A control method for AR-HUD projection large screen of automobile

Abstract

The application discloses a control method for AR-HUD projection large screen of a car, and belongs to the technical field of vehicle head-up display (HUD) and augmented reality (AR) display. The method comprises the following steps: S1, system initialization and calibration, loading optical calibration data and completing initial calibration of a vehicle body posture, an inertial measurement unit (IMU) inertial navigation, and a camera; S2, real-time acquisition of multi-sensor data, acquiring a vehicle body posture signal, an environment signal, a driver state signal, and a driving signal; and S3, vehicle body motion prediction and picture inverse compensation, predicting posture changes at the next moment according to acquired vehicle body motion data, and performing inverse motion pre-compensation on a projection picture, so that a virtual picture is stably anchored on a road surface. The application is used for solving the problems of limited field of view of an existing AR-HUD single light machine, dynamic picture drift and jitter, edge distortion ghosting, multi-light machine splicing misplacement, and poor environmental adaptability.

Description

Technical Field

[0001] This invention relates to the field of vehicle head-up display (HUD) and augmented reality (AR) display technology, and particularly to a control method for a large AR-HUD projection screen for automobiles. Background Technology

[0002] AR-HUD systems can overlay virtual image information such as navigation guidance, vehicle speed, lane warnings, and obstacle alerts onto the real road in front of the car in real time, allowing drivers to obtain key driving information without looking down, thus significantly improving driving safety and interactive experience. As automotive intelligence continues to increase, the industry's performance requirements for AR-HUDs are also constantly upgrading, especially the desire for a larger field of view, a wider display screen, and a greater distance for the virtual image, in order to create a stronger sense of immersion and a seamless blend of virtual and real elements.

[0003] Currently, existing AR-HUD systems typically employ a single optical engine paired with a fixed optical scheme to achieve large-screen displays. Some solutions attempt to expand the field of view by increasing the optical engine resolution or enlarging the optical module size, while others utilize a single optical engine for software-level screen partitioning. Furthermore, to address changes in vehicle posture during movement, existing systems generally adjust the projected image's position statically based on simple vehicle tilt signals. For differences in driver height and seating posture, manual adjustment or preset eye box positions are often used. In multi-screen display scenarios, a few solutions attempt to use dual optical engines for screen expansion, but these typically only achieve simple side-by-side display.

[0004] The aforementioned existing technologies have significant shortcomings in practical applications. Single-optical engines, limited by optical components and size, have limited field-of-view expansion capabilities, making it difficult to meet the demands of ultra-large screens and wide field-of-view displays. Even software partitioning cannot overcome physical hardware limitations. Existing systems mostly perform static or delayed adjustments for vehicle motion compensation, failing to achieve real-time and accurate anchoring of the virtual image to the real road under dynamic conditions such as bumps, steering, acceleration, and deceleration. This causes the displayed image to drift and jitter with vehicle movement, severely affecting the fusion of virtual and real images. For image distortion correction, existing methods primarily target the central area of ​​the image, while pincushion and barrel distortions and ghosting blurring are prominent in the edge areas, resulting in poor overall screen clarity. Drivers of different heights and seating positions often need to manually adjust to obtain a suitable field of vision, lacking dynamic real-time adaptation capabilities based on eye position. Regarding environmental adaptability, existing systems struggle to automatically optimize brightness and anti-glare effects based on complex lighting conditions such as strong light, nighttime, tunnels, and sunlight incidence angles. Summary of the Invention

[0005] To address the problems of limited field of view of existing AR-HUD single-optical-engines, dynamic image drift and jitter, edge distortion and ghosting, misalignment of multi-optical-engine splicing, and poor environmental adaptability, this invention provides a control method for large-screen AR-HUD projection in automobiles.

[0006] The technical solution adopted by this invention to solve its technical problem is: a control method for a large AR-HUD projection screen for automobiles. Preferably, the method comprises the following steps: S1: System initialization and calibration, loading optical calibration data and completing the initial calibration of vehicle attitude, IMU inertial navigation, and camera; S2: Real-time acquisition of multi-sensor data, obtaining vehicle attitude signals, environmental signals, driver status signals, and driving signals; S3: Vehicle motion prediction and inverse image compensation, predicting the attitude change in the next moment based on the acquired vehicle motion data, and performing inverse motion pre-compensation on the projected image to stabilize the virtual image on the road surface; S4: Large-screen partition rendering and distortion correction, dividing the large image to be displayed into a central main display area, an edge auxiliary area, and a bottom information area for differentiated processing, and calling an optical distortion mapping table. S5: Real-time correction; Dynamic image adaptation based on human eye position: Adjusts the position of the projected image, the output angle of the optical engine, and the field of view in real time according to the spatial coordinates of the driver's eyes detected by the DMS camera; S6: Ambient light adaptation and anti-glare control: Dynamically adjusts the brightness and contrast of the image according to the ambient light intensity and the angle of sunlight incidence; S7: Layered display of large-screen content and safety strategy: Divides the displayed content into at least three levels according to the importance of driving, and displays it differently in different areas of the image; S8: Multi-screen splicing and fusion and seam elimination: By selecting the splicing mode, the overlapping areas of the images are fused while maintaining synchronization of multiple optical engines; S9: Output and closed-loop feedback: The processed image is sent to the AR-HUD optical engine projection, and the image effect is sampled back in real time for closed-loop correction.

[0007] Preferably, the differentiated processing in S4 specifically includes: the central main display area displays navigation guidance, lane lines and obstacle warnings with high resolution and high contrast; the edge auxiliary area calls the optical distortion mapping table to correct pincushion or barrel distortion in real time; and the bottom information area displays vehicle speed, speed limit and signage auxiliary information.

[0008] Preferably, the splicing mode selection in S8 includes: horizontal splicing of two optical engines to expand the horizontal field of view, vertical splicing of multiple optical engines to achieve layered splicing of far-field AR images and near-field instrument images, or software partitioning splicing of a single optical engine.

[0009] Preferably, the fusion processing of the overlapping area in S8 includes: setting a 1% to 3% overlapping transition band, and eliminating the seam through brightness gradient feathering, pixel interpolation and grayscale uniformization processing; the multi-optical-machine synchronization adopts the same source clock synchronization to ensure the frame refresh synchronization of the multi-optical-machine.

[0010] Preferably, the reverse motion pre-compensation in S3 specifically includes: performing vertical reverse compensation on the image based on the vehicle body pitch change; performing horizontal reverse compensation on the image based on the vehicle body roll change; and dynamically correcting the image position and zoom based on the vehicle body bumps and jumps.

[0011] Preferably, the ambient light adaptive and anti-glare control in S6 specifically includes: automatically increasing the brightness and contrast of the central area in a strong daytime environment; reducing the overall brightness and turning off unnecessary high-brightness elements in a nighttime or tunnel environment; and dynamically adjusting the backlight according to the angle of sunlight incidence.

[0012] Preferably, the content layering display and security strategy in S7 is as follows: the first-level core information is displayed in the center with high brightness; the second-level regular information is displayed stably at the bottom of the screen; and the third-level extended information is displayed in the corner of the screen with low brightness.

[0013] Preferably, the dynamic image adaptation based on human eye position in S5 specifically includes adjusting the vertical position, horizontal position, optical engine output angle and offset, and cropping and scaling of the field of view of the projected image in real time according to the spatial coordinates of the driver's eyes detected by the DMS camera.

[0014] Preferably, the closed-loop feedback in S9 specifically includes: real-time re-acquisition of the stability of the image, splicing effect and alignment error, and entering closed-loop correction based on the re-acquisition results to continuously maintain the display effect.

[0015] Preferably, the multi-sensor data collected in S2 specifically includes: vehicle posture signals composed of acceleration and angular velocity, four-wheel height, suspension travel, and steering angle from the IMU; environmental signals composed of light intensity, sunlight incidence angle, and day / night status; driver status signals composed of the human eye position detected by the DMS camera; and driving signals composed of vehicle speed, gear, navigation path, ADAS obstacle and lane information.

[0016] The beneficial effects of this invention are as follows: It overcomes the inherent limitations of the physical field of view of a single optical engine. By supporting various splicing modes such as single-optical-engine software partitioning, dual-optical-engine horizontal splicing, and multi-optical-engine vertical and horizontal splicing, it can achieve large-field-of-view, large-size AR display images at a lower hardware cost, meeting the needs of wide-field-of-view immersive driving. By setting a 1%~3% transition band in the overlapping area of ​​the image and employing brightness gradient feathering, pixel interpolation, and grayscale uniformity processing, combined with multi-optical-engine synchronous clock control, it effectively eliminates splicing seams, misalignments, ghosting, and image tearing, making the spliced ​​area visually seamless and achieving imperceptible splicing. Regarding image stability, this invention collects vehicle posture signals and performs motion prediction, performing reverse motion pre-compensation for pitch, tilt, and bumpy / jumping motions on the projected image. This ensures that the virtual image remains stably anchored to the road surface under various dynamic conditions, without drifting or shaking with vehicle movement, significantly improving the accuracy of virtual-real fusion. In terms of image quality, by dividing the large screen into a central main display area, an edge auxiliary area, and a bottom information area for differentiated rendering, and calling the optical distortion map table to perform real-time pixel-level correction on the edge areas, pincushion distortion, barrel distortion, and ghosting blur are effectively eliminated, ensuring that the entire screen is clear and uniform. Attached Figure Description

[0017] To more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely exemplary, and those skilled in the art can derive other embodiments based on the provided drawings without creative effort.

[0018] Figure 1 This is a schematic diagram of the optimal embodiment of the control method for a large AR-HUD projection screen in a car according to the present invention. Detailed Implementation

[0019] 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 embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] This invention provides a control method for a large-screen AR-HUD projection in automobiles. The following detailed description, using a complete specific embodiment, outlines each S element of the method and its preferred implementation.

[0021] S1: System initialization and calibration, loading optical calibration data and completing the initial calibration of vehicle attitude, IMU inertial navigation, and camera.

[0022] After the vehicle is powered on, the AR-HUD control system automatically loads the optical calibration data pre-stored in non-volatile memory. This optical calibration data includes, but is not limited to, distortion mapping tables, brightness gamma tables, virtual image distance parameters, and multi-optical-mechanical stitching alignment parameters. Simultaneously, the system completes initial vehicle attitude calibration, i.e., by reading the vehicle's levelness, pitch angle, and roll angle at the initial moment as the reference zero point. Zero-bias calibration is performed on the IMU inertial navigation system, i.e., continuously acquiring gyroscope and accelerometer data in a stationary state, calculating and subtracting constant zero-bias. Intrinsic and extrinsic parameter calibrations are performed on the camera, including intrinsic parameter calibration (focal length, principal point coordinates, distortion coefficients) and extrinsic parameter calibration (camera mounting position and angle relative to the vehicle coordinate system). Through the above initialization, the system ensures that it acquires accurate reference data.

[0023] S2: Real-time acquisition of multi-sensor data, obtaining vehicle attitude signals, environmental signals, driver status signals, and driving signals. The system acquires these three types of signals in real time at a frequency of no less than 100 Hz. The first type is vehicle attitude signals, specifically including three-axis acceleration and three-axis angular velocity from the IMU inertial navigation system, four-wheel height sensor values ​​from the suspension system, suspension travel values ​​from the active suspension or body control module, and steering angle from the steering wheel angle sensor. The second type is environmental signals, including ambient light intensity collected by a light intensity sensor, solar incidence angle and azimuth angle collected by a sunlight sensor, and day / night status determined by a combination of time and light intensity. The third type is driver status signals, mainly captured by the driver monitoring system (DMS) camera, which uses image processing algorithms to detect the precise coordinates of the pupils in three-dimensional space. The fourth type is driving signals, including vehicle speed, gear position signals, navigation route planning data read from the CAN bus, and obstacle target list and lane line fitting parameters from the Advanced Driver Assistance System (ADAS).

[0024] S3: Vehicle Motion Prediction and Inverse Image Compensation. Based on the collected vehicle motion data, the system predicts the attitude change in the next moment and performs inverse motion pre-compensation on the projected image to stabilize the virtual image on the road surface. The system inputs the latest set of vehicle motion data collected in S2, including pitch angular velocity, roll angular velocity, vertical acceleration, and four-wheel height change rate, into a Kalman filter or motion prediction model to predict the vehicle attitude change within a time window of 10 to 50 milliseconds. Based on the prediction results, inverse motion pre-compensation is performed on the virtual image to be projected. Specifically, when a pitch change is predicted (i.e., the front of the vehicle dips or rises), the image is compensated for in the vertical direction; if the front of the vehicle dips, the image is moved upwards, and vice versa. When a roll change is predicted (i.e., the left and right suspension heights are inconsistent), the image is compensated for in the horizontal direction; if the vehicle tilts to the left, the image is moved to the right. When a sudden change in vertical acceleration is predicted that the vehicle body will experience bumps and jumps, the position and scaling factor of the image are dynamically corrected. The specific correction coefficient is calculated based on the bump amplitude and vehicle speed, so that the virtual image is always stably anchored to the target position on the real road surface. For example, the navigation arrow always points to the lane entrance, and does not drift up and down or shake left and right with the vehicle's movement.

[0025] S4: Large-screen partitioned rendering and distortion correction. The large screen to be displayed is divided into a central main display area, an edge auxiliary area, and a bottom information area for differentiated processing, and an optical distortion mapping table is called for real-time correction.

[0026] This invention divides the final large-screen display into three functional areas. The first area is the central main display area, located directly in front of the driver's line of sight, occupying approximately 50% to 70% of the screen width. This area is assigned the highest rendering resolution, such as 1920×720 pixels, and the highest contrast ratio, for displaying navigation guide arrows, lane line enhancement lines, and obstacle warning boxes—information most critical to driving safety. The second area is the edge auxiliary area, covering approximately 15% of the width of each of the left and right edges of the screen. This area uses a pre-stored optical distortion mapping table, obtained through offline calibration, which records the offset of each pixel from its ideal position to its actual projected position. For each pixel in the edge auxiliary area, the system performs a reverse lookup interpolation based on the distortion mapping table to correct pincushion or barrel distortion introduced by the HUD optical system in real time, while eliminating ghosting and blurring. The third area is the bottom information area, located at approximately 10% of the screen height at the bottom edge, for displaying auxiliary information such as vehicle speed, speed limit signs, and traffic signs. Through the aforementioned zone-based differentiated rendering and pixel-level distortion correction, a clear, uniform, and distortion-free display effect is achieved across the entire screen.

[0027] S5: Dynamic image adaptation based on human eye position. Based on the driver's binocular spatial coordinates detected by the DMS camera, the system adjusts the position of the projected image, the optical engine output angle, and the field of view in real time. The system obtains the driver's binocular spatial coordinates from the DMS camera in S2. These coordinates are referenced to the vehicle coordinate system, for example, with the steering wheel center as the origin, forward as the positive X-axis, right as the positive Y-axis, and upward as the positive Z-axis. Based on the average position of the binocular coordinates, i.e., the center points of the left and right eyes, the system calculates the following adjustments in real time: vertical offset of the projected image to align the image centerline with the driver's eye level; horizontal offset of the projected image to align the horizontal center of the image directly in front of the driver; horizontal and vertical deflection angles of the optical engine output angle, achieved by controlling the angles of micromirrors or reflectors within the optical engine; and cropping and scaling of the field of view. If the driver's eyes are far from the windshield, the displayed content is appropriately reduced to avoid exceeding the boundaries; if the driver's eyes are close, the content is appropriately enlarged to ensure complete visibility of the information. For example, when the system detects that a driver who is 1.6 meters tall is sitting too low, it automatically raises the view by 3 degrees and increases the vertical field of view output to ensure that the driver can see the entire screen without having to look up or down.

[0028] S6: Ambient light adaptive and anti-glare control, dynamically adjusts screen brightness and contrast according to ambient light intensity and sun incidence angle.

[0029] The system monitors the ambient light intensity (in lux) obtained from S2 in real time, as well as the angle of sunlight (the angle between the sunlight and the vehicle's windshield normal). Based on these parameters, it executes three adaptive control strategies. The first strategy: When the ambient light intensity exceeds 20,000 lux and is determined to be strong daylight, it automatically increases the backlight brightness of the central main display area to 90% to 100% of its maximum value, while simultaneously increasing the contrast to a preset high-contrast level to ensure clear visibility of road information such as navigation lane lines. The second strategy: When the ambient light intensity is below 100 lux or the tunnel sensor is triggered (i.e., at night or in a tunnel environment), it reduces the overall screen brightness to 30% to 50% of its maximum value and forcibly disables all unnecessary dynamic high-brightness elements, such as flashing decorative lighting effects, retaining only key warning information flashing at a lower brightness. The third strategy: It dynamically adjusts the local backlight brightness according to the angle of sunlight. When sunlight shines obliquely from the side windows, causing localized areas of the windshield to appear white, it effectively suppresses this whitening phenomenon by increasing the backlight brightness of the corresponding display area by 20% to 30% and reducing the contrast of that area, ensuring screen readability.

[0030] S7: Large screen content layering and safety strategy, which divides the displayed content into at least three levels according to the importance of driving, and displays them differently in different areas of the screen.

[0031] This invention divides all displayed content into three levels based on the importance of the driving task. Level 1 core information includes navigation guide arrows, lane keeping assist lines, and forward collision warning boxes. This information is displayed in the center of the main display area with high brightness, high saturation, and dynamic flashing to ensure the driver perceives it immediately. Level 2 routine information includes current vehicle speed, road speed limit signs, and traffic sign recognition results. This information is continuously displayed in the bottom information area with stable, medium brightness, not actively attracting attention but always readable. Level 3 extended information includes multimedia entertainment information such as song titles, air conditioning adjustment feedback, and vehicle setting prompts. This information is only briefly displayed in the lower left or right corner of the screen with low brightness and low saturation when the driver actively requests it or under specific conditions; it is hidden by default. Through this layered and zoned display, information overload is avoided, ensuring the driver's visual attention is always focused on the road ahead and the level 1 core information, conforming to automotive-grade safety design principles.

[0032] S8: Multi-screen splicing and seam elimination. By selecting the splicing mode, the overlapping areas of the images are processed for fusion, while maintaining synchronization of multiple optical cameras.

[0033] This embodiment selects a splicing mode based on hardware configuration and display requirements. In one mode, two optical engines are used for horizontal splicing, with each engine responsible for half of the horizontal field of view. The total horizontal field of view after splicing reaches 20 to 30 degrees, breaking through the typical 12 to 15 degree limitation of a single optical engine. In another mode, multiple optical engines are used for vertical splicing, with one engine handling the far-field AR image at a distance of over 10 meters, displaying navigation and warnings; the other engine handles the near-field instrument panel image at a distance of approximately 2.5 meters, displaying vehicle speed and driving data. In yet another mode, only a single high-resolution optical engine is used. The image source is logically divided into left and right regions by software, rendered separately, and then combined for output, achieving virtual splicing within a single optical engine. Regardless of the splicing mode used, an overlapping transition band is set at the seam between adjacent images. The width of this transition band is one percent to three percent of the width of a single image. For example, for an image 1920 pixels wide, the overlapping area width is 19 to 58 pixels. Within the overlapping area, the system performs fusion processing: pixel values ​​from both the left and right images are feathered with a gradual brightness gradient, meaning the brightness decreases linearly or non-linearly from 100% at one edge of the overlapping area to 0% at the other edge, and vice versa; bilinear interpolation is performed on the pixels in the overlapping area, taking the weighted average of the pixels from both images, with the weighting coefficient proportional to the pixel's position within the overlapping area; grayscale uniformity processing is performed to ensure that the average brightness and contrast of the entire image after stitching does not jump. Simultaneously, to ensure complete synchronization of each frame output by the multi-optical engine, the system adopts a common-source clock synchronization scheme, where all optical engines share a high-precision clock source, and frame synchronization signals are generated based on this clock source, ensuring that the frame refresh time error of each optical engine is less than 1 microsecond, fundamentally eliminating image misalignment and tearing. All optical engines participating in the stitching share the same set of attitude compensation parameters; that is, the pitch, tilt, and turbulence compensation amounts calculated in S3 are simultaneously applied to each image, ensuring overall image synchronization and stability.

[0034] S9: Output and closed-loop feedback, sending the processed image to the AR-HUD optical engine projection and sampling the image effect in real time for closed-loop correction.

[0035] The final image data processed by steps S1-S8 is then transmitted to the display driver module of the AR-HUD optical engine via a video transmission interface such as LVDS or GMSL for projection. Simultaneously, the system activates a closed-loop feedback mechanism: using a miniature camera installed in the vehicle or utilizing the optical sensors built into the HUD, it continuously samples the actual display effect of the virtual image projected onto the windshield. The sampled data includes quantitative indicators of image stability, such as the number of pixels shifted at the edges of adjacent frames; quantitative indicators of stitching effect, such as the maximum brightness difference within the overlapping area; and alignment errors, such as the lateral deviation between the virtual lane lines and the actual lane lines during virtual-real fusion. The sampled data is compared with the expected set values. If the stability error exceeds a threshold (e.g., image jitter greater than 2 pixels), the inverse compensation parameters in step S3 are automatically fine-tuned; if the stitching brightness difference is greater than 5%, the feathering curve is recalibrated; if the alignment error exceeds 0.1 degrees of viewing angle, the recalibration in step S1 or the correction of the eye position coordinates in step S5 is triggered. Through this real-time sampling and closed-loop correction, the system can maintain optimal display performance during long-term vehicle operation.

[0036] The specific implementation of the present invention has been described in detail above with reference to a preferred embodiment. It should be understood that those skilled in the art can adjust the specific parameter values ​​in each step according to actual system requirements, such as sampling frequency, overlap area width, brightness threshold, etc., without departing from the protection scope of the present invention. The core idea of ​​the present invention lies in systematically solving the problems of image stability, clarity, splicing, and security in large field-of-view AR-HUD displays through a series of methods such as vehicle motion prediction and inverse compensation, partitioned rendering and distortion correction, dynamic adaptation of human eye position, multi-sensor fusion, multi-screen splicing, and closed-loop feedback.

[0037] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. The above descriptions are only preferred embodiments of this application. It should be noted that due to the limitations of written expression, while there are objectively infinite specific structures, those skilled in the art can make several improvements, modifications, or changes without departing from the principles of this application, and can also combine the above technical features in an appropriate manner. These improvements, modifications, changes, or combinations, or the direct application of the inventive concept and technical solution to other situations without modification, should all be considered within the scope of protection of this application.

Claims

1. A control method for a large-screen AR-HUD projection in a car, characterized in that, include: S1: System initialization and calibration, loading optical calibration data and completing the initial calibration of vehicle attitude, IMU inertial navigation, and camera; S2: Real-time acquisition of multi-sensor data to obtain vehicle attitude signals, environmental signals, driver status signals and driving signals; S3: Vehicle motion prediction and image inverse compensation. Based on the collected vehicle motion data, predict the attitude change of the next moment and perform inverse motion pre-compensation on the projected image to make the virtual image stably anchored on the road surface. S4: Large-screen partition rendering and distortion correction, which divides the large screen to be displayed into a central main display area, an edge auxiliary area and a bottom information area for differentiated processing, and calls the optical distortion mapping table for real-time correction. S5: Dynamic image adaptation based on human eye position. Based on the driver's eye spatial coordinates detected by the DMS camera, the position of the projected image, the optical engine output angle and the field of view are adjusted in real time. S6: Ambient light adaptive and anti-glare control, dynamically adjusts screen brightness and contrast according to ambient light intensity and sun incidence angle; S7: Layered display of large-screen content and safety strategy, which divides the displayed content into at least three levels according to the importance of driving, and displays them differently in different areas of the screen; S8: Multi-screen splicing and seam elimination. By selecting the splicing mode, the overlapping areas of the images are processed for fusion, while maintaining synchronization of multiple optical engines. S9: Output and closed-loop feedback, sending the processed image to the AR-HUD optical engine projection and sampling the image effect in real time for closed-loop correction.

2. The control method for a large AR-HUD projection screen in a car according to claim 1, characterized in that, The differentiated processing in S4 specifically includes: the central main display area displays navigation guidance, lane lines, and obstacle warnings with high resolution and high contrast; the edge auxiliary area calls the optical distortion mapping table to correct pincushion or barrel distortion in real time; and the bottom information area displays vehicle speed, speed limit, and signage auxiliary information.

3. The control method for a large AR-HUD projection screen in a car according to claim 1, characterized in that, The splicing mode selection in S8 includes: horizontal splicing of dual optical engines to expand the horizontal field of view, vertical splicing of multiple optical engines to achieve layered splicing of far-field AR images and near-field instrument images, or software partitioning splicing of single optical engines.

4. The control method for a large AR-HUD projection screen in a car according to claim 1, characterized in that, The fusion processing of the overlapping area in S8 includes: setting a 1% to 3% overlapping transition band, and eliminating the seam through brightness gradient feathering, pixel interpolation and grayscale uniformization; the multi-optical-machine synchronization adopts the same source clock synchronization to ensure the frame refresh synchronization of the multi-optical-machine.

5. The control method for a large AR-HUD projection screen in a car according to claim 1, characterized in that, The reverse motion pre-compensation in S3 specifically includes: performing vertical reverse compensation on the image based on the vehicle's pitch change; performing horizontal reverse compensation on the image based on the vehicle's roll change; and dynamically correcting the image position and zoom based on the vehicle's bumps and jumps.

6. The control method for a large AR-HUD projection screen in a car according to claim 1, characterized in that, The ambient light adaptive and anti-glare control in S6 specifically includes: automatically increasing the brightness and contrast of the central area in a strong daytime environment; reducing the overall brightness and turning off unnecessary high-brightness elements in a nighttime or tunnel environment; and dynamically adjusting the backlight according to the angle of sunlight incidence.

7. The control method for a large AR-HUD projection screen in a car according to claim 1, characterized in that, The content layering display and security strategy in S7 are as follows: the first-level core information is displayed in the center with high brightness; the second-level regular information is displayed stably at the bottom of the screen; and the third-level extended information is displayed in the corner of the screen with low brightness.

8. The control method for a large AR-HUD projection screen in a car according to claim 1, characterized in that, The dynamic image adaptation based on human eye position in S5 specifically includes adjusting the vertical position, horizontal position, optical engine output angle and offset, and cropping and scaling of the field of view of the projected image in real time according to the spatial coordinates of the driver's eyes detected by the DMS camera.

9. A control method for a large AR-HUD projection screen in a car according to claim 1, characterized in that, The closed-loop feedback in S9 specifically refers to: the stability of the real-time captured image, the splicing effect and alignment error, and the closed-loop correction based on the captured image results to continuously maintain the display effect.

10. A control method for a large AR-HUD projection screen in a car according to claim 1, characterized in that, The multi-sensor data collected in S2 specifically includes: vehicle attitude signals composed of acceleration and angular velocity, four-wheel height, suspension travel, and steering angle from the IMU; environmental signals composed of light intensity, sunlight incidence angle, and day / night status; driver status signals composed of the position of the human eye detected by the DMS camera; and driving signals composed of vehicle speed, gear, navigation path, ADAS obstacle and lane information.

Images