Highway fog scene elastic speed limit and vehicle distance cooperative regulation method and system

Abstract

The application discloses a method and system for cooperative regulation of elastic speed limit and vehicle distance in a fog gathering scene on an expressway, and particularly relates to the technical field of intelligent traffic control, and is used to solve the problem of secondary accident risk induced by the existing dynamic speed limit method due to the neglect of traffic flow phase change characteristics; the traffic density, vehicle distance and visibility space-time distribution data are collected in real time; the traffic density is weighted and amplified based on the consistency of the visibility gradient direction and the traffic flow direction to generate a corrected traffic density; the phase change margin index representing the flow state stability is calculated in combination with the corrected traffic density and the optical safety deviation; when the phase change margin is lower than a threshold value, the maximum value of the speed limit drop is dynamically constrained according to the moving trend of the fog gathering, and the maximum value is kept positively correlated with the phase change margin; the target speed limit value is generated based on the constrained drop value, and the safety vehicle distance instruction is calculated by fusing the visibility data; the instruction is issued to the vehicles in the dynamic influence area of the fog gathering, and the cooperative optimization of the traffic system stability and the vehicle safety is realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent traffic control, more particularly, the present application relates to a method and system for elastic speed limit and vehicle distance cooperative regulation in a fog scene on a highway. BACKGROUND

[0002] Traffic safety prevention and control in a fog scene on a highway mainly relies on dynamic speed limit and vehicle distance control. Existing methods usually obtain visibility data in real time through roadside meteorological monitoring equipment, and adjust the speed limit value and safe vehicle distance command based on preset threshold rules, aiming to prevent rear-end accidents in the fog area by reducing speed and increasing vehicle spacing, and the regulation logic focuses on the instantaneous safety condition of a single vehicle.

[0003] However, the existing regulation method has systematic risks in the implementation process: when a large amplitude speed limit is taken on a fog section, although the safety of the local area can be improved, the dynamic characteristics of the traffic flow will be significantly changed, leading to deterioration of traffic flow stability and inducing upstream congestion and secondary accident risk. Since only the single vehicle safety threshold is considered, the deep-seated contradiction between traffic system state mutation and safety regulation is not solved, so the actual effect of elastic speed limit and vehicle distance control is restricted. SUMMARY

[0004] In order to overcome the above-mentioned defects of the prior art, the present application provides a method and system for elastic speed limit and vehicle distance cooperative regulation in a fog scene on a highway to solve the problems raised in the background art.

[0005] To achieve the above-mentioned purpose, the present application provides the following technical scheme:

[0006] The method for elastic speed limit and vehicle distance cooperative regulation in a fog scene on a highway comprises:

[0007] S1, real-time collect traffic running state parameters and visibility spatio-temporal distribution data of the fog affected section, the traffic running state parameters including traffic density and vehicle spacing;

[0008] S2, calculate the visibility gradient direction at the current time point according to the visibility spatio-temporal distribution data, and when the visibility gradient direction is consistent with the traffic direction, the traffic density is weighted and enlarged to generate a corrected traffic density;

[0009] S3, calculate the phase transition margin index of the traffic flow based on the corrected traffic density and the optical safety deviation, and the optical safety deviation is calculated according to the vehicle spacing and the visibility spatio-temporal distribution data;

[0010] S4, when the phase transition margin index is lower than a preset threshold, dynamically constrain the maximum speed reduction value according to the fog moving trend and the corrected traffic density, so that the maximum speed reduction value is positively correlated with the phase transition margin index;

[0011] S5, generating a target speed limit value based on the maximum speed limit reduction after the constraint, and calculating a safe distance instruction in combination with the visibility spatiotemporal distribution data;

[0012] S6, issuing the target speed limit value and the safe distance instruction to vehicles in the fog-affected road section.

[0013] Further, the traffic operation state parameters and the visibility spatiotemporal distribution data of the fog-affected road section are collected in real time, including:

[0014] The traffic density and the vehicle spacing are collected in real time by the roadside sensors arranged in the fog-affected road section, and the roadside sensors include video detection equipment and microwave radar equipment;

[0015] The visibility spatiotemporal distribution data is collected in real time by the meteorological monitoring station network arranged in the fog-affected road section, and the meteorological monitoring station network is composed of multiple meteorological sensors distributed at a preset interval;

[0016] The traffic density, vehicle spacing, and visibility spatiotemporal distribution data are transmitted to the edge computing node for timestamp alignment processing through a special communication protocol.

[0017] Further, the visibility gradient direction at the current time point is calculated according to the visibility spatiotemporal distribution data, and the traffic density is weighted and enlarged when the visibility gradient direction is consistent with the traffic flow direction to generate a corrected traffic density, including:

[0018] The visibility spatial distribution data at the current time point is extracted based on the visibility spatiotemporal distribution data;

[0019] The visibility gradient direction is determined by calculating the visibility change rate between adjacent meteorological monitoring stations;

[0020] The vector angle between the visibility gradient direction and the preset traffic flow direction is calculated;

[0021] When the vector angle is less than or equal to a preset angle threshold, it is determined that the visibility gradient direction is consistent with the traffic flow direction;

[0022] The traffic density is enlarged by a preset weighting coefficient to generate a corrected traffic density.

[0023] Further, the traffic density is enlarged by a preset weighting coefficient to generate a corrected traffic density, including:

[0024] The real-time detected traffic density is obtained; and the visibility gradient direction change feature is extracted;

[0025] The value of the preset weighting coefficient is determined according to the visibility gradient direction change feature;

[0026] The real-time detected traffic density is multiplied by the preset weighting coefficient to generate a corrected traffic density.

[0027] Further, the phase transition margin index of the traffic flow is calculated based on the corrected traffic density and the optical safety deviation degree, including:

[0028] extracting the visibility value at the current time point from the spatiotemporal distribution data of the visibility;

[0029] determining the optical safety deviation degree based on the spatial conflict relationship between the vehicle distance and the visibility value;

[0030] generating a flow state instability risk value according to the relative difference between the corrected traffic density and the critical traffic density;

[0031] establishing a dynamic game matrix of the optical safety deviation degree and the flow state instability risk value;

[0032] outputting a system risk coupling coefficient through a preset coordination rule of the dynamic game matrix;

[0033] comparing the system risk coupling coefficient with a preset phase transition threshold to generate a phase transition margin index, which represents the degree of deviation of the traffic flow metastable state from the phase transition boundary.

[0034] Further, the dynamic game matrix of the optical safety deviation degree and the flow state instability risk value includes:

[0035] defining the grade division interval of the optical safety deviation degree;

[0036] setting the risk level threshold of the flow state instability risk value;

[0037] constructing a decision matrix with the optical safety deviation degree grade as the row and the flow state instability risk value grade as the column;

[0038] presetting a coordination rule output value in the cell of the decision matrix.

[0039] Further, when the phase transition margin index is lower than the preset threshold, the maximum speed reduction value is positively correlated with the phase transition margin index according to the fog cluster movement trend and the dynamic constraint of the corrected traffic density, including:

[0040] when the phase transition margin index is lower than the preset threshold, obtaining the spatial position migration data of the fog cluster movement trend;

[0041] extracting the road spatial distribution characteristics corresponding to the corrected traffic density;

[0042] establishing a dynamic coupling constraint mechanism of the fog cluster movement trend and the corrected traffic density;

[0043] determining the constraint boundary of the maximum speed reduction value based on the dynamic coupling constraint mechanism;

[0044] The speed limit reduction amplitude proportional coefficient is obtained by deviating the phase change margin index from the preset threshold value;

[0045] The speed limit reduction amplitude proportional coefficient is applied to the constraint boundary to generate a final speed limit reduction amplitude maximum value.

[0046] Further, the target speed limit value is generated based on the constrained speed limit reduction amplitude maximum value, and the safety distance instruction is calculated in combination with the visibility spatiotemporal distribution data, including:

[0047] The current reference speed limit value is obtained;

[0048] The target speed limit value is generated by subtracting the speed limit reduction amplitude maximum value from the current reference speed limit value;

[0049] The visibility value at the current time point is extracted from the visibility spatiotemporal distribution data;

[0050] The minimum safety distance threshold value is calculated based on the target speed limit value and the visibility value;

[0051] The safety distance instruction is generated according to the minimum safety distance threshold value.

[0052] Further, the target speed limit value and the safety distance instruction are issued to vehicles in the fog group affected road section, including:

[0053] The dynamic spatial range of the fog group affected road section is determined;

[0054] The target speed limit value and the safety distance instruction are encapsulated as a cooperative control instruction set;

[0055] The cooperative control instruction set is broadcast to vehicles in the dynamic spatial range through a cooperative control communication protocol;

[0056] The parameter value of the broadcast instruction is dynamically adjusted according to the vehicle position information.

[0057] In another aspect, the present application provides a flexible speed limit and distance cooperative control system in a fog group scenario on an expressway, including:

[0058] A traffic environment perception module is used to collect traffic running state parameters and visibility spatiotemporal distribution data of the fog group affected road section in real time, and the traffic running state parameters include traffic density and vehicle distance;

[0059] A density dynamic correction module is used to calculate the visibility gradient direction at the current time point according to the visibility spatiotemporal distribution data, and the traffic density is weighted and amplified when the visibility gradient direction is consistent with the traffic flow direction, to generate a corrected traffic density;

[0060] A flow state margin evaluation module is used to calculate a phase change margin index of the traffic flow based on the corrected traffic density and an optical safety deviation, and the optical safety deviation is calculated according to the vehicle distance and the visibility spatiotemporal distribution data;

[0061] The amplitude dynamic constraint module is configured to dynamically constrain the speed reduction amplitude maximum value according to the group fog moving trend and the corrected traffic density when the phase transition margin index is lower than a preset threshold, so that the speed reduction amplitude maximum value is positively correlated with the phase transition margin index.

[0062] The instruction coordination generation module is configured to generate a target speed limit value based on the constrained speed reduction amplitude maximum value, and calculate a safety distance instruction in combination with the visibility spatiotemporal distribution data.

[0063] The instruction broadcast execution module is configured to publish the target speed limit value and the safety distance instruction to vehicles in the group fog affected road section.

[0064] Compared with the prior art, the present application has the following beneficial effects:

[0065] Firstly, the traffic flow phase transition margin is used to dynamically constrain the speed limit regulation intensity, so as to realize the systematic balance between safety and efficiency. The phase transition margin index is introduced to quantify the degree of traffic flow metastable state deviating from the phase transition boundary, and the speed reduction amplitude maximum value is dynamically constrained based on the index: when the traffic flow approaches the instability critical point, the reduction amplitude boundary is automatically expanded to strengthen the safety guarantee; when the flow state is stable, the reduction amplitude boundary is contracted to retain the traffic capacity. This positive correlation regulation resolves the contradiction between "over-speed reduction inducing congestion and congestion exacerbating accident risk" from the system level, reduces the invalid speed limit amplitude under the same visibility condition, and reduces the secondary accident rate.

[0066] Secondly, the spatiotemporal motion characteristics of the group fog and the traffic density correction mechanism are fused to improve the accuracy and timeliness of the coordinated control. The visibility gradient direction and the traffic flow direction are analyzed in coordination: when the visibility gradient direction is consistent with the traffic flow direction, the traffic density is weighted and enlarged to generate a corrected traffic density, so as to capture the traffic compression effect caused by the group fog movement in advance. Then, the speed reduction amplitude is dynamically constrained according to the group fog moving trend, so that the target speed limit value and the safety distance instruction are dynamically adapted to the group fog spatial migration. This breaks through the limitation of static zoning regulation, and the instruction publishing can accurately cover the actual influence range of the group fog, avoiding the early speed reduction of upstream vehicles or the response delay of downstream vehicles. Under the same equipment conditions, the timeliness of risk prevention and control is improved. BRIEF DESCRIPTION OF DRAWINGS

[0067] Figure 1 The flowchart of the present application is a flexible speed limit and distance coordination regulation method for the group fog scene on the expressway.

[0068] Figure 2 The structural schematic diagram of the present application is a flexible speed limit and distance coordination regulation system for the group fog scene on the expressway. DETAILED DESCRIPTION

[0069] With reference to the accompanying drawings, the technical solutions in the embodiments of the present application will be described clearly and completely. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all the other embodiments obtained by those of ordinary skill in the art without creative work fall within the scope of the present application.

[0070] Embodiment 1 Figure 1 The present application provides a method for cooperative regulation of elastic speed limit and vehicle distance in a fog group scene on an expressway, including:

[0071] S1, real-time collection of traffic running state parameters and visibility spatiotemporal distribution data of a fog group affected road section, the traffic running state parameters including traffic density and vehicle distance;

[0072] S2, calculation of a visibility gradient direction at a current time point according to the visibility spatiotemporal distribution data, weighted amplification of the traffic density when the visibility gradient direction is consistent with the traffic flow direction, and generation of a corrected traffic density;

[0073] S3, calculation of a phase transition margin index of the traffic flow based on the corrected traffic density and an optical safety deviation, the optical safety deviation being calculated according to the vehicle distance and the visibility spatiotemporal distribution data;

[0074] S4, when the phase transition margin index is lower than a preset threshold, dynamic constraint of a maximum speed reduction value according to a fog group movement trend and the corrected traffic density, so that the maximum speed reduction value is positively correlated with the phase transition margin index;

[0075] S5, generation of a target speed limit value based on the constrained maximum speed reduction value, and calculation of a safe vehicle distance instruction in combination with the visibility spatiotemporal distribution data;

[0076] S6, issuance of the target speed limit value and the safe vehicle distance instruction to vehicles in the fog group affected road section.

[0077] S1, real-time collection of traffic running state parameters and visibility spatiotemporal distribution data of a fog group affected road section, the traffic running state parameters including traffic density and vehicle distance, and the specific implementation being as follows:

[0078] Real-time traffic density and vehicle spacing are collected by roadside sensors deployed on the fog bank affected road section. The roadside sensor is composed of a video detection device and a microwave radar device. The video detection device takes overhead images of the road surface at a preset period, and extracts the headway between adjacent vehicles in the same lane as the measured value of vehicle spacing through image recognition algorithms. At the same time, the number of vehicles per kilometer in the lane is counted as the instantaneous detection value of traffic density. The microwave radar device transmits a frequency-modulated continuous wave at a preset frequency, and calculates the vehicle spacing according to the phase difference of the reflected wave. The detection results of the two devices are weighted and fused at the edge computing node. The fusion weight is dynamically adjusted according to the visibility condition: when the visibility value is higher than 200 meters, the weight of the video detection device is set to 0.7, and the weight of the microwave radar device is set to 0.3; when the visibility value decreases to below 200 meters, the weight of the video detection device is linearly reduced to 0.3, and the weight of the microwave radar device is correspondingly increased to 0.7. The unit of traffic density detection value is unified as vehicle per kilometer, and the unit of vehicle spacing detection value is unified as meter.

[0079] Real-time spatial and temporal distribution data of visibility are collected by a network of meteorological monitoring stations deployed on the fog bank affected road section. The network of meteorological monitoring stations is composed of multiple meteorological sensors deployed on both sides of the road at a preset interval. Each meteorological sensor is equipped with a forward scattering visibility meter to measure the horizontal atmospheric transmittance at a preset frequency, and the visibility value is obtained according to the conversion relationship between atmospheric transmittance and visibility. The unit of visibility value is unified as meter. After receiving the measurement data of all meteorological sensors, the edge computing node generates a visibility distribution surface based on the road centerline through a spatial interpolation algorithm. This surface covers the entire road section in the spatial dimension and is updated at fixed intervals in the time dimension, forming a spatial and temporal distribution dataset of visibility. The meteorological sensors are installed on poles at a certain height above the road surface, and the optical lens of the sensor is perpendicular to the direction of vehicle flow.

[0080] Traffic density, vehicle spacing, and spatial and temporal distribution data of visibility are transmitted to the edge computing node for timestamp alignment processing through a special communication protocol. The special communication protocol adopts a layered transmission architecture. After receiving the data, the edge computing node performs time synchronization operations: extracts the time tag of traffic density detection data, which is generated by the time module built-in the video detection device; extracts the time tag of visibility data, which is generated by the real-time clock inside the meteorological sensor; unifies the time reference of the two types of data to the standard time system through a time calibration algorithm. The calibration process includes calculating the clock deviation of the two time sources, establishing a linear compensation model, and correcting the time tag of the visibility data. After timestamp alignment processing, a synchronized dataset is generated, which contains three core fields: a unified time tag field; a traffic parameter field containing traffic density and vehicle spacing values; a visibility value field containing the visibility distribution value of the road section at that time point.

[0081] In the deployment of the meteorological monitoring station network, the installation position of the meteorological sensor needs to meet certain technical requirements: it is installed on a column at an appropriate height from the road surface to avoid interference from vehicle wake; the optical lens of the sensor is directed at a suitable angle to the vehicle flow direction to prevent direct light from vehicle headlights from affecting measurement accuracy; the spacing between adjacent sensors is determined according to the spatial variation characteristics of visibility. The measurement range of the visibility value is set to a reasonable interval, and when the measured value is below a certain threshold, it automatically switches to a high-sensitivity mode to improve the sampling frequency.

[0082] In the timestamp alignment process, the time calibration algorithm uses a linear fitting method: a plurality of time synchronization signal points are collected; the cumulative deviation of the meteorological sensor clock is calculated based on the time of the time service module; a deviation compensation equation is established, which is a linear function, the slope represents the clock drift rate, and the intercept represents the initial deviation; the equation is applied to real-time correction of the time label of subsequent visibility data. The time synchronization accuracy is verified by synchronization error detection, and the verification operation is performed regularly.

[0083] In the traffic parameter collection process, the weighted fusion of video detection equipment and microwave radar equipment uses an adaptive weight distribution strategy: when the visibility conditions are good, the video detection equipment is set to a higher weight; when the visibility conditions deteriorate, the weight of the video detection equipment is linearly reduced, and the weight of the microwave radar equipment is correspondingly increased. The weight adjustment curve is a piecewise function, and the turning point is set at a certain visibility value. The function slope is determined according to the device calibration test data. The fused vehicle spacing value needs to be checked for reasonableness: if the value change rate of adjacent detection periods exceeds a reasonable threshold, the abnormal value filtering mechanism is started, and the current value is replaced by the moving average value.

[0084] The spatial interpolation algorithm for visibility spatiotemporal distribution data uses the inverse distance weighting method: the position of each meteorological sensor is taken as the control point, and the road is divided into grid cells; for the visibility value of any grid cell point, the weight is calculated according to the distance from the adjacent control points; the weight coefficient is inversely proportional to the square of the distance; when the distance between the grid cell and the control point exceeds a certain range, the data reliability decay mechanism is started, and an error compensation amount is added to the calculation result. The generated visibility distribution surface is stored in matrix form, with the row number corresponding to the road stake number, the column number corresponding to the time sequence, and the matrix element value being the visibility value at that position and time.

[0085] The data transmission special communication protocol includes a designed frame structure: each frame of data includes a frame header, a payload, and a check code, the frame header includes a data type identifier and a time label; the payload uses binary encoding, the traffic density value is represented by an integer, the vehicle spacing value is represented by an integer, and the visibility value is represented by an integer. The check code uses the cyclic redundancy check method to ensure data transmission reliability.

[0086] In terms of device deployment, video detection devices are installed on poles at a certain height on the side of the road to monitor the target lane; microwave radar devices are installed at an angle that has been calibrated to ensure that the beam covers the detection area. The meteorological sensor network is laid out at a specific density, for example, with a larger spacing on straight sections and a smaller spacing on curved sections. The edge computing node is configured with a data preprocessing module to filter the raw detection data and eliminate random fluctuations.

[0087] In time synchronization operations, the timing module uses the time reference provided by the satellite positioning system, with a time tag accuracy of milliseconds. The parameters of the linear compensation model are determined through historical data analysis and recalibrated every 24 hours. The search radius of the spatial interpolation algorithm is set according to the characteristics of the road section, for example, a larger search radius is used on straight sections and a smaller search radius is used on complex sections.

[0088] In the weighted fusion strategy, the turning point position of the weight adjustment curve is determined through experiments, for example, by comparing the detection error rates of two devices under different visibility conditions. The trigger threshold of the outlier filtering mechanism is set according to the characteristics of the traffic flow, for example, a smaller threshold is used during stable traffic periods and a larger threshold is used during traffic fluctuations. The compensation coefficient of the data reliability decay mechanism is determined through error analysis experiments, and the compensation coefficient increases with distance.

[0089] In frame structure design, the binary encoding scheme of the payload is optimized, for example, the traffic density value is represented by 16 bits, with an accuracy that meets the detection requirements. The check code generation algorithm uses a standard implementation to ensure error detection capability. The trigger condition of the data discard mechanism is set according to the real-time requirements of the system, for example, it is started when the time deviation exceeds 200 milliseconds.

[0090] S2, calculate the visibility gradient direction at the current time point according to the visibility spatio-temporal distribution data, and when the visibility gradient direction is consistent with the traffic direction, the traffic density is weighted and enlarged to generate the corrected traffic density, which is implemented as follows:

[0091] Based on the visibility spatio-temporal distribution data, the visibility spatial distribution data at the current time point is extracted. The visibility spatio-temporal distribution data is stored in the edge computing node in matrix form, where the first dimension represents road location information and the second dimension represents time series. The extraction operation locates the data slice at the current time through the time index, which contains the visibility values of different road locations. Road locations are divided at fixed intervals, and visibility values are stored in meters. For locations without direct monitoring points, interpolation algorithms are used to supplement data based on the data of adjacent meteorological monitoring stations.

[0092] The visibility gradient direction is determined by calculating the visibility change rate between adjacent weather monitoring stations. First, identify the adjacent weather monitoring station pairs of the current road section, and each monitoring station pair has a preset distance. For each monitoring station pair, calculate the visibility change rate: subtract the visibility value of the upstream station from the visibility value of the downstream station, and then divide by the road distance between the two stations. Then calculate the gradient vector: take the road extension direction as the reference axis, convert the visibility change rate to a direction vector, and the vector direction points to the direction of increasing visibility. The final gradient direction is determined by the vector composition of the results of all monitoring station pairs, and the composition method is vector addition operation.

[0093] The visibility gradient direction and the preset traffic flow direction are calculated by vector angle calculation. The preset traffic flow direction is determined by the road geometry, and is stored as a fixed direction vector. The angle calculation uses the vector dot product formula: calculate the dot product of the visibility gradient direction vector and the preset traffic flow direction vector, divide by the product of the lengths of the two vectors, and then take the inverse cosine function value. After converting the calculation result to an angle value, the angle value is obtained.

[0094] When the vector angle is less than or equal to the preset angle threshold, it is determined that the visibility gradient direction is consistent with the traffic flow direction. The determination logic is: compare the calculated angle value with the preset angle threshold, if the calculated value is less than or equal to the threshold, trigger the density correction mechanism; if it is greater than the threshold, keep the original traffic density value.

[0095] The determination method of the preset angle threshold is: by collecting the gradient direction and traffic flow direction angle data in historical fog events, combining with the traffic flow stability index (such as speed variation coefficient), analyzing the flow state instability probability change curve under different angles, and selecting the angle value corresponding to the probability surge point as the threshold. The value is verified by 90% confidence interval, and the typical range is 15° to 45°, and the specific value does not require creative labor to implement.

[0096] The traffic density is amplified by a preset weighting coefficient to generate a corrected traffic density. First, obtain the real-time detected traffic density value, which comes from the synchronous data set of the data acquisition step. At the same time, extract the visibility gradient direction change characteristics, including the gradient direction stability index and the gradient intensity change rate. The gradient direction stability index calculation method is: statistics the change degree of the gradient direction in the recent several minutes. The gradient intensity change rate calculation method is: the ratio of the current gradient length to the previous period gradient length.

[0097] The value of the preset weighting coefficient is determined according to the visibility gradient direction change characteristics. Establish the weighting coefficient decision rule: when the gradient direction stability is high and the gradient intensity change rate is large, the weighting coefficient is set to a larger value; when the gradient direction stability is low and the gradient intensity change rate is small, the weighting coefficient is set to a smaller value. The weighting coefficient value range is controlled within a reasonable interval.

[0098] The real-time detected traffic density is multiplied by a preset weighting coefficient to generate a corrected traffic density. The multiplication calculation is completed at the edge computing node, and the calculation formula is: corrected traffic density equals original traffic density multiplied by weighting coefficient. The calculation result is kept at a proper precision, and the unit is consistent with the original traffic density. The corrected traffic density value is stored in the processing queue for subsequent steps.

[0099] In the gradient direction calculation process, the calculation of the visibility change rate needs to consider the data quality: when the data of a certain meteorological monitoring station is missing, the monitoring station pair containing the station is automatically excluded; when more than a certain proportion of monitoring station pair data is unavailable, the data recovery mechanism is started. The vector synthesis adopts the weighted average method, and the closer the monitoring station pair to the current road position, the higher the weight.

[0100] In the angle calculation, the determination method of the preset traffic flow direction is: the direction vector of each position point is stored in the road database, and the direction vector is calculated from the road design information. The inverse cosine function calculation adopts the standard mathematical method, and the precision is controlled within a reasonable range.

[0101] The gradient direction change feature analysis adopts a sliding time window mechanism: the window length is set to a fixed time length, and is periodically slid. The statistical characteristics of the gradient direction sequence in the window are calculated as stability indicators. The gradient strength change rate calculation adopts the adjacent period comparison method.

[0102] The weighting coefficient decision rule is adjusted by an optimization method: relevant data in historical fog events are collected, a prediction model is trained to optimize the decision rule. The decision rule update period is regular maintenance to ensure adaptation to environmental changes. The weighting coefficient needs to be range checked before application.

[0103] After the corrected traffic density is generated, a reasonableness check is performed: the change rate before and after correction is calculated, and if the change rate exceeds a reasonable threshold, an audit flag is started; compared with historical data, if there is a significant deviation, it is marked as an abnormal value. All correction records are stored in a log file.

[0104] In special scene processing, when the visibility gradient direction suddenly changes, an emergency processing program is started: the weighting coefficient update is frozen, and the previous value is maintained for a fixed time; when the fog quickly dissipates, the weighting coefficient is automatically reset to the baseline value.

[0105] In the gradient direction synthesis process, the vector addition adopts the component accumulation method: the gradient vector of each monitoring station pair is decomposed into coordinate components, and the final vector is synthesized after summation. The component calculation adopts the plane coordinate system, and the coordinate origin is set at the beginning of the road section.

[0106] The weighting coefficient determination process introduces traffic flow state feedback: when the real-time traffic density approaches the road capacity, the weighting coefficient is automatically reduced to avoid distortion caused by excessive amplification, and this mechanism is realized by real-time monitoring of traffic density values.

[0107] S3, calculate the phase transition margin index of traffic flow based on the corrected traffic density and the optical safety deviation degree, and the implementation is as follows:

[0108] The visibility value at the current time point is extracted from the visibility spatiotemporal distribution data. The visibility spatiotemporal distribution data is stored in the edge computing node in the form of a time series matrix, and the row index of the matrix corresponds to the road location identifier, and the column index corresponds to the time point. The extraction operation is performed by locating the column index where the current timestamp is located, and obtaining all row element values in the column to form a set of visibility spatial distribution data. For the target vehicle position, the visibility value of the corresponding position is extracted from the data set by matching the road position. The unit of the visibility value is meter, and when there is no direct monitoring point at the target position, the visibility values of the adjacent two meteorological monitoring stations are used for interpolation calculation.

[0109] The optical safety deviation degree is determined based on the spatial conflict relationship between the vehicle distance and the visibility value. The current vehicle distance detection value is obtained, which comes from the synchronous data set in the data acquisition step. The spatial conflict relationship quantification method is to calculate the ratio of the vehicle distance to the visibility value, which reflects the degree of spatial safety redundancy. The optical safety deviation degree is defined as the output of the conversion function of the ratio: the original ratio is input into the piecewise function, and the output range is standardized to between 0 and 1, and the value increases indicates that the safety deviation degree increases. The piecewise function sets the turning point: when the ratio reaches a certain threshold, the output value changes in steps.

[0110] The flow state instability risk value is generated according to the relative difference between the corrected traffic density and the critical traffic density. The corrected traffic density comes from the output result of the previous step. The critical traffic density is determined by the road design parameters, which are stored in the road segment attribute database. The relative difference is calculated in percentage form: the difference between the corrected traffic density and the critical traffic density is taken, divided by the critical traffic density, and multiplied by the percentage coefficient. The flow state instability risk value is defined as the mapping result of the percentage value, and the mapping process adopts a piecewise linear conversion, and the output value range is controlled between 0 and 1.

[0111] A dynamic game matrix of optical safety deviation degree and flow state instability risk value is established. The level division of optical safety deviation degree is defined: the continuous optical safety deviation value is divided into three levels, and the level boundary value is set. The risk level of flow state instability risk value is set: the risk value is divided into three levels, and the risk level boundary value is set. A three-row and three-column decision matrix is constructed, with the row title as the optical safety deviation degree level and the column title as the flow state instability risk level. The output value of the coordination rule is preset in the matrix unit, and the output value is the system risk coupling coefficient between 0 and 1. The matrix unit assignment rule follows the risk superposition principle: when the row and column risk levels are both high, the maximum value is output.

[0112] The system risk coupling coefficient is output by preset coordination rules of dynamic game matrix. The optical safety deviation value is calculated in real time to determine its corresponding grade. The flow state instability risk value is calculated to determine its risk grade. The unit at the intersection of the row and column in the decision matrix is located, and the preset output value is read as the system risk coupling coefficient. When the input value is at the boundary of the grade, the preset rule is used to determine the corresponding grade.

[0113] The phase transition margin index is generated by comparing the system risk coupling coefficient with the preset phase transition threshold. The preset phase transition threshold is determined by analyzing historical traffic flow data. The phase transition margin index is calculated as the preset phase transition threshold minus the system risk coupling coefficient, and the result value can be positive or negative. A positive value indicates that there is a margin from the phase transition boundary in the current state, and a negative value indicates that the phase transition boundary has been exceeded. The absolute value of the index reflects the degree of deviation.

[0114] The phase transition margin index quantifies the difference between the system risk coupling coefficient and the preset phase transition threshold to represent the degree of deviation of the traffic flow metastable state from the phase transition boundary. The specific representation relationship is:

[0115] Positive deviation: When the phase transition margin index is positive, it indicates that the system risk coupling coefficient is lower than the phase transition threshold, and the traffic flow is in a metastable state and far from the phase transition boundary. The larger the index value, the greater the buffer space from the phase transition boundary, and the higher the stability. For example, when the index value is 0.3, the traffic flow can withstand about 30% additional risk disturbance without phase transition.

[0116] Critical deviation: When the index is close to zero (e.g., in the ±0.05 interval), it indicates that the system risk coupling coefficient is approaching the phase transition threshold, and the traffic flow is in a phase transition critical state. At this time, a small disturbance may trigger a flow state mutation, and the primary warning needs to be activated.

[0117] Negative deviation: When the index is negative, it indicates that the system risk coupling coefficient has exceeded the phase transition threshold, and the traffic flow has broken through the metastable state boundary and entered the instability phase transition process. The larger the absolute value of the index, the deeper the instability. For example, when the index is -0.2, the highest level of control needs to be implemented immediately.

[0118] This index converts the abstract phase transition boundary into a quantifiable numerical scale, directly reflecting the traffic flow stability grade and deviation direction through the sign and numerical value, providing accurate basis for hierarchical control.

[0119] The determination method of the preset phase transition threshold is as follows: the corresponding relationship between the traffic flow parameters and the accident data in the historical fog event is statistically analyzed, the correlation coefficients of the warning accuracy and the false alarm rate under different thresholds are calculated, and the value that makes the comprehensive evaluation index optimal is selected as the threshold. This value is obtained by traversing the test method in a reasonable range, and a typical value is a specific value in the interval of 0.2 to 0.5.

[0120] In the calculation of optical safety deviation, the spatial conflict relationship considers the dynamic characteristics of vehicles: when special driving behavior is detected, a safety compensation factor is added to the original ratio. The turning points of the piecewise function are calibrated by experiments.

[0121] The determination of the critical traffic density introduces a dynamic adjustment mechanism: according to real-time environmental conditions, a specific proportion is floated based on the benchmark value. The floating rules consider meteorological factors and road surface conditions. The adjusted critical value needs to be range checked.

[0122] The construction of the dynamic game matrix uses a combination of experience and data: historical data is collected to optimize the matrix unit assignment. The matrix is regularly updated and maintained.

[0123] The reading process of the system risk coupling coefficient includes boundary processing: when the input value is close to the boundary of the level, interpolation calculation is performed according to the value of the adjacent unit. The interpolation weight is determined according to the deviation distance proportion.

[0124] After the generation of the phase change margin index, data smoothing processing is performed: the moving average method is used to eliminate random fluctuations. When the index value of consecutive multiple periods is below the warning threshold, the subsequent control mechanism is triggered.

[0125] In terms of abnormal data processing, when the visibility value is abnormal, the value of the previous valid period is automatically used to replace it; when the vehicle spacing data is missing, the theoretical value is calculated according to the traffic density. The calculation method considers the vehicle conversion coefficient.

[0126] The interval of the optical safety deviation level is divided by analysis and optimization: the boundary value is adjusted to observe the changes in the warning effect, and the optimal setting is selected.

[0127] The calculation of the flow instability risk value introduces a time accumulation effect: when the high-density state continues for more than a certain period of time, the risk value is accumulated at a preset rate, and an upper limit value is set.

[0128] The unit assignment of the decision matrix is set to have a nonlinear characteristic: in the double-high-risk area, a super-linear growth rule is used.

[0129] The setting of the phase change threshold considers the differences in road characteristics: according to the road geometric alignment, the parameters are automatically matched and adjusted.

[0130] The output value of the system risk coupling coefficient is subjected to data quality checking: when the sensor integrity rate of the input data source is insufficient, an error mark is added to the output value for subsequent judgment.

[0131] S4, when the phase change margin index is lower than the preset threshold, the maximum speed reduction value is dynamically constrained according to the fog movement trend and the corrected traffic density, so that the maximum speed reduction value is positively correlated with the phase change margin index, and the implementation is as follows:

[0132] When the phase transition margin index is lower than the preset threshold, obtain the spatial position migration data of the fog cluster moving trend. The phase transition margin index comes from the calculation result of the previous step, and the preset threshold is determined by analyzing historical traffic flow phase transition data. The fog cluster moving trend is obtained by analyzing the visibility spatiotemporal distribution data at consecutive time points: extract a set of visibility distribution data for multiple time periods, calculate the visibility change vector of each location unit; identify the area with visibility lower than a certain value as the core area of the fog cluster; track the spatial displacement of the core area centroid, and the displacement direction is the moving trend direction, and the displacement speed is calculated by dividing the displacement distance by the time interval. The spatial position migration data includes three elements: moving direction angle, moving speed, and core area coverage area.

[0133] The method for determining the preset threshold is as follows: by statistically analyzing the traffic flow phase transition critical point data in historical fog cluster events, analyzing the correlation between accident occurrence rate and traffic flow parameters, and selecting the value that makes the warning accuracy the highest as the threshold. This method is based on a large amount of measured data optimization, ensuring the scientificity of the threshold setting.

[0134] Extract the road spatial distribution characteristics corresponding to the corrected traffic density. The corrected traffic density comes from the output result of the previous step, and the road spatial distribution characteristics are extracted by the following method: divide the target road section into subintervals of fixed length; assign the corrected traffic density value according to the vehicle positioning information; generate a density-spatial distribution curve, and calculate the characteristic quantities of the curve, including the density peak position, the density change gradient, and the continuous length of the high-density area. The high-density area is defined as the area where the density value is greater than a certain proportion of the critical density.

[0135] Establish a dynamic coupling constraint mechanism between the fog cluster moving trend and the corrected traffic density. The coupling mechanism includes two constraint functions: the first function relates the fog cluster moving speed to the speed limit reduction amplitude sensitivity coefficient; the second function relates the traffic density distribution to the speed limit reduction amplitude spatial range. Dynamic coupling is achieved by weighted combination: set the moving speed weight and the density distribution weight; when the angle between the fog cluster moving direction and the traffic flow direction is less than a certain angle, adjust the weight proportion. The constraint mechanism outputs a constraint intensity coefficient.

[0136] Determine the constraint boundary of the maximum speed limit reduction amplitude based on the dynamic coupling constraint mechanism. The constraint boundary is determined by the reference reduction amplitude and the constraint intensity coefficient: the reference reduction amplitude is determined according to the road type; the upper limit of the constraint boundary is the reference reduction amplitude multiplied by the constraint intensity coefficient; the lower limit of the constraint boundary is a fixed proportion of the reference reduction amplitude. The constraint boundary is represented in the form of a numerical interval and is updated regularly.

[0137] The speed limit reduction amplitude proportionality coefficient is calculated by the deviation of the phase change margin index from the preset threshold. The deviation is calculated as the difference between the preset threshold and the phase change margin index. The proportionality coefficient is defined as a piecewise function of the deviation: a smaller coefficient is taken when the deviation is smaller; a medium coefficient is taken when the deviation is medium; and a larger coefficient is taken when the deviation is larger. The proportionality coefficient is set with an upper limit to prevent excessive speed reduction. A time decay factor is introduced in the calculation process: if the deviation state lasts for a certain time, the proportionality coefficient is increased at a fixed rate.

[0138] The final speed limit reduction amplitude maximum value is generated by applying the speed limit reduction amplitude proportionality coefficient to the constraint boundary. The generation algorithm is: multiply the upper limit of the constraint boundary by the speed limit reduction amplitude proportionality coefficient, and the calculation result is taken as the final speed limit reduction amplitude maximum value. The calculation result is smoothed: compared with the previous period value, if the change amplitude exceeds a reasonable threshold, a weighted average processing is adopted. The final value is stored in the control instruction cache area.

[0139] In the spatial position migration analysis, the fog cluster core area identification adopts double criteria: the visibility value is lower than a certain threshold and the area size is larger than a minimum limit. The centroid displacement calculation considers the boundary effect: when the fog cluster contacts the road boundary, the moving direction is automatically corrected. The moving speed calculation adopts the difference method.

[0140] The road space distribution feature extraction includes abnormality processing: when the data of a certain sub-interval is missing, the average value of the adjacent sub-interval is used to fill in. The density change gradient calculation adopts the numerical differentiation method.

[0141] The weight distribution of the dynamic coupling constraint mechanism is optimized through data analysis: historical fog cluster event data is collected to establish a model to fit the weight combination. The range of the constraint strength coefficient is set with reasonable boundaries.

[0142] The constraint boundary determination process considers the influence of road alignment: a reduction coefficient is added to the curved road section; the lower limit value is adjusted on the road section with large longitudinal slope. Reasonability check is performed before the boundary value is output: compared with the adjacent road section value, if the difference is too large, smooth transition is started.

[0143] The turning point of the piecewise function of the speed limit reduction amplitude proportionality coefficient is calibrated through experiments: the safety braking demand is tested under different deviation conditions. The growth rate of the time decay factor is adjusted according to the traffic flow state.

[0144] The generation of the final speed limit reduction amplitude maximum value increases the traffic state feedback: when an emergency is detected, the proportionality coefficient is automatically increased by a certain value. The smoothing processing adopts the filtering algorithm.

[0145] In special scene processing, when the fog cluster dissipates quickly, the constraint strength coefficient is forced to decrease; when the traffic density drops sharply, the proportionality coefficient is reset. These mechanisms avoid drastic fluctuations in control instructions.

[0146] The space adaptation is performed when the constraint boundary is applied: according to the local density value of the location where the vehicle is located, the actual reduction value is dynamically adjusted within the constraint boundary range. The adaptation is realized through the vehicle positioning system, and each vehicle receives differentiated speed limit instructions. The adaptation rule is: the reduction value in high-density areas takes a higher proportion of the upper limit of the boundary, and the reduction value in low-density areas takes a lower proportion of the lower limit of the boundary. The space adaptation is updated every 10 seconds to ensure matching with the real-time traffic state.

[0147] Steps S3 and S4 construct a meteorological-traffic coupling risk dynamic evaluation and adaptive speed limit control chain. Compared with the traditional independent monitoring of visibility or density: S3 establishes a dynamic game matrix that couples the optical safety deviation (representing the spatial conflict between visibility and vehicle spacing) and the flow instability risk value (reflecting the deviation between the corrected density and the critical density), solving the defect of fragmented evaluation of meteorological risk and traffic flow risk. The system risk coupling coefficient is quantified through the matrix unit preset rule, accurately capturing the phase transition critical characteristics of traffic flow in a fog environment. S4 dynamically constrains the speed limit reduction based on the phase transition margin index, breaking through the existing technology of fixed threshold speed limit mode. By establishing a dynamic coupling constraint mechanism of fog movement trend and corrected traffic density, the maximum speed limit reduction is positively adjusted with the phase transition margin: when the margin decreases, the reduction boundary is automatically expanded, and vice versa, realizing risk classification response and minimizing the loss of traffic efficiency under the premise of safety. By extracting the spatial migration data of the fog movement trend and the spatial distribution characteristics of the traffic density, the space-time adaptation of the speed limit instruction is realized in S4. Unlike the existing technology of uniform speed limit on the road section, differentiated speed limit values are generated according to the local density and the approaching state of the fog, solving the control mismatch problem caused by the mobility of the fog and the spatial and temporal heterogeneity of the traffic flow.

[0148] S5, based on the maximum speed limit reduction after constraint, generates a target speed limit value, and calculates a safe vehicle distance instruction combined with the spatial and temporal distribution data of visibility, and the specific implementation is as follows:

[0149] The current reference speed limit value is obtained. The current reference speed limit value is stored in the road attribute database and is dynamically adjusted according to the weather conditions according to the road design standard. The specific acquisition process is as follows: determine the road number and stake number range where the target vehicle is located; query the reference speed limit value of the road section, which is set according to the design speed in sunny conditions; superimpose the meteorological correction term: when the visibility is less than 1000 meters, the correction program is started, and the correction amount is determined according to the visibility classification table. For example, when the visibility is 200-500 meters, the reference speed limit value of the expressway is reduced from 120 kilometers / hour to 100 kilometers / hour. This value is updated every 30 seconds, and the data validity is checked when updating: if the meteorological sensor fails, the previous value is used and an abnormal state is marked.

[0150] The current benchmark speed limit value is reduced by the speed limit reduction maximum value to generate a target speed limit value. The speed limit reduction maximum value comes from the output result of step S4. Data alignment is performed before the subtraction calculation: when the benchmark speed limit value update period and the reduction value update period are inconsistent, take the effective value with the latest timestamp. The calculation process includes boundary protection: the result value cannot be lower than the legal minimum speed limit (such as 60 km / h on a highway); when the calculation result is lower than the lower limit, it is automatically corrected to the lower limit value. For example, when the benchmark speed limit is 100 km / h and the reduction maximum value is 30 km / h, the target speed limit value is 70 km / h. The calculation result is temporarily stored in the control command buffer, waiting for subsequent processing.

[0151] The visibility value at the current time point is extracted from the visibility spatiotemporal distribution data. The visibility spatiotemporal distribution data comes from the spatiotemporal matrix generated in step S1, and the extraction method is: locate the column index of the current timestamp in the time series matrix; obtain all row element values of the column to form a spatial distribution set; match the nearest road grid cell according to the target vehicle GPS coordinates; when the vehicle is at the junction of two grid cells, take the arithmetic mean of the visibility values of the two cells. The unit of the visibility value is meters, and data verification is performed when extracting: if the value exceeds the reasonable range (0-2000 meters), start the abnormal processing program: use the 10-minute moving average value to replace and trigger the sensor calibration instruction.

[0152] The minimum safe following distance threshold is calculated based on the target speed limit value and the visibility value. The calculation principle is: the safe following distance needs to meet the braking and stopping of the following vehicle within the visibility limit distance. The specific implementation steps are: first, determine the braking reaction time, which includes the driver's cognitive delay and the braking system response delay, which is calibrated as a fixed value through real vehicle testing; second, calculate the braking distance, using the physical kinematics formula: the braking distance is equal to the square of the target speed limit value divided by twice the deceleration; the deceleration value is determined according to the road adhesion coefficient, which comes from the real-time monitoring data of the roadside sensor. The minimum safe following distance threshold is equal to the sum of the braking reaction time travel distance and the braking distance, multiplied by a safety factor. The safety factor is set according to the visibility classification: when the visibility is less than 100 meters, take 1.5; when the visibility is between 100 and 200 meters, take 1.2; when the visibility is more than 200 meters, take 1.1.

[0153] The safe following distance instruction is generated according to the minimum safe following distance threshold. The instruction generation includes data conversion and packaging: convert the calculated safe following distance threshold to the standard instruction format; add the timestamp and road segment location code; set the instruction priority: set the highest priority when the visibility is less than 50 meters. Logical verification is performed before the instruction is transmitted: compare the safe following distance threshold with the current actual following distance, and add a warning marker when the threshold is greater than the actual following distance. The instruction is broadcast through the vehicle-infrastructure cooperation communication unit, and the data packet structure includes: instruction type code, recommended following distance value, valid time length, and applicable road segment code.

[0154] In the benchmark speed limit acquisition, the implementation details of the meteorological correction term are as follows: an visibility-speed limit mapping table is established, and the speed limit decreases by a fixed step for each certain interval of decrease in visibility. The mapping table is calibrated by combining expert experience and historical accident data, and a new version is updated once a year. When updating, the correlation between the newly added accident data and the speed limit setting is analyzed.

[0155] The target speed limit value calculation adds a smoothing process: compared with the target speed limit value of the previous period, if the change amplitude exceeds a reasonable threshold (for example, 15%), a gradual adjustment mechanism is adopted. The adjustment rate is set to not more than 5% of the speed limit value per minute, to avoid sudden changes in vehicle speed causing danger.

[0156] Optimization of visibility value extraction space matching: when there is a deviation between the vehicle position and the monitoring grid, a bilinear interpolation algorithm is used. Four adjacent grid points are taken as the center of the vehicle, and the final visibility value is calculated by inversely weighting the distance. The interpolation weight is calculated as the inverse normalized value of the square of the distance between each grid point and the vehicle.

[0157] Dynamic compensation is introduced in the calculation of safe following distance: when it is detected that it is raining or snowing, an additional road slip compensation term is added in the calculation of braking distance. The compensation amount is determined according to the intensity of precipitation: 10% for light rain, 20% for moderate rain, and 30% for heavy rain. The compensation coefficient is determined through vehicle dynamics simulation tests.

[0158] The instruction generation mechanism includes a feedback channel: it receives the instruction execution rate data returned by the vehicle terminal, and automatically enhances the instruction intensity when the execution rate continuously falls below a certain threshold. The enhancement methods include: increasing the instruction sending frequency, adding sound and light warning signs, and raising the warning level. The execution rate threshold is set to 70%, and if it is below this value for three consecutive periods, the enhancement mechanism is triggered.

[0159] In special scenarios, when the visibility changes sharply (the change rate of adjacent periods exceeds 30%), the safe following distance calculation starts the fast response mode: it directly uses the latest data without smoothing, and the instruction validity period is shortened to 50% of the normal value. When the traffic flow is in a synchronous flow state, the safe following distance threshold is additionally multiplied by a coordination coefficient of 0.8 to promote coordinated driving of vehicle platoon.

[0160] The safe following distance instruction implementation is differentiated in space: adjustment coefficients are set according to vehicle types. The coefficient for large trucks is 1.2, for passenger cars it is 1.0, and for emergency vehicles it is 0.9. The coefficient is set according to the vehicle braking performance parameter library, which integrates the vehicle model braking test data recorded by the Ministry of Industry and Information Technology. The final instruction is automatically matched with the vehicle type code to achieve personalized push.

[0161] S6, target speed limit value and safe following distance instructions are issued to vehicles in fog-affected road sections, and the specific implementation is as follows:

[0162] A dynamic spatial range of the fog-affected road section is determined. The dynamic spatial range is determined by fusing the fog movement trend and the real-time visibility distribution: first, the spatial position migration data of the fog movement trend generated in the S4 step is obtained, including the moving direction angle and the moving speed; the position of the fog front in the future set time period is predicted based on the moving speed; and a polygon coordinate set of the dynamic spatial range is generated in combination with the region boundary where the visibility is lower than the critical value (for example, 100 meters) in the current visibility spatio-temporal distribution data. The range is updated every 30 seconds, and the fog diffusion effect is considered when updating: when the visibility change rate exceeds the set threshold, the predicted front position is increased by a diffusion compensation distance.

[0163] The target speed limit value and the safe distance instruction are packaged into a cooperative control instruction set. The packaging process includes data structure conversion and information fusion: the target speed limit value comes from the calculation result of the S5 step, and the safe distance instruction comes from the output result of the S5 step; the instruction set includes three core fields: the speed limit instruction field stores the target speed limit value and the effective position range, the distance instruction field stores the minimum safe distance threshold and the applicable condition, and the timestamp field records the instruction generation time. A check code is added when packaging: a cyclic redundancy check calculation is performed on the instruction content to generate an 8-bit check code attached to the end of the data packet. The instruction set adopts a lightweight design, and the single packet data volume is controlled within 200 bytes.

[0164] The cooperative control instruction set is broadcast to vehicles within the dynamic spatial range through a vehicle-road cooperative communication protocol. The broadcast implementation process is: activating the roadside communication unit according to the dynamic spatial range; the communication protocol adopts the LTE-V2X standard mode; the broadcast period is set according to the traffic risk level: the broadcast is performed once every 5 seconds when the risk level is low, and the broadcast is performed once every 1 second when the risk level is high. The broadcast range is dynamically controlled: only the communication units located within the dynamic spatial range are allowed to send instructions, and the units outside the range remain silent. A priority label is added to the broadcast data packet: when the visibility is lower than 50 meters, it is set to the highest priority, and the non-safety communication is interrupted.

[0165] The parameter values of the broadcast instruction are dynamically adjusted according to the vehicle position information. The dynamic adjustment mechanism includes: real-time receiving of the latitude and longitude coordinates uploaded by the vehicle; spatial matching of the coordinates with the dynamic spatial range; when the vehicle approaches the fog boundary (for example, within 500 meters of the boundary), a preparatory speed reduction amount is added to the target speed limit value; when the vehicle is in the core area of the fog (the visibility is lower than 50 meters), the safe distance instruction value is multiplied by an emergency correction coefficient. The parameter adjustment result is issued in real time through instruction re-packaging to ensure that each vehicle receives personalized instructions.

[0166] The determination of dynamic spatial range includes exception handling: when the group fog moving speed detection is abnormal, start the conservative estimation mode: expand the current visibility distribution boundary by a fixed distance as the temporary range. The expansion distance is set according to historical data statistical analysis, for example, the typical value of highway scene is 1000 meters. The temporary range lasts until the detection data returns to normal.

[0167] The instruction encapsulation process implements data compression: difference coding is used for the target speed limit value, and the reference value is the road design speed; and graded coding is used for the safe distance, which is divided into intervals of 10 meters. The compression algorithm reduces the data transmission volume by 40%, and adds a decompression identifier for the vehicle terminal to identify.

[0168] The broadcast communication establishes a feedback mechanism: monitor the signal receiving strength value returned by the vehicle terminal; when the receiving strength is lower than the set threshold, automatically increase the transmission power or shorten the broadcast period. The feedback data is analyzed every 15 seconds, and a communication quality heat map is generated to optimize the roadside unit layout.

[0169] The parameter dynamic adjustment introduces a vehicle type factor: after identifying the vehicle type code, the safe distance instruction value of truck type vehicles is increased by a type compensation amount, and the compensation coefficient is set according to the vehicle total mass classification: less than 10 tons by 1.1 times, 10-20 tons by 1.2 times, and more than 20 tons by 1.3 times. The compensation coefficient is derived from the vehicle braking performance database.

[0170] In the fast-moving scene of group fog, the dynamic spatial range prediction uses a second-order extrapolation algorithm: based on the displacement data of the previous two periods, the acceleration is calculated, and the predicted position formula is the current position plus the moving speed multiplied by time plus half the acceleration multiplied by the square of time. The extrapolation time window is dynamically adjusted according to the visibility change rate: when the change rate is large, it is shortened to 1 minute, and when the change rate is small, it is extended to 3 minutes.

[0171] The instruction broadcast implements air interface resource scheduling: when the communication is congested, start the time division multiplexing mechanism, divide the dynamic spatial range into multiple sub-regions, and broadcast in a set time slice. The time slice allocation algorithm prioritizes high-risk sub-regions, and the risk level is calculated according to the weighted sum of visibility value and traffic density.

[0172] The parameter adjustment process considers the inter-vehicle communication capability: when the inter-vehicle communication delay is detected to be lower than the set threshold, the cooperative following control parameter is integrated into the safe distance instruction. This parameter enables the following vehicle to predict the braking behavior of the leading vehicle, reducing the safe distance requirement to 90% of the standard value. The cooperative parameter is updated in real time through the inter-vehicle communication link.

[0173] For large convoy scenarios, the broadcast mechanism adds a relay forwarding function: designating the lead vehicle of the convoy as a relay node to forward instructions to following vehicles that are outside the direct communication range. The relay selection algorithm is based on vehicle position sequence and communication capability index to ensure that the instruction synchronization rate of the entire convoy exceeds 95%.

[0174] Version management is performed during instruction updates: each frame of broadcast data contains a version sequence number, and the onboard terminal compares this number with the locally stored version number to determine whether to update the instruction. The version conflict handling rules are as follows: higher-priority instructions forcibly overwrite lower-priority instructions; when priorities are the same, the instruction with the latest timestamp is used. This mechanism avoids control jitter caused by frequent instruction switching.

[0175] Example 2: Figure 2 A schematic diagram of the elastic speed limit and vehicle distance coordinated control system for highway fog scenarios is provided. The elastic speed limit and vehicle distance coordinated control system for highway fog scenarios includes:

[0176] The traffic environment perception module is used to collect real-time traffic operation status parameters and visibility spatiotemporal distribution data of road sections affected by fog. Traffic operation status parameters include traffic density and vehicle spacing.

[0177] The density dynamic correction module is used to calculate the visibility gradient direction at the current time point based on the spatiotemporal distribution data of visibility. When the visibility gradient direction is consistent with the traffic flow direction, the traffic density is weighted and amplified to generate the corrected traffic density.

[0178] The flow margin assessment module is used to calculate the phase change margin index of traffic flow based on the corrected traffic density and optical safety deviation. The optical safety deviation is calculated based on vehicle spacing and spatiotemporal distribution data of visibility.

[0179] The speed limit dynamic constraint module is used to dynamically constrain the maximum speed limit reduction based on the movement trend of the fog and the traffic density correction when the phase change margin index is lower than the preset threshold, so that the maximum speed limit reduction is positively correlated with the phase change margin index.

[0180] The instruction co-generation module is used to generate the target speed limit value based on the maximum value of the constrained speed limit reduction, and to calculate the safe distance instruction by combining visibility spatiotemporal distribution data.

[0181] The instruction broadcasting execution module is used to issue target speed limits and safe following distance instructions to vehicles in road sections affected by fog.

[0182] All calculations involved in the embodiments are dimensionless numerical calculations, and the preset parameters and thresholds in the calculations are set by those skilled in the art according to the actual situation.

[0183] It should be noted that the present application can be deployed in the device itself to realize embedded application, or run on PC terminal or other terminal with user interface, so as to meet various hardware environment and use requirements.

[0184] The above-described embodiments can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented by software, the above-described embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the processes or functions described in the embodiments of the present application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another, for example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center through wireless or wired direction. The wired transmission mode includes optical fiber, twisted pair, coaxial cable, etc. The wireless transmission includes infrared, microwave, etc. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. containing one or more available medium collections. The available medium can be a magnetic medium (for example, floppy disk, hard disk, magnetic tape), optical medium (for example, DVD), or semiconductor medium. The semiconductor medium can be a solid state disk.

[0185] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the above-described system, device and module can refer to the corresponding process in the foregoing method embodiments, which will not be described here.

[0186] In several embodiments provided in the present application, it should be understood that the disclosed system, device and method can be implemented in other ways. For example, the above-described device embodiments are only schematic, for example, the division of the modules is only a logical function division, and actual implementation can have another division manner, for example, a plurality of modules or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the displayed or discussed each other can be indirect coupling or communication connection through some interfaces, devices or modules, which can be electrical, mechanical or other forms.

[0187] The modules described as separate components may or may not be physically separate, and the components shown as modules may or may not be physical modules, and may be located in one place or distributed on multiple network modules. Part or all of the modules can be selected according to actual needs to achieve the purpose of the embodiment of the present application.

[0188] In addition, the functional modules in each embodiment of the present application can be integrated into one processing module, or each module can exist physically alone, or two or more modules can be integrated into one module.

[0189] The functions, if realized in the form of software function modules and sold or used as independent products, can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application essentially or the part that contributes to the prior art or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a number of instructions for causing a computer device (which can be a personal computer, a server or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various program code storage media.

[0190] The above is only a specific embodiment of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or replacements within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

[0191] Finally: the above is only the preferred embodiment of the present application, and is not used to limit the present application, any modification, equivalent replacement, improvement, etc. within the spirit and principles of the present application, should be included in the protection scope of the present application.

Claims

1. A method for coordinated regulation of elastic speed limit and vehicle distance in a fog scenario on an expressway, characterized in that, The method comprises the following steps: S1, real-time collect traffic running state parameters and visibility spatiotemporal distribution data of a fog-affected road section, the traffic running state parameters including traffic density and vehicle spacing; S2, calculate the visibility gradient direction at the current time point according to the visibility spatiotemporal distribution data, and when the visibility gradient direction is consistent with the traffic flow direction, the traffic density is weighted and enlarged to generate a corrected traffic density; S3, calculate a phase transition margin index of the traffic flow based on the corrected traffic density and the optical safety deviation, including: extract the visibility value at the current time point from the visibility spatiotemporal distribution data, and obtain the current vehicle spacing detection value; determine the optical safety deviation based on the spatial conflict relationship between the vehicle spacing and the visibility value, and the spatial conflict relationship quantification method is: calculate the ratio of the vehicle spacing to the visibility value, and the optical safety deviation is defined as the output of the conversion function of the ratio, and the increase of the output value indicates the increase of the safety deviation degree; generate a flow state instability risk value according to the relative difference between the corrected traffic density and the critical traffic density, and the relative difference is calculated in percentage form: take the difference between the corrected traffic density and the critical traffic density, divide the critical traffic density by the percentage coefficient, and the flow state instability risk value is defined as the mapping result of the percentage value; establish a dynamic game matrix of the optical safety deviation and the flow state instability risk value, including: defining the grade division of the optical safety deviation; setting the risk level of the flow state instability risk value; constructing a decision matrix with the optical safety deviation level as the row and the flow state instability risk value level as the column; preset the coordination rule output value as the system risk coupling coefficient in the unit of the decision matrix, and the matrix unit preset coordination rule output value follows the risk superposition principle; output the system risk coupling coefficient through the preset coordination rule of the dynamic game matrix, and the preset coordination rule includes: positioning the matrix unit according to the optical safety deviation level and the flow state instability risk value level, and outputting the preset system risk coupling coefficient; compare the system risk coupling coefficient with the preset phase transition threshold to generate the phase transition margin index, and the phase transition margin index represents the degree of deviation of the traffic flow metastable state from the phase transition boundary; the phase transition margin index is calculated by subtracting the system risk coupling coefficient from the preset phase transition threshold, and the index value is positive, indicating that it is away from the phase transition boundary, and negative, indicating that it exceeds the phase transition boundary; S4, when the phase transition margin index is lower than the preset threshold, dynamically constrain the maximum speed reduction value according to the fog group movement trend and the corrected traffic density, so that the maximum speed reduction value is positively correlated with the phase transition margin index, including: when the phase transition margin index is lower than the preset threshold, obtain the spatial position migration data of the fog group movement trend; extract the road space distribution characteristics corresponding to the corrected traffic density; establish a dynamic coupling constraint mechanism of the fog group movement trend and the corrected traffic density; determine the constraint boundary of the maximum speed reduction value based on the dynamic coupling constraint mechanism; calculate the speed reduction proportion coefficient through the deviation of the phase transition margin index and the preset threshold, and the deviation is the difference between the preset threshold and the phase transition margin index, wherein the larger the deviation is, the larger the speed reduction proportion coefficient is; apply the speed reduction proportion coefficient to the constraint boundary to generate the final maximum speed reduction value; S5, generating a target speed limit value based on the constrained maximum speed reduction range, and calculating a safe distance instruction in combination with the visibility spatiotemporal distribution data; S6, issuing the target speed limit value and the safe distance instruction to vehicles in the fog-affected road section.

2. The method of claim 1, wherein the method is characterized by, Real-time collection of traffic operation state parameters and visibility spatiotemporal distribution data in the fog-affected road section, including: Real-time collection of traffic density and vehicle spacing through roadside sensors arranged in the fog-affected road section, the roadside sensors including video detection equipment and microwave radar equipment; Real-time collection of visibility spatiotemporal distribution data through a meteorological monitoring station network arranged in the fog-affected road section, the meteorological monitoring station network being composed of multiple meteorological sensors distributed at a preset interval; Transmission of the traffic density, vehicle spacing, and visibility spatiotemporal distribution data to an edge computing node for timestamp alignment processing through a special communication protocol.

3. The method of claim 2, wherein the method further comprises: determining a target speed limit for the vehicle based on the vehicle's current speed, the vehicle's current distance to the vehicle ahead, and the vehicle's current distance to the vehicle behind. According to the visibility spatiotemporal distribution data, the visibility gradient direction at the current time point is calculated, and when the visibility gradient direction is consistent with the traffic flow direction, the traffic density is weighted and enlarged to generate a corrected traffic density, including: Extracting the visibility spatial distribution data at the current time point based on the visibility spatiotemporal distribution data; Determining the visibility gradient direction by calculating the visibility change rate between adjacent meteorological monitoring stations; Calculating the vector angle between the visibility gradient direction and the preset traffic flow direction; When the vector angle is less than or equal to a preset angle threshold, it is determined that the visibility gradient direction is consistent with the traffic flow direction; The traffic density is enlarged by a preset weighting coefficient to generate a corrected traffic density.

4. The method of claim 3, wherein, The preset weighting coefficient is determined according to the visibility gradient direction change characteristics, and the real-time detected traffic density is multiplied by the preset weighting coefficient to generate the corrected traffic density. Generating a target speed limit value based on the constrained maximum speed reduction range, and calculating a safe distance instruction in combination with the visibility spatiotemporal distribution data, including: Obtaining a current reference speed limit value; Generating a target speed limit value by subtracting the maximum speed reduction range from the current reference speed limit value; 5. The method of claim 4, wherein the method further comprises: Extracting a visibility value at the current time point from the visibility spatiotemporal distribution data; Calculating a minimum safe distance threshold based on the target speed limit value and the visibility value; Generating a safe distance instruction according to the minimum safe distance threshold. Issuing the target speed limit value and the safe distance instruction to vehicles in the fog-affected road section, including: Determining the dynamic spatial range of the fog-affected road section; Packaging the target speed limit value and the safe distance instruction into a cooperative control instruction set; 6. The method of claim 5, wherein the method further comprises: Broadcasting the cooperative control instruction set to vehicles within the dynamic spatial range through a vehicle-road cooperative communication protocol; Dynamically adjusting the parameter values of the broadcast instruction according to the vehicle position information. Including: A traffic environment perception module for real-time collection of traffic operation state parameters and visibility spatiotemporal distribution data in the fog-affected road section, the traffic operation state parameters including traffic density and vehicle spacing; A density dynamic correction module for calculating the visibility gradient direction at the current time point according to the visibility spatiotemporal distribution data, and weighting and enlarging the traffic density when the visibility gradient direction is consistent with the traffic flow direction to generate a corrected traffic density; 7. The system for cooperative regulation of elastic speed limit and vehicle distance under the expressway fog group scene, used for realizing the method for cooperative regulation of elastic speed limit and vehicle distance under the expressway fog group scene according to any one of claims 1-6, characterized in that, ​ ​ ​ The flow state margin evaluation module is configured to calculate a phase transition margin index of the traffic flow based on the corrected traffic density and the optical safety deviation, and the optical safety deviation is calculated according to the vehicle distance and the spatiotemporal distribution data of the visibility. The amplitude-limited dynamic constraint module is configured to dynamically constrain the maximum speed reduction amplitude according to the moving trend of the fog cluster and the corrected traffic density when the phase transition margin index is lower than a preset threshold, so that the maximum speed reduction amplitude is positively correlated with the phase transition margin index. The instruction coordination generation module is configured to generate a target speed limit value based on the constrained maximum speed reduction amplitude, and calculate a safe distance instruction in combination with the spatiotemporal distribution data of the visibility. The instruction broadcast execution module is configured to publish the target speed limit value and the safe distance instruction to vehicles in the fog-affected road section.

Images