Road congestion condition prediction system and method based on intelligent traffic

Abstract

The invention discloses a road congestion condition prediction system and method based on intelligent traffic, relates to the technical field of road congestion prediction, and aims to solve the technical problems that most prediction models are insufficient in adaptability to dynamic environment factors such as weather and analysis dimensions are not dynamically adjusted in combination with real-time changes of weather in the prior art. According to the invention, through the architecture design of'platform overall planning-multi-unit cooperation-fine analysis-dynamic decision ', the whole-process optimization of road congestion prediction and control is realized, and through the deep fusion of lane-level fine data and real-time weather data, a dynamic sliding time window and sub-scene threshold judgment are combined, so that the road congestion prediction and control can be realized. Compared with the prior art, the method has the advantages that the prediction lead is obviously increased, the misjudgment rate is greatly reduced, reliable decision support is provided for traffic control, the pertinence and initiative of congestion control are realized, the dynamic configuration efficiency of lane resources is optimized, and a full-process collaborative closed-loop management system is constructed.

Description

Technical Field

[0001] This invention relates to the field of road congestion prediction technology, specifically to a road congestion prediction system and method based on intelligent transportation. Background Technology

[0002] With the acceleration of urbanization and the continuous increase in the number of motor vehicles, road congestion has become a core problem that restricts the efficiency of urban traffic operation, affects residents' travel experience, and increases environmental pollutant emissions. Intelligent Transportation Systems (ITS) are a key technology for solving traffic congestion. One of their core requirements is to achieve accurate prediction of road congestion conditions, so as to provide advance support for traffic control and diversion decisions.

[0003] Currently, most existing road congestion prediction technologies in the industry are based on historical traffic flow data, fixed-time period monitoring data, or single-dimensional real-time traffic parameters (such as average vehicle speed and traffic volume). These technologies mainly exhibit the following technical characteristics and limitations: From the perspective of data collection, it relies heavily on traditional equipment such as coil detectors and video surveillance. Some use microwave or laser detectors but fail to fully utilize lane-level fine data. The data granularity is relatively coarse, making it difficult to accurately capture subtle changes in the driving status of vehicles within a single lane. From the perspective of influencing factors, most prediction models are not adaptable enough to dynamic environmental factors such as weather. They use fixed thresholds or static correction coefficients to handle the impact of weather, without dynamically adjusting the analysis dimensions in combination with real-time changes in weather. From the perspective of tracing the source of congestion, the focus is mostly on identifying the state after congestion occurs, with a weak ability to identify the pre-congestion causes (i.e., the sources of slowdown), and an inability to distinguish between the regional and non-regional attributes of the sources of slowdown, resulting in a lack of targeted diversion measures. From the perspective of lane resource optimization, the application of switchable lanes such as tidal lanes and reversible lanes is already quite common. However, the switching time is mostly set in a fixed mode, without adaptive adjustment based on dynamic parameters such as real-time traffic flow changes and lane change demands, resulting in insufficient adaptability and efficiency of lane switching.

[0004] To address the aforementioned technical shortcomings, we propose an architecture design of "platform coordination - multi-unit collaboration - refined analysis - dynamic decision-making" to construct a road congestion prediction system and method that integrates congestion prediction, source identification, and lane optimization, thereby achieving targeted and proactive congestion management. Summary of the Invention

[0005] The purpose of this invention is to solve the problems mentioned above by proposing a road congestion prediction system and method based on intelligent transportation.

[0006] The objective of this invention can be achieved through the following technical solution: a road congestion prediction system based on intelligent transportation, including a road prediction platform, which is communicatively connected to a driving condition analysis unit, a speed reduction source identification unit, and a lane switching decision setting unit. The driving condition analysis unit analyzes the driving conditions of the road monitoring area and predicts road congestion based on the driving condition analysis and the current driving environment. The speed reduction source identification unit identifies the sources of speed reduction in the road monitoring area to facilitate congestion prediction and targeted traffic management. After receiving the lane switching decision setting signal, the lane switching decision setting unit sets the lane switching decision for the road monitoring area.

[0007] Furthermore, the process of the driving condition analysis unit is as follows: It acquires real-time, continuous lane-level data from microwave or laser detectors on expressways, including vehicle speed, headway, and vehicle length; it also accesses real-time, refined weather data from the meteorological bureau, including visibility, precipitation, and road surface temperature. Set a sliding time window, which is shortened in real time according to the continuous changes in weather data; calculate the average headway and speed standard deviation of each road segment in the road monitoring area within the sliding time window; compare the weather data with the baseline data under historical normal weather to obtain the weather influence coefficient of following distance; set safe following distance thresholds and speed variance risk thresholds under different visibility and road conditions.

[0008] Furthermore, during the period when the weather impact coefficient continues to increase, if the average headway value continues to approach the safe following distance threshold and the standard deviation of vehicle speed continues to increase, it is inferred that the increase in the weather impact coefficient corresponds to a continuous increase in road traffic impact, generating a weather interference aggravation signal and sending it to the road prediction platform; after receiving the weather interference aggravation signal, the road prediction platform determines that road congestion has aggravated and diverts traffic and limits speed in the road monitoring area, and clears the road when reducing the traffic flow in the current section; If the average headway does not consistently approach the safe following distance threshold, or the standard deviation of vehicle speed does not continuously increase, it is inferred that the increase in the weather impact coefficient does not continuously increase the impact on road traffic. A low-impact weather interference signal is generated and sent to the road prediction platform. After receiving the low-impact weather interference signal, the road prediction platform determines that the road is about to become congested, issues a safety warning to the road monitoring area, and sets countermeasures according to the type of weather interference.

[0009] Furthermore, during the non-continuous increase phase of the weather impact coefficient, if the average headway value continues to approach the safe following distance threshold and the standard deviation of vehicle speed continues to increase, it is inferred that the increase in the weather impact coefficient corresponds to an unavoidable impact on road traffic, generating a weather interference signal and sending it to the road prediction platform; after receiving the weather interference signal, the road prediction platform determines that the road is congested and driving is difficult, stops the increase in traffic flow, and conducts voice broadcasts for each road segment to reduce the following distance; If the average headway does not consistently approach the safe following distance threshold, or the speed standard deviation does not continuously increase, it is inferred that the increase in the weather impact coefficient has a gradually decreasing impact on road traffic, generating a weather interference dissipation signal and sending it to the road prediction platform. After receiving the weather interference dissipation signal, the road prediction platform will conduct traffic management in the road monitoring area and gradually open the road for vehicles to merge into intersections.

[0010] Furthermore, the process of the deceleration source identification unit is as follows: The system identifies road sections within the road monitoring area where vehicle speeds decrease and determines the source of the speed reduction based on monitoring of vehicles returning to zero. Based on traffic flow monitoring within the road monitoring area, it obtains the percentage of vehicles pulling over when the traffic flow reaches the source of the speed reduction, and the ratio of the frequency of adjacent vehicles waiting to stop at zero speed when there are vehicles at the source of the speed reduction to the frequency of vehicles overtaking by using other lanes. The system then compares the percentage of vehicles pulling over when the traffic flow reaches the source of the speed reduction, and the ratio of the frequency of adjacent vehicles waiting to stop at zero speed when there are vehicles at the source of the speed reduction to the frequency of vehicles overtaking by using other lanes, with the vehicle percentage threshold and the frequency ratio threshold, respectively.

[0011] Furthermore, if the proportion of vehicles that pull over when the traffic flow reaches the source of the deceleration exceeds the vehicle proportion threshold, or if the ratio of the frequency of adjacent vehicles waiting to stop at zero speed to the frequency of vehicles overtaking when there are vehicles at the source of the deceleration exceeds the frequency ratio threshold, it is inferred that the source of the deceleration in the current road monitoring area is affected by regional factors, and a regional impact signal is generated and sent to the road prediction platform. If the proportion of vehicles pulling over when traffic reaches the source of the deceleration does not exceed the vehicle proportion threshold, and the ratio of the frequency of adjacent vehicles waiting to stop at zero speed to the frequency of vehicles overtaking when there are vehicles at the source of the deceleration does not exceed the frequency ratio threshold, then it is inferred that the source of the deceleration in the current road monitoring area is not affected by regional factors, and a non-regional influence signal is generated and sent to the road prediction platform.

[0012] Furthermore, the process of the lane switching decision setting unit is as follows: The switching time period of the switchable lanes is obtained, based on the completed switching traffic flow of the switchable lanes within the road monitoring area during the switching time period and the lane-changing traffic flow of the switchable lanes before switching in the non-switchable lanes; If the completed switching traffic flow of a switchable lane exceeds the switching traffic flow threshold, and the lane-changing traffic flow before the switchable lane in a non-switchable lane exceeds the set lane-changing traffic flow threshold, it is inferred that the current switching period is set to be adapted, but adaptive settings are required. An adaptive setting signal is generated and sent to the road prediction platform. After receiving the adaptive setting signal, the road prediction platform, based on the current road traffic demand, i.e., the lane-changing demand or traffic flow increase span demand of the set road type, uses the lane-changing frequency or traffic flow increase span setting as the trigger condition, and sets the switchable lane to a switchable state after the trigger condition is generated. If the traffic flow of vehicles completing lane switching exceeds the lane switching traffic flow threshold, and the traffic flow of vehicles changing lanes before switching lanes in non-lane switching lanes does not exceed the set lane changing traffic flow threshold, it is inferred that the current switching time period setting is adapted and there is no need for adaptive setting. A single time period setting signal is generated and sent to the road prediction platform. After receiving the signal, the road prediction platform sets the switching time period as a single trigger condition.

[0013] Furthermore, if the completed switching traffic flow of the switchable lane does not exceed the switching traffic flow threshold, and the lane-changing traffic flow before the switchable lane in the non-switchable lane exceeds the set lane-changing traffic flow threshold, it is inferred that the current switching time period setting is incompatible. A time period adjustment setting signal is generated and sent to the road prediction platform. After receiving the signal, the road prediction platform resets the switching time period. If the completed switching traffic flow of the switchable lane does not exceed the switching traffic flow threshold, and the lane-changing traffic flow before the switchable lane in the non-switchable lane does not exceed the set lane-changing traffic flow threshold, it is inferred that the current switching time period setting is incompatible. A lane-switching suspension setting signal is generated and sent to the road prediction platform. After receiving the signal, the road prediction platform suspends the switching of the lane.

[0014] This invention also proposes a method for predicting road congestion based on intelligent transportation, with the following specific steps: Step 1: Traffic Condition Analysis. Analyze the traffic conditions in the road monitoring area and predict road congestion based on the analysis and the current driving environment. Step 2: Identify the source of speed reduction. Identify the source of speed reduction in the road monitoring area to facilitate congestion prediction and targeted traffic management. Step 3: Lane switching decision setting. Lane switching decisions are set for the road monitoring area.

[0015] Compared with the prior art, the beneficial effects of the present invention are: 1. Lane-level data can accurately capture the micro-state of vehicle movement within a single lane, laying a data foundation for subsequent precise analysis. The introduction of refined weather data compensates for the shortcomings of existing technologies in considering dynamic environmental factors, enabling congestion prediction to fully adapt to the impact of weather changes. The dynamic window design enables real-time adaptation of analysis dimensions to weather changes, avoiding the problem of analysis lag in scenarios of sudden weather changes when using a fixed window. Furthermore, through the collaborative calculation of average headway, vehicle speed standard deviation, and weather influence coefficient, weather factors are quantitatively correlated with vehicle driving status, providing accurate quantitative basis for subsequent congestion trend judgment.

[0016] Achieving refined and differentiated congestion prediction can accurately distinguish between four scenarios: aggravated by weather interference, low impact, obstruction, and dissipation. This avoids the "one-size-fits-all" prediction model of existing technologies, and allows for proactive and targeted control measures. For example, it can divert traffic and limit speeds in advance before weather interference intensifies, and issue warnings and prepare for salting / drainage when weather interference has a low impact. This effectively reduces the probability of congestion or alleviates its severity. Furthermore, by linking "prediction and control" in advance, it enhances the initiative of traffic management and breaks away from the passive "post-congestion response" model.

[0017] 2. By identifying road sections where vehicles slow down, determining the source of the slowdown, and collecting parameters such as the percentage of vehicles parked on the side of the road and the ratio of waiting and overtaking frequencies of adjacent vehicles, the system can accurately trace the precursors of congestion. Specifically, it shifts the focus of analysis from "congestion status" to "the source of slowdown," thereby solving the problem of not being able to predict the process of congestion formation and providing core evidence for avoiding congestion in advance. The process of differentiating the sources of speed reduction and implementing targeted traffic management involves comparing the collected quantitative parameters with corresponding thresholds to distinguish between the regional and non-regional attributes of the sources of speed reduction and implementing differentiated control measures. This achieves a "precise profile" of traffic management measures, avoiding the problem of generalization of existing technical traffic management measures. For example, to address congestion around schools, additional traffic management personnel can be deployed in advance during school hours, and to address unreasonable intersection settings, traffic signs can be optimized to improve traffic management efficiency and effectiveness.

[0018] 3. By acquiring real-time parameters such as the completed switching traffic flow of switchable lanes and the lane-changing traffic flow in non-switchable lanes, dynamic quantitative basis is provided for the evaluation of switching decisions. This breaks the static mode of existing technologies that rely on fixed time periods, enabling lane switching decisions to conform to real-time traffic operation status and improve the resource utilization rate of switchable lanes. For example, adaptive settings can enable lane switching to respond to dynamic demands such as the frequency of lane changes or the increase in traffic flow, avoiding the interference of unreasonable lane switching on traffic operation. For example, by pausing settings or adjusting time periods, local congestion caused by improper switching timing can be avoided, realizing a dynamic closed loop of lane resource optimization and ensuring that lane switching is always adapted to traffic operation needs. Attached Figure Description

[0019] To facilitate understanding by those skilled in the art, the present invention will be further described below with reference to the accompanying drawings.

[0020] Figure 1 This is a system principle block diagram of the present invention; Figure 2 This is a flowchart of the method of the present invention. Detailed Implementation

[0021] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. 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.

[0022] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0023] Please see Figure 1 As shown, the road congestion prediction system based on intelligent transportation includes a road prediction platform, which is connected to a driving condition analysis unit, a speed reduction source identification unit, and a lane switching decision setting unit. The road prediction platform generates driving condition analysis signals and sends them to the driving condition analysis unit. After receiving the driving condition analysis signal, the driving condition analysis unit performs driving condition analysis on the road monitoring area and predicts road congestion based on the driving condition analysis and the current driving environment. It acquires real-time, continuous lane-level data from microwave or laser detectors on expressways, including vehicle speed, headway, and vehicle length, while also accessing real-time, refined weather data from the meteorological bureau, including visibility, precipitation, and road surface temperature. A sliding time window is set, which is shortened in real time according to the continuous changes in weather data. The average headway and speed standard deviation of each road segment in the road monitoring area are calculated within the sliding time window. The weather data is compared with the baseline data under historical normal weather conditions to obtain the weather influence coefficient of following distance. Set safe following distance thresholds and speed variance risk thresholds under different visibility and road conditions; During a period of continuous increase in the weather impact coefficient, if the average headway value continues to approach the safe following distance threshold and the standard deviation of vehicle speed continues to increase, it is inferred that the increase in the weather impact coefficient has a continuous increase in its impact on road traffic, generating a weather interference aggravation signal and sending it to the road prediction platform, such as during a period of continuous snowfall; after receiving the weather interference aggravation signal, the road prediction platform determines that road congestion has intensified and implements traffic diversion and speed limits in the road monitoring area, and clears the road when reducing the traffic flow on the current road section; If the average headway does not consistently approach the safe following distance threshold, or the standard deviation of vehicle speed does not continuously increase, it is inferred that the increase in the weather impact coefficient has not continuously increased its impact on road traffic. A low-impact weather interference signal is generated and sent to the road prediction platform, such as at the beginning of snowfall. After receiving the low-impact weather interference signal, the road prediction platform determines that the road will soon be congested, issues a safety warning to the road monitoring area, and sets countermeasures according to the type of weather interference, such as spreading salt in snowy weather and draining water in rainy weather. During the non-continuous increase phase of the weather impact coefficient, if the average headway value continues to approach the safe following distance threshold and the standard deviation of vehicle speed continues to increase, it is inferred that the increase in the weather impact coefficient has an unavoidable impact on road traffic, generating a weather interference signal and sending it to the road prediction platform; after receiving the weather interference signal, the road prediction platform determines that the road is congested and driving is difficult, stops the increase in traffic flow, and conducts voice broadcasts for each road segment to reduce the following distance; If the average headway does not consistently approach the safe following distance threshold, or the speed standard deviation does not continuously increase, it is inferred that the increase in the weather impact coefficient has a gradually decreasing impact on road traffic, generating a weather interference dissipation signal and sending it to the road prediction platform; after receiving the weather interference dissipation signal, the road prediction platform will conduct traffic management in the road monitoring area and gradually open the road for vehicles to merge into intersections; The road prediction platform generates a speed reduction source identification signal and sends it to the speed reduction source identification unit; After receiving the speed reduction source identification signal, the speed reduction source identification unit identifies the speed reduction source in the road monitoring area to facilitate congestion prediction and targeted traffic management. The system identifies road sections where vehicles slow down within the road monitoring area and determines the source of the slowdown based on monitoring of vehicles returning to zero speed. Based on traffic flow monitoring within the road monitoring area, it obtains the percentage of vehicles that pull over when the traffic flow reaches the source of the slowdown. It also obtains the ratio of the frequency of adjacent vehicles waiting to return to zero speed to the frequency of vehicles overtaking when there are vehicles returning to zero speed at the source of the slowdown. The percentage of vehicles pulling over when traffic reaches the source of the deceleration, and the ratio of the frequency of adjacent vehicles waiting to stop when there are vehicles at zero speed at the source of the deceleration to the frequency of vehicles overtaking by using the lane, are compared with the vehicle number percentage threshold and the frequency ratio threshold, respectively. If the proportion of vehicles pulling over when traffic reaches the source of the slowdown exceeds a certain threshold, or if the ratio of the frequency of adjacent vehicles waiting to stop at zero speed to the frequency of vehicles overtaking when there are vehicles at the source of the slowdown exceeds a certain frequency ratio threshold, it is inferred that the source of the slowdown in the current road monitoring area is affected by regional factors. This generates a regional impact signal and sends it to the road prediction platform. After receiving the regional impact signal, the road prediction platform determines the type of buildings around the current source of the slowdown. If there are buildings such as shopping malls or schools, it records the current time period and determines the congestion period through multiple monitoring sessions, making congestion predictions in advance. At the same time, it sets traffic diversion measures based on the building type to reduce the persistence of the source of the slowdown. If the proportion of vehicles pulling over when traffic reaches the source of the speed reduction does not exceed the vehicle proportion threshold, and the ratio of the frequency of adjacent vehicles waiting to stop at zero speed to the frequency of vehicles overtaking when there are vehicles at the source of the speed reduction does not exceed the frequency ratio threshold, then it is inferred that the source of the speed reduction in the current road monitoring area is not affected by geographical location. A non-geographical influence signal is generated and sent to the road prediction platform. After receiving the non-geographical influence signal, the road prediction platform conducts a road setting assessment of the source of the speed reduction and infers the cause of the speed reduction based on multiple monitoring data, such as intersection settings or traffic flow adaptability. Targeted road traffic control is then implemented based on the cause of the speed reduction. The road prediction platform generates lane switching decision setting signals and sends them to the lane switching decision setting unit; After receiving the lane switching decision setting signal, the lane switching decision setting unit sets the lane switching decision for the road monitoring area so that switchable lanes can make advantageous and targeted switching decisions, such as tidal lanes and reversible lanes. The switching time period of the switchable lanes is obtained, based on the completed switching traffic flow of the switchable lanes within the road monitoring area during the switching time period and the lane-changing traffic flow of the switchable lanes before switching in the non-switchable lanes; If the completed switching traffic flow of a switchable lane exceeds the switching traffic flow threshold, and the lane-changing traffic flow before the switchable lane in a non-switchable lane exceeds the set lane-changing traffic flow threshold, it is inferred that the current switching time period is set to be compatible, but adaptive settings are required. An adaptive setting signal is generated and sent to the road prediction platform. After receiving the adaptive setting signal, the road prediction platform, based on the current road traffic demand, i.e., the lane-changing demand or traffic flow increase span demand of the set road type, uses the lane-changing frequency or traffic flow increase span setting as the trigger condition, and sets the switchable lane to a switchable state after the trigger condition is generated. If the traffic flow of vehicles completing the lane switching exceeds the lane switching traffic flow threshold, and the traffic flow of vehicles changing lanes before the lane switching in the non-lane switching lane does not exceed the set lane changing traffic flow threshold, it is inferred that the current lane switching time period is set to be compatible and there is no need to perform adaptive setting. A single time period setting signal is generated and sent to the road prediction platform. After receiving the signal, the road prediction platform sets the lane switching time period as a single trigger condition. If the traffic flow of the completed lane switching does not exceed the switching traffic flow threshold, and the traffic flow of the lane changing before the switchable lane in the non-switching lane exceeds the set lane changing traffic flow threshold, it is inferred that the current switching time period setting is not suitable, a time period adjustment setting signal is generated and sent to the road prediction platform, and the road prediction platform resets the switching time period after receiving it. If the traffic flow of the completed lane switching does not exceed the lane switching traffic flow threshold, and the traffic flow of the lane changing before the lane switching in the non-lane switching lane does not exceed the set lane changing traffic flow threshold, it is inferred that the current lane switching time setting is not suitable, and a lane switching pause setting signal is generated and sent to the road prediction platform. After receiving the signal, the road prediction platform pauses the lane switching.

[0024] Please see Figure 2 As shown, this invention also proposes a road congestion prediction method based on intelligent transportation, the specific steps of which are as follows: Step 1: Traffic Condition Analysis. Analyze the traffic conditions in the road monitoring area and predict road congestion based on the analysis and the current driving environment. Step 2: Identify the source of speed reduction. Identify the source of speed reduction in the road monitoring area to facilitate congestion prediction and targeted traffic management. Step 3: Lane switching decision setting. Lane switching decisions are set for the road monitoring area.

[0025] In summary, this invention, through the coordinated efforts of a road prediction platform, a driving condition analysis unit, a speed reduction source identification unit, and a lane switching decision setting unit, integrates lane-level refined data and real-time weather data. It employs dynamic sliding time windows and scenario-specific threshold judgments to accurately predict congestion trends, distinguish between the regional and non-regional attributes of speed reduction sources, and achieve adaptive decision-making for lane switching. This addresses the shortcomings of existing technologies, such as insufficient adaptation to dynamic environmental factors, weak congestion source tracing, and poor lane switching adaptability. It significantly improves prediction lead time, reduces misjudgment rates, and provides reliable decision support for traffic management.

[0026] Thresholds, preset values, or preset ranges are set for result comparison and analysis to determine whether they are good or bad. The value of these thresholds is determined by a combination of large-scale model analysis of sample data and human experience. They can also be adjusted appropriately based on seasonal or rational factors.

[0027] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to any specific implementation. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A road congestion prediction system based on intelligent transportation, comprising a road prediction platform, characterized in that, The road prediction platform has communication connections to a driving condition analysis unit, a speed reduction source identification unit, and a lane switching decision setting unit. The driving condition analysis unit analyzes the driving conditions of the road monitoring area and predicts road congestion based on the driving condition analysis and the current driving environment. The speed reduction source identification unit identifies the sources of speed reduction in the road monitoring area to facilitate congestion prediction and targeted traffic management. The lane switching decision setting unit sets lane switching decisions for the road monitoring area.

2. The road congestion prediction system based on intelligent transportation according to claim 1, characterized in that, The process of the driving condition analysis unit is as follows: It acquires real-time, continuous lane-level data from microwave or laser detectors on expressways, including vehicle speed, headway, and vehicle length; it also accesses real-time, refined weather data from the meteorological bureau, including visibility, precipitation, and road surface temperature. Set a sliding time window, which is shortened in real time according to the continuous changes in weather data; calculate the average headway and speed standard deviation of each road segment in the road monitoring area within the sliding time window; compare the weather data with the baseline data under historical normal weather to obtain the weather influence coefficient of following distance; set safe following distance thresholds and speed variance risk thresholds under different visibility and road conditions.

3. The road congestion prediction system based on intelligent transportation according to claim 2, characterized in that, During the period when the weather impact coefficient continues to increase, if the average headway value continues to approach the safe following distance threshold and the standard deviation of vehicle speed continues to increase, it is inferred that the increase in the weather impact coefficient corresponds to a continuous increase in road traffic impact, generating a weather interference aggravation signal and sending it to the road prediction platform; after receiving the weather interference aggravation signal, the road prediction platform determines that road congestion has aggravated and diverts traffic and limits speed in the road monitoring area, and clears the road when reducing the traffic flow in the current section; If the average headway does not consistently approach the safe following distance threshold, or the standard deviation of vehicle speed does not continuously increase, it is inferred that the increase in the weather impact coefficient does not continuously increase the impact on road traffic. A low-impact weather interference signal is generated and sent to the road prediction platform. After receiving the low-impact weather interference signal, the road prediction platform determines that the road is about to become congested, issues a safety warning to the road monitoring area, and sets countermeasures according to the type of weather interference.

4. The road congestion prediction system based on intelligent transportation according to claim 3, characterized in that, During the non-continuous increase phase of the weather impact coefficient, if the average headway value continues to approach the safe following distance threshold and the standard deviation of vehicle speed continues to increase, it is inferred that the increase in the weather impact coefficient corresponds to an unavoidable impact on road traffic, generating a weather interference signal and sending it to the road prediction platform; after receiving the weather interference signal, the road prediction platform determines that the road is congested and driving is difficult, stops the increase in traffic flow, and conducts voice broadcasts for each road segment to reduce the following distance; If the average headway does not consistently approach the safe following distance threshold, or the speed standard deviation does not continuously increase, it is inferred that the increase in the weather impact coefficient has a gradually decreasing impact on road traffic, generating a weather interference dissipation signal and sending it to the road prediction platform. After receiving the weather interference dissipation signal, the road prediction platform will conduct traffic management in the road monitoring area and gradually open the road for vehicles to merge into intersections.

5. The road congestion prediction system based on intelligent transportation according to claim 1, characterized in that, The process of the deceleration source identification unit is as follows: The system identifies road sections within the road monitoring area where vehicle speeds decrease and determines the source of the speed reduction based on monitoring of vehicles returning to zero. Based on traffic flow monitoring within the road monitoring area, it obtains the percentage of vehicles pulling over when the traffic flow reaches the source of the speed reduction, and the ratio of the frequency of adjacent vehicles waiting to stop at zero speed when there are vehicles at the source of the speed reduction to the frequency of vehicles overtaking by using other lanes. The system then compares the percentage of vehicles pulling over when the traffic flow reaches the source of the speed reduction, and the ratio of the frequency of adjacent vehicles waiting to stop at zero speed when there are vehicles at the source of the speed reduction to the frequency of vehicles overtaking by using other lanes, with the vehicle percentage threshold and the frequency ratio threshold, respectively.

6. The road congestion prediction system based on intelligent transportation according to claim 5, characterized in that, If the proportion of vehicles that pull over when the traffic flow reaches the source of the deceleration exceeds the vehicle proportion threshold, or if the ratio of the frequency of adjacent vehicles waiting to stop at zero speed to the frequency of vehicles overtaking when there are vehicles at the source of the deceleration exceeds the frequency ratio threshold, it is inferred that the source of the deceleration in the current road monitoring area is affected by the region, and a regional influence signal is generated and sent to the road prediction platform. If the proportion of vehicles pulling over when traffic reaches the source of the deceleration does not exceed the vehicle proportion threshold, and the ratio of the frequency of adjacent vehicles waiting to stop at zero speed to the frequency of vehicles overtaking when there are vehicles at the source of the deceleration does not exceed the frequency ratio threshold, then it is inferred that the source of the deceleration in the current road monitoring area is not affected by regional factors, and a non-regional influence signal is generated and sent to the road prediction platform.

7. The road congestion prediction system based on intelligent transportation according to claim 1, characterized in that, The process of the lane switching decision setting unit is as follows: The switching time period of the switchable lanes is obtained, based on the completed switching traffic flow of the switchable lanes within the road monitoring area during the switching time period and the lane-changing traffic flow of the switchable lanes before switching in the non-switchable lanes; If the traffic flow of the completed lane switching exceeds the switching traffic flow threshold, and the traffic flow of the lane changing before the switchable lane in the non-switching lane exceeds the set lane changing traffic flow threshold, it is inferred that the current switching time period is set to be adapted, but adaptive setting is required. An adaptive setting signal is generated and sent to the road prediction platform. After receiving the adaptive setting signal, the road prediction platform increases the span requirement according to the current road traffic demand, that is, according to the lane changing demand or traffic flow of the set road type. The trigger condition is set based on the frequency of lane changes or the increase in traffic volume, and the switchable lane is set to a switchable state after the trigger condition is triggered. If the traffic flow of vehicles completing lane switching exceeds the lane switching traffic flow threshold, and the traffic flow of vehicles changing lanes before switching lanes in non-lane switching lanes does not exceed the set lane changing traffic flow threshold, it is inferred that the current switching time period setting is adapted and there is no need for adaptive setting. A single time period setting signal is generated and sent to the road prediction platform. After receiving the signal, the road prediction platform sets the switching time period as a single trigger condition.

8. The road congestion prediction system based on intelligent transportation according to claim 7, characterized in that, If the traffic flow completing the lane change does not exceed the lane change traffic flow threshold, and the traffic flow before the lane change in the non-lane change lane exceeds the set lane change traffic flow threshold, it is inferred that the current lane change time period setting is incompatible. A time period adjustment setting signal is generated and sent to the road prediction platform. After receiving the signal, the road prediction platform resets the lane change time period. If the traffic flow completing the lane change does not exceed the lane change traffic flow threshold, and the traffic flow before the lane change in the non-lane change lane does not exceed the set lane change traffic flow threshold, it is inferred that the current lane change time period setting is incompatible. A lane change suspension setting signal is generated and sent to the road prediction platform. After receiving the signal, the road prediction platform suspends the lane change.

9. A road congestion prediction method based on intelligent transportation, employing the road congestion prediction system based on intelligent transportation as described in any one of claims 1-8, characterized in that, The specific steps are as follows: Step 1: Traffic Condition Analysis. Analyze the traffic conditions in the road monitoring area and predict road congestion based on the analysis and the current driving environment. Step 2: Identify the source of speed reduction. Identify the source of speed reduction in the road monitoring area to facilitate congestion prediction and targeted traffic management. Step 3: Lane switching decision setting. Lane switching decisions are set for the road monitoring area.

Images