A control method of a special long tunnel driving fatigue awakening LED wall washing lamp

Abstract

The present application relates to the field of public transport related equipment control method, and particularly relates to a long tunnel driving fatigue wake-up LED wall washing lamp control method, comprising: calibrating the driving fatigue sensitive section of each space unit of the whole tunnel, completing the alignment; updating the space unit where the vehicle is located and collecting parameters in real time, completing the grading determination of the real-time fatigue state of the driver; generating the fatigue wake-up light signal coding parameter set corresponding to the current driving tracking ID vehicle, the current fatigue state and the current space unit; driving the wall washing lamp to generate the light flow signal according to the set parameters. The present application constructs the light flow signal framework deeply bound with the driving state, synchronously realizes the double functions of fatigue wake-up and direction navigation, binds the real-time fatigue state grading of the driver with the ladder stimulation level, makes the light flow signal only act on the corresponding peripheral visual area of the tunnel side wall, does not interfere with the driver observation, and eliminates the interference of the dynamic light on the normal driving observation from the spatial distribution.

Description

Technical Field

[0001] This invention relates to the field of control methods for public transportation equipment, specifically to a control method for LED wall washer lights to wake up drivers from driver fatigue in long tunnels. Background Technology

[0002] With the continuous expansion of my country's highway network and the sustained increase in the mileage of mountainous expressways, the number and operational mileage of extra-long highway tunnels, as key structures in the highway network, are also rising year by year. Extra-long tunnels are characterized by enclosed spaces, monotonous driving environments, significant differences in internal and external lighting, and high difficulty in emergency rescue. Drivers traveling long distances through tunnels are constantly exposed to a constant and monotonous visual environment, leading to a lack of visual stimulation that easily causes inattention, visual fatigue, and even deep driving fatigue. This can result in lane departures, rear-end collisions, and other traffic accidents, making extra-long tunnels high-accident and high-risk sections of the expressway network.

[0003] In existing technologies, lighting control schemes for extra-long tunnels mostly focus on meeting basic indicators such as road surface illuminance, uniformity, and glare threshold as stipulated in current highway tunnel lighting design specifications. They generally employ static dimming modes with fixed zones, time periods, or seasonal variations, addressing only the basic visibility needs of drivers within the tunnel. They completely neglect the physiological fatigue evolution characteristics of drivers during long-distance travel, failing to proactively intervene in or alleviate driver fatigue. The constant and monotonous tunnel lighting environment can actually accelerate the accumulation of driver visual fatigue, exacerbating driving safety risks within extra-long tunnels.

[0004] Currently, some fatigue wake-up methods have emerged that combine lighting systems, but they all use dynamic lighting stimulation with uniform light flashing, color temperature switching, or fixed patterns throughout the tunnel. They fail to consider the tunnel alignment, spatial light environment distribution, and driver fatigue evolution patterns. Furthermore, they use fixed lighting parameters and stimulation patterns without considering the vehicle's real-time speed and spatial location. They also fail to achieve dynamic perception and classification of the driver's real-time fatigue state and ignore the functional boundaries of the driver's central and peripheral vision. Dynamic changes in lighting often directly affect the driver's central visual area for observing the road surface and lane lines, seriously interfering with the driver's normal judgment of the road conditions ahead. Summary of the Invention

[0005] This invention addresses the technical problems existing in the prior art by providing a method for controlling LED wall washer lights to wake up drivers from fatigue in long tunnels.

[0006] The technical solution of this invention to solve the above-mentioned technical problems is as follows: A method for controlling LED wall washer lights to wake up drivers from fatigue in long tunnels, comprising the following steps: Step S1: After standardizing the spatial unit division based on the extra-long tunnel, obtain the continuous and equidistant standardized spatial units of the entire tunnel. Based on the preset driving behavior and physiological data sample set, identify the driving fatigue sensitive sections of each spatial unit of the entire tunnel, statistically obtain the fatigue characteristics and fatigue characteristic frequency ratio of each spatial unit, and simultaneously calculate the average fatigue characteristic frequency ratio of the entire tunnel to complete the benchmarking and alignment. Step S2: The fatigue sensitivity level of each spatial unit is classified and bound to the label. The spatial visual background parameter set of each spatial unit is matched with the label one by one to build a full tunnel calibration database. A unique driving tracking ID is generated for each vehicle throughout the tunnel journey. The perception unit continuously and dynamically tracks the vehicle with the bound tracking ID, updates the spatial unit where the vehicle is located and the collected parameters in real time, and completes the classification and determination of the driver's real-time fatigue status. Step S3: Based on the collected parameters of the vehicle's current spatial unit, combined with the vehicle's real-time driving speed and the driver's fatigue state, a basic optical flow signal framework with orientation guidance function is built. The stepped stimulation level corresponding to the driver's current fatigue state is matched, and the brightness and color temperature control parameters of the corresponding steps are set. The stepped progressive control mechanism adapted to the fatigue level is embedded into the basic optical flow signal framework. Then, the visual constraint boundary and correction parameters of the optical flow signal are calculated. After the introduction of the blind avoidance strategy is completed, the fatigue wake-up optical signal encoding parameter set is generated for the vehicle corresponding to the current driving tracking ID, the current fatigue state, and the current spatial unit through dual calibration based on the optical flow signal propulsion speed, brightness, and color temperature parameters. Step S4: Bind the LED wall washer lights in the tunnel to control zones according to spatial units, set a unique address code for the wall washer light control zone corresponding to each spatial unit, build a wall washer light control database, calculate the precise timing of control command issuance based on the fatigue wake-up light signal encoding parameter set, combined with the wall washer light signal response delay and command transmission delay, and issue control commands to the corresponding wall washer light control zone according to the preset timing, driving the wall washer lights to generate a progressive light flow signal for fatigue wake-up along the vehicle's driving direction according to the set parameters.

[0007] In a preferred embodiment, step S1 involves standardizing the spatial unit division of the tunnel, specifically as follows: Taking the starting point of the tunnel entrance as the zero point, along the vehicle travel direction, the entire tunnel is continuously divided into equal intervals according to the calibration length of the minimum space control unit of the tunnel, resulting in several space units. The last section that is less than the length of a single unit is merged with the previous space unit to ensure that the length of each space unit is uniform. The calibration length of the minimum space control unit of the tunnel comes from the matching result of the tunnel's completed alignment parameters and the length of the minimum independent control section of the wall washer light. Each divided spatial unit is bound to a unique tunnel mileage station interval, straight-line distance from the tunnel entrance, straight-line distance from the tunnel exit, and tunnel alignment parameters, where the tunnel alignment parameters refer to the horizontal curve radius and longitudinal slope within the spatial unit. For each spatial unit, using a Class I metrology device that conforms to the national verification procedures for luminance meters and color temperature meters, the baseline parameters including horizontal illuminance, vertical illuminance, and correlated color temperature were measured at the standard human eye height when the vehicle was driving and at the center line position of each lane. Each baseline parameter was collected through multiple measuring points that were equally spaced within the spatial unit, and the arithmetic mean of all measuring points was taken as the measured value of the baseline parameter. Each spatial unit generates a corresponding set of spatial visual background parameters, which includes the unit's mileage station range, straight distance from the entrance / exit, horizontal curve radius, longitudinal slope, main lighting background horizontal illuminance, background vertical illuminance, and background correlated color temperature. The identification of driver fatigue-sensitive areas is based on a sample set of driving behavior and physiological data, specifically as follows: The sample set of driving behavior and physiological data was extracted. In each spatial unit, there were fatigue characteristics such as the duration of a single eye closure reaching the single eye closure duration threshold, the lane deviation rate reaching the lane deviation rate threshold, the standard deviation of the steering wheel angle reaching the steering wheel angle standard deviation threshold, and the heart rate variability feature ratio reaching the heart rate variability feature ratio threshold. The thresholds mentioned above are all derived from the general judgment thresholds of the fatigue monitoring standard. For example, the lane deviation threshold is determined based on the distance between the vehicle tires and the driving marking line. The number of samples exhibiting fatigue characteristics in the sample set for each spatial unit is counted and recorded as the fatigue sample number. The ratio of the number of fatigue samples to the total number of valid samples is used as the proportion of fatigue feature frequency in the corresponding spatial unit. The ratio of the summation of the fatigue characteristic frequency proportions of all spatial units in the entire tunnel to the total number of spatial units in the entire tunnel is taken as the average fatigue characteristic frequency proportion of the entire tunnel.

[0008] In a preferred embodiment, step S2, which compares the fatigue characteristic frequency ratio of each spatial unit with the average fatigue characteristic frequency ratio of the entire tunnel, includes: If the proportion of fatigue characteristic frequency in any spatial unit is greater than or equal to the proportion of fatigue characteristic frequency in the entire tunnel multiplied by the first-level sensitivity threshold, it is marked as a first-level fatigue sensitive section. If the proportion of fatigue characteristic frequency in any spatial unit is greater than or equal to the proportion of fatigue characteristic frequency in the entire tunnel multiplied by the secondary sensitivity threshold, but less than the proportion of fatigue characteristic frequency in the entire tunnel multiplied by the primary sensitivity threshold, it is marked as a secondary fatigue sensitive section. The remaining spatial units are uniformly marked as regular sections; Step S2 binds each spatial unit to a corresponding tag, and then binds the spatial visual baseline parameter set and tag of each spatial unit one by one to generate a full tunnel calibration database.

[0009] In a preferred embodiment, step S2 involves initial identification of vehicles within the tunnel and generation of tracking IDs, including: A sensing unit, including millimeter-wave radar, OBU on-board unit reading and writing equipment, and high-definition area array camera, is installed at the tunnel entrance. The unique identification, vehicle model parameters, and vehicle rated driving speed stored in the vehicle's OBU are read and written by the OBU on-board unit to complete the initial identification of the vehicle. Millimeter-wave radar is used to collect real-time vehicle speed, lane number, vehicle outline length and width parameters to complete the initial collection of vehicle driving status. Using a high-definition area array camera, the driver's frontal face region image parameters are acquired at a fixed acquisition frame rate that meets the requirements of face detection frame rate. A face detection algorithm based on face features is used to locate and extract the driver's face region from the image parameters. Then, a standardized set of face feature points is extracted from the face region, and the feature coordinates of the eye region are locked to complete the initial locking of the driver's face features. Each vehicle in the tunnel will be given a unique driving tracking ID that remains unchanged throughout its entire journey in the tunnel, and the collected parameters will be bound to the tracking ID; In step S2, the interval sensing units are deployed along the longitudinal direction of the tunnel at fixed intervals, and each vehicle bound to the tracking ID is continuously and dynamically tracked, thereby updating the current spatial unit and collected parameters of each vehicle in real time. Based on preset eye movement feature indicators and driving behavior feature indicators, the collected parameters are standardized and mapped to scores, and fatigue status is graded based on the obtained standardized scores.

[0010] In a preferred embodiment, step S3 involves constructing a basic optical flow signal framework with orientation guidance functionality using the spatial visual background parameter set and the real-time vehicle speed acquired by the sensing unit, including: Based on the current spatial unit where the vehicle is located, and taking the corresponding spatial unit as the starting point, the next spatial unit is continuously selected in the vehicle's driving direction as the basic action segment of the optical flow signal. The optical flow signal is extended in the basic action segment according to the current fatigue state of the vehicle to obtain the complete action segment of the optical flow signal. For example, in the case of mild fatigue, the corresponding number of spatial units are extended on the basis of the basic action segment. If it is in the case of moderate fatigue, the corresponding number of spatial units are extended. The extended spatial units are determined according to the overall length of the tunnel. The mileage station interval and spatial visual background parameter set of all spatial units in the complete action segment are used as the action boundary of the optical flow signal. Based on the extracted real-time vehicle speed, the same optical flow propulsion speed is set, and the main illumination background vertical illuminance in the spatial visual background parameter set of the current vehicle spatial unit is used as the initial brightness of the optical flow. The initial color temperature of the optical flow is kept consistent with the background correlated color temperature in the spatial visual background parameter set of the current unit to ensure that the basic optical flow signal will not conflict with the tunnel main illumination and avoid glare interference. Finally, a basic optical flow signal framework containing the optical flow action section, optical flow propulsion speed, initial brightness of the optical flow, and initial color temperature of the optical flow is generated.

[0011] In a preferred embodiment, after building the basic optical flow signal framework, step S3 classifies the fatigue state of the vehicle driver into a ladder level including a zero-level ladder, a first-level ladder, a second-level ladder, and a third-level ladder, and sets corresponding stimulus intensity difference coefficients for the zero-level ladder, a first-level ladder, a second-level ladder, and a third-level ladder, and the stimulus intensity difference coefficient is a normalized value of the corresponding standardized score. The brightness progression is set based on the tiered levels. The maximum output brightness of the wall washer light is multiplied by the stimulation intensity difference coefficient of the corresponding tier, and then multiplied by ten. The brightness progression cycle is set based on the tiered levels. The real-time acquisition cycle is multiplied by the reciprocal of the corresponding tier number and then multiplied by ten to ensure that the progression cycle decreases as the tiered level increases. The upper limit of the brightness peak is set based on the tiered levels. The maximum output brightness of the wall washer light is multiplied by the stimulation intensity difference coefficient of the corresponding tier, and then multiplied by ten, while not exceeding the background vertical illuminance of the main lighting in the current space unit. Using the brightness progression cycle as the time unit, the brightness of the optical flow signal increases from the current value by one brightness progression step within each brightness progression cycle until it reaches the upper limit of the brightness peak of the corresponding step level. After reaching the upper limit of the brightness peak, the brightness drops back to the initial brightness of the optical flow with the same brightness progression step and the same brightness progression cycle, thus forming a continuous step-like brightness change curve. It should be noted that for the second and third steps, a small continuous fluctuation in brightness is superimposed within each progression cycle. The fluctuation amplitude is five percent of the current progressive brightness, and the fluctuation form is a continuous sinusoidal fluctuation to ensure the gradualness of the stimulus intensity and avoid abrupt changes. The color temperature progression control rules for the optical flow signal are set based on the tiered levels. Specifically, the product of the difference between the maximum and minimum color temperatures specified by the wall washer lamp at the factory and the corresponding stimulus intensity difference coefficient is used as the color temperature progression. The color temperature progression cycle is set to be consistent with the brightness progression cycle. The initial color temperature of the optical flow is added to the difference between the maximum and minimum color temperatures specified by the wall washer lamp at the factory, multiplied by ten times the stimulus intensity difference coefficient, as the upper limit of the color temperature peak. Using the color temperature progression cycle as the time unit, the color temperature of the optical flow signal increases from the current value by one color temperature progression step within each color temperature progression cycle until it reaches the upper limit of the color temperature peak of the corresponding step level. After reaching the upper limit of the peak, the color temperature falls back to the initial color temperature of the optical flow with the same color temperature progression step and the same color temperature progression cycle, forming a step-like progressive color temperature change curve that is synchronized with the brightness curve. The step-by-step progressive control mechanism was embedded into the basic optical flow signal framework.

[0012] In a preferred embodiment, step S3 uses the driver's eye position as the origin of the coordinate system, the vehicle's driving direction as the central axis, and the center line of the vehicle's driving lane as the longitudinal baseline. It reads the tunnel's lane width and wall washer light installation position parameters, calculates the central visual range boundary and the peripheral visual range boundary, and obtains correction parameters. The wall washer light installation position parameters include the wall washer light installation height range and the horizontal distance between the wall washer light and the lane center line. The central visual range boundary refers to the horizontal viewing angle corresponding to the lane's edge lines on both sides, with the driver's eye position as the apex and the vehicle's driving direction as the central axis. This range completely corresponds to the road surface and the lane area in front of the tunnel and is the core visual area for the driver to observe the road surface and lane. The peripheral visual range boundary refers to the area defined by the driver's eye position as the apex, the vehicle's direction of travel as the central axis, with the inner boundary being the outer boundary of the central visual range and the outer boundary being the horizontal viewing angle corresponding to the left and right sidewalls of the tunnel. This range completely corresponds to the left and right sidewall areas of the tunnel, and the installation position of the wall washer light is entirely within this area, including: The optical flow signal is only presented within the installation height range of the wall washer light, and the gradual change of the optical flow signal only advances along the longitudinal direction of tunnel travel. The brightness and color temperature parameters of the wall washer lights at the same lateral height remain consistent at the same time, thus establishing the spatial distribution constraint of the optical flow signal. Based on the boundaries of the central visual range and the peripheral visual range, combined with the brightness increment, the background vertical illuminance of the main lighting, and the real-time vehicle speed, the peripheral visual motion sensitive angular velocity range and the minimum perceptible difference threshold of human eye brightness are calculated. The angular velocity obtained by dividing the spatial unit calibration length by the real-time vehicle speed and converting it through trigonometric functions is used as the lower limit of the peripheral visual motion sensitive angular velocity range. The angular velocity obtained by converting the maximum brightness adjustment speed of the wall washer light at the factory calibration through trigonometric functions is used as the upper limit of the range. The ratio of the minimum brightness adjustment step size of the wall washer light at the factory calibration to the background vertical illuminance of the tunnel main lighting is used as the minimum perceptible difference threshold of human eye brightness. The maximum permissible speed ratio for optical flow propulsion is twice the real-time speed of the vehicle, and the minimum permissible speed ratio for optical flow propulsion is the actual real-time speed of the vehicle.

[0013] In a preferred embodiment, step S3 performs a calibration based on the acquired correction parameters, specifically: acquiring the actual angular velocity of the optical flow signal within the driver's peripheral vision, comparing the actual angular velocity with the peripheral visual motion-sensitive angular velocity range, performing boundary constraint judgment, if the actual angular velocity is within the peripheral visual motion-sensitive angular velocity range, it is determined that the velocity is compliant and no adjustment is needed, if the actual angular velocity is less than the lower limit of the peripheral visual motion-sensitive angular velocity range, it is determined that the velocity is too low and is prone to causing inattentive blindness, and linear fitting is performed to increase it, if the actual angular velocity is greater than the upper limit of the peripheral visual motion-sensitive angular velocity range, it is determined that the velocity is too high, and linear fitting is performed to decrease it, finally obtaining the calibrated progressive propulsion speed of the optical flow; When the actual angular velocity is within the range of peripheral visual motion-sensitive angular velocities, and the relative value of brightness change is greater than or equal to the minimum perceptible difference threshold of human eye brightness, the secondary calibration is determined to be complete. All calibrated brightness control curves, color temperature control curves, brightness fluctuation parameters, and change period parameters are output, and a fatigue wake-up light signal encoding parameter set is generated for the vehicle corresponding to the current driving tracking ID, the current fatigue marker tag, and the current spatial unit.

[0014] In a preferred embodiment, step S4 is based on the standardized division results of tunnel space units, binding the wall washer light control zones to the space units. The LED wall washer lights in the tunnel are bound to the space units according to the space units. Each space unit corresponds to an independent control zone with a unique address code. The wall washer lights in each control zone are divided into two independent control units: the left side wall control group and the right side wall control group. The left side wall control group and the right side wall control group are set with independent sub-address codes, which are respectively bound to the left lane and the right lane in the tunnel to form a wall washer light control group number, thus constructing a wall washer light control database. Read the signal response time of the wall washer light as set by the factory, add the fixed delay of the instruction transmission of the roadside edge computing unit, and obtain the advance issuance time of the control instruction to ensure that the optical flow change of the wall washer light is completely synchronized with the vehicle's driving position. Extract the optical flow action section and the wall washer light control group number from the optical signal encoding parameter set and match them with the wall washer light control database. Extract the vehicle's real-time driving speed, the calibrated optical flow progressive propagation speed, and the distance between the vehicle's current location and the starting point of the first control zone to be activated. Divide the distance between the vehicle's current location and the starting point of the first control zone to be activated by the vehicle's real-time driving speed to obtain the estimated arrival time. Add the difference between the estimated arrival time and the advance issuance time of the control command to the current time as the command issuance time node, and issue the control command to the wall washer light of the corresponding control zone. In a preferred embodiment, step S4, based on the fatigue wake-up light signal encoding parameter set, sends the corresponding brightness control curve, color temperature control curve, brightness fluctuation parameter, and change period parameter to the wall washer light corresponding to the control zone. After receiving the control command, the wall washer light generates a light flow signal that gradually advances along the driving direction according to the set parameters and a determined fixed control period.

[0015] The beneficial effects of this invention are as follows: It constructs an optical flow signal framework deeply bound to the driving state, simultaneously realizing the dual functions of fatigue awakening and directional navigation. Taking the vehicle's real-time spatial unit as the starting point and the driving direction as the axis, it dynamically adjusts the optical flow action section in combination with the fatigue level. At the same time, it sets the optical flow propulsion speed based on the vehicle's real-time driving speed and sets the initial brightness and color temperature of the optical flow based on the tunnel's background light environment parameters, ensuring the matching of the optical flow signal with the vehicle's driving trajectory and spatial position. It also binds the driver's real-time fatigue level to the stepped stimulation level one by one, sets the stimulation intensity difference coefficient based on the normalized value of the fatigue standardization score, and then matches the differentiated brightness, color temperature progression, change period, and peak upper limit to form a synchronous and linked stepped change curve adapted to the fatigue level. It establishes a full-process constraint and blind avoidance mechanism based on the physiological characteristics of human vision. Taking the driver's eye position as the coordinate origin, it accurately divides the functional boundaries of central vision and peripheral vision, so that the optical flow signal only acts on the peripheral vision area corresponding to the tunnel sidewall, without interfering with the driver's central vision core area for observing the road surface and lane lines, thus avoiding the interference of dynamic lights on normal driving observation from a spatial distribution perspective. Attached Figure Description

[0016] Figure 1 This is a flowchart of the present invention. Detailed Implementation

[0017] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0018] In the description of this application, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the stated features. In the description of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0019] In the description of this application, the term "for example" is used to mean "used as an example, illustration, or description." Any embodiment described as "for example" in this application is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is provided to enable any person skilled in the art to make and use the invention. Details are set forth in the following description for purposes of explanation. It should be understood that those skilled in the art will recognize that the invention can be made without using these specific details. In other instances, well-known structures and processes will not be described in detail to avoid obscuring the description of the invention with unnecessary detail. Therefore, the invention is not intended to be limited to the embodiments shown, but is consistent with the broadest scope of the principles and features disclosed in this application.

[0020] As attached Figure 1 As shown, this embodiment provides: a method for controlling LED wall washer lights to wake up drivers in a long tunnel. Step S1: After standardizing the spatial unit division based on the long tunnel, obtain continuous and equidistant standardized spatial units throughout the tunnel. Based on a preset sample set of driving behavior and physiological data, identify the driving fatigue sensitive sections of each spatial unit in the tunnel, statistically obtain the fatigue characteristics and frequency ratio of each spatial unit, and simultaneously calculate the average frequency ratio of fatigue characteristics throughout the tunnel to complete the alignment. Step S1 involves standardizing the spatial unit division of the tunnel, specifically as follows: Taking the starting point of the tunnel entrance as the zero point, along the vehicle travel direction, the entire tunnel is continuously divided into equal intervals according to the calibration length of the minimum space control unit of the tunnel, resulting in several space units. The last section that is less than the length of a single unit is merged with the previous space unit to ensure that the length of each space unit is uniform. The calibration length of the minimum space control unit of the tunnel comes from the matching result of the tunnel's completed alignment parameters and the length of the minimum independent control section of the wall washer light. Each divided spatial unit is bound to a unique tunnel mileage station interval, straight-line distance from the tunnel entrance, straight-line distance from the tunnel exit, and tunnel alignment parameters, where the tunnel alignment parameters refer to the horizontal curve radius and longitudinal slope within the spatial unit. For each spatial unit, using a Class I metrology device that conforms to the national verification procedures for luminance meters and color temperature meters, the baseline parameters including horizontal illuminance, vertical illuminance, and correlated color temperature were measured at the standard human eye height when the vehicle was driving and at the center line position of each lane. Each baseline parameter was collected through multiple measuring points that were equally spaced within the spatial unit, and the arithmetic mean of all measuring points was taken as the measured value of the baseline parameter. Each spatial unit generates a corresponding set of spatial visual background parameters, which includes the unit's mileage station range, straight distance from the entrance / exit, horizontal curve radius, longitudinal slope, main lighting background horizontal illuminance, background vertical illuminance, and background correlated color temperature. Step S1 involves identifying driver fatigue-sensitive areas based on a sample set of driving behavior and physiological data. Specifically: The sample set of driving behavior and physiological data was extracted. In each spatial unit, there were fatigue characteristics such as the duration of a single eye closure reaching the single eye closure duration threshold, the lane deviation rate reaching the lane deviation rate threshold, the standard deviation of the steering wheel angle reaching the steering wheel angle standard deviation threshold, and the heart rate variability feature ratio reaching the heart rate variability feature ratio threshold. The thresholds mentioned above are all derived from the general judgment thresholds of the fatigue monitoring standard. For example, the lane deviation threshold is determined based on the distance between the vehicle tires and the driving marking line. The number of samples exhibiting fatigue characteristics in the sample set for each spatial unit is counted and recorded as the fatigue sample number. The ratio of the number of fatigue samples to the total number of valid samples is used as the proportion of fatigue feature frequency in the corresponding spatial unit. The ratio of the summation of the fatigue characteristic frequency proportions of all spatial units in the entire tunnel to the total number of spatial units in the entire tunnel is taken as the average fatigue characteristic frequency proportion of the entire tunnel.

[0021] Step S2: The fatigue sensitivity level of each spatial unit is classified and bound to the label. The spatial visual background parameter set of each spatial unit is matched with the label one by one to build a full tunnel calibration database. A unique driving tracking ID is generated for each vehicle throughout the tunnel journey. The perception unit continuously and dynamically tracks the vehicle with the bound tracking ID, updates the spatial unit where the vehicle is located and the collected parameters in real time, and completes the classification and determination of the driver's real-time fatigue status. Step S2 compares the frequency proportion of fatigue characteristics in each spatial unit with the average frequency proportion of fatigue characteristics across the entire tunnel, including: If the proportion of fatigue characteristic frequency in any spatial unit is greater than or equal to the proportion of fatigue characteristic frequency in the entire tunnel multiplied by the first-level sensitivity threshold, it is marked as a first-level fatigue sensitive section. If the proportion of fatigue characteristic frequency in any spatial unit is greater than or equal to the proportion of fatigue characteristic frequency in the entire tunnel multiplied by the secondary sensitivity threshold, but less than the proportion of fatigue characteristic frequency in the entire tunnel multiplied by the primary sensitivity threshold, it is marked as a secondary fatigue sensitive section. The remaining spatial units are uniformly marked as regular sections; Step S2 binds each spatial unit to a corresponding tag, and then binds the spatial visual baseline parameter set and tag of each spatial unit one by one to generate a full tunnel calibration database.

[0022] In some other specific implementations, the Level 1 sensitivity threshold multiple refers to the determination coefficient of the Level 1 fatigue sensitive section in the calibrated tunnel. It represents the proportion of fatigue characteristic frequency of the spatial unit and needs to reach the minimum multiple of the average fatigue characteristic frequency of the entire tunnel. The minimum multiple mentioned here is the lowest value of the proportion of fatigue characteristic frequency in the higher / major risk level section defined by road traffic. The Level 2 sensitivity threshold multiple refers to the general risk level section. This application further discloses the calibration method for the first-level fatigue-sensitive section and the second-level fatigue-sensitive section in this embodiment: The test selected the most common vehicle type used in tunnel operations, followed the tunnel's speed limit, maintained free-flow driving conditions, and ensured no abnormal driving behaviors such as overtaking, acceleration, deceleration, or stopping, which was completely consistent with the normal driving conditions of daily tunnel operation. The tunnel's main lighting was on in the standard daily operation mode, without extreme weather or traffic control conditions, to ensure that the test data fully matched the daily operation conditions and covered drivers with different driving experience, age groups, and tunnel driving frequencies. The collected indicators include eye movement characteristics (blinking frequency, duration of single eye closure, and pupil diameter fluctuation rate), driving behavior characteristics (lane deviation rate, standard deviation of steering wheel angle, and following distance fluctuation rate), and physiological characteristics (heart rate variability-related parameters). The collection frequency of all indicators is completely consistent with the real-time collection frequency of this scheme, and all data are bound one-to-one with the mileage marker of the tunnel space unit to ensure that each data point can be accurately mapped to a unique tunnel space unit.

[0023] Based on the above basic settings, the collected valid samples are labeled with fatigue characteristics according to the fatigue characteristic judgment rules, forming a binary label result of fatigue characteristics; the number of samples with fatigue characteristics among all valid samples in the spatial unit is counted, that is, the number of fatigue samples; the number of fatigue samples is divided by the total number of valid samples in the spatial unit to obtain the frequency ratio of fatigue characteristics in the spatial unit; the frequency ratio of fatigue characteristics of all spatial units in the whole tunnel is summed and divided by the total number of spatial units in the whole tunnel to obtain the average frequency ratio of fatigue characteristics in the whole tunnel, which is used as the benchmark value for threshold calculation; Construct a cumulative distribution function for the proportion of fatigue characteristic frequencies of spatial units in the entire tunnel. Plot the cumulative distribution curve with the ratio of the proportion of fatigue characteristic frequencies of spatial units to the average proportion of the entire tunnel as the x-axis and the proportion of the cumulative number of spatial units corresponding to the ratio to the total number of units in the entire tunnel as the y-axis.

[0024] The point where the curve slope first shows a sharp increase corresponds to the critical value where fatigue risk abruptly changes from a normal level to a high level. The ratio corresponding to this critical value is the initial value of the first-level sensitivity threshold multiple. The point where the curve slope first shows a significant increase corresponds to the critical value where fatigue risk abruptly changes from a low level to a medium level. The ratio corresponding to this critical value is the initial value of the second-level sensitivity threshold multiple. In some other specific implementations, clustering methods other than cumulative distribution function inflection point identification can also be used, which will not be elaborated here.

[0025] Step S2 performs initial identification of vehicles within the tunnel and generates a tracking ID, including: A sensing unit, including millimeter-wave radar, OBU on-board unit reading and writing equipment, and high-definition area array camera, is installed at the tunnel entrance. The unique identification, vehicle model parameters, and vehicle rated driving speed stored in the vehicle's OBU are read and written by the OBU on-board unit to complete the initial identification of the vehicle. Millimeter-wave radar is used to collect real-time vehicle speed, lane number, vehicle outline length and width parameters to complete the initial collection of vehicle driving status. Using a high-definition area array camera, the driver's frontal face region image parameters are acquired at a fixed acquisition frame rate that meets the requirements of face detection frame rate. A face detection algorithm based on face features is used to locate and extract the driver's face region from the image parameters. Then, a standardized set of face feature points is extracted from the face region, and the feature coordinates of the eye region are locked to complete the initial locking of the driver's face features. Each vehicle in the tunnel will be given a unique driving tracking ID that remains unchanged throughout its entire journey in the tunnel, and the collected parameters will be bound to the tracking ID; Step S2 involves deploying interval sensing units at fixed intervals along the longitudinal direction of the tunnel and continuously and dynamically tracking each vehicle bound to a tracking ID, thereby updating the current spatial unit and collected parameters of each vehicle in real time. The specific process of dynamic tracking involves: data relay matching between adjacent millimeter-wave radar intervals. The millimeter-wave radar in the previous interval transmits the vehicle's driving tracking ID, real-time location, driving speed, and driving lane data to the millimeter-wave radar in the next interval. The radar in the next interval predicts the trajectory based on the vehicle's position and speed, completing the relay identification of vehicles with the same driving tracking ID, thus achieving continuous tracking of the vehicle throughout its entire journey within the tunnel. At a fixed data update cycle that meets the real-time requirements of vehicle tracking, the current spatial unit, real-time driving speed, driving lane number, and following distance of the vehicle corresponding to the driving tracking ID are updated in real time. The vehicle distance is determined by matching the vehicle's real-time mileage marker with the spatial unit marker interval calibrated in step one. The high-definition area array camera in the interval continuously acquires images of the driver's face region at a fixed acquisition frame rate that meets the requirements of face detection frame rate, based on the initially locked set of driver's face feature points. The image preprocessing algorithm is used to remove backlight and reflection interference caused by the tunnel main lighting. Then, the driver's real-time eye movement features are extracted from the preprocessed images, including blink frequency, duration of single eye closure, pupil diameter change rate, and gaze point distribution range. All eye movement features are bound to the driving tracking ID. Based on preset eye movement feature indicators and driving behavior feature indicators, the collected parameters are standardized and mapped to scores, and fatigue status is graded based on the obtained standardized scores.

[0026] In some other specific implementations, based on the relevant standards for driver fatigue monitoring, two primary indicators are set: eye movement characteristic indicators and driving behavior characteristic indicators. The eye movement characteristic indicators are further divided into three secondary indicators: blinking frequency, duration of single eye closure, and pupil diameter fluctuation rate. The driving behavior characteristic indicators are further divided into three secondary indicators: lane departure rate, standard deviation of steering wheel angle, and following distance fluctuation rate.

[0027] Pre-set baseline thresholds for alertness and severe fatigue for each secondary indicator. The baseline threshold for alertness corresponds to the indicator characteristic value when the driver is in a normal, alert driving state, while the baseline threshold for severe fatigue corresponds to the indicator characteristic value when the driver is in a state of severe fatigue.

[0028] If the real-time collected parameters are between the baseline threshold for a conscious state and the baseline threshold for a severely fatigued state, the standardized score is the difference between the real-time collected parameters and the baseline threshold for a conscious state, divided by the difference between the baseline threshold for a severely fatigued state and the baseline threshold for a conscious state. If the real-time collected parameters are less than or equal to the baseline threshold for a conscious state, the standardized score is fixed at zero. If the real-time collected parameters are greater than or equal to the baseline threshold for a severely fatigued state, the standardized score is fixed at one hundred. The higher the standardized score of all secondary indicators, the more severe the fatigue level in the corresponding dimension.

[0029] Therefore, standardized scores between zero and the mild fatigue threshold are marked as a conscious state; scores above the mild fatigue threshold and between the moderate fatigue threshold are marked as a mild fatigue state; scores above the moderate fatigue threshold and between the severe fatigue threshold are marked as a moderate fatigue state; and scores above the severe fatigue threshold and between one percent are marked as a severe fatigue state. The mild, moderate, and severe fatigue thresholds are all derived from industry-standard fatigue grading and real-vehicle test verification results.

[0030] Step S3: Based on the collected parameters of the vehicle's current spatial unit, combined with the vehicle's real-time driving speed and the driver's fatigue state, a basic optical flow signal framework with orientation guidance function is built. The stepped stimulation level corresponding to the driver's current fatigue state is matched, and the brightness and color temperature control parameters of the corresponding steps are set. The stepped progressive control mechanism adapted to the fatigue level is embedded into the basic optical flow signal framework. Then, the visual constraint boundary and correction parameters of the optical flow signal are calculated. After the introduction of the blind avoidance strategy is completed, the fatigue wake-up optical signal encoding parameter set is generated for the vehicle corresponding to the current driving tracking ID, the current fatigue state, and the current spatial unit through dual calibration based on the optical flow signal propulsion speed, brightness, and color temperature parameters. Step S3 involves constructing a basic optical flow signal framework with orientation guidance capabilities using the spatial visual background parameter set and the real-time vehicle speed acquired through the sensing unit, including: Based on the current spatial unit where the vehicle is located, and taking the corresponding spatial unit as the starting point, the next spatial unit is continuously selected in the vehicle's driving direction as the basic action segment of the optical flow signal. The optical flow signal is extended in the basic action segment according to the current fatigue state of the vehicle to obtain the complete action segment of the optical flow signal. For example, in the case of mild fatigue, the corresponding number of spatial units are extended on the basis of the basic action segment. If it is in the case of moderate fatigue, the corresponding number of spatial units are extended. The extended spatial units are determined according to the overall length of the tunnel. The mileage station interval and spatial visual background parameter set of all spatial units in the complete action segment are used as the action boundary of the optical flow signal. Based on the extracted real-time vehicle speed, the same optical flow propulsion speed is set, and the main illumination background vertical illuminance in the spatial visual background parameter set of the current vehicle spatial unit is used as the initial brightness of the optical flow. The initial color temperature of the optical flow is kept consistent with the background correlated color temperature in the spatial visual background parameter set of the current unit to ensure that the basic optical flow signal will not conflict with the tunnel main illumination and avoid glare interference. Finally, a basic optical flow signal framework containing the optical flow action section, optical flow propulsion speed, initial brightness of the optical flow, and initial color temperature of the optical flow is generated.

[0031] Step S3: After building the basic optical flow signal framework, the fatigue state of the vehicle driver is classified into different levels, including level 0, level 1, level 2, and level 3. The corresponding stimulus intensity difference coefficients are set for level 0, level 1, level 2, and level 3, and the stimulus intensity difference coefficients are normalized values ​​of the corresponding standardized scores. The brightness progression is set based on the tiered levels. The maximum output brightness of the wall washer light is multiplied by the stimulation intensity difference coefficient of the corresponding tier, and then multiplied by ten. The brightness progression cycle is set based on the tiered levels. The real-time acquisition cycle is multiplied by the reciprocal of the corresponding tier number and then multiplied by ten to ensure that the progression cycle decreases as the tiered level increases. The upper limit of the brightness peak is set based on the tiered levels. The maximum output brightness of the wall washer light is multiplied by the stimulation intensity difference coefficient of the corresponding tier, and then multiplied by ten, while not exceeding the background vertical illuminance of the main lighting in the current space unit. Using the brightness progression cycle as the time unit, the brightness of the optical flow signal increases from the current value by one brightness progression step within each brightness progression cycle until it reaches the upper limit of the brightness peak of the corresponding step level. After reaching the upper limit of the brightness peak, the brightness drops back to the initial brightness of the optical flow with the same brightness progression step and the same brightness progression cycle, thus forming a continuous step-like brightness change curve. It should be noted that for the second and third steps, a small continuous fluctuation in brightness is superimposed within each progression cycle. The fluctuation amplitude is five percent of the current progressive brightness, and the fluctuation form is a continuous sinusoidal fluctuation to ensure the gradualness of the stimulus intensity and avoid abrupt changes. The color temperature progression control rules for the optical flow signal are set based on the tiered levels. Specifically, the product of the difference between the maximum and minimum color temperatures specified by the wall washer lamp at the factory and the corresponding stimulus intensity difference coefficient is used as the color temperature progression. The color temperature progression cycle is set to be consistent with the brightness progression cycle. The initial color temperature of the optical flow is added to the difference between the maximum and minimum color temperatures specified by the wall washer lamp at the factory, multiplied by ten times the stimulus intensity difference coefficient, as the upper limit of the color temperature peak. Using the color temperature progression cycle as the time unit, the color temperature of the optical flow signal increases from the current value by one color temperature progression step within each color temperature progression cycle until it reaches the upper limit of the color temperature peak of the corresponding step level. After reaching the upper limit of the peak, the color temperature falls back to the initial color temperature of the optical flow with the same color temperature progression step and the same color temperature progression cycle, forming a step-like progressive color temperature change curve that is synchronized with the brightness curve. The step-by-step progressive control mechanism was embedded into the basic optical flow signal framework.

[0032] Step S3 uses the driver's eye position as the origin of the coordinate system, the vehicle's driving direction as the central axis, and the center line of the vehicle's driving lane as the longitudinal baseline. It reads the tunnel's lane width and wall washer light installation position parameters, calculates the central visual range boundary and the peripheral visual range boundary, and obtains correction parameters. The wall washer light installation position parameters include the wall washer light installation height range and the horizontal distance between the wall washer light and the lane center line. The central visual range boundary refers to the horizontal viewing angle corresponding to the lane's edge lines on both sides, with the driver's eye position as the vertex and the vehicle's driving direction as the central axis. This range completely corresponds to the road surface and the lane area in front of the tunnel and is the core visual area for the driver to observe the road surface and lane. The peripheral visual range boundary refers to the area defined by the driver's eye position as the apex, the vehicle's direction of travel as the central axis, with the inner boundary being the outer boundary of the central visual range and the outer boundary being the horizontal viewing angle corresponding to the left and right sidewalls of the tunnel. This range completely corresponds to the left and right sidewall areas of the tunnel, and the installation position of the wall washer light is entirely within this area, including: The optical flow signal is only presented within the installation height range of the wall washer light, and the gradual change of the optical flow signal only advances along the longitudinal direction of tunnel travel. The brightness and color temperature parameters of the wall washer lights at the same lateral height remain consistent at the same time, thus establishing the spatial distribution constraint of the optical flow signal. Based on the boundaries of the central visual range and the peripheral visual range, combined with the brightness increment, the background vertical illuminance of the main lighting, and the real-time vehicle speed, the peripheral visual motion sensitive angular velocity range and the minimum perceptible difference threshold of human eye brightness are calculated. The angular velocity obtained by dividing the spatial unit calibration length by the real-time vehicle speed and converting it through trigonometric functions is used as the lower limit of the peripheral visual motion sensitive angular velocity range. The angular velocity obtained by converting the maximum brightness adjustment speed of the wall washer light at the factory calibration through trigonometric functions is used as the upper limit of the range. The ratio of the minimum brightness adjustment step size of the wall washer light at the factory calibration to the background vertical illuminance of the tunnel main lighting is used as the minimum perceptible difference threshold of human eye brightness. The maximum permissible speed ratio for optical flow propulsion is twice the real-time speed of the vehicle, and the minimum permissible speed ratio for optical flow propulsion is the actual real-time speed of the vehicle.

[0033] Step S3 performs a calibration based on the acquired correction parameters. Specifically, it acquires the actual angular velocity of the optical flow signal within the driver's peripheral vision. The acquisition logic is based on the basic progressive propagation speed of the optical flow and the horizontal distance between the wall washer light and the driver's eyes. The actual angular velocity is calculated using trigonometric functions. The actual angular velocity is compared with the peripheral visual motion-sensitive angular velocity range. Boundary constraint judgment is performed. If the actual angular velocity is within the peripheral visual motion-sensitive angular velocity range, it is determined that the speed is in compliance and no adjustment is needed. If the actual angular velocity is less than the lower limit of the peripheral visual motion-sensitive angular velocity range, it is determined that the speed is too low and may easily cause inattentive blindness. Linear fitting is performed to increase the speed. If the actual angular velocity is greater than the upper limit of the peripheral visual motion-sensitive angular velocity range, it is determined that the speed is too high and may easily cause visual interference. Linear fitting is performed to decrease the speed. Finally, the calibrated optical flow progressive propagation speed is obtained. Furthermore, in some other specific implementations, the boundary constraint linear fitting adjustment is as follows: when the actual angular velocity is less than the lower limit of the interval, the lower limit of the interval is used as the target value, and the calibrated optical flow progressive propulsion speed is calculated in reverse. At the same time, the upper limit of the calibrated propulsion speed is set as the ratio of the vehicle's real-time driving speed to the maximum allowable speed, ensuring that the optical flow propulsion speed is not too fast and avoiding visual dizziness; when the actual angular velocity is greater than the upper limit of the interval, the upper limit of the interval is used as the target value, and the calibrated optical flow progressive propulsion speed is calculated in reverse. At the same time, the lower limit of the calibrated propulsion speed is set as the ratio of the vehicle's real-time driving speed to the minimum allowable speed, ensuring that the optical flow propulsion speed is always faster than the vehicle's driving speed and maintaining the orientation guidance function; When the actual angular velocity is within the range of peripheral visual motion-sensitive angular velocities, and the relative value of brightness change is greater than or equal to the minimum perceptible difference threshold of human eye brightness, the secondary calibration is determined to be complete. All calibrated brightness control curves, color temperature control curves, brightness fluctuation parameters, and change period parameters are output, and a fatigue wake-up light signal encoding parameter set is generated for the vehicle corresponding to the current driving tracking ID, the current fatigue marker tag, and the current spatial unit.

[0034] Step S4: Bind the LED wall washer lights in the tunnel to control zones according to spatial units, set a unique address code for the wall washer light control zone corresponding to each spatial unit, build a wall washer light control database, calculate the precise timing of control command issuance based on the fatigue wake-up light signal encoding parameter set, combined with the wall washer light signal response delay and command transmission delay, and issue control commands to the corresponding wall washer light control zone according to the preset timing, driving the wall washer lights to generate a progressive light flow signal for fatigue wake-up along the vehicle's driving direction according to the set parameters.

[0035] Step S4, based on the standardized division results of tunnel space units, binds the wall washer light control zones to the space units. The LED wall washer lights in the tunnel are bound to the space units according to the space units. Each space unit corresponds to an independent control zone with a unique address code. The wall washer lights in each control zone are divided into two independent control units: the left side wall control group and the right side wall control group. The left side wall control group and the right side wall control group are set with independent sub-address codes, which are bound to the left lane and the right lane in the tunnel respectively to form the wall washer light control group number, thus constructing the wall washer light control database. Read the signal response time of the wall washer light as set by the factory, add the fixed delay of the instruction transmission of the roadside edge computing unit, and obtain the advance issuance time of the control instruction to ensure that the optical flow change of the wall washer light is completely synchronized with the vehicle's driving position. Extract the optical flow action section and the wall washer light control group number from the optical signal encoding parameter set and match them with the wall washer light control database. Extract the vehicle's real-time driving speed, the calibrated optical flow progressive propagation speed, and the distance between the vehicle's current location and the starting point of the first control zone to be activated. Divide the distance between the vehicle's current location and the starting point of the first control zone to be activated by the vehicle's real-time driving speed to obtain the estimated arrival time. Add the difference between the estimated arrival time and the advance issuance time of the control command to the current time as the command issuance time node, and issue the control command to the wall washer light of the corresponding control zone.

[0036] Step S4, based on the fatigue wake-up light signal encoding parameter set, sends the corresponding brightness control curve, color temperature control curve, brightness fluctuation parameter, and change period parameter to the wall washer light in the control zone. After receiving the control command, the wall washer light generates a light flow signal that gradually advances along the driving direction according to the set parameters and a determined fixed control period.

[0037] It should be noted that the descriptions of each embodiment in the above embodiments have different focuses. For parts that are not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0038] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0039] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0040] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0041] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0042] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0043] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for controlling LED wall washer lights to wake up drivers from driver fatigue in long tunnels, characterized in that, Includes the following steps: Step S1: Standardize the spatial unit division based on the long tunnel. Using the starting point of the tunnel entrance as the zero point, complete the continuous equidistant division of the entire tunnel along the vehicle driving direction to obtain standardized spatial units with continuous equidistant spacing throughout the tunnel. Each spatial unit is bound to a unique tunnel mileage station interval and tunnel alignment parameters, and a spatial visual baseline parameter set including horizontal illuminance, vertical illuminance, and correlated color temperature is generated through actual measurement. Based on a preset driving behavior and physiological data sample set, the driving fatigue sensitive sections of each spatial unit in the entire tunnel are calibrated. Four types of fatigue features are extracted from each spatial unit: duration of single eye closure, lane departure rate, standard deviation of steering wheel angle, and heart rate variability. The frequency ratio of fatigue features corresponding to each spatial unit is statistically obtained, and the average frequency ratio of fatigue features throughout the entire tunnel is calculated simultaneously to complete the calibration alignment. Based on the average frequency ratio of fatigue features throughout the entire tunnel, each spatial unit is divided into first-level fatigue sensitive sections, second-level fatigue sensitive sections, and regular sections through first-level and second-level sensitivity threshold multiples, and corresponding labels are bound. The spatial visual baseline parameter set of each spatial unit is matched one-to-one with the label to construct a calibration database for the entire tunnel. Step S2: At the tunnel entrance, a sensing unit consisting of millimeter-wave radar, OBU vehicle-mounted unit reading and writing equipment, and high-definition area array camera is set up to complete vehicle identification, driving status collection, and initial locking of driver facial features. A unique driving tracking ID is generated for each vehicle throughout the entire tunnel journey, and the collected parameters are bound to the tracking ID. Interval sensing units are deployed along the longitudinal direction of the tunnel. Through the relay matching mechanism of adjacent interval sensing units, vehicles with bound tracking IDs are continuously and dynamically tracked, and the spatial unit where the vehicle is located and the collected parameters are updated in real time. Based on preset eye movement feature indicators and driving behavior feature indicators, the collected parameters are standardized and mapped to complete the graded determination of the driver's real-time fatigue state. Step S3: Based on the collected parameters of the current spatial unit of the vehicle, combined with the real-time driving speed of the vehicle and the driver's fatigue state, select and extend the complete optical flow action section along the driving direction with the current spatial unit of the vehicle as the starting point, set the optical flow propulsion speed consistent with the real-time driving speed of the vehicle, set the initial brightness and initial color temperature of the optical flow based on the visual background parameters of the current unit space, and build a basic optical flow signal framework with orientation guidance function. The system matches the driver's current fatigue level with a stepped stimulation level. Based on the stimulation intensity difference coefficient of the corresponding step, it sets the progressive growth, progressive period, and peak upper limit of brightness and color temperature, generating synchronous stepped progressive brightness change curves and color temperature change curves. The stepped progressive control mechanism adapted to the fatigue level is embedded into the basic optical flow signal framework. Then, the visual constraint boundary and correction parameters of the optical flow signal are calculated. The central vision and peripheral vision range boundaries are divided with the driver's eye position as the coordinate origin. The optical flow is limited to the peripheral vision area of ​​the tunnel sidewall. The peripheral vision motion sensitive angular velocity range and the minimum perceptible difference threshold of human eye brightness are calculated. After the introduction of the blind avoidance strategy, the actual optical flow angular velocity is compared with the sensitive angular velocity range and linear fitting calibration is performed to verify that the relative value of brightness change meets the minimum perceptible difference threshold of human eye brightness. Through dual calibration based on the optical flow signal propulsion speed, brightness, and color temperature parameters, a fatigue wake-up optical signal encoding parameter set is generated for the vehicle corresponding to the current driving tracking ID, the current fatigue state, and the current spatial unit. Step S4: Bind the LED wall washer lights in the tunnel to control zones according to spatial units, set a unique address code for the wall washer light control zone corresponding to each spatial unit, and divide each control zone into the left side wall control group and the right side wall control group of the corresponding lane respectively and set an independent sub-address code to build the wall washer light control database. Based on the fatigue wake-up optical signal encoding parameter set, combined with the signal response delay and command transmission delay of the wall washer light, the advance issuance time of the control command is calculated. Combined with the estimated arrival time of the vehicle, the precise issuance sequence is deduced, and the control command is issued to the corresponding wall washer light control zone according to the preset sequence. The wall washer light is driven to generate an optical flow signal for fatigue wake-up that gradually advances along the vehicle's driving direction according to the brightness control curve, color temperature control curve, brightness fluctuation parameters, and change period parameters.

2. The method for controlling LED wall washer lights to wake up drivers from driver fatigue in long tunnels according to claim 1, characterized in that, Step S1 involves standardizing the spatial unit division of the tunnel, specifically as follows: Using the starting point of the tunnel entrance as the zero point, along the vehicle travel direction, the entire tunnel is continuously divided into equal intervals according to the calibrated length of the tunnel's minimum space control unit, thus obtaining several space units; Each divided spatial unit is bound to a unique tunnel mileage station interval, straight-line distance from the tunnel entrance, straight-line distance from the tunnel exit, and tunnel alignment parameters; For each spatial unit, the baseline parameters including horizontal illuminance, vertical illuminance, and correlated color temperature were measured using a first-level metrology device at the standard human eye height when the vehicle is in motion and at the center line position of each lane. Each baseline parameter was collected by multiple measuring points arranged at equal intervals within the spatial unit, and the arithmetic mean of all measuring points was taken as the measured value of the baseline parameter. Generate a corresponding set of spatial visual background parameters for each spatial unit; The identification of driver fatigue-sensitive areas is based on a sample set of driving behavior and physiological data, specifically as follows: The extracted driving behavior and physiological data sample set shows fatigue characteristics in each spatial unit, such as the duration of a single eye closure reaching a single eye closure duration threshold, the lane deviation rate reaching a lane deviation rate threshold, the standard deviation of the steering wheel angle reaching a steering wheel angle standard deviation threshold, and the ratio of heart rate variability characteristics reaching a heart rate variability characteristic threshold. The number of samples exhibiting fatigue characteristics in the sample set for each spatial unit is counted and recorded as the fatigue sample number. The ratio of the number of fatigue samples to the total number of valid samples is used as the proportion of fatigue feature frequency in the corresponding spatial unit. The ratio of the summation of the fatigue characteristic frequency proportions of all spatial units in the entire tunnel to the total number of spatial units in the entire tunnel is taken as the average fatigue characteristic frequency proportion of the entire tunnel.

3. The method for controlling LED wall washer lights to wake up drivers from driver fatigue in long tunnels according to claim 2, characterized in that, Step S2 compares the fatigue characteristic frequency ratio of each spatial unit with the average fatigue characteristic frequency ratio of the entire tunnel, including: If the proportion of fatigue characteristic frequency in any spatial unit is greater than or equal to the proportion of fatigue characteristic frequency in the entire tunnel multiplied by the first-level sensitivity threshold, it is marked as a first-level fatigue sensitive section. If the proportion of fatigue characteristic frequency in any spatial unit is greater than or equal to the proportion of fatigue characteristic frequency in the entire tunnel multiplied by the secondary sensitivity threshold, but less than the proportion of fatigue characteristic frequency in the entire tunnel multiplied by the primary sensitivity threshold, it is marked as a secondary fatigue sensitive section. The remaining spatial units are uniformly marked as regular sections; Step S2 binds each spatial unit to a corresponding tag, and then binds the spatial visual baseline parameter set and tag of each spatial unit one by one to generate a full tunnel calibration database.

4. The method for controlling LED wall washer lights to wake up drivers from driver fatigue in long tunnels according to claim 1, characterized in that, Step S2 involves initial identification of vehicles within the tunnel and generation of tracking IDs, including: A sensing unit, including millimeter-wave radar, OBU on-board unit reading and writing equipment, and high-definition area array camera, is installed at the tunnel entrance. The unique identification, vehicle model parameters, and vehicle rated driving speed stored in the vehicle's OBU are read and written by the OBU on-board unit to complete the initial identification of the vehicle. Millimeter-wave radar is used to collect real-time vehicle speed, lane number, vehicle outline length and width parameters to complete the initial collection of vehicle driving status. Using a high-definition area array camera, image parameters of the driver's frontal face area are acquired. A face detection algorithm based on facial features is used to locate and extract the driver's face area from the image parameters. Then, a standardized set of facial feature points is extracted from the face area, and the feature coordinates of the eye area are locked to complete the initial locking of the driver's facial features. Each vehicle in the tunnel will be given a unique driving tracking ID that remains unchanged throughout its entire journey in the tunnel, and the collected parameters will be bound to the tracking ID; In step S2, the interval sensing units are deployed along the longitudinal direction of the tunnel at fixed intervals, and each vehicle bound to the tracking ID is continuously and dynamically tracked, thereby updating the current spatial unit and collected parameters of each vehicle in real time. Based on preset eye movement feature indicators and driving behavior feature indicators, the collected parameters are standardized and mapped to scores, and fatigue status is graded based on the obtained standardized scores.

5. The method for controlling LED wall washer lights to wake up drivers from driver fatigue in long tunnels according to claim 4, characterized in that, Step S3 involves constructing a basic optical flow signal framework using the spatial visual background parameter set and the real-time vehicle speed obtained through the sensing unit, including: Based on the current spatial unit of the vehicle, and taking the corresponding spatial unit as the starting point, the next spatial unit is continuously selected in the vehicle's driving direction as the basic action segment of the optical flow signal. The optical flow signal is then extended in the basic action segment according to the current fatigue state of the vehicle to obtain the complete action segment of the optical flow signal. The mileage station interval and spatial visual background parameter set of all spatial units in the complete action segment are used as the action boundary of the optical flow signal. Based on the extracted real-time vehicle speed, the same optical flow propulsion speed is set, and the main illumination background vertical illuminance in the spatial visual background parameter set of the current vehicle spatial unit is used as the initial brightness of the optical flow. The initial color temperature of the optical flow is kept consistent with the background correlated color temperature in the spatial visual background parameter set of the current unit. Finally, a basic optical flow signal framework containing the optical flow action section, optical flow propulsion speed, initial brightness of the optical flow, and initial color temperature of the optical flow is generated.

6. The method for controlling LED wall washer lights to wake up drivers from driver fatigue in long tunnels according to claim 5, characterized in that, In step S3, after building the basic optical flow signal framework, the fatigue state of the vehicle driver is classified into different levels, including zero-level, first-level, second-level, and third-level steps. Corresponding stimulation intensity difference coefficients are set for the zero-level, first-level, second-level, and third-level steps, and the stimulation intensity difference coefficients are normalized values ​​of the corresponding standardized scores. The brightness progression is set based on the tiered levels by multiplying the factory-calibrated maximum output brightness of the wall washer by the stimulation intensity difference coefficient of the corresponding tier, and then multiplying by ten. The brightness progression cycle is set based on the tiered levels by multiplying the real-time acquisition cycle by the reciprocal of the corresponding tier number and then multiplying by ten. The peak brightness upper limit is set based on the tiered levels by multiplying the factory-calibrated maximum output brightness of the wall washer by the stimulation intensity difference coefficient of the corresponding tier, and then multiplying by ten. Using the brightness progression cycle as the time unit, the brightness of the optical flow signal increases from the current value by one brightness progression step within each brightness progression cycle until it reaches the upper limit of the brightness peak of the corresponding step level. After reaching the upper limit of the brightness peak, the brightness falls back to the initial brightness of the optical flow with the same brightness progression step and the same brightness progression cycle, thus forming a continuous step-like progressive brightness change curve. The color temperature progression control rules for the optical flow signal are set based on the tiered levels. Specifically, the product of the difference between the maximum and minimum color temperatures specified by the wall washer lamp at the factory and the corresponding stimulus intensity difference coefficient is used as the color temperature progression. The color temperature progression cycle is set to be consistent with the brightness progression cycle. The initial color temperature of the optical flow is added to the difference between the maximum and minimum color temperatures specified by the wall washer lamp at the factory, multiplied by ten times the stimulus intensity difference coefficient, as the upper limit of the color temperature peak. Using the color temperature progression cycle as the time unit, the color temperature of the optical flow signal increases from the current value by one color temperature progression step within each color temperature progression cycle until it reaches the upper limit of the color temperature peak of the corresponding step level. After reaching the upper limit of the peak, the color temperature falls back to the initial color temperature of the optical flow with the same color temperature progression step and the same color temperature progression cycle, forming a step-like progressive color temperature change curve that is synchronized with the brightness curve. The step-by-step progressive control mechanism was embedded into the basic optical flow signal framework.

7. The method for controlling LED wall washer lights to wake up drivers from driver fatigue in long tunnels according to claim 1, characterized in that, Step S3 uses the driver's eye position as the coordinate origin, the vehicle's direction of travel as the central axis, and the center line of the vehicle's lane as the longitudinal baseline. It reads the tunnel lane width and wall washer light installation position parameters, calculates the central visual range boundary and the peripheral visual range boundary, and obtains correction parameters, including: The optical flow signal is only presented within the installation height range of the wall washer light, and the gradual change of the optical flow signal only advances along the longitudinal direction of tunnel travel. The brightness and color temperature parameters of the wall washer lights at the same lateral height remain consistent at the same time, thus establishing the spatial distribution constraint of the optical flow signal. Based on the boundaries of the central visual range and the peripheral visual range, combined with the brightness increment, the background vertical illuminance of the main lighting, and the real-time vehicle speed, the peripheral visual motion sensitive angular velocity range and the minimum perceptible difference threshold of human eye brightness are calculated. The angular velocity obtained by dividing the spatial unit calibration length by the real-time vehicle speed and converting it through trigonometric functions is used as the lower limit of the peripheral visual motion sensitive angular velocity range. The angular velocity obtained by converting the maximum brightness adjustment speed of the wall washer light at the factory calibration through trigonometric functions is used as the upper limit of the range. The ratio of the minimum brightness adjustment step size of the wall washer light at the factory calibration to the background vertical illuminance of the tunnel main lighting is used as the minimum perceptible difference threshold of human eye brightness. The maximum permissible speed ratio for optical flow propulsion is twice the real-time speed of the vehicle, and the minimum permissible speed ratio for optical flow propulsion is the actual real-time speed of the vehicle.

8. The method for controlling LED wall washer lights to wake up drivers from driver fatigue in long tunnels according to claim 7, characterized in that, Step S3 performs a calibration based on the acquired correction parameters. Specifically, it involves: acquiring the actual angular velocity of the optical flow signal within the driver's peripheral vision, comparing the actual angular velocity with the peripheral vision motion-sensitive angular velocity range, performing boundary constraint judgment, determining that the actual angular velocity is within the peripheral vision motion-sensitive angular velocity range if it is within the range, determining that the velocity is in compliance if it is less than the lower limit of the range, and performing linear fitting to increase the velocity, and determining that the actual angular velocity is greater than the upper limit of the range, and performing linear fitting to decrease the velocity, thus obtaining the calibrated progressive propulsion speed of the optical flow. When the actual angular velocity is within the range of peripheral visual motion-sensitive angular velocities, and the relative value of brightness change is greater than or equal to the minimum perceptible difference threshold of human eye brightness, the secondary calibration is determined to be complete. All calibrated brightness control curves, color temperature control curves, brightness fluctuation parameters, and change period parameters are output, and a fatigue wake-up light signal encoding parameter set is generated for the vehicle corresponding to the current driving tracking ID, the current fatigue marker tag, and the current spatial unit.

9. The method for controlling LED wall washer lights to awaken drivers from driver fatigue in long tunnels according to claim 1, characterized in that, Step S4, based on the standardized division results of tunnel space units, binds the wall washer light control zones to the space units. The LED wall washer lights in the tunnel are bound to the space units according to the space units. Each space unit corresponds to an independent control zone with a unique address code. The wall washer lights in each control zone are divided into two independent control units: the left side wall control group and the right side wall control group. The left side wall control group and the right side wall control group are set with independent sub-address codes, which are respectively bound to the left lane and the right lane in the tunnel to form a wall washer light control group number, thus constructing a wall washer light control database. Read the signal response time set by the wall washer light at the factory, add the fixed delay of the instruction transmission from the roadside edge computing unit to obtain the advance issuance time of the control instruction, extract the optical flow action section and the wall washer light control group number from the optical signal encoding parameter set, and match them with the wall washer light control database; Extract the vehicle's real-time driving speed, the calibrated optical flow progressive propagation speed, and the distance between the vehicle's current location and the starting point of the first control zone to be activated. Divide the distance between the vehicle's current location and the starting point of the first control zone to be activated by the vehicle's real-time driving speed to obtain the estimated arrival time. Add the difference between the estimated arrival time and the advance issuance time of the control command to the current time as the command issuance time node, and issue the control command to the wall washer light of the corresponding control zone.

10. The method for controlling LED wall washer lights to wake up drivers from driver fatigue in long tunnels according to claim 1, characterized in that, Step S4, based on the fatigue wake-up light signal encoding parameter set, sends the corresponding brightness control curve, color temperature control curve, brightness fluctuation parameter, and change period parameter to the wall washer light in the control zone. After receiving the control command, the wall washer light generates a light flow signal that gradually advances along the driving direction according to the set parameters and a determined fixed control period.

Images