A method and system for adaptive control of continuity light blanket in a dynamic scene

Abstract

This invention discloses an adaptive control method and system for continuous light carpet in dynamic scenes, relating to the field of vehicle lighting technology. The method includes: first, locking the unique identifier of the target lane to complete target locking; second, comparing the lane line list corresponding to each frame of the video stream with the target lane identifier in real time; third, in case of lane line data loss, activating a timing and fault-tolerance mechanism to freeze the control parameters of the previous frame of the light carpet; and fourth, executing light carpet safety shutdown and cooling control after a timeout. This method establishes a stable benchmark by locking the unique identifier of the target lane, avoiding target drift caused by perceived jitter; for temporary lane line loss, it adopts "timing + parameter freezing" to overcome the rigidity defects of traditional methods; and through timeout judgment and cooling period control, it forms a complete control closed loop of "locking-comparison-fault tolerance-stable output," significantly improving the continuity and anti-interference capability of the light carpet display, and is suitable for light carpet guidance control in dynamic driving scenarios.

Description

Technical Field

[0001] This invention relates to the field of vehicle lighting technology, and in particular to an adaptive control method and system for a continuous light carpet in dynamic scenes. Background Technology

[0002] High-pixel intelligent headlights based on DLP (Digital Light Processing) or Micro-LED technology can project high-resolution patterns, or "light carpets," onto the road surface. Light carpet technology can provide drivers with functions such as lane keeping assist, lane change guidance, and lane departure warning, significantly improving driving safety and experience. Currently, most light carpet control solutions on the market adopt a "single trigger-response" mode. Specifically, when the vehicle detects a specific signal (such as a turn signal being activated), it directly invokes the corresponding light carpet pattern for projection. When the perception system does not return valid lane line data, the system immediately shuts off the light carpet projection. This basic logic is commonly found in mass-produced vehicles with ADB (Adaptive High Beam) functionality.

[0003] However, the above-mentioned traditional solutions have significant drawbacks in complex actual working conditions such as dynamic lane changes: poor continuity of light carpet projection and easy occurrence of high-frequency visual flicker, which seriously affects driving safety.

[0004] Specifically, when a vehicle performs a dynamic lane change, the lane line information captured by the forward-looking vision camera is highly susceptible to interference from factors such as road surface reflection, worn lane markings, shadows, vehicle swaying, and curve deformation. This leads to frequent momentary loss of lane line detection, confidence jumps, and ID errors. Traditional control logic directly and forcefully binds the validity of the perceived data to the on / off state of the light carpet, employing a rigid strategy of "disabling upon failure of one frame and turning on upon recovery of the next frame." This results in the light carpet exhibiting high-frequency, irregular, and uncontrollable flickering on the road surface.

[0005] Existing technologies typically only employ simple methods such as trajectory holding, brightness holding, and data caching to handle short-term sensor failures. Therefore, when there are short-term sensor jitters, data jumps, or frequent lane changes, it is impossible to guarantee continuous, stable, flicker-free, and misaligned light carpet rendering.

[0006] The human eye is highly sensitive to sudden brightening and fading of light spots within the field of vision. Such flickering not only fails to provide stable and reliable lane change guidance assistance, but also creates strong visual interference, seriously distracts the driver's attention, and may even cause secondary safety risks such as glare and misjudgment, which violates the original design intention of the light carpet as a driving assistance device. Summary of the Invention

[0007] The technical problem this invention aims to solve is: to address issues such as light carpet flickering, shape distortion, position misalignment, state oscillation, and ping-pong effect caused by sensor noise, short-term data failure, frequent state switching, and projection jumps during dynamic lane changes. This invention provides an adaptive control method and system for continuous light carpets in dynamic scenes, achieving continuous, stable, flicker-free, misalignment-free, jump-free, and distortion-free light carpets throughout the entire dynamic lane change process from five dimensions: rendering preservation, state management, lane locking, cooling and anti-shake, and preemption arbitration.

[0008] The technical solution adopted by this invention to solve its technical problem is: an adaptive control method for a continuous light carpet in a dynamic scene, comprising the following steps: S1. Target Lock: At the instant the lane change triggering conditions of the light carpet are met, read and lock the unique identifier of the target lane.

[0009] S2. Real-time comparison: In each frame of the lane change process, continuously compare the lane line list currently output by the vehicle (containing the unique identifier of the lane) with the unique identifier of the target lane in real time.

[0010] S3. Abnormal timing and fault tolerance: If a matching target lane unique identifier is not found in the latest frame, the target lane data is determined to be lost. Instead of performing the light carpet closing action, a data loss timer is immediately started or accumulated.

[0011] During the period before the data loss timer reaches the preset threshold, the system is forcibly frozen and the valid control parameters of the previous frame are maintained.

[0012] S4. Timeout Exit: When the accumulated duration of the data loss timer exceeds the threshold, it indicates that an irreversible substantial change has occurred in the perceived environment, and it will be judged as "environment failure" and the safe shutdown procedure of the light carpet will be executed.

[0013] After exiting the dynamic lane-changing state, a cooling-off period mechanism is adopted to avoid jitter caused by uninterrupted switching.

[0014] This method locks the target lane during dynamic lane changes, providing a stable baseline for the lane-changing process and preventing frequent drift of the target lane due to sensor jitter. It overcomes the rigidity of traditional "disconnect upon loss" solutions by not immediately exiting when lane lines are briefly lost, ensuring the lane-changing process is not interrupted by momentary interference. By maintaining the execution of parameters from the previous frame within the timing cycle, it achieves inertial continuation of the lane-changing operation, significantly improving continuity and stability. By only determining environmental failure after a timeout, it avoids accidental and frequent exits, enhancing the system's anti-interference capability. By entering a cooling-off period after exiting and prohibiting re-entry into the same lane-changing state, it fundamentally avoids repeated state switching caused by lane line fluctuations (the ping-pong effect). This forms a complete, stable, and anti-interference lane-changing control closed loop, achieving a significantly different approach from the passive fallback and simple hold methods of existing technologies.

[0015] In some embodiments, in step S3, the effective control parameters of the previous frame include lane line parameters and light carpet rendering parameters.

[0016] This method integrates the inertial holding mechanism with both lane change trajectory control and light carpet visual guidance, achieving synchronous and continuous vehicle actions and visual cues. This avoids driver misjudgment caused by asynchronous lane change actions and light carpet display, further improving driving safety.

[0017] In some embodiments, the light carpet rendering parameters are generated based on a smooth curve determined according to the light carpet emission point (starting anchor point) at the current position of the vehicle, the target position point (ending anchor point) of the target lane, and intermediate constraint control points.

[0018] The edges of the light carpet are constructed using smooth curves, replacing the traditional rectangular and mechanically shifted light spots, resulting in a more natural visual appearance that better meets driving expectations. The curves are generated based on the position and tangent direction, allowing the shape of the light carpet to change naturally and gradually with the vehicle's movement, without abrupt changes or abrupt transitions. This improves the readability and comfort of the light carpet guidance, reduces visual interference, and truly achieves assistance rather than glare.

[0019] In some embodiments, the smooth curve is preferably a third-order Bézier curve, which generates the edge shape of the lane change light carpet and further improves the visual quality of the light carpet.

[0020] In some embodiments, in step S1, a preemptive priority state machine triggers lane changes according to priority to achieve orderly switching of the light carpet function state under different driving scenarios. Priority, from highest to lowest, is as follows: (1.) Dynamic lane change status: handles lane change guidance, with the highest priority.

[0021] (2). Following vehicle status: Maintain distance to the vehicle in front and cut off the light blanket.

[0022] (3). Width indication state: Narrow road passage assistance.

[0023] (4) Straight ahead: Basic lane keeping guidance.

[0024] (5) Default state: No light blanket or basic lighting.

[0025] By using a preemptive priority state machine, the highest execution authority for lane change guidance is guaranteed, and lane change requests can be responded to immediately without delay. After exiting with high priority, the system automatically and smoothly returns to normal, enabling orderly switching of light carpets in multiple scenarios, avoiding state conflicts and chaos, and making the switching of light carpet modes such as lane change, following, width indication, and straight driving natural, stable, and safe.

[0026] This invention also discloses an adaptive control system for a continuous light carpet in a dynamic scene, comprising: The perception data acquisition module acquires real-time vehicle status data and environmental data. The state machine logic control module determines the current state of the light carpet function to be activated based on the preemptive priority state machine according to the sensing data, manages the transition between states, and executes fault-tolerant logic within the state. The rendering strategy calculation module calculates the light carpet rendering parameters based on the light carpet target parameters output by the preemptive priority state machine, and then calculates the projected image based on the light carpet rendering parameters through perspective transformation and outputs it to the vehicle's light carpet projection execution module.

[0027] This system adopts a modular architecture to achieve a complete anti-interference closed loop with five layers: locking, timing, inertial holding, exiting, and cooling. Each module works together to achieve uninterrupted, flicker-free, oscillating, and non-frequent switching throughout the dynamic lane changing process. It can be directly integrated into intelligent headlights, ADAS, or domain controllers, with strong hardware adaptability and easy mass production.

[0028] In some embodiments, the vehicle status data includes vehicle speed, steering signal, steering wheel angle, etc.; the vehicle status data serves as the triggering condition for the light carpet function, used to determine whether the lane change triggering condition is met. The environmental data includes a list of lane lines (lane line polynomial coefficients, lane unique identifiers), distance to the vehicle in front, etc.

[0029] In some embodiments, the management status includes priority arbitration; The management status also includes the light blanket lifecycle and anti-shake control.

[0030] In some embodiments, the light carpet rendering parameters are calculated based on the light carpet target parameters; The target parameters of the light carpet are determined based on the locked target lane information.

[0031] The beneficial effects of this invention are: 1. Significantly improved visual continuity: Through inertial retention, it covers most of the instantaneous sensor vibration scenarios in actual driving (road reflection, bumps, lane line wear, etc.), transforming the original "data loss → light off → data back → light on" flickering into a visually imperceptible smooth transition; solving the industry pain point of high-frequency flickering and frequent switching of the light carpet during dynamic lane changes. 2. Improved safety: Eliminates the safety hazard of driver distraction caused by the flickering of the light carpet, making the light carpet a truly effective driving assistance tool; 3. Enhanced system robustness: The cooling period mechanism further prevents frequent system oscillations when sensor data is near boundary values. Attached Figure Description

[0032] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0033] Figure 1 This is a framework diagram of the control system of the present invention.

[0034] Figure 2 This is a flowchart of the control method of the present invention.

[0035] Figure 3 This is a flowchart of the light blanket function state switching process of the present invention. Detailed Implementation

[0036] The present invention will now be described in further detail with reference to the accompanying drawings. These drawings are simplified schematic diagrams, illustrating only the basic structure of the invention, and therefore only show the components relevant to the invention.

[0037] Example 1: like Figure 1 , Figure 3 The image shows an embodiment of the present invention, an adaptive control system for a continuous light carpet in a dynamic scene, comprising a sensing data acquisition module, a state machine logic control module, and a rendering strategy calculation module.

[0038] In this embodiment, the perception data acquisition module acquires real-time vehicle status data, including vehicle speed, steering signal, and steering wheel angle, as well as environmental data, including lane line polynomial coefficients, lane unique identifiers (Target_Lane_ID), and distance to the vehicle in front, from the vehicle's CAN bus and vision perception system. Specifically, the vehicle CAN bus acquires real-time physical signals such as vehicle speed, steering signal, and steering wheel angle collected by the vehicle speed sensor and turn signal stalk. The forward-looking vision camera and perception computing platform in the vehicle's vision perception system acquire and process lane line data (lane line polynomial coefficients, lane unique identifiers, and distance to the vehicle in front). The vehicle's intelligent headlight domain controller and its RAM are used to instantaneously write and persistently store the locked Target_Lane_ID.

[0039] In this embodiment, the state machine logic control module, as the core of the system, is responsible for determining the current state of the light carpet function to be activated based on the perception data (vehicle status data and environmental data) acquired by the perception data acquisition module, managing the transitions between states, and executing fault-tolerant logic within each state. Specifically, the state machine logic control module employs a multi-level state machine coordination mechanism when determining the current state of the light carpet function to be activated, both to support the orderly switching of the light carpet in different driving scenarios, such as... Figure 3 As shown, a set of states with priorities is predefined, ordered from highest to lowest priority as follows: (1) Dynamic lane change status (DynamicChange): handles lane change guidance, with the highest priority.

[0040] (2). Follow status: Maintain distance to the vehicle in front and cut off the light blanket.

[0041] (3). Width indication (Body): Narrow road passage assistance.

[0042] (4) Straight: Basic lane keeping guidance.

[0043] (5) Default state: No light blanket or basic lighting. The above priority ranking strictly follows the urgency of driving safety. For example, if active lane changing involves a high risk of lateral collision, trajectory guidance must be provided first; if following another vehicle involves anti-glare regulations, the light carpet must be cut off at any time to avoid affecting the vehicle in front; the need for near-end anti-scratching for width indication takes precedence over far-end cruise control for straight-ahead driving. The ranking method perfectly matches the shift in human attention caused by sudden road conditions, ensuring that the light carpet presentation best aligns with current driving intuition and safety logic. The state switching logic in the preemptive priority state machine adopts a "preemptive" mechanism to execute a multi-level state machine coordination mechanism. For example, when in "straight-ahead state," if the turn signal is detected and the lane changing conditions are met, it immediately transitions to "dynamic lane changing state." When a high-priority state exits, the system automatically falls back to the lower-priority state that meets the conditions.

[0044] The management state, as a mechanism for the coordinated scheduling and conflict resolution of various light carpet functions (such as straight driving, following, lane changing, etc.), specifically includes two aspects: Priority arbitration: Compared to the existing technology that uses simple condition stacking (if-else) for light carpet state switching (which often results in overlapping, chaotic images or stiff back-and-forth flickering when multiple functions are triggered simultaneously in complex road conditions, lacking a smooth "preemption and state fallback" mechanism), this solution preemptively switches according to a preset priority (such as lane changing > following > straight driving) when the vehicle simultaneously meets multiple light carpet triggering conditions, avoiding light carpet crashes or distorted forms when functions are mutually exclusive. Lifecycle and anti-shake control: Responsible for scheduling the "entry, rendering, and exit" of each light carpet state. Through the "multi-frame confirmation" mechanism upon entry and the "cooling-off period" mechanism after exit, data glitches from sensors (a combination of onboard sensors responsible for vehicle state data and environmental data, including vehicle speed sensors, forward vision cameras, etc.) are filtered to prevent the light carpet from flickering or jumping frequently under critical indicators.

[0045] In this embodiment, the rendering strategy calculation module calculates the light carpet rendering parameters based on the light carpet target parameters output by the preemptive priority state machine, and then calculates the projected image based on the light carpet rendering parameters through perspective transformation and outputs it to the light carpet projection execution module.

[0046] For example, the specific rules and parameterized configurations for entering and switching the light carpet function state are as follows: 1. Dynamic lane change light blanket (highest priority, most rigorous implementation) Entry requirements configuration: Three hard security criteria must be met simultaneously: Speed ​​threshold: enter_speed_threshold_kmh = 30 (intercepts invalid triggers when crawling at low speeds).

[0047] Steering wheel angle: less than enter_steering_wheel_max_angle = 30 degrees (excluding misjudgments of large curvature steering behaviors such as U-turns).

[0048] Target lane line exists: The sensor detected that the Target_Lane_ID of the target lane is valid.

[0049] Switching rules: When in any basic light carpet (straight / following) state, if the above three hard safety indicators are all met, since the dynamic lane change has the highest preemption level, the preemption priority state machine will immediately suspend the current state (output light carpet target parameters), and the rendering strategy calculation module will execute the third-order Bézier curve to cut into the target lane for rendering (obtain light carpet rendering parameters).

[0050] 2. Straight-line light carpet and following light carpet (basic functions) Straight driving (basic cruise): requires vehicle speed >30km / h, vehicle not to cross the line and steering wheel angle to be very small (<15 degrees).

[0051] Follow mode (anti-glare forced intervention): The trigger threshold is lower (enter_speed_threshold_kmh = 20). When the vehicle in front is detected in the environmental data, the end of the light carpet (end_x) will automatically and dynamically retract to always maintain a safe distance of 1 meter from the rear of the vehicle in front (distacne_between_car_ahead_and_carpet_tail_m = 1 meter).

[0052] Example 2: like Figure 2 , Figure 3 The image shows Embodiment 2 of the present invention. Based on Embodiment 1, it provides an adaptive control method for a continuous light carpet in a dynamic scene, comprising the following steps: S1. Target Lock: Dynamic lane change status is activated. Based on locking onto the target lane, the system anchors relevant information about the target lane, continuously anchoring the core target of the lane change. The specific execution is as follows: Scene and Priority Arbitration: In the state machine logic control module, the preemptive priority state machine ranks the scenes from highest to lowest priority as follows: Dynamic Lane Change > Following Vehicle > Wiring Warning > Straight Ahead > Default State. This arbitrates the scenes, ensuring that dynamic lane change is the highest priority guiding task, lane change intentions can be responded to instantly, and are not blocked by lower priority functions; multi-scene light carpets automatically and orderly switch, and high-priority exits smoothly, without state conflicts, logical confusion, or rendering errors.

[0053] Dynamic lane change triggering: When the perception data acquisition module detects that data such as vehicle speed, steering signal, steering wheel angle, lane line polynomial coefficient, lane unique identifier (Target_Lane_ID), and distance to the vehicle in front meet the lane change triggering conditions (vehicle speed meets the standard, turn signal is on, target lane is valid), the system immediately reads and locks the unique identifier (Target_Lane_ID) of the target lane. All subsequent rendering logic is based on this locked Target_Lane_ID, rather than directly using the real-time detection results that may change in the next frame. By providing a stable and unchanging reference benchmark for lane changes, frequent drift of the lane change target caused by lane line jumps and perception jitter is avoided, ensuring clear direction and controllable trajectory throughout the lane change process, thus improving the reliability and safety of lane changes.

[0054] S2. Real-time Comparison: Based on the locked target lane, during dynamic lane changing, the perception data acquisition module and the state machine logic control module work together to perform high-frequency real-time comparison operations in each frame to ensure the effectiveness of perception and the stability of light carpet rendering. The specific execution is as follows: In the perception data acquisition module, the forward-looking vision camera continuously collects video streams at a frame rate, and the perception computing platform outputs the latest lane line list frame by frame (including lane line polynomial coefficients and lane unique identifier IDs; the lane line list is generated by the perception computing platform based on the video streams collected by the forward-looking vision camera).

[0055] In each frame of the lane change process, the state machine logic control module accurately compares the latest lane line list output by the perception data acquisition module with the locked target lane Target_Lane_ID to determine whether the lane line Target_Lane_ID matches.

[0056] S3. Abnormal Timing and Fault Tolerance: If no matching target lane Target_Lane_ID is found in the latest frame comparison, the state machine logic control module executes fault tolerance logic, which does not immediately shut down the light carpet rendering or terminate the lane change intention operation. Specifically, it executes as follows: Data loss response: Immediately start or increment a "data loss timer" (Loss_Timer), which is executed with high-precision microsecond-level timing by the hardware timer / clock crystal oscillator built into the vehicle lighting domain controller; Parameter Freeze and Inertial Rendering: While the Loss_Timer has not reached the preset threshold (e.g., 500ms, adapting to the time characteristics of the lane change process and avoiding premature termination or misjudgment), the state machine logic control module forcibly freezes and maintains the valid lane line parameters (curvature, heading angle, lateral displacement offset) of the previous frame, continuing light carpet rendering. This step utilizes the continuity of vehicle physical motion, effectively masking instantaneous jitter in sensor data (road surface reflection, occlusion, wear, etc.), overcoming the rigidity of traditional solutions that "turn off as soon as it's lost and turn on as soon as it's good." By keeping the valid light carpet rendering parameters of the previous frame unchanged, smooth, flicker-free, and abrupt light carpet rendering is achieved, ensuring consistent execution of lane change intentions and improving the availability and stability of the assistance system.

[0057] To further enhance the visual quality of the light carpet, the rendering strategy calculation module in this embodiment uses a third-order Cubic Bezier curve based on the target parameters of the light carpet. , , , The guide trajectory and shape contour of the light carpet (i.e., light carpet rendering parameters) are generated for the light carpet target parameters in the dynamic lane-changing state. Then, the projection image is calculated based on the light carpet rendering parameters through perspective transformation and output to the light carpet projection execution module.

[0058] The third-order Bézier curve is based on the light blanket emission point at the current position of the vehicle. Target location of the target lane and intermediate constraint control points , Calculate the guiding trajectory and morphological contour of the generated light carpet, specifically: (Starting Anchor Point): Always anchored at the origin of the vehicle's coordinate system. As the vehicle moves laterally, This represents the vehicle's current absolute physical location.

[0059] (End point anchor): The longitudinal pre-aiming distance anchored in front of the target lane. Place. Will Substitute the target lane line equation obtained from each frame refresh Find the horizontal coordinate Therefore .

[0060] (Automotive heading constraint control point): strictly constrained to within the range of... Starting from the vehicle's current real-time heading angle, the ray is emanating from that angle. The quantification calculation formula is: (in This is the near-end gravity weighting coefficient; The vehicle's real-time heading angle vector, collected and output in real time by the vehicle's body sensors, reflects the vehicle's real-time driving attitude and is the core reference for constraining the position of P1. This rule ensures that the light blanket is absolutely aligned with the vehicle's front direction when it is first emitted, eliminating base shape distortion.

[0061] (Target lane tangent constraint control points): For the target lane equation in Find the first derivative at that point to obtain the tangent vector. . Strictly bound to On the extension of the reverse tangent line starting from the given point. The quantification calculation formula is: (in (This is a weighting factor for merging at the far end). This rule ensures that the light carpet can smoothly merge into the target lane at a very small angle when it extends to the far end.

[0062] It should be noted that, , For longitudinal aiming distance The output value is proportionally linked and is adjusted in real time using the vehicle speed and lateral displacement offset as adjustment factors.

[0063] Typical operating conditions: ; High operating conditions: When the vehicle speed is high (e.g., >80km / h), the system automatically increases the speed. (You can't turn the steering wheel sharply at high speeds, so the light carpet should initially be "straight enough" to give you a sense of security.) Large lateral offset adaptation: When the target lane is far away (i.e., the lateral displacement offset is large), increase... , (If the lateral displacement between two points is too large, an extremely distorted acute angle is likely to appear in the middle of the curve. Increasing the weighting coefficient can forcibly stretch the curve to both ends and straighten it, thus alleviating the severe distortion in the middle section.)

[0064] S4. Timeout Exit: Monitor the data loss timer status in real time, and execute the corresponding exit operation based on the timing result and the perception recovery status. At the same time, activate the cooldown period anti-jitter mechanism, as follows: Failure detection and exit: If the accumulated duration of Loss_Timer exceeds a threshold (e.g., 500ms) and lane line perception has not returned to normal, it indicates an irreversible substantial change in the perception environment. The system determines that the current driving environment has failed and can no longer meet the perception requirements for lane changing. It then executes the safe shutdown procedure for the light carpet, and the state machine logic control module completes the light carpet lifecycle control. This strategy avoids misjudgments, false exits, and frequent exits, transforming occasional "flickering" into stable "maintenance." Exit only occurs when the environment is truly unusable, making the logic more robust and significantly improving visual stability.

[0065] Cooling-off anti-shake mechanism: Upon exiting dynamic lane change mode, the system initiates a dedicated cooling-off period for lane changes, with a set cooling-off time threshold (e.g., 1000ms). During the cooling-off period, even if lane line perception returns to normal, the system is prohibited from re-entering dynamic lane change mode targeting the same lane. The state machine logic control module performs anti-shake control. After the cooling-off period ends, the system resumes normal lane change intent triggering function, awaiting new lane change commands. This eliminates "state ping-pong oscillation" at its source, preventing lane line fluctuations from causing repeated start-stop of the lane change function; it also prevents the light carpet from repeatedly switching on and off and flashing, further improving visual comfort and driving safety.

[0066] This solution, through the organic combination of preemptive state management, target lane locking, timing buffering, rendering parameter inertia retention, Bezier smooth deformation, and cooling-period anti-shake, differs from the rigid switch and passive fallback of existing technologies. It achieves a high-quality guidance effect of light carpet without flickering, interruption, jump, or oscillation during dynamic lane changes, significantly improving the safety, smoothness, and user experience of intelligent driving assistance.

[0067] Based on the above-described preferred embodiments of the present invention, and through the foregoing description, those skilled in the art can make various changes and modifications without departing from the inventive concept. The technical scope of this invention is not limited to the contents of the specification, but must be determined according to the scope of the claims.

Claims

1. An adaptive control method for a continuous light carpet in a dynamic scene, characterized in that, Includes the following steps: S1. Target Lock: When the lane change triggering conditions of the light carpet are met, read and lock the unique identifier of the target lane; S2. Real-time comparison: During the lane change process, analyze each frame of the video stream and continuously match and compare the current lane line list output based on the video stream with the unique identifier of the target lane in real time. S3. Abnormal timing and fault tolerance: If the current frame does not match a lane line that matches the unique identifier of the target lane, it is determined that the target lane data is lost, the light carpet is not turned off, and the data loss timer is started or accumulated. During the period before the data loss timer reaches the preset threshold, the system will forcibly freeze and maintain the valid control parameters of the previous frame. S4. Timeout Exit: When the data loss timer exceeds the preset threshold, the environment is determined to be faulty, the light blanket safety shutdown process is executed and the dynamic lane change state is exited; After exiting the dynamic lane change state, it enters a cooling-off period, during which the light blanket is not activated.

2. The adaptive control method for a continuous light carpet in a dynamic scene according to claim 1, characterized in that, In step S3, the effective control parameters of the previous frame include lane line parameters and light carpet rendering parameters.

3. The adaptive control method for a continuous light carpet in a dynamic scene according to claim 1, characterized in that, In step S1, the channel change trigger of the light blanket is arbitrated by a preemptive priority state machine according to priority of the light blanket function state. The priority of the light blanket's functional states, from highest to lowest, is as follows: (1) Dynamic lane change status: handles lane change guidance, with the highest priority; (2). Following vehicle status: Maintain distance to the vehicle in front and cut off the light blanket; (3). Width indication status: Narrow road passage assistance; (4) Straight-ahead driving: Basic lane keeping guidance; (5) Default state: No light blanket or basic lighting.

4. The adaptive control method for a continuous light carpet in a dynamic scene according to claim 2, characterized in that, The light carpet rendering parameters are generated based on smooth curves.

5. The adaptive control method for a continuous light carpet in a dynamic scene according to claim 4, characterized in that, The smooth curve is a third-order Bézier curve; The third-order Bézier curve is based on the starting anchor point. End point anchor Near-end heading constraint control points and far-end tangent constraint control points Calculate the guiding trajectory and shape contour of the generated light carpet.

6. The adaptive control method for a continuous light carpet in a dynamic scene according to claim 5, characterized in that, The , The calculation formula is: in, , These are the weighting coefficients; This is the vehicle's real-time heading angle vector; For the target lane equation in The tangent vector at the first derivative; End point anchor The longitudinal aiming distance.

7. An adaptive control system for a continuous light carpet in a dynamic scene according to the control method of any one of claims 1-6, characterized in that, include: The perception data acquisition module acquires real-time vehicle status data and environmental data. The state machine logic control module is based on a preemptive priority state machine to determine the current state of the light carpet function that should be activated based on the sensing data, manages the transition between states, and executes the fault-tolerant logic of step S3 within the state. The rendering strategy calculation module is used to calculate and generate light carpet rendering parameters based on the light carpet target parameters output by the preemptive priority state machine, and then generate a projection image based on the light carpet rendering parameters and output it to the vehicle's light carpet projection execution module.

8. The adaptive control system for a continuous light carpet in a dynamic scene according to claim 7, characterized in that, The vehicle status data includes vehicle speed, steering signal, and steering wheel angle, which serve as trigger conditions for the light carpet function. The environmental data includes a lane list and the distance to the vehicle ahead; the lane list includes a lane unique identifier.

9. The adaptive control system for a continuous light carpet in a dynamic scene according to claim 7, characterized in that, The light carpet rendering parameters are calculated using the third-order Bézier curve based on the light carpet target parameters. The target parameters of the light carpet are determined based on the locked target lane information.

10. The adaptive control system for a continuous light carpet in a dynamic scene according to claim 7, characterized in that, The management status includes management priority arbitration, timeout exit, and cooldown period control.

Images