A laser radar-based intelligent control method and system for vehicle lighting

Abstract

This invention discloses an intelligent control method and system for vehicle lighting based on lidar, relating to the field of vehicle lighting. It includes: a data acquisition module for acquiring three-dimensional point cloud data of the road ahead detected by the vehicle-mounted lidar; and a processing module for receiving the three-dimensional point cloud data, performing noise reduction, coordinate calibration, and multi-dimensional data fusion processing on the three-dimensional point cloud data to generate a standardized road environment perception dataset. This invention uses lidar to detect three-dimensional point cloud data of the road ahead, processes the data to generate a standardized road environment perception dataset, accurately identifies road types, distribution of traffic participants, and their relative motion relationships, and dynamically outputs control parameters for the shape, brightness, and projection angle of the laser headlight beam to drive the actuator to adjust the lighting state.

Description

Technical Field

[0001] This invention relates to the field of vehicle lighting technology, specifically to a vehicle lighting intelligent control method and system based on lidar. Background Technology

[0002] Vehicle lighting is a key component for ensuring driving safety, encompassing headlights (high / low beam), turn signals, brake lights, and side marker lights.

[0003] Patent application number 202411136496.5 discloses an intelligent control method for LED vehicle lights based on environmental perception, including: acquiring environmental state data during vehicle operation at monitoring time points; inputting the current environmental state data into a vehicle light control strategy output model for processing, and outputting an LED vehicle light control strategy; intelligently controlling the LED vehicle lights on the vehicle through the LED vehicle light control strategy; and also including real-time training of the vehicle light control strategy output model, the specific steps of which are as follows: at the monitoring time point, inputting the environmental state data into a target vehicle light control strategy output model for processing, outputting a target vehicle light control strategy, and then inputting the environmental state data into the target vehicle light control strategy output model. The process involves splicing data to construct value assessment data, feeding this data into a value assessment network model for processing, outputting policy value data, calculating the policy gradient value of the policy value data with respect to the target headlight control policy, and then using the gradient ascent method to adjust the parameters of the headlight control policy output model based on the policy gradient value, thereby achieving real-time training of the headlight control policy output model. Simultaneously, it also includes real-time training of the value assessment network model. This application aims to address the problem that "traditional LED headlight systems often lack the ability to perceive the surrounding environment and driving conditions, which may prevent them from making optimal lighting adjustments in complex driving scenarios, requiring manual adjustments by the driver and affecting the driving experience."

[0004] However, existing technologies for intelligent control of vehicle headlights are mostly limited to intelligent control of opening and closing, without any intelligent control measures for the beam shape, brightness, and projection angle of the headlights.

[0005] To address this, we propose a method and system for intelligent control of vehicle lighting based on lidar. Summary of the Invention

[0006] In view of the above-mentioned shortcomings of the existing technology, the present invention provides a vehicle lighting intelligent control method and system based on lidar, which can effectively solve the problems of the existing technology.

[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions;

[0008] This invention discloses an intelligent vehicle lighting control system based on lidar, comprising:

[0009] The system comprises the following modules: an acquisition module for collecting 3D point cloud data of the road ahead detected by the vehicle-mounted LiDAR; a processing module for receiving the 3D point cloud data, performing noise reduction, coordinate calibration, and multi-dimensional data fusion to generate a standardized road environment perception dataset; a recognition module for acquiring the standardized road environment perception dataset, extracting scene features from the dataset, classifying them, and identifying the current road type, distribution of traffic participants, and their relative motion relationships; a calculation module for receiving the recognition results from the recognition module, and outputting dynamic control parameters for the laser headlight beam shape, brightness, and projection angle based on the recognition results and preset lighting control logic; a control module for driving the actuators of the vehicle-mounted laser headlights to adjust the beam shape, brightness, and projection angle deflection according to the dynamic control parameters; and an optimization module for collecting real-time operating status data from the control module and environmental feedback data from secondary LiDAR detection, and iteratively optimizing the dynamic control parameters output by the calculation module.

[0010] The three-dimensional point cloud data of the road ahead includes the road's geometric contour, the spatial coordinates of dynamic targets, and motion state parameters.

[0011] Furthermore, the dynamic target motion state parameters acquired by the acquisition module are obtained through inter-frame point cloud matching calculations using the lidar:

[0012] Inter-frame point cloud matching: Acquire two consecutive frames of point cloud data from the LiDAR, denoted as frame t and frame (t+1), with a frame interval of 0.1s. For a dynamic target point cloud cluster A in frame t, find the corresponding point cloud cluster A′ in frame (t+1) using Euclidean distance matching, and determine the center point of A. The center point of A′ ;

[0013] Calculation of the velocity of a dynamic target: , Indicates the frame interval time;

[0014] Calculation of motion acceleration: Obtain the dynamic target point cloud cluster in frame t-1. center point Then calculate the motion velocity of the t-th frame, denoted as . Then the acceleration is The unit is ;

[0015] If the matching degree of two point cloud clusters is less than 0.8, the motion state parameters of the previous frame are used, and the matching degree is determined by the ratio of the number of overlapping point clouds to the total number.

[0016] Furthermore, when the processing module performs noise reduction processing on the 3D point cloud data, it includes the following steps:

[0017] Traversing each point in the 3D point cloud data Let i = 1, 2, ..., N, where N is the total number of point clouds, and each... Let r be the center of a sphere, and define a spherical neighborhood with radius r. Count the number of other point clouds contained within this spherical neighborhood. ;

[0018] Calculate the average number of neighborhoods for all point clouds. Simultaneously set the noise reduction threshold , This represents the density adjustment coefficient, with a value ranging from 0.3 to 0.6. When the lidar detection distance is long, resulting in a sparse overall point cloud distribution, the density adjustment coefficient... The value should be as large as possible; when the lidar detection distance is short, resulting in a dense overall distribution of the point cloud, the density adjustment coefficient should be increased. The smaller the value, the better;

[0019] Number of neighboring point clouds < point Points identified as noise are removed, while others are retained. ≥ The point cloud is used to obtain denoised point cloud data;

[0020] The radius r of the spherical neighborhood is dynamically determined based on the real-time detection distance d of the lidar. The calculation formula is r = d × 0.02, where d ranges from 0 to 200 meters, and 0.02 is an initial preset constant term, with a value range of 0.01 to 0.05.

[0021] Furthermore, when the processing module performs coordinate calibration on the 3D point cloud data, the processing objective is to complete a three-level coordinate transformation between the lidar coordinate system, the vehicle body coordinate system, and the world coordinate system. The process is as follows:

[0022] Transformation from LiDAR coordinate system to vehicle coordinate system:

[0023] Point cloud coordinates in the lidar coordinate system Convert to vehicle coordinate system The conversion formula is: , This represents the 3×3 rotation matrix from the lidar coordinate system to the vehicle coordinate system. This represents the 3×1 translation vector from the lidar coordinate system to the vehicle coordinate system;

[0024] It is determined by the pitch angle α, roll angle β, and heading angle γ when the lidar is installed;

[0025] ;

[0026] The three-dimensional offset of the lidar installation position relative to the vehicle's origin Sure,

[0027] Right now ;

[0028] Transformation from vehicle coordinate system to world coordinate system:

[0029] coordinates in the vehicle coordinate system Convert to world coordinate system coordinates The conversion formula is: , This represents a 3×3 rotation matrix from the vehicle coordinate system to the world coordinate system. This represents a 3×1 translation vector from the vehicle coordinate system to the world coordinate system;

[0030] , Indicates the vehicle's current GPS heading angle;

[0031] , Indicates the vehicle's current GPS coordinates;

[0032] When correcting for deviations in the transformed W-system coordinates, the following applies:

[0033] ;

[0034] In the formula: These are the corrected coordinates; , and The calculation logic is consistent;

[0035] The coordinates of the j-th frame of 3D point cloud data after transformation from the LiDAR coordinate system to the vehicle coordinate system and then to the world coordinate system are the x-axis coordinates in the world coordinate system, where j is the frame number. Similarly.

[0036] Furthermore, when the recognition module extracts scene features and performs classification, it executes sequential operations of extraction, fusion, and determination:

[0037] Extracting core scene features:

[0038] Linear fitting is performed on the road edge point cloud in the standardized road environment perception dataset to obtain the slope of the road edge. The absolute value of the slope is taken and denoted as the road edge feature S1.

[0039] For each traffic participant, identify the minimum bounding box of the point cloud cluster, and denote the volume of the bounding box as the traffic participant volume feature S2;

[0040] Calculate the relative speed of traffic participants relative to vehicles, and take the absolute value of the relative speed, which is denoted as the relative motion speed characteristic S3;

[0041] Feature-weighted fusion: , As weight, The sum of the three is 1, and all three are positive numbers, with values ​​ranging from [0.2, 0.3], [0.4, 0.5], and [0.2, 0.3] respectively.

[0042] Scene classification determination:

[0043] When F < F1, it is determined to be: urban branch road with no large traffic participants;

[0044] When F1≤F<F2, it is determined that: it is an urban arterial road and there are small traffic participants;

[0045] When F≥F2, it is determined that: it is a highway and there are large traffic participants;

[0046] Where [F1, F2] are preset thresholds, and the units are defined as follows: .

[0047] Furthermore, when the calculation module outputs dynamic control parameters based on the recognition results, the applied lighting control logic is as follows:

[0048] Beam shape control includes control of the horizontal width W and the vertical height H:

[0049] The horizontal width W is determined based on the standard lane width for the current road type:

[0050] When on urban side roads, , This indicates the standard lane width for the current road type, and 0.5 indicates the reserved safety distance from the roadside. The unit is meters (m).

[0051] When on the main urban road, On the highway, ;

[0052] The vertical height H is determined based on the maximum height of traffic participants:

[0053] , H represents the maximum height of traffic participants, 0.3 represents the reserved height safety distance, and H∈[1.8,3.0]m;

[0054] Brightness control includes determining the reference brightness and controlling the actual brightness:

[0055] The ambient light intensity I is set as follows: when I < 200 lux, the reference brightness is 8000 cd, corresponding to nighttime; when 200 lux ≤ I < 500 lux, the reference brightness is 5000 cd, corresponding to dusk or dawn; when I ≥ 500 lux, the reference brightness is 2000 cd, corresponding to daytime.

[0056] Based on the relative distance between traffic participants and vehicles correction, , Indicates actual brightness. Indicates the reference brightness. This represents the baseline distance, initially set to 50m, and ∈[1000,10000]cd;

[0057] Projection angle control includes the control of horizontal deflection angle and vertical deflection angle:

[0058] Horizontal offset of traffic participants Sure, , Indicates the horizontal deflection angle;

[0059] The vertical height h of traffic participants and Sure, , Indicates the vertical deflection angle, and Constraints Within the range.

[0060] Furthermore, when the control module drives the actuator to adjust the projection angle deflection, it simultaneously employs PID closed-loop control logic to control the adjustment accuracy. The process includes:

[0061] The target projection angle is set, and its source is the calculation module, which collects the current actual angle of the actuator and calculates the angle deviation.

[0062] The output control quantity of the PID controller is obtained by the following formula:

[0063] ;

[0064] In the formula: This refers to the control quantity output by the PID controller to the actuator for projecting the laser headlight at the current control cycle time t. This is the proportionality coefficient; This refers to the angle deviation at the current control cycle time t. The integral coefficient; For any point in time The corresponding angular deviation; These are the differential coefficients; This represents the integral value of the angular deviation over the time interval [0, t]. express The derivative with respect to time t;

[0065] The actuator is based on the control quantity The drive angle adjustment motor has a control cycle duration of 0.02 seconds, until... < Stop adjusting at that time;

[0066] in, ∈[0.8,1.2], the larger the value is when the actual value of the projection angle of the vehicle laser headlight deviates from the target value and the deviation needs to be reduced quickly, the smaller the value is when the actual value of the projection angle deviates from the target value and the overshoot needs to be avoided; ∈[0.1,0.3], when the static deviation between the actual projection angle and the target value lasts for a long time and the deviation amplitude is large, The larger the value, the shorter the duration and the smaller the deviation amplitude between the actual projection angle and the target value. The smaller the value; The value is larger when the rate of change of the projection angle deviation of the vehicle laser headlight is fast and it is necessary to suppress the oscillation during the angle adjustment process, and smaller when the rate of change of the projection angle deviation is slow and it is necessary to ensure the stability of the adjustment process.

[0067] Furthermore, when the optimization module iteratively optimizes the dynamic control parameters, it uses the deviation between the actual lighting parameters output by the actuator and the dynamic control parameters output by the calculation module as the core optimization index. The process includes:

[0068] Calculate the optimization deviation E to quantify the difference between the actual lighting effect and the theoretical control target:

[0069] ;

[0070] In the formula: The actual output brightness, actual projection angle, and actual beam width of the actuator used for secondary detection and inversion by lidar; The calculation module outputs dynamic control parameters for brightness, angle, and width.

[0071] Set an optimization coefficient λ, which is based on the rate of environmental change. Dynamic adjustment;

[0072] , This represents the current light intensity. The light intensity for the previous optimization cycle; To optimize the cycle, =1s;

[0073] when When λ < 50 lux / s, take 0.1; when 50 lux / s ≤ When λ is <100 lux / s, it is taken as 0.3. When the value is ≥100 lux / s, λ is taken as 0;

[0074] When the value is ≥100 lux / s, λ is taken as 0;

[0075] The optimized control parameters are:

[0076] ;

[0077] in, To optimize brightness, projection angle, and beam width, The feedback is sent to the calculation module to replace the original parameters, completing one iteration.

[0078] Furthermore, the acquisition module is interconnected with the processing module and the identification module via a wireless network, the identification module is interconnected with the calculation module via a wireless network, the calculation module is interconnected with the control module via a wireless network, and the acquisition module and the control module are interconnected with the optimization module via a wireless network.

[0079] On the other hand, a method for intelligent control of vehicle lighting based on lidar includes:

[0080] The system collects 3D point cloud data of the road ahead detected by the vehicle-mounted LiDAR. The collected 3D point cloud data undergoes denoising, coordinate calibration, and multi-dimensional data fusion processing to generate a standardized road environment perception dataset. Coordinate calibration requires a three-level transformation between the LiDAR coordinate system, the vehicle's coordinate system, and the world coordinate system. Scene features are extracted from the standardized road environment perception dataset and weighted fusion classification is performed to identify the current road type, the distribution of traffic participants, and their relative motion relationships. Based on the identified road type, traffic participant distribution, and relative motion relationships, combined with preset lighting control logic, dynamic control parameters for the laser headlight beam shape, brightness, and projection angle are output. The actuator driving the vehicle-mounted laser headlight adjusts the beam shape, brightness, and projection angle deflection according to the output dynamic control parameters. PID closed-loop control logic is applied simultaneously during the projection angle deflection adjustment process. Real-time operating status data of the actuator and environmental feedback data from secondary LiDAR detection are collected. Using the deviation between the actual lighting parameters and the dynamic control parameters as the core indicator, the dynamic control parameters are iteratively optimized and fed back to the parameter calculation stage to replace the original parameters.

[0081] Compared with the known prior art, the technical solution provided by this invention has the following beneficial effects:

[0082] This invention provides a vehicle lighting intelligent control method and system based on lidar. During execution, the method and system detect three-dimensional point cloud data of the road ahead using lidar, process the data to generate a standardized road environment perception dataset, accurately identify road type, distribution of traffic participants, and their relative motion relationships, and dynamically output control parameters for the shape, brightness, and projection angle of the laser headlight beam. This drives the actuator to adjust the lighting state. Furthermore, the control parameters are iteratively optimized by combining real-time operating status and secondary environmental feedback, enabling adaptation to different road scenarios and traffic conditions. This avoids excessively strong or weak lighting, reduces visual interference to other traffic participants, improves driving visibility at night and under complex lighting conditions, and ensures driving safety. Simultaneously, PID closed-loop control ensures lighting adjustment accuracy, and parameters are optimized based on the rate of environmental change, further enhancing the adaptability and stability of lighting control and reducing unnecessary energy consumption. Attached Figure Description

[0083] 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.

[0084] Figure 1This is a schematic diagram of a vehicle lighting intelligent control system based on lidar;

[0085] Figure 2 This is a flowchart illustrating a method for intelligent control of vehicle lighting based on lidar. Detailed Implementation

[0086] 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.

[0087] The present invention will be further described below with reference to embodiments. Example 1

[0088] This embodiment presents a vehicle lighting intelligent control system based on lidar, such as... Figure 1 As shown, it includes:

[0089] The acquisition module is used to acquire three-dimensional point cloud data of the road ahead detected by the vehicle-mounted lidar;

[0090] The dynamic target motion state parameters acquired in the acquisition module are obtained through inter-frame point cloud matching calculations using the lidar:

[0091] Inter-frame point cloud matching: Acquire two consecutive frames of point cloud data from the LiDAR, denoted as frame t and frame (t+1), with a frame interval of 0.1s. For a dynamic target point cloud cluster A in frame t, find the corresponding point cloud cluster A′ in frame (t+1) using Euclidean distance matching, and determine the center point of A. The center point of A′ ;

[0092] Calculation of the velocity of a dynamic target: , Indicates the frame interval time;

[0093] Calculation of motion acceleration: Obtain the dynamic target point cloud cluster in frame t-1. center point Then calculate the motion velocity of the t-th frame, denoted as . Then the acceleration is The unit is ;

[0094] If the matching degree of two point cloud clusters is less than 0.8, the motion state parameters of the previous frame are used, and the matching degree is determined by the ratio of the number of overlapping point clouds to the total number.

[0095] The processing module is used to receive 3D point cloud data, perform noise reduction, coordinate calibration and multi-dimensional data fusion processing on the 3D point cloud data, and generate a standardized road environment perception dataset.

[0096] When the processing module performs noise reduction on 3D point cloud data, it includes the following steps:

[0097] Traversing each point in the 3D point cloud data Let i = 1, 2, ..., N, where N is the total number of point clouds, and each... Let r be the center of a sphere, and define a spherical neighborhood with radius r. Count the number of other point clouds contained within this spherical neighborhood. ;

[0098] Calculate the average number of neighborhoods for all point clouds. Simultaneously set the noise reduction threshold , This represents the density adjustment coefficient, with a value ranging from 0.3 to 0.6. When the lidar detection distance is long, resulting in a sparse overall point cloud distribution, the density adjustment coefficient... The value should be as large as possible; when the lidar detection distance is short, resulting in a dense overall distribution of the point cloud, the density adjustment coefficient should be increased. The smaller the value, the better;

[0099] Number of neighboring point clouds < point Points identified as noise are removed, while others are retained. ≥ The point cloud is used to obtain denoised point cloud data;

[0100] Among them, the radius r of the spherical neighborhood is dynamically determined according to the real-time detection distance d of the lidar. The calculation formula is r=d×0.02, where the value of d ranges from 0 to 200 meters, and 0.02 is an initial preset constant term, with the value of the preset constant term ranging from 0.01 to 0.05.

[0101] When the processing module performs coordinate calibration on the 3D point cloud data, the processing objective is to complete a three-level coordinate transformation between the lidar coordinate system, the vehicle body coordinate system, and the world coordinate system. The process is as follows:

[0102] Transformation from LiDAR coordinate system to vehicle coordinate system:

[0103] Point cloud coordinates in the lidar coordinate system Convert to vehicle coordinate system The conversion formula is: , This represents the 3×3 rotation matrix from the lidar coordinate system to the vehicle coordinate system. This represents the 3×1 translation vector from the lidar coordinate system to the vehicle coordinate system;

[0104] It is determined by the pitch angle α, roll angle β, and heading angle γ when the lidar is installed;

[0105] ;

[0106] The three-dimensional offset of the lidar installation position relative to the vehicle's origin Determined, that is ;

[0107] Transformation from vehicle coordinate system to world coordinate system:

[0108] coordinates in the vehicle coordinate system Convert to world coordinate system coordinates The conversion formula is: , This represents a 3×3 rotation matrix from the vehicle coordinate system to the world coordinate system. This represents a 3×1 translation vector from the vehicle coordinate system to the world coordinate system;

[0109] , Indicates the vehicle's current GPS heading angle;

[0110] , Indicates the vehicle's current GPS coordinates;

[0111] When correcting for deviations in the transformed W-system coordinates, the following applies:

[0112] ;

[0113] In the formula: These are the corrected coordinates; , and The calculation logic is consistent;

[0114] The coordinates of the j-th frame of 3D point cloud data after transformation from the LiDAR coordinate system to the vehicle coordinate system and then to the world coordinate system are the x-axis coordinates in the world coordinate system, where j is the frame number. Similarly;

[0115] In the above setup, the transformation formula from the lidar coordinate system to the vehicle coordinate system is achieved by combining a 3×3 rotation matrix and a 3×1 translation vector. The rotation matrix is ​​determined by the pitch angle α, roll angle β, and yaw angle γ of the lidar during installation, which can offset the influence of installation angle deviation on the coordinates. The translation vector is determined by the three-dimensional offset of the lidar relative to the vehicle origin, ensuring that the coordinate transformation can accurately reflect the relative positional relationship between the lidar and the vehicle.

[0116] The transformation formula from the vehicle coordinate system to the world coordinate system uses the vehicle's current GPS heading angle to construct a rotation matrix and the current GPS coordinates to construct a translation vector, thereby aligning the vehicle coordinates with the global coordinates. The deviation correction formula corrects the deviation by using the difference between the average value of the coordinates after multiple frames of transformation and the transformed coordinates of the current frame, effectively reducing coordinate deviations caused by GPS positioning errors and installation deviations, and ensuring the accuracy of 3D point cloud data in a unified coordinate system.

[0117] The identification module is used to acquire a standardized road environment perception dataset, extract scene features from the dataset and classify them, and identify the current road type, distribution of traffic participants and their relative motion relationships.

[0118] When the recognition module extracts scene features and performs classification, it executes the sequential operations of extraction, fusion, and judgment:

[0119] Extracting core scene features:

[0120] Linear fitting is performed on the road edge point cloud in the standardized road environment perception dataset to obtain the slope of the road edge. The absolute value of the slope is taken and denoted as the road edge feature S1.

[0121] For each traffic participant, identify the minimum bounding box of the point cloud cluster, and denote the volume of the bounding box as the traffic participant volume feature S2;

[0122] Calculate the relative speed of traffic participants relative to vehicles, and take the absolute value of the relative speed, which is denoted as the relative motion speed characteristic S3;

[0123] Feature-weighted fusion: , As weight, The sum of the three is 1, and all three are positive numbers, with values ​​ranging from [0.2, 0.3], [0.4, 0.5], and [0.2, 0.3] respectively.

[0124] Scene classification determination:

[0125] When F < F1, it is determined to be: urban branch road with no large traffic participants;

[0126] When F1≤F<F2, it is determined that: it is an urban arterial road and there are small traffic participants;

[0127] When F≥F2, it is determined that: it is a highway and there are large traffic participants;

[0128] Where [F1, F2] are preset thresholds, and the units are defined as follows: ;

[0129] The calculation module receives the recognition results from the recognition module and, based on the recognition results and the preset lighting control logic, outputs dynamic control parameters for the shape, brightness, and projection angle of the laser headlight beam.

[0130] When the calculation module outputs dynamic control parameters based on the recognition results, the applied lighting control logic is as follows:

[0131] Beam shape control includes control of the horizontal width W and the vertical height H:

[0132] The horizontal width W is determined based on the standard lane width for the current road type:

[0133] When on urban side roads, , This indicates the standard lane width for the current road type, and 0.5 indicates the reserved safety distance from the roadside. The unit is meters (m).

[0134] This indicates the standard lane width for the current road type, and 0.5 indicates the reserved safety distance from the roadside. The unit is meters (m).

[0135] When on the main urban road, On the highway, ;

[0136] The vertical height H is determined based on the maximum height of traffic participants:

[0137] , H represents the maximum height of traffic participants, 0.3 represents the reserved height safety distance, and H∈[1.8,3.0]m;

[0138] Brightness control includes determining the reference brightness and controlling the actual brightness:

[0139] The ambient light intensity I is set as follows: when I < 200 lux, the reference brightness is 8000 cd, corresponding to nighttime; when 200 lux ≤ I < 500 lux, the reference brightness is 5000 cd, corresponding to dusk or dawn; when I ≥ 500 lux, the reference brightness is 2000 cd, corresponding to daytime.

[0140] Based on the relative distance between traffic participants and vehicles correction, , Indicates actual brightness. Indicates the reference brightness. This represents the baseline distance, initially set to 50m, and ∈[1000,10000]cd;

[0141] Projection angle control includes the control of horizontal deflection angle and vertical deflection angle:

[0142] Horizontal offset of traffic participants Sure, , Indicates the vertical deflection angle, and Constraints Within the range;

[0143] The control module is used to drive the actuator of the vehicle laser headlight to adjust the beam shape, brightness and projection angle according to the dynamic control parameters of the laser headlight beam shape, brightness and projection angle.

[0144] When the control module drives the actuator to adjust the projection angle deflection, it simultaneously uses PID closed-loop control logic to control the adjustment accuracy. The process includes:

[0145] The target projection angle is set, and its source is the calculation module, which collects the current actual angle of the actuator and calculates the angle deviation.

[0146] The output control quantity of the PID controller is obtained by the following formula:

[0147] ;

[0148] In the formula: This refers to the control quantity output by the PID controller to the actuator for projecting the laser headlight at the current control cycle time t. This is the proportionality coefficient; This refers to the angle deviation at the current control cycle time t. The integral coefficient; For any point in time The corresponding angular deviation; These are the differential coefficients; This represents the integral value of the angular deviation over the time interval [0, t]. express The derivative with respect to time t;

[0149] The actuator is based on the control quantity The drive angle adjustment motor has a control cycle duration of 0.02 seconds, until... < Stop adjusting at that time;

[0150] in, ∈[0.8,1.2], the larger the value is when the actual value of the projection angle of the vehicle laser headlight deviates from the target value and the deviation needs to be reduced quickly, the smaller the value is when the actual value of the projection angle deviates from the target value and the overshoot needs to be avoided; ∈[0.1,0.3], when the static deviation between the actual projection angle and the target value lasts for a long time and the deviation amplitude is large, The larger the value, the shorter the duration and the smaller the deviation amplitude between the actual projection angle and the target value. The smaller the value; ∈[0.05,0.15], the larger the value is when the rate of change of the projection angle deviation of the vehicle laser headlight is fast and it is necessary to suppress the oscillation during the angle adjustment process, and the smaller the value is when the rate of change of the projection angle deviation is slow and it is necessary to ensure the stability of the adjustment process.

[0151] The above formula combines proportional, integral, and derivative terms. The proportional coefficient ranges from 0.8 to 1.2. A larger value can quickly reduce the deviation when it is large, while a smaller value can avoid overshoot when the deviation is small, ensuring the response speed and stability of the angle adjustment. The integral coefficient ranges from 0.1 to 0.3. A larger value can effectively eliminate accumulated errors when the static deviation is long and large, while a smaller value can avoid integral saturation when the deviation is small, ensuring the accuracy of angle control. The derivative coefficient ranges from 0.05 to 0.15. A larger value can suppress oscillations during the adjustment process when the deviation changes rapidly, while a smaller value can ensure the smoothness of the adjustment when the change is slow. The 0.02-second control cycle enables high-frequency adjustments until the angle deviation is less than the set threshold, ensuring that the projection angle of the vehicle laser headlight can quickly and accurately reach the target value.

[0152] The optimization module is used to collect real-time operating status data of the control module and environmental feedback data from the secondary detection of the lidar, and to iteratively optimize the dynamic control parameters output by the calculation module.

[0153] When the optimization module iteratively optimizes the dynamic control parameters, it uses the deviation between the actual lighting parameters output by the actuator and the dynamic control parameters output by the calculation module as the core optimization index. The process includes:

[0154] Calculate the optimization deviation E to quantify the difference between the actual lighting effect and the theoretical control target:

[0155] ;

[0156] In the formula: The actual output brightness, actual projection angle, and actual beam width of the actuator used for secondary detection and inversion by lidar; The calculation module outputs dynamic control parameters for brightness, angle, and width.

[0157] The above formula is calculated by summing the absolute values ​​of the actual brightness, projection angle, and beam width retrieved by the secondary detection of the lidar and the theoretical parameters output by the calculation module. It can comprehensively quantify the difference between the actual lighting effect and the theoretical control target in key dimensions, providing a clear direction for optimization.

[0158] Set an optimization coefficient λ, which is based on the rate of environmental change. Dynamic adjustment;

[0159] , This represents the current light intensity. The light intensity for the previous optimization cycle; To optimize the cycle, =1s;

[0160] when When λ < 50 lux / s, take 0.1; when 50 lux / s ≤ When λ is <100 lux / s, it is taken as 0.3. When the value is ≥100 lux / s, λ is taken as 0;

[0161] In the above formula, the optimization coefficient λ is calculated by dividing the difference in light intensity between the current and previous optimization cycles by 1 second optimization cycle. When the light intensity change rate is less than 50 lux / s, it is 0.1; when it is 50~100 lux / s, it is 0.3; and when it is greater than or equal to 100 lux / s, it is 0. It can dynamically adjust the optimization intensity according to the speed of change of ambient light. When the light intensity changes quickly, the optimization amplitude is increased to quickly adapt to the environment; when the change is slow, the amplitude is decreased to maintain stable control.

[0162] The optimized control parameters are:

[0163] ;

[0164] in, To optimize brightness, projection angle, and beam width, The feedback is sent to the calculation module to replace the original parameters, completing one iteration;

[0165] Among them, the three-dimensional point cloud data of the road ahead includes the road geometric contour, the spatial coordinates of dynamic targets, and motion state parameters;

[0166] The acquisition module is interconnected with the processing module and the identification module via a wireless network. The identification module is interconnected with the calculation module via a wireless network. The calculation module is interconnected with the control module via a wireless network. The acquisition module and the control module are interconnected with the optimization module via a wireless network.

[0167] In this embodiment, the acquisition module collects 3D point cloud data of the road ahead detected by the vehicle-mounted LiDAR. The processing module receives the 3D point cloud data, performs noise reduction, coordinate calibration, and multi-dimensional data fusion processing on the 3D point cloud data to generate a standardized road environment perception dataset. The recognition module then acquires the standardized road environment perception dataset, extracts scene features from the dataset, classifies them, and identifies the current road type, distribution of traffic participants, and relative motion relationships. The calculation module receives the recognition results from the recognition module and, based on the recognition results and preset lighting control logic, outputs dynamic control parameters for the shape, brightness, and projection angle of the laser headlight beam. The control module synchronously drives the actuator of the vehicle-mounted laser headlight to adjust the beam shape, brightness, and projection angle deflection according to the dynamic control parameters of the laser headlight beam shape, brightness, and projection angle. Finally, the optimization module collects the real-time working status data of the control module and the environmental feedback data from the secondary detection of the LiDAR, and iteratively optimizes the dynamic control parameters output by the calculation module.

[0168] In the above embodiments, the system can accurately perceive the geometry of the road ahead and the status of traffic participants in actual driving scenarios, and dynamically adjust the shape, brightness and projection angle of the laser headlight beam.

[0169] For example, at night or under complex lighting conditions, it can adapt to different road widths, avoid glare to other road users, optimize brightness according to target distance, and continuously optimize lighting effects with real-time feedback, thereby improving driving visibility and safety and reducing driving risks caused by improper lighting. Example 2

[0170] At the implementation level, based on Example 1, this example refers to... Figure 2 A further detailed description of the LiDAR-based intelligent vehicle lighting control system in Example 1 is provided below:

[0171] A method for intelligent control of vehicle lighting based on lidar, comprising:

[0172] Collect 3D point cloud data of the road ahead detected by vehicle-mounted lidar;

[0173] The collected 3D point cloud data is sequentially denoised, calibrated, and fused with multi-dimensional data to generate a standardized road environment perception dataset. Coordinate calibration requires three-level transformation between the lidar coordinate system, the vehicle coordinate system, and the world coordinate system.

[0174] Scene features are extracted from a standardized road environment perception dataset and weighted fusion classification is performed to identify the current road type, the distribution of traffic participants, and the relative motion relationship between them.

[0175] Based on the identified road type, distribution of traffic participants, and relative motion relationships, combined with the preset lighting control logic, dynamic control parameters for the shape, brightness, and projection angle of the laser headlight beam are output.

[0176] The actuator that drives the vehicle laser headlights adjusts the beam shape, brightness, and projection angle deflection according to the output dynamic control parameters. The projection angle deflection adjustment process simultaneously applies PID closed-loop control logic.

[0177] The system collects real-time operating status data of the actuator and environmental feedback data from secondary detection by lidar. Using the deviation between actual lighting parameters and dynamic control parameters as the core indicator, it iteratively optimizes the dynamic control parameters and feeds them back to the parameter calculation stage to replace the original parameters.

[0178] In summary, during the execution of the methods and systems described in the above embodiments, the three-dimensional point cloud data of the road ahead is detected by LiDAR, and a standardized road environment perception dataset is generated through data processing. This dataset accurately identifies the road type, the distribution of traffic participants, and their relative motion relationships. Based on this, the laser headlight beam shape, brightness, and projection angle control parameters are dynamically output to drive the actuator to adjust the lighting state. Furthermore, the control parameters are iteratively optimized by combining real-time working status and secondary detection environment feedback. This allows the system to adapt to different road scenarios and traffic conditions, avoiding excessively strong or weak lighting, reducing visual interference to other traffic participants, improving the clarity of driving vision at night and under complex lighting conditions, and ensuring driving safety. At the same time, PID closed-loop control ensures the accuracy of lighting adjustment, and the parameters are optimized based on the rate of environmental change, further enhancing the adaptability and stability of lighting control and reducing unnecessary energy consumption.

[0179] 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 laser radar-based intelligent control system for vehicle lighting, characterized by, include: The acquisition module is used to acquire three-dimensional point cloud data of the road ahead detected by the vehicle-mounted lidar; The dynamic target motion state parameters collected by the acquisition module are obtained through inter-frame point cloud matching calculations using lidar: Inter-frame point cloud matching: Acquire two consecutive frames of point cloud data from the LiDAR, denoted as frame t and frame (t+1), with a frame interval of 0.1s. For a dynamic target point cloud cluster A in frame t, find the corresponding point cloud cluster A′ in frame (t+1) using Euclidean distance matching, and determine the center point of A. The center point of A′ ; Calculation of the velocity of a dynamic target: , Indicates the frame interval time; Calculation of motion acceleration: Obtain the dynamic target point cloud cluster in frame t-1. center point Then calculate the motion velocity of the t-th frame, denoted as . Then the acceleration is The unit is ; If the matching degree of two point cloud clusters is less than 0.8, the motion state parameters of the previous frame are used, and the matching degree is determined by the ratio of the number of overlapping point clouds to the total number. The processing module is used to receive 3D point cloud data, perform noise reduction, coordinate calibration and multi-dimensional data fusion processing on the 3D point cloud data, and generate a standardized road environment perception dataset. The identification module is used to acquire a standardized road environment perception dataset, extract scene features from the dataset and classify them, and identify the current road type, distribution of traffic participants and their relative motion relationships. The calculation module receives the recognition results from the recognition module and, based on the recognition results and the preset lighting control logic, outputs dynamic control parameters for the shape, brightness, and projection angle of the laser headlight beam. The control module is used to drive the actuator of the vehicle laser headlight to adjust the beam shape, brightness and projection angle according to the dynamic control parameters of the laser headlight beam shape, brightness and projection angle. The optimization module is used to collect real-time operating status data of the control module and environmental feedback data from the secondary detection of the lidar, and to iteratively optimize the dynamic control parameters output by the calculation module. The three-dimensional point cloud data of the road ahead includes the road's geometric contour, the spatial coordinates of dynamic targets, and motion state parameters.

2. The intelligent vehicle lighting control system based on lidar according to claim 1, characterized in that, When the processing module performs noise reduction on the 3D point cloud data, it includes the following steps: Traversing each point in the 3D point cloud data Let i = 1, 2, ..., N, where N is the total number of point clouds, and each... Let r be the center of a sphere, and define a spherical neighborhood with radius r. Count the number of other point clouds contained within this spherical neighborhood. : Calculate the average number of neighborhoods for all point clouds. Simultaneously set the noise reduction threshold , This represents the density adjustment coefficient, with a value ranging from 0.3 to 0.

6. When the lidar detection distance is long, resulting in a sparse overall point cloud distribution, the density adjustment coefficient... The value should be as large as possible; when the lidar detection distance is short, resulting in a dense overall distribution of the point cloud, the density adjustment coefficient should be increased. The smaller the value, the better; the elevation angle α during the installation of the LiDAR should be. Number of neighboring point clouds point Points identified as noise are removed, while others are retained. The point cloud is used to obtain denoised point cloud data; The radius r of the spherical neighborhood is dynamically determined based on the real-time detection distance d of the lidar. The calculation formula is r = d × 0.02, where d ranges from 0 to 200 meters, and 0.02 is an initial preset constant term, with a value range of 0.01 to 0.

05.

3. The intelligent vehicle lighting control system based on lidar according to claim 1, characterized in that, When the processing module performs coordinate calibration on the 3D point cloud data, the processing objective is to complete a three-level coordinate transformation between the lidar coordinate system, the vehicle body coordinate system, and the world coordinate system. The process is as follows: Transformation from LiDAR coordinate system to vehicle coordinate system: Point cloud coordinates in the lidar coordinate system Convert to vehicle coordinate system The conversion formula is: , This represents the 3×3 rotation matrix from the lidar coordinate system to the vehicle coordinate system. This represents the 3×1 translation vector from the lidar coordinate system to the vehicle coordinate system; It is determined by the pitch angle α, roll angle β, and heading angle γ when the lidar is installed; ; The three-dimensional offset of the lidar installation position relative to the vehicle's origin Determined, that is ; Transformation from vehicle coordinate system to world coordinate system: coordinates in the vehicle coordinate system Convert to world coordinate system coordinates The conversion formula is: , This represents a 3×3 rotation matrix from the vehicle coordinate system to the world coordinate system. This represents a 3×1 translation vector from the vehicle coordinate system to the world coordinate system; , Indicates the vehicle's current GPS heading angle; , Indicates the vehicle's current GPS coordinates; When correcting for deviations in the transformed W-system coordinates, the following applies: ； In the formula: These are the corrected coordinates; , and The calculation logic is consistent; The coordinates of the j-th frame of 3D point cloud data after transformation from the LiDAR coordinate system to the vehicle coordinate system and then to the world coordinate system are the x-axis coordinates in the world coordinate system, where j is the frame number. Similarly.

4. The intelligent vehicle lighting control system based on lidar according to claim 1, characterized in that, When the recognition module extracts scene features and performs classification, it executes sequential operations of extraction, fusion, and judgment: Extracting core scene features: Linear fitting is performed on the road edge point cloud in the standardized road environment perception dataset to obtain the slope of the road edge. The absolute value of the slope is taken and denoted as the road edge feature S1. For each traffic participant, identify the minimum bounding box of the point cloud cluster, and denote the volume of the bounding box as the traffic participant volume feature S2; Calculate the relative speed of traffic participants relative to vehicles, and take the absolute value of the relative speed, which is denoted as the relative motion speed characteristic S3; Feature-weighted fusion: , As weight, The sum of the three is 1, and all three are positive numbers, with values ​​ranging from [0.2, 0.3], [0.4, 0.5], and [0.2, 0.3] respectively. Scene classification determination: When F < F1, it is determined to be: urban branch road with no large traffic participants; When F1≤F<F2, it is determined that: it is an urban arterial road and there are small traffic participants; When F≥F2, it is determined that: it is a highway and there are large traffic participants; Wherein, [F1,F2] are preset thresholds.

5. The intelligent vehicle lighting control system based on lidar according to claim 1, characterized in that, When the calculation module outputs dynamic control parameters based on the recognition results, the applied lighting control logic is as follows: Beam shape control includes control of the horizontal width W and the vertical height H: The horizontal width W is determined based on the standard lane width for the current road type: When on urban side roads, , This indicates the standard lane width for the current road type, and 0.5 indicates the reserved safety distance from the roadside. The unit is meters (m). When on the main urban road, On the highway, ; The vertical height H is determined based on the maximum height of traffic participants: , H represents the maximum height of traffic participants, 0.3 represents the reserved height safety distance, and H∈[1.8,3.0]m; Brightness control includes determining the reference brightness and controlling the actual brightness: The ambient light intensity I is set as follows: when I < 200 lux, the reference brightness is 8000 cd, corresponding to nighttime; when 200 lux ≤ I < 500 lux, the reference brightness is 5000 cd, corresponding to dusk or dawn; when I ≥ 500 lux, the reference brightness is 2000 cd, corresponding to daytime. Based on the relative distance between traffic participants and vehicles correction, , Indicates actual brightness. Indicates the reference brightness. This represents the baseline distance, initially set to 50m. ; Projection angle control includes the control of horizontal deflection angle and vertical deflection angle: Horizontal offset of traffic participants Sure, , Indicates the horizontal deflection angle; The vertical height h of traffic participants and Sure, , Indicates the vertical deflection angle, and Constraints Within the range.

6. The intelligent vehicle lighting control system based on lidar according to claim 1, characterized in that, When the control module drives the actuator to adjust the projection angle deflection, it simultaneously uses PID closed-loop control logic to control the adjustment accuracy. The process includes: The target projection angle is set, and its source is the calculation module, which collects the current actual angle of the actuator and calculates the angle deviation. The output control quantity of the PID controller is obtained by the following formula: ； In the formula: This refers to the control quantity output by the PID controller to the actuator for projecting the laser headlight at the current control cycle time t. This is the proportionality coefficient; This refers to the angle deviation at the current control cycle time t. The integral coefficient; For any point in time The corresponding angular deviation; These are the differential coefficients; This represents the integral value of the angular deviation over the time interval [0, t]. express The derivative with respect to time t; The actuator is based on the control quantity The drive angle adjustment motor has a control cycle duration of 0.02 seconds, until... Stop adjusting at that time; Stop adjusting at that time; The value is larger when the actual projection angle of the vehicle laser headlight deviates significantly from the target value and the deviation needs to be reduced quickly; the value is smaller when the actual projection angle deviates significantly from the target value and overshoot needs to be avoided. When the static deviation between the actual projection angle and the target angle persists for a long time and the deviation is large, The larger the value, the shorter the duration and the smaller the deviation amplitude between the actual projection angle and the target value. The smaller the value; The value is larger when the rate of change of the projection angle deviation of the vehicle laser headlight is fast and it is necessary to suppress the oscillation during the angle adjustment process; the value is smaller when the rate of change of the projection angle deviation is slow and it is necessary to ensure the stability of the adjustment process.

7. The intelligent vehicle lighting control system based on lidar according to claim 1, characterized in that, When the optimization module iteratively optimizes the dynamic control parameters, it uses the deviation between the actual lighting parameters output by the actuator and the dynamic control parameters output by the calculation module as the core optimization index. The process includes: Calculate the optimization deviation E to quantify the difference between the actual lighting effect and the theoretical control target: ； In the formula: The actual output brightness, actual projection angle, and actual beam width of the actuator used for secondary detection and inversion by lidar; The calculation module outputs dynamic control parameters for brightness, angle, and width. Set an optimization coefficient λ, which is based on the rate of environmental change. Dynamic adjustment; , This represents the current light intensity. The light intensity for the previous optimization cycle; To optimize the cycle, =1s; when When λ < 50 lux / s, take 0.1; when 50 lux / s ≤ When λ is <100 lux / s, it is taken as 0.

3. When the value is ≥100 lux / s, λ is taken as 0; The optimized control parameters are: ； in, To optimize brightness, projection angle, and beam width, The feedback is sent to the calculation module to replace the original parameters, completing one iteration.

8. The intelligent vehicle lighting control system based on lidar according to claim 1, characterized in that, The acquisition module is interconnected with the processing module and the identification module via a wireless network. The identification module is interconnected with the calculation module via a wireless network. The calculation module is interconnected with the control module via a wireless network. The acquisition module and the control module are interconnected with the optimization module via a wireless network.

9. A method for intelligent control of vehicle lighting based on lidar, wherein the method is an implementation method of the intelligent control system for vehicle lighting based on lidar as described in any one of claims 1-8, characterized in that, include: Collect 3D point cloud data of the road ahead detected by vehicle-mounted lidar; The collected 3D point cloud data is sequentially denoised, calibrated, and fused with multi-dimensional data to generate a standardized road environment perception dataset. Coordinate calibration requires three-level transformation between the lidar coordinate system, the vehicle coordinate system, and the world coordinate system. Scene features are extracted from a standardized road environment perception dataset and weighted fusion classification is performed to identify the current road type, the distribution of traffic participants, and the relative motion relationship between them. Based on the identified road type, distribution of traffic participants, and relative motion relationships, combined with the preset lighting control logic, dynamic control parameters for the shape, brightness, and projection angle of the laser headlight beam are output. The actuator that drives the vehicle laser headlights adjusts the beam shape, brightness, and projection angle deflection according to the output dynamic control parameters. The projection angle deflection adjustment process simultaneously applies PID closed-loop control logic. The system collects real-time operating status data of the actuator and environmental feedback data from secondary detection by lidar. Using the deviation between actual lighting parameters and dynamic control parameters as the core indicator, it iteratively optimizes the dynamic control parameters and feeds them back to the parameter calculation stage to replace the original parameters.

Images