A highway dynamic speed limit recommendation method and system based on car-road cooperation technology

Abstract

The application belongs to the technical field of intelligent traffic management and control, and discloses a highway dynamic speed limit recommendation method and system based on vehicle-road cooperation technology, which comprises the following steps: time-space matching of data of roadside equipment and vehicle-mounted terminals through a time stamp alignment algorithm to generate a synchronous traffic state data set; judging the high-risk area boundary according to the visibility sudden drop signal and the distribution density of the vehicle emergency braking behavior in the synchronous data set, and triggering the dynamic speed limit calculation module if the proportion of the emergency braking vehicles in the area exceeds a preset threshold; issuing a differentiated speed limit instruction through a vehicle-road cooperation communication link, adopting a basic safety vehicle speed for small passenger cars and a compensated vehicle speed for heavy trucks, and synchronously displaying the electronic road sign and the vehicle-mounted terminal; and monitoring the change trend of the average speed of the vehicle flow after the execution of the speed limit instruction in real time, and reacquiring the vehicle acceleration data of the road section and triggering the iteration optimization of the speed limit value if the speed variance of the local road section continuously increases.

Description

Technical Field

[0001] This invention belongs to the field of intelligent traffic management and control technology, specifically relating to a method and system for recommending dynamic speed limits on highways based on vehicle-road cooperative technology. Background Technology

[0002] As the core artery of modern transportation networks, highways directly impact regional economic development and social operations through their traffic efficiency and safety levels. The rise of vehicle-road cooperative technology has provided new ideas for improving highway management efficiency, with dynamic speed limit control being a crucial means of balancing traffic density and driving safety. Currently, mainstream speed limit management relies primarily on fixed signs or manual intervention. While this can handle routine scenarios, static speed limit schemes struggle to respond promptly to changes in road conditions under complex conditions such as sudden accidents and extreme weather. More importantly, existing systems lack the ability to fuse and process multi-source heterogeneous data. For example, they cannot correlate real-time vehicle braking behavior with the slippery condition of the road ahead, leading to delayed warnings or speed limit recommendations that do not match actual risks.

[0003] The core technical challenge in this field lies in the contradiction between the real-time nature of traffic condition perception and the accuracy of decision-making. Road environment information (such as visibility and road surface friction coefficient) needs to be collected at high frequency by roadside equipment, while vehicle operation data (such as vehicle speed and acceleration) is subject to communication delays and packet loss risks. The spatiotemporal alignment accuracy of these two data streams directly determines the reliability of speed limit recommendations. When heavy rain causes a sudden drop in visibility on a certain road section, if the system fails to simultaneously obtain speed reduction requests from all vehicles in that area, conflicting speed limit instructions may be generated, exacerbating local congestion. A deeper contradiction lies in the difficulty of balancing the global optimization of dynamic speed limits with individual adaptability. Traditional algorithms often focus on overall traffic flow balance but neglect the differentiated speed limit requirements of heavy trucks and passenger cars, a contradiction that is particularly prominent on special road sections such as long downhill slopes.

[0004] Establishing a dynamic speed limit mechanism that balances real-time response capabilities with multi-dimensional risk assessment has become a key issue in improving the effectiveness of proactive traffic management on highways. Especially in complex road sections such as continuous curves and tunnel complexes, when sudden fog causes some vehicles to brake suddenly, the system needs to quickly calculate safe speed thresholds while avoiding cascading deceleration of following traffic due to excessive speed limits. This technical requirement of ensuring millisecond-level decision-making speed while achieving deep coupling of vehicle-road data constitutes a critical business bottleneck that urgently needs to be overcome in the current field of intelligent transportation. Summary of the Invention

[0005] To address the problems existing in the prior art, this invention provides a method and system for recommending dynamic speed limits on highways based on vehicle-road cooperative technology, which effectively reduces traffic accident rates and improves road traffic efficiency.

[0006] To achieve the above objectives, the present invention provides the following solution: A method for recommending dynamic speed limits on highways based on vehicle-road cooperative technology, the method comprising: The system acquires visibility and road surface friction coefficient data collected by roadside equipment, and simultaneously receives real-time vehicle speed and acceleration information uploaded by vehicle-mounted terminals. A timestamp alignment algorithm is used to perform spatiotemporal matching of the data collected by roadside equipment and the data uploaded by vehicle-mounted terminals to obtain a spatiotemporally synchronized traffic status dataset. Based on the visibility drop signal and vehicle emergency braking behavior distribution density in the spatiotemporally synchronized traffic status dataset, the boundary of high-risk areas is determined. If the proportion of emergency braking vehicles in the area exceeds a preset threshold, the dynamic speed limit calculation module is triggered. A hierarchical decision-making model is adopted. The first layer calculates the basic safe speed based on the road friction coefficient and visibility. The second layer superimposes the traffic flow density correction coefficient. If the proportion of heavy trucks exceeds 30%, an additional braking distance compensation value is added. Differentiated speed limit commands are issued through the vehicle-road cooperative communication link, using a basic safe speed for passenger cars and a compensated speed for heavy trucks, and are displayed synchronously on electronic road signs and vehicle terminals; The system monitors the average speed change trend of traffic flow after the speed limit command is executed in real time. If the speed variance of a local road segment continues to increase, the system reacquires the vehicle acceleration data for that road segment and triggers iterative optimization of the speed limit value.

[0007] Preferably, the method for acquiring visibility and road surface friction coefficient data collected by roadside equipment, simultaneously receiving real-time vehicle speed and acceleration information uploaded by vehicle-mounted terminals, and using a timestamp alignment algorithm to perform spatiotemporal matching of the data collected by roadside equipment and the data uploaded by vehicle-mounted terminals to obtain a spatiotemporally synchronized traffic state dataset includes: Acquire visibility and friction coefficient data collected by roadside equipment and label them as the first dataset; Receive real-time vehicle speed and acceleration information uploaded by the vehicle terminal and mark it as the second dataset; A timestamp alignment algorithm is used to process the first and second datasets, matching data points with the same timestamp. If the timestamps match successfully, the visibility, friction, vehicle speed, and acceleration data are merged to generate a third dataset. Based on the visibility and friction values ​​in the third dataset, the road surface condition level is determined and labeled as the fourth dataset; The fourth dataset is analyzed using a random forest model, and traffic state prediction results are output.

[0008] Preferably, the method for determining the boundary of a high-risk area based on the visibility drop signal and the distribution density of vehicle emergency braking behavior in the spatiotemporally synchronized traffic state dataset, and triggering the dynamic speed limit calculation module if the proportion of emergency braking vehicles in the area exceeds a preset threshold, includes: Acquire real-time visibility values, timestamps of vehicle emergency braking signals, and vehicle location data; Spatial clustering algorithm is used to divide the high-density areas of sudden braking and generate a distribution map; If the spatiotemporal overlap between a sudden drop point and a number of sudden braking events within a region exceeds a threshold, it is marked as a region edge. Based on the mapping relationship between the percentage value and the preset speed limit value, the dynamic speed limit calculation module is triggered to output the dynamic speed limit value; The speed limit value of the region edge is dynamically updated by fusing the visibility change trend in the signal source with the vehicle displacement using Kalman filtering and then sent to the vehicle terminal.

[0009] Preferably, a hierarchical decision-making model is adopted. The first layer calculates the basic safe vehicle speed based on the road surface friction coefficient and visibility. The second layer superimposes a traffic flow density correction coefficient. If the proportion of heavy trucks exceeds 30%, an additional braking distance compensation value is added. Obtain current road surface friction coefficient and visibility data, and use a linear regression model to calculate the basic safe vehicle speed; Based on real-time traffic density data, the correction coefficient is determined through a preset density-speed mapping table; The adjusted safe speed is obtained by adding the correction factor to the base safe speed. Obtain the percentage of heavy trucks on the current road section. If the percentage exceeds 30%, calculate the braking distance compensation value. The braking distance compensation value is converted into a vehicle speed adjustment amount and then added to the adjusted safe vehicle speed. Output the final recommended safe speed.

[0010] Preferably, the method of issuing differentiated speed limit commands through the vehicle-road cooperative communication link, using a basic safe speed for passenger cars and a compensated speed for heavy trucks, and displaying the results synchronously on electronic road signs and vehicle terminals includes: Obtain real-time vehicle type data based on the communication link; If the vehicle type is a passenger car, the preset safety value is used as the speed limit; If the vehicle type is a heavy vehicle, the speed limit value is obtained by adjusting the safety value based on the compensation value; Speed ​​limits are transmitted to electronic road signs and vehicle terminals via road cooperation; A time synchronization protocol is used to keep the electronic road signs and the vehicle terminal synchronized in terms of vehicle speed; Monitor the status of the communication link, and if it is interrupted, switch to the redundant link maintenance command.

[0011] Preferably, the method of real-time monitoring of the average speed change trend of traffic flow after the execution of the speed limit command, and triggering iterative optimization of the speed limit value by re-acquiring the vehicle acceleration data of that road segment if the speed variance of a local road segment continues to increase, includes: Obtain average traffic speed data for road segments after the speed limit order is executed; Calculate the speed variance within a continuous time window. If the variance exceeds a preset threshold and continues to rise, mark the road segment area. Acceleration data is collected from vehicle terminals in the marked road section area, and the average value is taken after filtering out outliers. A linear regression model is used to establish the mapping relationship between acceleration and the current speed limit value, and the speed limit adjustment coefficient is output. If the absolute value of the adjustment coefficient is greater than the set tolerance, a new speed limit value is generated and sent to the target road segment area; Update the "Historical Rate Limit Iteration Count" field, overwriting the original record after rate limit execution.

[0012] The present invention also provides a dynamic speed limit recommendation system for highways based on vehicle-road cooperative technology. The system is used to implement the aforementioned method and includes: a data acquisition and spatiotemporal matching module, a high-risk area identification module, a hierarchical speed limit decision module, an instruction issuance and display module, and an effect monitoring and iterative optimization module. The data acquisition and spatiotemporal matching module is used to acquire visibility and road surface friction coefficient data collected by roadside equipment, synchronously receive real-time vehicle speed and acceleration information uploaded by vehicle terminal, and use a timestamp alignment algorithm to perform spatiotemporal matching of the two types of data to obtain a spatiotemporally synchronized traffic status dataset. The high-risk area identification module is used to determine the boundary of the high-risk area based on the visibility drop signal and the distribution density of vehicle emergency braking behavior in the synchronous dataset. If the proportion of emergency braking vehicles in the area exceeds a preset threshold, the dynamic speed limit calculation module is triggered. The layered speed limit decision module is used to adopt a layered decision model. The first layer calculates the basic safe speed based on the road friction coefficient and visibility. The second layer superimposes the traffic flow density correction coefficient. If the proportion of heavy trucks exceeds 30%, an additional braking distance compensation value is added. The instruction issuance and display module is used to issue differentiated speed limit instructions through the vehicle-road cooperative communication link, using a basic safe speed for passenger cars and a compensated speed for heavy trucks, and displaying them synchronously on electronic road signs and vehicle terminals. The effect monitoring and iterative optimization module is used to monitor the average speed change trend of traffic flow after the speed limit command is executed in real time. If the speed variance of a local road segment continues to increase, the vehicle acceleration data of that road segment will be reacquired to trigger iterative optimization of the speed limit value.

[0013] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention discloses a dynamic speed limit intelligent decision-making method based on vehicle-road cooperation. Addressing the risk aggregation of vehicles braking suddenly due to a sharp drop in visibility and abrupt changes in road surface friction coefficient under adverse weather conditions, the method fuses environmental data collected by roadside equipment with vehicle behavior data uploaded by onboard terminals using a spatiotemporal matching algorithm to construct a high-risk area boundary recognition model. When the density of vehicles braking suddenly exceeds a threshold, a hierarchical speed limit decision-making mechanism is triggered. An innovative compensation algorithm based on vehicle type differences is proposed, with heavy trucks receiving additional friction coefficient compensation due to their braking distance requirements. Finally, speed limit commands are synchronously issued between electronic road signs and onboard terminals via vehicle-road cooperative communication. A speed variance feedback mechanism is used to dynamically optimize the speed limit strategy, forming a closed loop of "perception-decision-control-optimization," effectively reducing traffic accident rates and improving road traffic efficiency. Attached Figure Description

[0014] To more clearly illustrate the technical solution of the present invention, the drawings used in the embodiments are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0015] Figure 1 This is a schematic diagram of a method for recommending dynamic speed limits on highways based on vehicle-road cooperative technology according to an embodiment of the present invention; Figure 2 This is a schematic diagram of a dynamic speed limit recommendation system for highways based on vehicle-road cooperative technology, according to an embodiment of the present invention. Detailed Implementation

[0016] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0017] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0018] Example 1 like Figure 1 As shown, this invention provides a method for recommending dynamic speed limits on highways based on vehicle-road cooperative technology, the method comprising: The system acquires visibility and road surface friction coefficient data collected by roadside equipment, and simultaneously receives real-time vehicle speed and acceleration information uploaded by vehicle-mounted terminals. A timestamp alignment algorithm is used to perform spatiotemporal matching of the data collected by roadside equipment and the data uploaded by vehicle-mounted terminals to obtain a spatiotemporally synchronized traffic status dataset. Based on the visibility drop signal and vehicle emergency braking behavior distribution density in the spatiotemporally synchronized traffic status dataset, the boundary of high-risk areas is determined. If the proportion of emergency braking vehicles in the area exceeds a preset threshold, the dynamic speed limit calculation module is triggered. A hierarchical decision-making model is adopted. The first layer calculates the basic safe speed based on the road friction coefficient and visibility. The second layer superimposes the traffic flow density correction coefficient. If the proportion of heavy trucks exceeds 30%, an additional braking distance compensation value is added. Differentiated speed limit commands are issued through the vehicle-road cooperative communication link, using a basic safe speed for passenger cars and a compensated speed for heavy trucks, and are displayed synchronously on electronic road signs and vehicle terminals; The system monitors the average speed change trend of traffic flow after the speed limit command is executed in real time. If the speed variance of a local road segment continues to increase, the system reacquires the vehicle acceleration data for that road segment and triggers iterative optimization of the speed limit value.

[0019] In this embodiment, the method for acquiring visibility and road surface friction coefficient data collected by roadside equipment, synchronously receiving real-time vehicle speed and acceleration information uploaded by vehicle-mounted terminals, and using a timestamp alignment algorithm to perform spatiotemporal matching of the data collected by roadside equipment and the data uploaded by vehicle-mounted terminals to obtain a spatiotemporally synchronized traffic state dataset includes: Visibility and friction coefficient data collected by roadside equipment were acquired and labeled as the first dataset. Specifically, the roadside equipment collected visibility data using infrared sensors and lidar, with a measurement range of 50 to 1000 meters and an accuracy of ±5%. The friction coefficient was obtained using embedded pavement sensors, with a sampling frequency of 10Hz and a measurement error of less than 0.05. This data was stored in JSON format, with timestamps using the UTC standard, forming the first dataset.

[0020] The system receives real-time vehicle speed and acceleration information uploaded by the vehicle-mounted terminal and marks it as the second dataset. Specifically, the vehicle-mounted terminal acquires vehicle speed and acceleration data via the CAN bus, with a vehicle speed sampling interval of 100ms and acceleration data collected by a three-axis inertial sensor at a frequency of 50Hz. Typical data in the second dataset, such as the timestamp 2023-05-15T08:30:45Z, corresponds to a vehicle speed of 72km / h and a longitudinal acceleration of -0.3m / s².

[0021] A timestamp alignment algorithm is used to process the first and second datasets, matching data points with the same timestamp. If the timestamps match successfully, the visibility, friction coefficient, vehicle speed, and acceleration data are merged to generate a third dataset. Specifically, the timestamp alignment algorithm uses a sliding window matching method with a window width of ±500ms. A successful match is determined when the time difference between the two datasets is within the window. For example, the time difference between the two data points mentioned above is only 120ms, so the merged data generates the following third dataset entries: visibility 320 meters, friction coefficient 0.62, vehicle speed 72 km / h, and acceleration -0.3 m / s².

[0022] Based on the visibility and friction values ​​in the third dataset, the road surface condition level is determined and labeled as the fourth dataset; A random forest model is used to analyze the fourth dataset and output traffic state prediction results. Specifically, the road surface condition level classification is based on the industry standard JT / T715-2008. When visibility is 200-500 meters and the friction coefficient is 0.4-0.6, it is judged as a level 2 warning state. Taking the aforementioned merged data as an example, the system automatically marks this item as a level 2 warning, forming the fourth dataset. The random forest model is pre-trained with 100 decision trees. The input features include visibility level, friction coefficient level, vehicle speed level, and acceleration direction. The output is a traffic state prediction of level 0-3. For example, when a level 2 warning state is input and continuous negative acceleration is detected, the model may predict that level 3 congestion will occur in the next 5 minutes.

[0023] In this embodiment, the method for determining the boundary of a high-risk area based on the visibility drop signal and the distribution density of vehicle emergency braking behavior in the spatiotemporally synchronized traffic state dataset, and triggering the dynamic speed limit calculation module if the proportion of emergency braking vehicles in the area exceeds a preset threshold, includes: The system acquires real-time visibility values, timestamps of vehicle emergency braking signals, and vehicle location data. Specifically, real-time visibility data is collected by roadside lidar at a sampling frequency of 10Hz, while emergency braking signals are obtained from ABS trigger events on the vehicle terminal, with timestamp accuracy at the millisecond level.

[0024] A spatial clustering algorithm was used to divide high-density areas of sudden braking and generate a distribution map. Specifically, the DBSCAN spatial clustering algorithm aggregated sudden braking events according to latitude and longitude coordinates, setting a neighborhood radius of 50 meters and a minimum sample size of 3, thereby identifying high-density areas of sudden braking. For example, if 15 sudden braking events were detected on a certain road segment within 10 minutes, three core regions were generated after clustering, with region A having a sudden braking density of 8 times per 100 square meters.

[0025] If the spatiotemporal overlap between sudden descent points and emergency braking events within a region exceeds a threshold, it is marked as a region boundary. Specifically, a sudden descent point is defined as a location where visibility drops by more than 100 meters within 5 seconds. If more than 70% of the sudden descent points and emergency braking events within a certain region overlap within a time window of ±2 seconds, it is determined to be a high-risk region boundary. For example, if 10 sudden descent points are detected in region B, and 8 of them match emergency braking timestamps, the overlap is 80%, exceeding the preset threshold of 75%, so it is marked as a speed limit adjustment zone.

[0026] Based on the mapping relationship between the percentage value and the preset speed limit value, the dynamic speed limit calculation module is triggered to output the dynamic speed limit value. Specifically, the dynamic speed limit calculation model adopts a piecewise linear mapping to associate visibility with the proportion of emergency braking. For example, when the visibility is less than 200 meters and the proportion of emergency braking exceeds 60%, the speed limit value is reduced from 80km / h to 50km / h.

[0027] The system uses Kalman filtering to fuse visibility trends and vehicle displacement data from the signal source, dynamically updates speed limits at the edges of the affected area, and then sends these updates to the vehicle-mounted terminals. Specifically, Kalman filtering is used to fuse multi-vehicle data. For example, if visibility continues to decrease and the average vehicle speed fluctuates by more than 15% during a certain period, the system will further reduce the speed limit by 10 km / h and broadcast it to the vehicle-mounted terminals in that area via V2X.

[0028] In this embodiment, a hierarchical decision-making model is adopted. The first layer calculates the basic safe vehicle speed based on the road surface friction coefficient and visibility. The second layer superimposes the traffic flow density correction coefficient. If the proportion of heavy trucks exceeds 30%, an additional braking distance compensation value is added. The system acquires current road surface friction coefficient and visibility data, and uses a linear regression model to calculate a baseline safe speed. Specifically, the road surface friction coefficient can be collected in real time by onboard sensors or road monitoring equipment; for example, the friction coefficient is measured to be 0.3 on a wet road surface and 0.7 on a dry road surface. Visibility data relies on weather sensors or camera detection, such as visibility dropping to 100 meters in foggy weather. The linear regression model is trained using historical accident data to establish the correlation between friction coefficient, visibility, and safe speed. If a friction coefficient of 0.4 and visibility of 200 meters are input, the model may output a baseline safe speed of 60 km / h.

[0029] Based on real-time traffic density data, a correction coefficient is determined using a preset density-speed mapping table; the correction coefficient is then superimposed onto the base safe speed to obtain the adjusted safe speed. Obtain the percentage of heavy trucks on the current road section. If the percentage exceeds 30%, calculate the braking distance compensation value. The braking distance compensation value is converted into a vehicle speed adjustment amount and added to the adjusted safe speed. Specifically, traffic density is detected by cameras or radar, and a preset density-speed mapping table defines correction coefficients for different densities. For example, the correction coefficient is 1.0 for low density and drops to 0.8 for high density. Assuming the current density corresponds to a correction coefficient of 0.9, after adding it to a base speed of 60 km / h, the adjusted speed is 54 km / h.

[0030] The system outputs the final recommended safe speed. Specifically, the proportion of heavy trucks is determined using vehicle type recognition technology. If trucks account for 35% of a road segment, a braking distance compensation value needs to be calculated. Trucks typically have a 20% longer braking distance than passenger cars, and this compensation value can be converted into a speed adjustment. For example, if the original speed is 54 km / h, it needs to be reduced by 10%, resulting in a final recommended safe speed of 49 km / h.

[0031] In this embodiment, the method of issuing differentiated speed limit commands through the vehicle-road cooperative communication link, using a basic safe speed for passenger cars and a compensated speed for heavy trucks, and displaying the results synchronously on electronic road signs and vehicle terminals includes: Obtain real-time vehicle type data based on the communication link; If the vehicle type is a passenger car, the preset safety value is used as the speed limit; If the vehicle type is a heavy vehicle, the speed limit value is obtained by adjusting the safety value based on the compensation value; Speed ​​limits are transmitted to electronic road signs and vehicle terminals via road cooperation; A time synchronization protocol is used to keep the electronic road signs and the vehicle terminal synchronized in terms of vehicle speed; Monitor the status of the communication link, and if it is interrupted, switch to the redundant link maintenance command.

[0032] Specifically, real-time vehicle type data obtained through the communication link can be collected based on onboard OBUs or roadside sensing devices.

[0033] When the roadside millimeter-wave radar detects a vehicle with a wheelbase exceeding 12 meters, it is identified as a heavy vehicle, triggering the compensation value adjustment logic.

[0034] The braking distance compensation value for heavy vehicles is usually set to 1.5 times that of passenger cars. For example, if the preset safety value for passenger cars is 80 km / h, the speed limit for heavy vehicles needs to be reduced to 60 km / h.

[0035] When issuing speed limits using vehicle-road cooperative systems, it is necessary to ensure synchronization between the electronic license plate and the vehicle-mounted terminal.

[0036] Employing the IEEE 1588 time synchronization protocol, the electronic sign display delay is controlled to within 50 milliseconds by broadcasting timestamps through the master clock node. If the communication link is interrupted, the redundant link can switch to 4G / 5G dual-mode backup to ensure that the command issuance success rate is not less than 99.9%.

[0037] It should be noted that the impact of the proportion of heavy vehicles on safe speed is related to the hierarchical decision-making model in historical dialogues. When the proportion of heavy vehicles on a certain road segment reaches 40%, in addition to the speed limit based on vehicle type alone, the braking distance compensation value in the historical rules also needs to be superimposed to form a double guarantee.

[0038] While the electronic speed limit is dynamically displayed, the vehicle terminal will also provide voice prompts to the driver, enhancing the safety warning effect.

[0039] The roadside unit sends speed limit commands to vehicles via DSRC short-range communication. For example, when the electronic sign displays a speed limit of 70 km / h, the onboard HUD simultaneously projects a red warning box. If the vehicle's actual speed exceeds the speed limit by 10% for 5 seconds, active braking intervention is triggered. This design can effectively reduce the risk of rear-end collisions caused by speed differences in mixed-traffic areas.

[0040] In this embodiment, the method of real-time monitoring of the average speed change trend of traffic flow after the speed limit command is executed, and triggering iterative optimization of the speed limit value by re-acquiring vehicle acceleration data for that road segment if the speed variance of a local road segment continues to increase, includes: Obtain average traffic speed data for road segments after the speed limit order is executed; Calculate the speed variance within a continuous time window. If the variance exceeds a preset threshold and continues to rise, mark the road segment area. Acceleration data is collected from vehicle terminals in the marked road section area, and the average value is taken after filtering out outliers. A linear regression model is used to establish the mapping relationship between acceleration and the current speed limit value, and the speed limit adjustment coefficient is output. If the absolute value of the adjustment coefficient is greater than the set tolerance, a new speed limit value is generated and sent to the target road segment area; Update the "Historical Rate Limit Iteration Count" field, overwriting the original record after rate limit execution.

[0041] Specifically, when acquiring average traffic speed data for a road segment, vehicle speed information can be collected in real time using roadside sensing devices (such as millimeter-wave radar) and combined with positioning data reported by vehicle terminals to eliminate outliers caused by GPS drift. For example, if speed samples of 100 vehicles are collected on a certain road segment within 10 minutes, after removing data that deviate from the mean by ±3σ, the average speed of the remaining 95 vehicles is 80 km / h, which is used as the current traffic flow speed benchmark.

[0042] Specifically, when calculating the speed variance, a 5-minute time window can be set, and speed fluctuations within three consecutive windows can be statistically analyzed. If the variance values ​​of a certain road segment in the three windows are 12, 18, and 25 (unit: km² / h²), respectively, and the preset threshold is 20, then the road segment is marked due to the continuous upward trend. At this time, the system automatically filters the acceleration data of 50 heavy trucks in the area. After removing outliers such as sudden braking or acceleration, the average acceleration of the remaining 45 vehicles is -0.3 m / s², indicating that the vehicles are generally in a deceleration state.

[0043] For example, the linear regression model can be trained based on historical data, with the current speed limit (e.g., 70 km / h) and the mean acceleration (-0.3 m / s²) as inputs, and an output adjustment coefficient of -0.15. If the tolerance threshold is 0.1, a new speed limit of 70 × (1 - 0.15) = 59.5 km / h is generated and prioritized for distribution to heavy truck terminals on that road segment. Simultaneously, the system updates the iteration count in the original speed limit record from 3 to 4, overwriting the results of the previous execution round.

[0044] It should be noted that the redundant communication link continuously monitors the status of the main link during this process. For example, when the main link packet loss rate exceeds 5%, it automatically switches to the 5G backup link to ensure that the speed limit command is sent to the electronic road sign with a delay of less than 200ms. The vehicle terminal calibrates the displayed content through a time synchronization protocol to avoid asynchrony between the speed limit information of passenger cars and trucks due to transmission delays.

[0045] This system uses dynamic feedback to adjust the speed of heavy-duty trucks in advance on long downhill slopes, reducing the risk of brake overheating caused by frequent braking. Meanwhile, passenger cars continue to travel at a basic safe speed, balancing traffic efficiency with the safety of mixed-traffic lanes.

[0046] Example 2 like Figure 2 As shown, the present invention also provides a dynamic speed limit recommendation system for highways based on vehicle-road cooperative technology. The system is used to implement the aforementioned method. The system includes: a data acquisition and spatiotemporal matching module, a high-risk area identification module, a hierarchical speed limit decision module, an instruction issuance and display module, and an effect monitoring and iterative optimization module. The data acquisition and spatiotemporal matching module is used to acquire visibility and road surface friction coefficient data collected by roadside equipment, and synchronously receive real-time vehicle speed and acceleration information uploaded by vehicle terminals. The two types of data are spatiotemporally matched using a timestamp alignment algorithm to obtain a spatiotemporally synchronized traffic status dataset. The high-risk area identification module is used to determine the boundary of high-risk areas based on the visibility drop signal and the distribution density of vehicle emergency braking behavior in the synchronous dataset. If the proportion of emergency braking vehicles in the area exceeds a preset threshold, the dynamic speed limit calculation module is triggered. The layered speed limit decision module is used to adopt a layered decision model. The first layer calculates the basic safe speed based on the road friction coefficient and visibility. The second layer superimposes the traffic flow density correction coefficient. If the proportion of heavy trucks exceeds 30%, an additional braking distance compensation value is added. The instruction issuance and display module is used to issue differentiated speed limit instructions through the vehicle-road cooperative communication link. It adopts the basic safe speed for passenger cars and the compensated speed for heavy trucks, and displays them synchronously on the electronic road sign and the vehicle terminal. The effect monitoring and iterative optimization module is used to monitor the trend of average speed change of traffic flow after the speed limit command is executed in real time. If the speed variance of a local road segment continues to increase, the vehicle acceleration data of that road segment will be reacquired to trigger iterative optimization of the speed limit value.

[0047] The embodiments described above are merely preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Various modifications and improvements made to the technical solutions of the present invention by those skilled in the art without departing from the spirit of the present invention should fall within the protection scope defined by the claims of the present invention.

Claims

1. A method for recommending dynamic speed limits on highways based on vehicle-road cooperative technology, characterized in that, The method includes: The system acquires visibility and road surface friction coefficient data collected by roadside equipment, and simultaneously receives real-time vehicle speed and acceleration information uploaded by vehicle-mounted terminals. A timestamp alignment algorithm is used to perform spatiotemporal matching of the data collected by roadside equipment and the data uploaded by vehicle-mounted terminals to obtain a spatiotemporally synchronized traffic status dataset. Based on the visibility drop signal and vehicle emergency braking behavior distribution density in the spatiotemporally synchronized traffic status dataset, the boundary of high-risk areas is determined. If the proportion of emergency braking vehicles in the area exceeds a preset threshold, the dynamic speed limit calculation module is triggered. A hierarchical decision-making model is adopted. The first layer calculates the basic safe speed based on the road friction coefficient and visibility. The second layer superimposes the traffic flow density correction coefficient. If the proportion of heavy trucks exceeds 30%, an additional braking distance compensation value is added. Differentiated speed limit commands are issued through the vehicle-road cooperative communication link, using a basic safe speed for passenger cars and a compensated speed for heavy trucks, and are displayed synchronously on electronic road signs and vehicle terminals; The system monitors the average speed change trend of traffic flow after the speed limit command is executed in real time. If the speed variance of a local road segment continues to increase, the system reacquires the vehicle acceleration data for that road segment and triggers iterative optimization of the speed limit value.

2. The method according to claim 1, characterized in that, The method for acquiring visibility and road surface friction coefficient data collected by roadside equipment, synchronously receiving real-time vehicle speed and acceleration information uploaded by vehicle-mounted terminals, and using a timestamp alignment algorithm to perform spatiotemporal matching of the data collected by roadside equipment and the data uploaded by vehicle-mounted terminals to obtain a spatiotemporally synchronized traffic state dataset includes: Acquire visibility and friction coefficient data collected by roadside equipment and label them as the first dataset; Receive real-time vehicle speed and acceleration information uploaded by the vehicle terminal and mark it as the second dataset; A timestamp alignment algorithm is used to process the first and second datasets, matching data points with the same timestamp. If the timestamps match successfully, the visibility, friction, vehicle speed, and acceleration data are merged to generate a third dataset. Based on the visibility and friction values ​​in the third dataset, the road surface condition level is determined and labeled as the fourth dataset; The fourth dataset is analyzed using a random forest model, and traffic state prediction results are output.

3. The method according to claim 1, characterized in that, Based on the visibility drop signal and vehicle emergency braking behavior distribution density in the spatiotemporally synchronized traffic state dataset, the boundary of high-risk areas is determined. If the proportion of emergency braking vehicles in the area exceeds a preset threshold, the dynamic speed limit calculation module is triggered using the following methods: Acquire real-time visibility values, timestamps of vehicle emergency braking signals, and vehicle location data; Spatial clustering algorithm is used to divide the high-density areas of sudden braking and generate a distribution map; If the spatiotemporal overlap between a sudden drop point and a number of sudden braking events within a region exceeds a threshold, it is marked as a region edge. Based on the mapping relationship between the percentage value and the preset speed limit value, the dynamic speed limit calculation module is triggered to output the dynamic speed limit value; The speed limit value of the region edge is dynamically updated by fusing the visibility change trend in the signal source with the vehicle displacement using Kalman filtering and then sent to the vehicle terminal.

4. The method according to claim 1, characterized in that, A hierarchical decision-making model is adopted. The first layer calculates the basic safe speed based on the road surface friction coefficient and visibility. The second layer superimposes a traffic density correction coefficient. If heavy trucks account for more than 30%, additional braking distance compensation values ​​are added. Obtain current road surface friction coefficient and visibility data, and use a linear regression model to calculate the basic safe vehicle speed; Based on real-time traffic density data, the correction coefficient is determined through a preset density-speed mapping table; The adjusted safe speed is obtained by adding the correction factor to the base safe speed. Obtain the percentage of heavy trucks on the current road section. If the percentage exceeds 30%, calculate the braking distance compensation value. The braking distance compensation value is converted into a vehicle speed adjustment amount and then added to the adjusted safe vehicle speed. Output the final recommended safe speed.

5. The method according to claim 1, characterized in that, The method of issuing differentiated speed limit commands through vehicle-road cooperative communication links, using a basic safe speed for passenger cars and a compensated speed for heavy trucks, and displaying them synchronously on electronic road signs and vehicle terminals includes: Obtain real-time vehicle type data based on the communication link; If the vehicle type is a passenger car, the preset safety value is used as the speed limit; If the vehicle type is a heavy vehicle, the speed limit value is obtained by adjusting the safety value based on the compensation value; Speed ​​limits are transmitted to electronic road signs and vehicle terminals via road cooperation; A time synchronization protocol is used to keep the electronic road signs and the vehicle terminal synchronized in terms of vehicle speed; Monitor the status of the communication link, and if it is interrupted, switch to the redundant link maintenance command.

6. The method according to claim 1, characterized in that, Real-time monitoring of the average speed change trend of traffic flow after the execution of speed limit commands; if the speed variance of a local road segment continues to increase, then re-acquiring vehicle acceleration data for that road segment; methods for triggering iterative optimization of speed limit values ​​include: Obtain average traffic speed data for road segments after the speed limit order is executed; Calculate the speed variance within a continuous time window. If the variance exceeds a preset threshold and continues to rise, mark the road segment area. Acceleration data is collected from vehicle terminals in the marked road section area, and the average value is taken after filtering out outliers. A linear regression model is used to establish the mapping relationship between acceleration and the current speed limit value, and the speed limit adjustment coefficient is output. If the absolute value of the adjustment coefficient is greater than the set tolerance, a new speed limit value is generated and sent to the target road segment area; Update the "Historical Rate Limit Iteration Count" field, overwriting the original record after rate limit execution.

7. A dynamic speed limit recommendation system for highways based on vehicle-road cooperative technology, the system being used to implement the method described in any one of claims 1-6, characterized in that, The system includes: a data acquisition and spatiotemporal matching module, a high-risk area identification module, a tiered speed limit decision module, an instruction issuance and display module, and an effect monitoring and iterative optimization module. The data acquisition and spatiotemporal matching module is used to acquire visibility and road surface friction coefficient data collected by roadside equipment, synchronously receive real-time vehicle speed and acceleration information uploaded by vehicle terminal, and use a timestamp alignment algorithm to perform spatiotemporal matching of the two types of data to obtain a spatiotemporally synchronized traffic status dataset. The high-risk area identification module is used to determine the boundary of the high-risk area based on the visibility drop signal and the distribution density of vehicle emergency braking behavior in the synchronous dataset. If the proportion of emergency braking vehicles in the area exceeds a preset threshold, the dynamic speed limit calculation module is triggered. The layered speed limit decision module is used to adopt a layered decision model. The first layer calculates the basic safe speed based on the road friction coefficient and visibility. The second layer superimposes the traffic flow density correction coefficient. If the proportion of heavy trucks exceeds 30%, an additional braking distance compensation value is added. The instruction issuance and display module is used to issue differentiated speed limit instructions through the vehicle-road cooperative communication link, using a basic safe speed for passenger cars and a compensated speed for heavy trucks, and displaying them synchronously on electronic road signs and vehicle terminals. The effect monitoring and iterative optimization module is used to monitor the average speed change trend of traffic flow after the speed limit command is executed in real time. If the speed variance of a local road segment continues to increase, the vehicle acceleration data of that road segment will be reacquired to trigger iterative optimization of the speed limit value.

Images