight adjustment strategy through the intelligent lighting control module, and responding to the data provided by the sensor in real time. For example, it automatically switches to low beam when encountering oncoming vehicles or pedestrians, and adjusts the brightness when visibility is reduced, which helps to improve driving safety and comfort when driving at night. The actuator can accurately control the brightness and direction of the lights, especially to provide better lighting effects in special road conditions such as turns and tunnels. The dynamic follow-up control of the lights not only improves the driving vision, but also reduces the driving risks caused by poor light. The intelligent headlight control can automatically reduce the light intensity when high brightness is not required, avoiding unnecessary energy waste and improving the energy efficiency of the vehicle.

[0134] Although the present invention has been disclosed as above with preferred embodiments, the embodiments are merely examples for the convenience of description and are not intended to limit the present invention. Those skilled in the art may make several changes and modifications without departing from the spirit and scope of the present invention. The scope of protection claimed by the present invention shall be based on the claims.

Claims

1. An intelligent vehicle lamp assembly, characterized in that: The intelligent vehicle lamp assembly includes a sensor module, an instrument prompt module, a processing module, an intelligent lighting control module and an actuator; Wherein, the sensor module is used to collect key data of vehicle status and external environment; The instrument prompt module is used to display the status of the vehicle lighting system and driving-related information in real time; The processing module is used to analyze key data of vehicle status and external environment and generate a vehicle light adjustment strategy; The intelligent lighting control module is used to receive the vehicle lighting adjustment strategy and convert the vehicle lighting adjustment strategy into a specific control instruction; The actuator is used to execute the control instruction and adjust the brightness and direction of the vehicle light based on the control instruction.

2. The intelligent vehicle lamp assembly according to claim 1, characterized in that: The sensor module includes a camera image acquisition module and a laser radar scanning module; The camera image acquisition module is used to capture the image data of the road ahead, traffic signs, lane lines and other vehicles in real time through a pre-configured camera, and detect weather conditions to obtain visual information data; The laser radar scanning module is used to perform a 360-degree scan of the front environment through a pre-configured laser radar, measure the distance to obstacles, the distance and shape of the road edge and other surrounding objects, and obtain environmental perception data.

3. The intelligent vehicle lamp assembly according to claim 1, characterized in that: The processing module analyzes key data of vehicle status and external environment and generates a headlight adjustment strategy including: Collect visual information data and environmental perception data, and convert the visual information data and environmental perception data into fuzzy data; Preset corresponding fuzzy rules according to different driving scenarios, and store the fuzzy rules in a fuzzy rule library; The fuzzified data is matched with the fuzzy rules in the fuzzy rule base to find the best rule that meets the current driving conditions and generate the corresponding headlight adjustment strategy.

4. The intelligent vehicle lamp assembly according to claim 3, characterized in that: The collecting of visual information data and environmental perception data and converting the visual information data and environmental perception data into fuzzy data includes: Preprocess the collected visual information data and environmental perception data; Extract feature data from the preprocessed visual information data and environmental perception data; The extracted feature data are fuzzified and fuzzy sets suitable for fuzzy logic reasoning are generated.

5. The intelligent vehicle lamp assembly according to claim 4, characterized in that: The preprocessing of the collected visual information data and environmental perception data includes: Define a state vector containing visual information data and environmental perception data; Initialize the state transfer matrix, observation matrix, and covariance matrix of process noise and observation noise; Set the initial measurement values ​​of the visual information data and the environmental perception data as the initial estimate of the state vector, and initialize the error covariance matrix; Use the state transfer matrix to predict the state value at the next moment based on the previous visual information data and environmental perception data; update the prediction error covariance matrix, consider the influence of process noise, and evaluate the error range between the visual information data and environmental perception data at the next moment; Calculate the Kalman gain based on the prediction error covariance matrix and the sensor's observation noise covariance matrix; The predicted state estimate is corrected using the current visual information data and environmental perception data, and the updated visual information data and environmental perception data estimates are generated in combination with the Kalman gain, and the error covariance matrix is ​​updated.

6. The intelligent vehicle lamp assembly according to claim 5, characterized in that: The expression for calculating the Kalman gain is: K k =P k - H k T (H k P k - H k T +R k ) -1 ; In the formula, K k represents the Kalman gain, which determines the relative weight of the predicted value and the observed value, the predicted value and the measured value of the equilibrium state estimation; P k - Represents the prediction error of visual information data and environmental perception data; H k represents the observation matrix; H k T represents the transpose of the observation matrix; R k Represents measurement noise in visual information data and environmental perception data.

7. The intelligent vehicle lamp assembly according to claim 6, characterized in that: The extracting feature data from the preprocessed visual information data and environmental perception data includes: Use edge detection algorithms to extract object edge contours from preprocessed visual information data and identify key visual elements; Identify specific objects from key visual elements, analyze the brightness values ​​of specific objects and their surroundings, extract light intensity features, and evaluate the lighting conditions of the environment; Use laser radar to calculate the distance between the obstacle in front and the vehicle and extract distance feature information; By performing time series analysis on the distance feature information of multi-frame lidar, the speed and acceleration of specific objects are calculated, the dynamic features of specific objects are extracted, and the collision risk is predicted.

8. The intelligent vehicle lamp assembly according to claim 7, characterized in that: The step of fuzzifying the extracted feature data and generating a fuzzy set suitable for fuzzy logic reasoning includes: The extracted feature data are defined as fuzzy variables, and a value range is set for each fuzzy variable; Create corresponding fuzzy sets for each fuzzy variable; Define membership functions for each fuzzy set; The corresponding fuzzy sets are input into the membership function, and the membership in each fuzzy set is calculated.

9. The intelligent vehicle lamp assembly according to claim 3, characterized in that: The fuzzified data is matched with the fuzzy rules in the fuzzy rule base to find the best rule that meets the current driving conditions and generate the corresponding headlight adjustment strategy, including: According to the membership degree of each fuzzy variable, the matching degree of each fuzzy rule is calculated using fuzzy reasoning method; An aggregation method is used to combine the results of multiple partial matching rules to form a comprehensive reasoning output; Defuzzification is performed based on the aggregated output results to generate a lighting adjustment strategy including turning on the high beam, switching to the low beam, or adjusting the brightness; The generated headlight adjustment strategy is passed to the headlight control system.

10. The intelligent vehicle lamp assembly according to claim 9, characterized in that: The method of calculating the matching degree of each fuzzy rule using the fuzzy reasoning method according to the membership degree of each fuzzy variable includes: For each fuzzy rule, the membership of each fuzzy variable in the antecedent condition of the fuzzy rule is calculated according to the membership value of the relevant fuzzy variable; Apply the minimum operation to the membership value of the antecedent condition of each rule to obtain the overall matching degree of the current fuzzy rule; According to the matching degree of each fuzzy rule, the fuzzy rule is applied to the corresponding output set to determine the matching degree of the rule.

Images