Intelligent self-closed-loop vehicle lamp bumpy road light control system and method

Abstract

The invention discloses an intelligent self-closed-loop automobile lamp bumpy road light control system and method, and belongs to the technical field of automobile intelligent lighting. The light control method comprises the following steps: S1, collecting vehicle state information through an inertial measurement unit; s2, road bumping feature analysis and vehicle body posture prediction are carried out; s3, making a vehicle lamp deflection decision; s4, the projection angle of the vehicle lamp is adjusted; s5, whether the vehicle-mounted camera is shielded by rainwater or not is judged, if yes, a prediction mode is started, and if not, the step S6 is executed; s6, collecting the adjusted current vehicle lamp illumination light spot image; step S7, judging whether the position of the illumination light spot of the vehicle lamp has deviation or not, if so, skipping to step S8, and if not, ending light control; and step S8, performing car lamp illumination light spot calibration, and then skipping to the step S6. According to the invention, a bumpy road illumination blind area can be eliminated, the risk of dazzling during meeting is reduced, the response speed is increased, and the night driving safety is enhanced.

Description

Technical Field

[0001] This invention relates to an intelligent self-closed-loop vehicle headlight control system and method for bumpy road surfaces, belonging to the field of automotive intelligent lighting technology. Background Technology

[0002] Currently, at night or in low light conditions, vehicles experience pitching motion when driving on bumpy roads due to uneven surfaces. Traditional headlights have a fixed projection angle and cannot adjust to the vehicle's condition, causing the illuminated area to jump violently during bumpy driving, resulting in numerous problems.

[0003] On the one hand, the light spot can easily drift out of the effective illumination area, creating blind spots. When the vehicle rises, the light spot shifts upward, resulting in insufficient illumination of the road surface further ahead; when it dips, the light spot shifts downward, failing to illuminate distant hazards in advance, increasing the risk of collision. On the other hand, the flickering headlights during oncoming traffic cause high-frequency glare. During bumpy driving, the direction and angle of the headlights frequently change, potentially shining suddenly into the eyes of oncoming drivers, obstructing their vision, making it difficult to avoid collisions, and easily leading to accidents. Furthermore, while existing dynamic headlight systems can adjust the illumination, they rely on suspension height sensors, resulting in a 200-300ms delay in signal processing. At high speeds, the vehicle has already traveled several meters within this delay; if road conditions change abruptly ahead, the system cannot adjust the illumination in time, affecting the driver's judgment and threatening nighttime driving safety.

[0004] In conclusion, the problems with traditional headlight illumination caused by vehicle pitch motion when driving on bumpy roads, as well as the response delay of existing dynamic headlight systems, seriously affect the safety of nighttime driving and urgently need to be solved through technological innovation and improvement. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to overcome the shortcomings of the prior art and provide an intelligent self-closed-loop vehicle headlight lighting control system and method for bumpy roads. This system can effectively eliminate blind spots on bumpy roads, reduce the risk of glare when meeting oncoming traffic, improve system response speed, realize millisecond-level dynamic adjustment of vehicle headlight patterns, adapt to various complex road conditions, and significantly enhance nighttime driving safety and driving comfort.

[0006] To solve the above-mentioned technical problems, the technical solution of the present invention is as follows:

[0007] This invention provides an intelligent self-closed-loop vehicle headlight control method for bumpy road surfaces, specifically including the following steps:

[0008] Step S1: Collect vehicle status information through the inertial measurement unit;

[0009] Step S2: Based on the vehicle state information collected by the inertial measurement unit, perform road bump feature analysis and vehicle attitude prediction;

[0010] Step S3: Based on the analysis of road surface bump characteristics and vehicle body posture prediction, make a headlight deflection decision;

[0011] Step S4: The stepper motor adjusts the headlight projection angle according to the headlight deflection decision;

[0012] Step S5: Determine whether the vehicle camera is blocked by rain. If so, enable the prediction mode to compensate for the headlight projection angle. If not, proceed to step S6.

[0013] Step S6: Acquire an image of the adjusted current headlight illumination spot using the vehicle-mounted camera;

[0014] Step S7: Identify the position information of the current headlight illumination spot based on the current headlight illumination spot image, and determine whether there is a deviation between the current headlight illumination spot position and the target headlight illumination spot position. If yes, proceed to step S8; otherwise, the headlight control ends.

[0015] Step S8: Perform headlight illumination spot calibration. The stepper motor adjusts the headlight projection angle according to the spot calibration result, and then jumps to step S6 to continue execution.

[0016] Furthermore, the feature extraction formula for the road surface bump feature analysis is as follows:

[0017]

[0018] Where A(f) is the dominant frequency of vehicle body vibration;

[0019] t represents time;

[0020] a(t) is the triaxial composite acceleration;

[0021] F{a(t)} represents the Fourier transform of the triaxial composite acceleration;

[0022] T represents the time window;

[0023] j is the imaginary unit;

[0024] f is the characteristic frequency;

[0025] a x (t) represents the X-axis acceleration value;

[0026] a y (t) represents the acceleration value along the Y-axis;

[0027] a z (t) represents the Z-axis acceleration value.

[0028] Furthermore, the formula for calculating the characteristic frequency energy ratio in the road surface bump feature analysis is as follows:

[0029]

[0030] Among them, E ratio The characteristic frequency energy ratio;

[0031] f c This is the center frequency of the vehicle body vibration.

[0032] A(f) is the dominant frequency of vehicle body vibration;

[0033] f is the characteristic frequency.

[0034] Furthermore, the calculation formula for the state equation of the vehicle body attitude prediction is as follows:

[0035]

[0036] Where, θ k Let k be the predicted vehicle pitch angle at time k;

[0037] Let be the predicted angular velocity at time k;

[0038] Δt is the sampling interval time;

[0039] θ k-1 The predicted vehicle pitch angle at time k-1;

[0040] The predicted angular velocity at time k-1;

[0041] a k The actual angular acceleration at time k;

[0042] w k The actual angular velocity at time k;

[0043] W k This is process noise.

[0044] Furthermore, the calculation formula for the observation equation of the vehicle body attitude prediction is as follows:

[0045]

[0046] Among them, z k These are actual observed values;

[0047] θ k Let k be the predicted vehicle pitch angle at time k;

[0048] Let be the predicted angular velocity at time k;

[0049] v k To observe noise.

[0050] Furthermore, the calculation formula for the headlight deflection decision is as follows:

[0051]

[0052] Where, β cmd The deflection angle of the headlight target;

[0053] K p For proportional gain;

[0054] K d This is the differential gain;

[0055] The vehicle's pitch angle for the predicted future moment;

[0056] t represents the current time.

[0057] Δt is the predicted time after the current moment;

[0058] This represents the rate of change of the pitch angle.

[0059] Furthermore, the calculation formula for the headlight illumination spot calibration is as follows:

[0060] Δβ=K fb ×(y ref -y act )×e -τs ;

[0061] Where Δβ is the light angle compensation amount;

[0062] K fb For adaptive feedback gain;

[0063] y ref The location of the target vehicle headlight illumination spot;

[0064] y act This indicates the actual location of the headlight illumination spot.

[0065] τ is the system delay compensation coefficient;

[0066] s is a complex frequency variable.

[0067] In another aspect, the present invention provides an intelligent self-closed-loop vehicle headlight lighting control system for bumpy roads, including an inertial measurement unit, an on-board camera, an edge computing unit, and a headlight module;

[0068] The inertial measurement unit is used to collect vehicle status information;

[0069] The vehicle-mounted camera is used to capture images of the light spots illuminated by the vehicle headlights;

[0070] The edge computing unit is used to perform road bump feature analysis and vehicle attitude prediction based on the vehicle state information collected by the inertial measurement unit, and to make headlight deflection decisions based on the road bump features and vehicle attitude prediction, and to perform headlight illumination spot calibration based on the headlight illumination spot image collected by the vehicle camera.

[0071] The headlight module is used to adjust the headlight projection angle based on headlight deflection decisions and headlight illumination spot calibration.

[0072] Furthermore, the vehicle lighting module includes a stepper motor drive module, a stepper motor, and an LED light assembly;

[0073] The stepper motor drive module is used to control the stepper motor based on the headlight deflection decision and the headlight illumination spot calibration.

[0074] The stepper motor is used to adjust the projection angle of the LED light assembly;

[0075] The LED light group is used for illumination.

[0076] Furthermore, the vehicle status information includes the vehicle pitch rate, vehicle acceleration, and vehicle vibration frequency.

[0077] By adopting the above technical solution, the present invention has the following beneficial effects:

[0078] This invention uses an inertial measurement unit (IMU) to collect vehicle state information, replacing the traditional suspension height sensor. This significantly reduces the delay in signal acquisition and processing, enabling timely detection of sudden changes in road conditions ahead. Edge computing units analyze road bump characteristics to identify different road types, providing a basis for subsequent control strategy selection. Edge computing units also predict vehicle attitude with high precision, providing a data foundation for advance decision-making on headlight deflection angles. The edge computing unit's headlight deflection decisions based on road bump characteristic analysis and vehicle attitude prediction effectively eliminate blind spots on bumpy roads, reduce the risk of glare from oncoming traffic, improve system response speed, and achieve millisecond-level dynamic adjustment of headlight patterns. This allows for adaptation to various complex road conditions, significantly enhancing nighttime driving safety and comfort. Headlight illumination spot calibration is performed based on the deviation between the current headlight illumination spot position and the target headlight illumination spot position. This addresses headlight illumination spot position deviations caused by sensor bias, system delays, and changes in ambient light intensity, ensuring illumination accuracy. Attached Figure Description

[0079] Figure 1 This is a schematic diagram of the intelligent self-closed-loop vehicle headlight lighting control system for bumpy roads according to the present invention.

[0080] Figure 2This is a flowchart of the intelligent self-closed-loop vehicle headlight lighting control method for bumpy road surfaces according to the present invention. Detailed Implementation

[0081] To make the content of this invention easier to understand, the invention will be further described in detail below with reference to specific embodiments and accompanying drawings.

[0082] Example 1

[0083] like Figure 1 As shown, this embodiment provides an intelligent self-closed-loop vehicle headlight control system for bumpy road surfaces. It includes an inertial measurement unit (IMU), an onboard camera, an edge computing unit, and a headlight module. The IMU and onboard camera form the perception layer, the edge computing unit forms the predictive control layer, and the headlight module forms the execution feedback layer. By constructing a three-layer closed-loop control system, it achieves fully localized closed-loop control of sensor data acquisition, algorithm computation by the localized edge computing chip, and actuator driving. This eliminates traditional CAN bus communication, completely eliminating bus latency and improving nighttime lighting accuracy.

[0084] Specifically, the inertial measurement unit (IMU) is used to collect vehicle state information, including vehicle pitch rate, vehicle acceleration, and vehicle vibration frequency. Using an IMU to replace the traditional suspension height sensor significantly reduces signal acquisition delay.

[0085] Specifically, the vehicle-mounted camera is used to capture images of the light spots illuminated by the vehicle's headlights.

[0086] Specifically, the edge computing unit is a controller containing a SOC, used to perform road bump feature analysis and vehicle attitude prediction based on the vehicle state information collected by the inertial measurement unit. In particular, based on the road bump feature analysis, the road surface can be classified into continuous bumpy road surface, high-frequency gravel road surface, sudden single pothole road surface, and flat road surface. The vehicle attitude prediction is to predict the vehicle's pitch angle and pitch velocity. When performing vehicle attitude prediction, a Kalman filter is introduced to suppress noise in the data collected by the inertial measurement unit, which helps to improve the accuracy of vehicle attitude prediction. Then, the headlight deflection decision is made based on the road surface bump characteristics and vehicle posture prediction, and the headlight illumination spot is calibrated based on the headlight illumination spot image captured by the vehicle camera. In particular, in the headlight deflection decision algorithm, the required headlight compensation angle when the vehicle is driving on bumpy roads is calculated in advance by calculating the target headlight deflection angle. This can keep the headlight spot position stable, prevent the headlight spot from leaving the effective illumination area, avoid high-frequency glare caused by headlight flickering, and improve the safety of night driving. After the headlight deflection decision is executed, the headlight illumination spot image after the deflection angle is adjusted is captured by the camera, and further headlight angle compensation calculation is performed based on the current headlight illumination spot position deviation, which improves the accuracy of headlight angle deflection.

[0087] Specifically, the headlight module is used to adjust the headlight projection angle based on headlight deflection decisions and headlight illumination spot calibration. The headlight module includes a stepper motor driver module, a stepper motor, and an LED light assembly. The stepper motor driver module controls the stepper motor based on headlight deflection decisions and headlight illumination spot calibration. The stepper motor adjusts the projection angle of the LED light assembly, which is used for illumination.

[0088] Example 2

[0089] like Figure 2 As shown, this embodiment provides a lighting control method for an intelligent self-closed-loop vehicle headlight system for bumpy road surfaces, as described in Embodiment 1, specifically including the following steps:

[0090] Step S1: Collect vehicle status information using an inertial measurement unit (IMU). This vehicle status information includes vehicle pitch rate, vehicle acceleration, and vehicle vibration frequency. Using an IMU to replace the traditional suspension height sensor significantly reduces the signal acquisition delay.

[0091] Step S2: Based on the vehicle state information collected by the inertial measurement unit, perform road bump feature analysis and vehicle attitude prediction. Road bump feature analysis includes bump feature extraction and characteristic frequency-energy ratio calculation. By extracting bump features, different road surface types can be identified, and the vehicle's current mode can be distinguished based on the road surface type, providing a basis for subsequent control strategy selection; by calculating the characteristic frequency-energy ratio, the severity of bumps can be quantified. Vehicle attitude prediction includes state equation calculation and observation equation calculation. The state equation is used to predict the vehicle attitude at the next moment, and the observation equation is used to correct the predicted value and reduce the influence of noise. The state equation and observation equation work together to achieve high-precision attitude prediction.

[0092] Specifically, the feature extraction algorithm for road bump feature analysis separates the dominant frequency of vehicle body vibration using Fast Fourier Transform, as shown in the following formula:

[0093]

[0094] Where A(f) is the dominant frequency of vehicle body vibration, which is the frequency domain representation obtained after Fourier transform of the time domain signal a(t);

[0095] t represents time;

[0096] a(t) is the triaxial composite acceleration. The triaxial composite acceleration refers to the total acceleration obtained by combining the acceleration components on three mutually perpendicular axes (usually labeled as X, Y, and Z axes) in three-dimensional space. This composite acceleration can more comprehensively describe the motion state of an object in space and represents the scalar value of the overall acceleration of the object.

[0097] F{a(t)} represents the Fourier transform of the triaxial composite acceleration;

[0098] T is the time window, preset to 100ms, which determines the frequency resolution. It describes the minimum frequency interval required to distinguish two signals with similar frequencies. The larger the time window, that is, the longer the analysis period, the finer the frequency details that can be distinguished, and therefore the higher the frequency resolution.

[0099] j is the imaginary unit;

[0100] f is the characteristic frequency, which reflects the main frequency of vehicle body vibration. It refers to the frequency of typical vibration, fluctuation or oscillation, and can be obtained through an inertial measurement unit.

[0101] a x (t) represents the X-axis acceleration value;

[0102] a y (t) represents the acceleration value along the Y-axis;

[0103] a z (t) represents the Z-axis acceleration value.

[0104] Specifically, the formula for calculating the characteristic frequency energy ratio in road surface bump feature analysis is as follows:

[0105]

[0106] Among them, E ratio The characteristic frequency energy ratio;

[0107] f c This is the center frequency of vehicle body vibration, typically 2–5 Hz, corresponding to suspension vibration. It is used to quantify vibration characteristics and can be obtained through an inertial measurement unit.

[0108] A(f) is the dominant frequency of vehicle body vibration, which is the frequency domain representation obtained after Fourier transform of the time domain signal a(t);

[0109] f is the characteristic frequency, which reflects the main frequency of vehicle body vibration. It refers to the frequency of typical vibration, fluctuation or oscillation, and can be obtained through an inertial measurement unit.

[0110] The main frequency band energy refers to the main energy of the signal being concentrated within a specific frequency band. In this embodiment, the frequency bandwidth is 1Hz.

[0111] Representing the total vibrational energy from 0 to 20 Hz, it refers to the total amount of energy possessed during the vibration process.

[0112] In this embodiment, based on the characteristic frequency and characteristic frequency energy ratio in the bumpy road surface characteristic analysis, the road surface can be divided into continuous bumpy road surface, high-frequency gravel road surface, sudden single pothole road surface, and flat road surface. When the characteristic frequency is 2-5Hz (corresponding to the vehicle suspension natural frequency) and the characteristic frequency energy ratio is >0.65 (the dominant frequency band energy accounts for more than 65% of the total vibration energy), it represents a continuous bumpy road surface, and the vehicle is in bumpy mode. When the characteristic frequency is >8Hz (high-frequency vibration dominates) and the characteristic frequency energy ratio is >0.5, it represents a high-frequency gravel road surface, and the vehicle is in gravel mode. When the vehicle acceleration is >1.5g and the duration is <0.1s (transient impact characteristics), the characteristic frequency is >8Hz and the characteristic frequency energy ratio is >0.5 (high-frequency vibration induced after transient impact), it represents a sudden single pothole road surface, and the vehicle is in pothole mode. When none of the above conditions are met, it represents a flat road surface, and the vehicle is in flat mode.

[0113] Specifically, when predicting vehicle attitude, a Kalman filter is introduced to suppress noise in the data collected by the inertial measurement unit, which helps improve the accuracy of vehicle attitude prediction. The formula for calculating the state equation for vehicle attitude prediction is as follows:

[0114]

[0115] Where, θ k Let k be the predicted vehicle pitch angle, with a positive value representing the pitch direction;

[0116] Let be the predicted angular velocity at time k;

[0117] Δt is the sampling interval, which is preset to 1ms;

[0118] θ k-1 The predicted vehicle pitch angle at time k-1;

[0119] The predicted angular velocity at time k-1;

[0120] a k The actual angular acceleration at time k is obtained by differential calculation from the data collected by the inertial measurement unit.

[0121] w k The actual angular velocity at time k is obtained by the inertial measurement unit.

[0122] W k The process noise follows a Gaussian distribution N(0,Q), where Q = 0.01 is the process noise covariance. In N(0,Q), 0 is the mean of the distribution, i.e., the average value, and Q is the variance of the distribution, which describes the dispersion of data points around the mean.

[0123] Specifically, the calculation formula for the observation equation of vehicle body attitude prediction is as follows:

[0124]

[0125] Among them, z k These are actual observed values;

[0126] θ k Let k be the predicted vehicle pitch angle at time k;

[0127] Let be the predicted angular velocity at time k;

[0128] v k The observation noise follows a Gaussian distribution N(0,R), where R = 0.1 is the observation noise covariance.

[0129] Step S3: Based on the analysis of road surface bump characteristics and vehicle posture prediction, a headlight deflection decision is made, i.e., the target headlight deflection angle is calculated. This decision is the core of the entire control system. By calculating the headlight deflection angle in advance, proactive control is achieved, avoiding violent fluctuations in the headlight illumination area when the vehicle is bumpy, preventing the headlight illumination spot from leaving the effective illumination area, effectively eliminating blind spots on bumpy roads, reducing the risk of glare from oncoming traffic, improving system response speed, and achieving millisecond-level dynamic adjustment of headlight pattern. This enables the system to adapt to various complex road conditions, significantly enhancing nighttime driving safety and driving comfort.

[0130] Specifically, the calculation formula for the headlight deflection decision is as follows:

[0131]

[0132] Where, β cmd The headlight target deflection angle refers to the deflection angle of the headlight beam direction relative to the direction the vehicle is traveling in a straight line when turning.

[0133] K p This is the proportional gain, calibrated based on sprung mass and suspension height, with a default value of 0.8. It is used to perform proportional compensation to follow the predicted angle and is responsible for eliminating static deviations.

[0134] K d The differential gain is calculated based on the dynamic model, and the calculation formula is as follows: Where β 阻尼 For damping torque, K is the rate of change of angular velocity. d Used to suppress oscillations caused by pitch angle motion, responsible for dynamic process stability, with a default value of 0.4;

[0135] The vehicle's pitch angle for the predicted future moment;

[0136] t represents the current time.

[0137] Δt is the prediction time after the current moment, with a default value of 50ms;

[0138] This represents the rate of change of the pitch angle.

[0139] In this embodiment, when the vehicle is in bumpy mode, gravel mode, and pothole mode, the corresponding PID parameters are matched. For example, when the vehicle is in bumpy mode, K is enabled. p =0.8, K d =0.4; When the vehicle is in gravel mode, K is enabled. p =0.5, K d =0.6; K is enabled when the vehicle is in pothole mode. p=1,K d =0.2. When the vehicle is in flat mode, the basic compensation is maintained, that is, when the vehicle is on normal uphill or downhill sections or level roads, the headlight system maintains the preset static compensation.

[0140] Step S4: The stepper motor adjusts the headlight projection angle according to the headlight deflection decision.

[0141] Step S5: The road image captured by the vehicle camera is detected by the edge computing unit. If the vehicle camera is detected to be blocked by rain, the prediction mode is activated to compensate for the headlight projection angle. If the vehicle camera is not blocked by rain, the process proceeds to step S6.

[0142] When the vehicle camera is detected to be obstructed by rain, in order to prevent rain interference from causing inaccurate information collected by the vehicle camera and affecting subsequent headlight projection angle compensation, it is necessary to pause the acquisition of headlight illumination spot images by the vehicle camera. That is, real-time visual feedback is turned off, and headlight projection angle compensation is performed entirely by relying on the Kalman filter prediction model. The control mode changes from closed-loop control to open-loop control, which can maintain the basic operation of the system and ensure the accuracy of light control.

[0143] The calculation formula for headlight projection angle compensation when using predictive mode is as follows, without visual feedback:

[0144] θ comp =θ pred +0.3(θ pred -θ meas );

[0145] Where, θ comp The vehicle pitch angle is corrected for errors. The headlight deflection angle is adjusted according to the corrected vehicle pitch angle to compensate for the headlight projection angle.

[0146] θ pred The vehicle pitch angle predicted by the Kalman filter is the future attitude predicted based on inertial measurement unit data;

[0147] θ meas The measured vehicle pitch angle is obtained from the inertial measurement unit.

[0148] 0.3 is the correction coefficient, which is the prediction error compensation weight. It is obtained through calibration and is mainly used to balance the reliability of prediction and actual measurement.

[0149] Step S6: Collect the adjusted current headlight illumination spot image using the vehicle-mounted camera, as the basis for subsequent spot position comparison.

[0150] Step S7: Identify the position information of the current headlight illumination spot based on the current headlight illumination spot image, and determine whether there is a deviation between the current headlight illumination spot position and the target headlight illumination spot position. If yes, proceed to step S8; otherwise, end the headlight control.

[0151] Step S8: Perform headlight illumination spot calibration. The stepper motor adjusts the headlight projection angle based on the spot calibration result, and then jumps to step S6 to continue execution. Headlight illumination spot calibration can address headlight illumination spot position deviations caused by sensor bias, system delay, and changes in ambient light intensity, ensuring illumination accuracy.

[0152] Specifically, the calculation formula for vehicle headlight illumination spot calibration is as follows:

[0153] Δβ=K fb ×(y ref -y act )×e -τs ;

[0154] Where Δβ is the light angle compensation amount;

[0155] y ref The location of the target vehicle headlight illumination spot;

[0156] y act The actual location of the vehicle headlight illumination spot is obtained by performing image processing such as threshold segmentation and edge detection on the vehicle headlight illumination spot image captured by the camera;

[0157] τ is the system delay compensation coefficient, used to compensate for the response delay of the actuator. It depends on the specific system characteristics and application scenario and can be obtained through repeated experiments and adjustments. In this embodiment, it is preset to 20ms.

[0158] s is a complex frequency variable, an important parameter in the Laplace transform, used to represent the frequency and phase of a signal;

[0159] K fb The core of adaptive feedback gain lies in dynamically adjusting K by monitoring the ambient light intensity in real time. fb This value, while ensuring control precision, can optimize energy efficiency.

[0160] K fb The mapping relationship with ambient light intensity is as follows: Among them, K min K represents the minimum gain value under strong light conditions, used to reduce system sensitivity and improve robustness. max is the maximum gain value in dark environments, used to improve calibration accuracy; 'a' is the S-curve steepness coefficient, used to control possible abrupt changes in gain, obtained through calibration; L vL1 represents the ambient light intensity, obtained through real-time measurement; L2 represents the threshold value for switching between night mode and day mode, obtained through actual measurement and calibration.

[0161] According to different ambient light intensities L v By classifying the environment types, the corresponding adaptive feedback gain K is obtained. fb Values, specific data are as follows: When L v When the temperature is >50°C, the environment is under strong sunlight, and at this time K0 fb =0.2; when 10 <L v When K is ≤50, the environment is cloudy or overcast. fb =0.4; when 1 <L v When K is ≤10, the environment is dusk or dawn, at which time K fb =0.7; when 0.1 <L v When K ≤ 1, the environment is nighttime with streetlights, and K is... fb =0.9; when L v When K is ≤0.1, the environment is completely dark. fb =1.

[0162] The working principle of this invention is as follows:

[0163] Vehicle status information is collected via an inertial measurement unit (IMU); based on this information, road surface bump characteristics are analyzed and vehicle attitude is predicted; headlight deflection decisions are made based on these analyses; a stepper motor adjusts the headlight projection angle according to the deflection decision; it is determined whether the onboard camera is obstructed by rain. If so, prediction mode is activated to compensate for the headlight projection angle; otherwise, the adjusted headlight illumination spot image is collected via the onboard camera; the current headlight illumination spot position is identified based on the image, and it is determined whether there is a deviation between the current and target headlight illumination spot positions. If no deviation is found, headlight control ends; otherwise, headlight illumination spot calibration is performed, and the stepper motor adjusts the headlight projection angle based on the calibration results.

[0164] This invention uses an inertial measurement unit (IMU) to collect vehicle state information, replacing the traditional suspension height sensor. This significantly reduces the delay in signal acquisition and processing, enabling timely detection of sudden changes in road conditions ahead. Edge computing units analyze road bump characteristics to identify different road types, providing a basis for subsequent control strategy selection. Edge computing units also predict vehicle attitude with high precision, providing a data foundation for advance decision-making on headlight deflection angles. The edge computing unit's headlight deflection decisions based on road bump characteristic analysis and vehicle attitude prediction effectively eliminate blind spots on bumpy roads, reduce the risk of glare from oncoming traffic, improve system response speed, and achieve millisecond-level dynamic adjustment of headlight patterns. This allows for adaptation to various complex road conditions, significantly enhancing nighttime driving safety and comfort. Headlight illumination spot calibration is performed based on the deviation between the current headlight illumination spot position and the target headlight illumination spot position. This addresses headlight illumination spot position deviations caused by sensor bias, system delays, and changes in ambient light intensity, ensuring illumination accuracy.

[0165] The specific embodiments described above further illustrate the technical problems, technical solutions, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for controlling vehicle headlights on bumpy roads using an intelligent self-closed-loop system, characterized in that, Specifically, the steps include the following: Step S1: Collect vehicle status information through the inertial measurement unit; Step S2: Based on the vehicle state information collected by the inertial measurement unit, perform road bump feature analysis and vehicle attitude prediction; Step S3: Based on the analysis of road surface bump characteristics and vehicle body posture prediction, make a headlight deflection decision; Step S4: The stepper motor adjusts the headlight projection angle according to the headlight deflection decision; Step S5: Determine whether the vehicle camera is blocked by rain. If so, enable the prediction mode to compensate for the headlight projection angle. If not, proceed to step S6. Step S6: Acquire an image of the adjusted current headlight illumination spot using the vehicle-mounted camera; Step S7: Identify the position information of the current headlight illumination spot based on the current headlight illumination spot image, and determine whether there is a deviation between the current headlight illumination spot position and the target headlight illumination spot position. If yes, proceed to step S8; otherwise, the headlight control ends. Step S8: Perform headlight illumination spot calibration. The stepper motor adjusts the headlight projection angle according to the spot calibration result, and then jumps to step S6 to continue execution.

2. The intelligent self-closed-loop vehicle headlight control method for bumpy road surfaces according to claim 1, characterized in that, The feature extraction formula for the road surface bump feature analysis is as follows: Where A(f) is the dominant frequency of vehicle body vibration; t represents time; a(t) is the triaxial composite acceleration; F{a(t)} represents the Fourier transform of the triaxial composite acceleration; T represents the time window; j is the imaginary unit; f is the characteristic frequency; a x (t) represents the X-axis acceleration value; a y (t) represents the acceleration value along the Y-axis; a z (t) represents the Z-axis acceleration value.

3. The intelligent self-closed-loop vehicle headlight control method for bumpy road surfaces according to claim 2, characterized in that, The formula for calculating the characteristic frequency energy ratio in the road surface bump feature analysis is as follows: Among them, E ratio The characteristic frequency energy ratio; f c This is the center frequency of the vehicle body vibration. A(f) is the dominant frequency of vehicle body vibration; f is the characteristic frequency.

4. The intelligent self-closed-loop vehicle headlight control method for bumpy road surfaces according to claim 1, characterized in that, The formula for calculating the state equation for vehicle body attitude prediction is as follows: Where, θ k Let k be the predicted vehicle pitch angle at time k; Let be the predicted angular velocity at time k; Δt is the sampling interval time; θ k-1 The predicted vehicle pitch angle at time k-1; The predicted angular velocity at time k-1; a k The actual angular acceleration at time k; w k The actual angular velocity at time k; W k This is process noise.

5. The intelligent self-closed-loop vehicle headlight control method for bumpy road surfaces according to claim 4, characterized in that, The calculation formula for the observation equation of the vehicle body attitude prediction is as follows: Among them, z k These are actual observed values; θ k Let k be the predicted vehicle pitch angle at time k; Let be the predicted angular velocity at time k; v k To observe noise.

6. The intelligent self-closed-loop vehicle headlight control method for bumpy road surfaces according to claim 1, characterized in that, The calculation formula for the headlight deflection decision is as follows: Where, β cmd The deflection angle of the headlight target; K p For proportional gain; K d This is the differential gain; The vehicle's pitch angle for the predicted future moment; t represents the current time. Δt is the predicted time after the current moment; This represents the rate of change of the pitch angle.

7. The intelligent self-closed-loop vehicle headlight control method for bumpy road surfaces according to claim 1, characterized in that, The calculation formula for the headlight illumination spot calibration is as follows: Δβ=K fb ×(y ref -y act )×e -τs ; Where Δβ is the light angle compensation amount; K fb For adaptive feedback gain; y ref The location of the target vehicle headlight illumination spot; y act This indicates the actual location of the headlight illumination spot. τ is the system delay compensation coefficient; s is a complex frequency variable.

8. A lighting control system applying the intelligent self-closed-loop vehicle lighting control method for bumpy road surfaces as described in any one of claims 1 to 7, characterized in that, It includes an inertial measurement unit, an in-vehicle camera, an edge computing unit, and a headlight module; The inertial measurement unit is used to collect vehicle status information; The vehicle-mounted camera is used to capture images of the light spots illuminated by the vehicle headlights; The edge computing unit is used to perform road bump feature analysis and vehicle attitude prediction based on the vehicle state information collected by the inertial measurement unit, and to make headlight deflection decisions based on the road bump features and vehicle attitude prediction, and to perform headlight illumination spot calibration based on the headlight illumination spot image collected by the vehicle camera. The headlight module is used to adjust the headlight projection angle based on headlight deflection decisions and headlight illumination spot calibration.

9. The lighting control system according to claim 8, characterized in that, The vehicle lighting module includes a stepper motor drive module, a stepper motor, and LED lights; The stepper motor drive module is used to control the stepper motor based on the headlight deflection decision and the headlight illumination spot calibration. The stepper motor is used to adjust the projection angle of the LED light assembly; The LED light group is used for illumination.

10. The lighting control system according to claim 8, characterized in that, The vehicle status information includes the vehicle pitch rate, vehicle acceleration, and vehicle vibration frequency.

Images