A future intelligent control method, system, and storage medium for coordinated highway entrances and exits.

By acquiring lane-level traffic data and vehicle operation data, identifying vehicles waiting to merge, to be diverted, and those going straight on the main line, and generating global control results, the shortcomings of traffic condition assessment and collaborative decision-making in highway entrance and exit control methods are solved. This achieves integrated collaborative management and control of the main line, entrance ramps, and exit ramps, improving traffic efficiency and safety.

CN123072202APending Publication Date: 2026-11-10RES INST OF HIGHWAY MINIST OF TRANSPORT +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611315322.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-08-27
Publication Date
2026-11-10

AI Technical Summary

Technical Problem

The existing highway entrance and exit control methods lack a unified traffic condition assessment and collaborative decision-making mechanism, resulting in a mismatch between entrance merging, exit diversion and mainline traffic, affecting traffic efficiency and driving safety, and easily exacerbating congestion, especially when traffic flow changes.

Method used

By acquiring lane-level traffic data and vehicle operation data, the system identifies vehicles waiting to merge, vehicles waiting to divert, and vehicles going straight on the main line, determines traffic status indicators, generates global control results, and issues traffic control instructions to intelligent connected vehicles. This enables integrated and coordinated management of the main line, entrance ramps, and exit ramps, ensuring orderly vehicle passage.

Benefits of technology

It improves traffic coordination and stability at entrances and exits, reduces the risk of conflicts between vehicles vying for lanes in weaving areas, and enhances traffic efficiency and safety, as well as the ability to adapt to changes in traffic conditions.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123072202A_ABST
    Figure CN123072202A_ABST
Patent Text Reader

Abstract

This invention discloses a future highway entrance and exit coordination intelligent control method, system, and storage medium, relating to the field of road vehicle traffic control technology. The method includes: acquiring lane-level traffic data, vehicle operation data, and vehicle traffic demand in the coordinated control area; identifying vehicles to merge, vehicles to divert, and mainline straight-through vehicles, determining mainline traffic saturation, entrance merging pressure, exit diversion load, and weaving zone conflict risk; determining merging and diversion target flow rates and mainline lane control parameters based on traffic state indicators; determining the traffic order, traffic period, and target speed using global control results as constraints, and generating lane-changing trajectories for vehicles to merge and divert, ensuring that conflicting vehicle traffic periods do not overlap; and issuing traffic control commands to intelligent connected vehicles to regulate merging, diversion, weaving, and mainline traffic. This invention is used for highway entrance and exit coordination control, reducing conflict and congestion risks, and improving traffic efficiency and safety.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of road vehicle traffic control technology, and in particular to a future highway entrance and exit coordination intelligent control method, a future highway entrance and exit coordination intelligent control system, and a computer-readable storage medium. Background Technology

[0002] Highway entrances and exits are critical sections where mainline traffic and ramp traffic merge, diverge, and weave. Vehicles on entrance ramps need to enter the mainline of the highway, while vehicles on exit ramps need to leave. Merging vehicles, diverging vehicles, and mainline through traffic influence each other within the entrance / exit weaving area. Especially when traffic flow changes, there is a strong spatiotemporal correlation between entrance merging, exit diverging, and mainline through traffic.

[0003] Current highway entrance and exit control methods typically involve separate control for entrance ramps, exit ramps, and the mainline. Entrance ramps primarily employ flow restriction or release control, exit ramps mainly use diversion guidance, and the mainline manages speed or lanes based on local traffic conditions. There is a lack of a unified traffic condition assessment and collaborative decision-making mechanism among these control links. When the mainline's capacity is high, if vehicles continue to merge at the original flow rate, it can easily exacerbate mainline congestion; conversely, when the mainline's capacity is not fully utilized, overly strict entrance controls can lead to insufficient merging capacity. Furthermore, the lack of coordination between the diversion load of exit ramps and the merging flow of entrance ramps makes it difficult to achieve overall traffic balance at entrances and exits.

[0004] At entrance / exit weaving zones, existing control methods rely heavily on road signs, lane markings, local speed limits, or single-vehicle guidance. These methods typically target vehicles traveling in a single direction or a specific group, lacking a unified sequence and timetable for merging vehicles, vehicles waiting to be diverted, and vehicles proceeding straight on the main road. When merging and diverting vehicles approach the weaving zone simultaneously, they may engage in lane-changing, crossing, or sudden speed changes within the same area. Existing control methods struggle to ensure that vehicles with conflicting relationships pass through the weaving zone in a coordinated sequence, thus impacting traffic efficiency and safety.

[0005] With the development of intelligent connected vehicles and vehicle-to-infrastructure (V2I) communication technologies, vehicles can upload their own operating status and receive roadside control commands, providing a data and execution foundation for refined control of highway entrances and exits. However, existing technologies still lack a coordinated control method that can jointly determine the inflow and outflow of traffic at entrances and exits based on lane-level traffic data, vehicle operation data, and vehicle traffic demand, and further convert the global control results into coordinated control methods for each vehicle's travel time, target speed, and lane-changing trajectory. Therefore, it is necessary to propose a future intelligent control method for coordinated highway entrances and exits to achieve global coordination of traffic flow at entrances and exits and orderly passage of vehicles at the individual vehicle level in weaving zones. Summary of the Invention

[0006] To address the aforementioned issues, this invention provides a future intelligent control method, system, and storage medium for coordinated access to highways. By acquiring lane-level traffic data, vehicle operation data, and vehicle traffic demand from the mainline, entrance ramps, exit ramps, and weaving zones within a preset coordinated control area, the system determines mainline traffic saturation, entrance merging pressure, exit diversion load, and weaving zone conflict risks. It then jointly generates target merging flow, target diversion flow, and mainline lane control parameters, using the global control results to constrain the passage order, passage time, target speed, lane-changing trajectory, and lane passage requirements of individual vehicles. Corresponding traffic control commands are then issued to intelligent connected vehicles, achieving integrated coordinated management of the mainline, entrance ramps, and exit ramps. This reduces the risk of vehicle lane-grabbing conflicts in weaving zones and queuing backflow at exit ramps, improving traffic efficiency, driving safety, and traffic stability in the access areas.

[0007] To achieve the above objectives, the present invention provides a future intelligent control method for coordinated access to highway entrances and exits, comprising: Acquire lane-level traffic data, vehicle operation data, and vehicle traffic demand for the main line, entrance ramps, exit ramps, and entrance / exit weaving areas of the expressway within the preset collaborative control area; Based on the vehicle operation data and the vehicle traffic demand, identify vehicles to merge, vehicles to be diverted, and mainline straight-through vehicles within the preset collaborative control area, wherein the vehicles to merge, vehicles to be diverted, and mainline straight-through vehicles are all intelligent connected vehicles. Based on the lane-level traffic data and the vehicle operation data, traffic status indicators including mainline traffic saturation, inlet merging pressure, outlet diversion load and weaving zone conflict risk are determined. Based on the traffic state indicators, the target merging flow of the entrance ramps and the target diversion flow of the exit ramps are jointly determined, along with the mainline lane control parameters, to form a global control result. The mainline lane control parameters include the mainline lane function configuration and the mainline lane speed constraint. Using the global control result as a single-vehicle scheduling constraint, based on the vehicle operation data of the vehicles to be merged, the vehicles to be diverted, and the mainline straight-through vehicles, the passage order and passage time of each vehicle entering the entrance and exit weaving area are determined, the target driving speed corresponding to each vehicle is generated, and the corresponding lane-changing trajectory is generated for the vehicles to be merged and the vehicles to be diverted, forming a single-vehicle scheduling result. Among them, the passage time of vehicles to be merged and vehicles to be diverted that have conflicting relationships do not overlap. Based on the global control results and the single-vehicle scheduling results, traffic control instructions containing corresponding travel time periods and target speeds are issued to the vehicles to be merged, the vehicles to be diverted, and the vehicles traveling straight on the main line. The traffic control instructions issued to the vehicles to be merged and the vehicles to be diverted also include corresponding lane-changing trajectories, and the traffic control instructions issued to the vehicles traveling straight on the main line also include corresponding lane passage requirements based on the main line lane function configuration. The traffic control commands regulate the merging of vehicles at the entrance ramps according to the target merging flow, regulate the diversion of vehicles at the exit ramps according to the target diversion flow, and allow vehicles with conflicting relationships to pass through the entrance / exit weaving area according to the passage order and passage time. In addition, the commands regulate the lane passage and driving speed of the mainline straight-through vehicles according to the mainline lane function configuration and the mainline lane speed constraints.

[0008] In the above technical solution, preferably, the lane-level traffic data is obtained by roadside sensing devices deployed on the main line of the expressway, entrance ramps, exit ramps and entrance / exit weaving areas, and the vehicle operation data and vehicle traffic demand are obtained by receiving data sent by vehicle terminals through vehicle-road communication equipment. The lane-level traffic data includes traffic density, average speed of each lane, lane occupancy, queue length, and traffic arrival rate; the vehicle operation data includes real-time vehicle location, speed, acceleration, trajectory, vehicle type, and estimated travel time. Before determining traffic status indicators based on lane-level traffic data and vehicle operation data, the process further includes acquiring road environment data and performing data fusion, cleaning, and spatiotemporal calibration on the lane-level traffic data, vehicle operation data, and road environment data to obtain processed lane-level traffic data and vehicle operation data. The traffic status indicators are determined based on the processed lane-level traffic data and vehicle operation data, wherein the road environment data includes road surface condition, weather conditions, light intensity, and temporary road control status.

[0009] In the above technical solution, preferably, the traffic status index further includes an entrance / exit queue overflow risk value; The determination of traffic state indicators includes predicting the trend of entrance and exit traffic flow changes within the next 5 to 30 seconds based on processed lane-level traffic data and vehicle operation data continuously acquired within a time window, and determining the conflict risk of the weaving area and the queuing overflow risk value of the entrance and exit based on the trend of entrance and exit traffic flow changes and the processed lane-level traffic data and vehicle operation data.

[0010] In the above technical solution, preferably, the joint determination of the target merging flow of the entrance ramp, the target diversion flow of the exit ramp, and the mainline lane control parameters includes: When the mainline traffic saturation is lower than the preset low saturation threshold, increase the target inflow of the entrance ramps; When the main line traffic saturation reaches the preset congestion threshold, the target inflow of the entrance ramps will be reduced. When the exit diversion load or the risk value of queuing overflow at the entrance or exit reaches a preset threshold, the target diversion flow of the exit ramp is reduced. Based on the traffic load of the mainline, entrance ramps, and exit ramps, the mainline lane function configuration and mainline lane speed constraints are determined. The mainline lane control parameters include the mainline lane function configuration and the mainline lane speed constraints, and the mainline lane function configuration and the mainline lane speed constraints serve as single-vehicle scheduling constraints for the mainline through traffic.

[0011] In the above technical solution, preferably, determining the passage order and passage time of each vehicle entering the entrance / exit weaving area includes: Based on the expected passage time and travel trajectory of each vehicle, identify the vehicles with conflicting relationships that enter the same conflict area within the entrance / exit weaving zone; Determine the passage order for the vehicles with conflicting relationships, and allocate non-overlapping dedicated passage time slots as the passage time periods according to the passage order; The target driving speed is determined based on the dedicated passage time slot corresponding to each vehicle, and the lane-changing trajectory of the vehicle to be merged and the vehicle to be diverted is determined.

[0012] In the above technical solution, preferably, the step of regulating vehicle merging at the entrance ramp according to the target merging flow includes: Vehicles waiting to merge at the entrance ramp are divided into vehicle platoons; Based on the target merging flow and the gap in the main traffic flow, the formation size, formation speed and merging timing of the vehicle formation are determined; When the mainline traffic saturation reaches a preset congestion threshold, flow control is implemented on the entrance ramps, and the entry of vehicles is restricted from entering the mainline by adjusting the platoon size and merging timing of the vehicle platoons.

[0013] In the above technical solution, preferably, the method of regulating the vehicle diversion of the exit ramp according to the target diversion flow includes: The diversion demand of the vehicles to be diverted is predicted based on the vehicle traffic demand. In the pre-designated lane-changing area upstream of the exit ramp, a lane-changing guidance instruction is issued to the vehicles to be diverted, guiding them to change lanes to the outer diversion lane. The vehicle diversion rate is determined based on the target diversion flow and the real-time capacity of the exit ramps to reduce the risk of queues at the exit ramps flowing back onto the main line.

[0014] In the above technical solution, preferably, after issuing the access control command, it further includes: The efficiency of traffic flow after control, the number of vehicle conflicts, the queue length and the time it takes for vehicles to pass through are obtained as feedback data for instruction execution. The instruction execution feedback data is compared with preset traffic efficiency threshold, preset vehicle conflict number threshold, preset queue length threshold, and preset vehicle passage time threshold. When the traffic flow efficiency is lower than the preset traffic flow efficiency threshold, the number of vehicle conflicts reaches the preset number of vehicle conflicts threshold, the queue length reaches the preset queue length threshold, or the vehicle passage time reaches the preset vehicle passage time threshold, the control parameters used to determine the global control result and the single-vehicle scheduling result are corrected, and the target merging flow of the entrance ramp, the target diversion flow of the exit ramp, the mainline lane control parameters, the passage order of each vehicle, the passage time period, the target driving speed, and the lane-changing trajectory of the vehicles to be merged and the vehicles to be diverted are re-determined. Based on the re-determined global control result and single-vehicle scheduling result, an updated traffic control command is issued.

[0015] This invention also proposes a future intelligent control system for highway entrance and exit coordination, comprising a data acquisition module, a traffic condition assessment module, a global decision-making module, a vehicle scheduling module, a control execution module, and a feedback optimization module, wherein: The data acquisition module is used to acquire lane-level traffic data, vehicle operation data and vehicle traffic demand within a preset cooperative control area through roadside sensing devices and vehicle-road communication devices. The traffic condition assessment module is used to determine traffic condition indicators based on the lane-level traffic data and vehicle operation data output by the data acquisition module. The global decision-making module is used to determine the target merging flow of the entrance ramp, the target diversion flow of the exit ramp, and the mainline lane control parameters based on the traffic state indicators output by the traffic state assessment module, and to form a global control result. The mainline lane control parameters include the mainline lane function configuration and the mainline lane speed constraint. The single-vehicle scheduling module is used to identify vehicles to merge, vehicles to be diverted, and mainline straight-through vehicles based on the vehicle operation data and vehicle traffic demand output by the data acquisition module, and to determine the passage order, passage time period, and target driving speed of each vehicle using the global control result output by the global decision module as constraints, as well as to generate the lane-changing trajectory of vehicles to merge and vehicles to be diverted, thus forming a single-vehicle scheduling result. The control execution module is used to issue passage control instructions to intelligent connected vehicles based on the global control results and the single vehicle scheduling results. The passage control instructions issued to vehicles traveling straight on the main line include the corresponding lane passage requirements based on the main line lane function configuration and the target driving speed based on the main line lane speed constraints. The feedback optimization module is used to acquire traffic flow efficiency, vehicle conflict frequency, queue length, and vehicle passage time as instruction execution feedback data, and compare the instruction execution feedback data with preset traffic flow efficiency thresholds, preset vehicle conflict frequency thresholds, preset queue length thresholds, and preset vehicle passage time thresholds. When the traffic flow efficiency is lower than the preset traffic flow efficiency threshold, the vehicle conflict frequency reaches the preset vehicle conflict frequency threshold, the queue length reaches the preset queue length threshold, or the vehicle passage time reaches the preset vehicle passage time threshold, the control parameters used to determine the global control result and the single-vehicle scheduling result are corrected, and the global decision module and the single-vehicle scheduling module are triggered to regenerate the global control result and the single-vehicle scheduling result, and the control execution module is triggered to issue updated traffic control instructions based on the regenerated global control result and single-vehicle scheduling result.

[0016] The present invention also proposes a computer-readable storage medium storing a computer program thereon, wherein when the computer program is executed by a processor, it implements any of the above-disclosed methods for coordinated intelligent control of future highway entrances and exits.

[0017] Compared with the prior art, the beneficial effects of the present invention are as follows: (1) By jointly determining the target merging flow, target diversion flow and mainline lane control parameters based on the mainline traffic saturation, entrance merging pressure, exit diversion load and weaving area conflict risk, and using the global control results as single-vehicle scheduling constraints, the integrated coordinated control of the mainline, entrance ramps and exit ramps is realized, which improves the mismatch between entrance merging, exit diversion and mainline traffic under the traditional decentralized control, and improves the overall coordination and traffic stability of the entrance and exit areas.

[0018] (2) By determining the passage order, passage time, target driving speed and lane changing trajectory for vehicles waiting to merge, vehicles waiting to divert and vehicles going straight on the main line, and allocating non-overlapping passage time periods for merging and diverting vehicles with conflicting relationships, the possibility of vehicles in the weaving area occupying conflict areas at the same time is reduced, the risks of cutting in, sudden acceleration and deceleration and frequent lane changing are reduced, and the safety and orderliness of vehicle passage in the weaving area are improved.

[0019] (3) By determining the merging size, merging speed and merging timing of the entrance vehicle platoon based on the target merging flow and the gap between the main line traffic flow, and implementing entrance flow restriction when the main line is congested; at the same time, by implementing advance lane change guidance and diversion rate control based on the target diversion flow and the real-time carrying capacity of the exit ramp, the smooth merging of entrance vehicles and the orderly diversion of exit vehicles are realized, reducing the interference of entrance merging on the main line traffic and the risk of queues at the exit ramp flowing back to the main line.

[0020] (4) By acquiring instruction execution feedback data such as traffic flow efficiency, number of vehicle conflicts, queue length and vehicle passage time, the control parameters of global control and single vehicle scheduling are corrected when the feedback data reaches the corresponding threshold, and updated passage control instructions are issued, realizing continuous response to traffic status changes and iterative control results, and improving the ability of entrance and exit coordination control to adapt to traffic flow fluctuations and congestion risk changes. Attached Figure Description

[0021] Figure 1 This is a flowchart illustrating a future highway entrance and exit coordination intelligent control method disclosed in one embodiment of the present invention. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, 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, 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.

[0023] The present invention will now be described in further detail with reference to the accompanying drawings: like Figure 1As shown, the present invention provides a future highway entrance and exit coordination intelligent control method, which is applied to a preset coordinated control area. This preset coordinated control area covers the highway mainline, entrance ramps, exit ramps, and the entrance / exit weaving zone where entrance traffic, exit traffic, and mainline traffic converge. This area can also extend upstream and downstream of the entrance and exit ramps according to the actual road structure, allowing vehicles the distance required to complete speed and lane adjustments before entering the weaving zone.

[0024] This implementation method applies to traffic scenarios where all vehicles merging, diverting, and proceeding straight on the main line are intelligent connected vehicles. Each intelligent connected vehicle can provide its own operating status and traffic requirements, and can receive corresponding traffic control commands, thereby transforming highway entrance and exit control from traditional fixed signal timing and roadside sign guidance to dynamic control tailored to specific vehicles.

[0025] First, lane-level traffic data, vehicle operation data, and vehicle traffic demand are acquired for the highway mainline, entrance ramps, exit ramps, and weaving areas within the pre-defined collaborative control area. Lane-level traffic data reflects the overall traffic flow status of each road segment and lane; vehicle operation data reflects the current position, speed, and movement trend of specific vehicles; and vehicle traffic demand characterizes the intentions of vehicles to enter the mainline, exit at a predetermined exit, or continue straight along the mainline. These three types of data collectively constitute the input for subsequent traffic state assessment and vehicle control, enabling the utilization of both overall traffic flow status and individual vehicle operation status within the same control process.

[0026] Based on vehicle operation data representing the vehicle's road location, lane, and direction of movement, and combined with traffic demand, vehicles within a pre-defined cooperative control area are classified and identified. Vehicles located at entrance ramps or merging areas and requiring entry into the main line are identified as vehicles awaiting merging; vehicles located on the main line and requiring exit via exit ramps are identified as vehicles awaiting diversion; and vehicles without current merging or diversion requirements and needing to continue along the main line are identified as vehicles traveling straight on the main line. These classification results are used to determine the direction of movement of different vehicles within the weaving zone and the types of control commands that need to be received.

[0027] Subsequently, based on lane-level traffic data and vehicle operation data, traffic status indicators were determined, including mainline traffic saturation, merging pressure at entrance ramps, diversion load at exit ramps, and conflict risk in weaving zones. Mainline traffic saturation characterizes the proximity between current traffic demand and the available capacity of the mainline; merging pressure at entrance ramps characterizes the demand of vehicles waiting to merge at entrance ramps and its impact on mainline traffic; diversion load at exit ramps characterizes the relationship between the number of vehicles scheduled to leave and the current capacity of exit ramps; and conflict risk in weaving zones characterizes the possibility of temporal and spatial overlap between vehicles moving in different directions within the weaving zone. By employing these indicators across different dimensions, it is possible to avoid relying solely on the local conditions of entrance or exit ramps for control.

[0028] Based on traffic state indicators, the target merging flow of entrance ramps and the target diversion flow of exit ramps are jointly determined, along with mainline lane control parameters, to form a global control result. Mainline traffic saturation characterizes the proximity between existing mainline traffic demand and mainline capacity; entrance merging pressure characterizes the demand of vehicles waiting to merge at entrance ramps and its impact on mainline traffic; exit diversion load characterizes the relationship between the number of vehicles scheduled to leave and the current capacity of exit ramps; weaving zone conflict risk characterizes the possibility of temporal and spatial overlap between vehicles moving in different directions within the weaving zone. By employing these indicators across different dimensions, it is possible to avoid relying solely on the local conditions of entrance or exit ramps for individual control.

[0029] The mainline lane control parameters include mainline lane function configuration and mainline lane speed constraints. The mainline lane function configuration is used to determine the passage requirements or passage priority relationships of different mainline lanes for straight, merging, or diverting traffic within the current control cycle. The mainline lane speed constraints are used to limit the permissible speed range of vehicles in each mainline lane. The resulting global control result reflects the overall traffic flow organization requirements of the preset cooperative control area within the current control cycle.

[0030] Using global control results as constraints for individual vehicle scheduling, when arranging the passage of vehicles waiting to merge, vehicles waiting to be diverted, and vehicles traveling straight on the main line, the number of vehicles merging within the current control cycle should be adapted to the target merging flow, the number of vehicles diverting should be adapted to the target diverting flow, the selected lanes should conform to the functional configuration of the main line lanes, and the target driving speed should conform to the speed constraints of the corresponding lanes. Through this constraint relationship, the global-level traffic flow control target is transmitted to the individual vehicle-level operation control. Therefore, based on the vehicle operation data of vehicles waiting to merge, vehicles waiting to be diverted, and vehicles traveling straight on the main line, the passage order and passage time of each vehicle entering the entrance / exit weaving area are determined. For vehicles whose movement paths do not conflict, they can be arranged to pass at the same or similar times; for vehicles waiting to merge and vehicles waiting to be diverted whose merging and diverting paths conflict within the weaving area, non-overlapping passage time periods are arranged respectively. The passage order is used to determine the order in which related vehicles enter the weaving area, and the passage time period is used to determine the time range within which each vehicle can occupy the corresponding weaving space.

[0031] Based on the vehicle's current location, current speed, and allocated travel time period, a target travel speed is generated for each vehicle, enabling vehicles to adjust their speed by accelerating, maintaining a constant speed, or decelerating to arrive at and pass through the weaving zone during the corresponding travel time period. For vehicles waiting to merge or to be diverted, a lane-changing trajectory is also generated based on the vehicle's current lane, target travel lane, and corresponding travel time period, allowing vehicles to complete merging or diverting while meeting the global lane function configuration and weaving zone timing requirements. The single-vehicle scheduling result is formed by the travel order, travel time period, target travel speed, and lane-changing trajectory.

[0032] Based on the overall control results and individual vehicle scheduling results, traffic control instructions are issued to vehicles waiting to merge, vehicles waiting to be diverted, and vehicles traveling straight on the main line. The traffic control instructions include the corresponding time period and target speed. The traffic control instructions issued to vehicles waiting to merge and vehicles waiting to be diverted also include the corresponding lane-changing trajectory, while the traffic control instructions issued to vehicles traveling straight on the main line also include the corresponding lane passage requirements based on the main line lane function configuration. After receiving the instructions, vehicles adjust their operating status according to the target speed and complete lane changes according to the corresponding lane-changing trajectory when lane changes are required.

[0033] By executing traffic control commands, at the overall level, the merging of vehicles at the entrance ramps is regulated according to the target merging flow, and the diversion of vehicles at the exit ramps is regulated according to the target diversion flow. At the weaving zone level, vehicles with conflicting relationships that are waiting to merge and vehicles waiting to be diverted pass through the entrance and exit weaving zones according to their passage order and passage time. At the mainline level, the lane passage and driving speed of vehicles going straight on the mainline are regulated according to the mainline lane function configuration and mainline lane speed constraints, forming a coordinated control method of traffic state assessment - global flow and lane decision-making - individual vehicle timing and trajectory scheduling - vehicle execution.

[0034] In this implementation, by jointly deciding on the inbound traffic flow, the outbound traffic flow, and the mainline lane control parameters, and further converting the global control results into single-vehicle travel time, target speed, and lane-changing trajectory, the imbalance in traffic flow matching caused by the independent control of the mainline, inbound ramps, and outbound ramps can be overcome. By ensuring that the travel time of conflicting vehicles does not overlap, the sudden acceleration and deceleration, lane-grabbing, and weaving conflicts caused by the simultaneous encroachment of inbound and outbound traffic on the weaving area can be reduced, thereby improving the traffic stability and safety of the entrance and exit areas.

[0035] In the above embodiments, preferably, roadside sensing equipment is deployed along the main line of the highway, entrance ramps, exit ramps, and intersections of entrances and exits. The roadside sensing equipment may include roadside lidar, high-definition cameras, and traffic flow sensors, with each device forming a continuous or interconnected sensing range facing its corresponding lane. The roadside lidar and high-definition cameras are used to obtain the distribution and movement of vehicles within the lanes, while the traffic flow sensors are used to obtain traffic flow statistics for each lane. Vehicle-to-infrastructure (V2I) communication equipment is deployed in the corresponding road areas. This V2I equipment can employ 5G-V2X communication units and establish an information exchange link with the vehicle terminals of intelligent connected vehicles.

[0036] Lane-level traffic data is obtained through roadside sensing devices. This data includes traffic density, average speed per lane, lane occupancy, queue length, and vehicle arrival rate. Traffic density reflects the number of vehicles per unit road length; average speed per lane reflects the speed of traffic flow; lane occupancy reflects the degree to which a lane is occupied by vehicles during the data collection period; queue length reflects the degree of vehicle accumulation at entrances, exits, or upstream of weaving zones; and vehicle arrival rate reflects the number of vehicles entering the corresponding road area per unit time. This data is correlated by road segment and lane to maintain lane-level spatial resolution.

[0037] Intelligent connected vehicles' terminals transmit vehicle operation data and traffic demand in real time or according to a preset update cycle. Vehicle operation data includes real-time vehicle location, speed, acceleration, trajectory, vehicle type, and estimated travel time. Real-time vehicle location and trajectory determine the vehicle's current road segment, lane, and subsequent path; speed and acceleration determine the vehicle's motion state and estimate the time to reach the weaving zone; vehicle type distinguishes vehicles with different dynamic characteristics or road traffic characteristics; estimated travel time indicates the expected time for the vehicle to enter the relevant control position or weaving zone. Traffic demand indicates vehicle merging at entrances, diversion at exits, or straight-through traffic on the main line.

[0038] In addition to the data mentioned above, road environment data is also acquired. This data includes road surface condition, weather conditions, light intensity, and temporary road closures. Road surface condition can include whether the road is dry, wet, or flooded; weather conditions can include rain, snow, and fog that affect vehicle operation; and temporary road closures can represent temporary lane closures, road construction, or other temporary traffic restrictions. This road environment data is used to correct judgments regarding road capacity and vehicle safety conditions.

[0039] Before determining traffic status indicators, lane-level traffic data, vehicle operation data, and road environment data undergo data fusion, cleaning, and spatiotemporal calibration. First, a correspondence is established between multi-source data based on data collection time, road segment, lane position, and vehicle identity. Data exceeding reasonable variation ranges due to equipment failure, communication anomalies, or extreme interference is identified and removed, and short-term missing or duplicate data is processed. Subsequently, the time and spatial coordinate references of different data sources are unified, ensuring that vehicle positions and lane status detected by roadside equipment are within the same spatiotemporal reference system as the positions and trajectories reported by vehicle terminals.

[0040] In one implementation, roadside sensing results and vehicle terminal self-reporting results can be used for bidirectional verification. For example, the corresponding vehicle is located in the roadside sensing data based on the location reported by the vehicle terminal, and the continuity of the reported vehicle speed is verified using speed changes detected by the roadside equipment. After verification, processed lane-level traffic data and vehicle operation data are obtained. The data update cycle can be set to 100 milliseconds to provide continuously updated data input for traffic condition assessment and vehicle dispatching.

[0041] The processed lane-level traffic data and vehicle operation data are transmitted to the traffic state indicator determination process. Road environment data is used to correct data validity judgments and interpret traffic conditions under different road conditions during processing, avoiding the interpretation of lane capacity as normal road conditions under rain, snow, fog, low light, or temporary traffic control conditions.

[0042] In this implementation, by fusing and verifying roadside sensing data and vehicle terminal self-reported data according to a unified time and space benchmark, data deviations caused by single sensing device failures, obstructions, or communication anomalies can be reduced. By introducing road environment data, traffic condition assessments can be adapted to actual road traffic conditions, providing a more real-time, complete, and consistent data foundation for subsequent traffic flow decisions and vehicle dispatching.

[0043] In the above implementation, the processed lane-level traffic data and vehicle operation data are organized according to continuous time windows. Each time window contains data from multiple continuously updated moments, enabling the identification of changes in traffic density, average speed of each lane, lane occupancy, queue length, vehicle arrival rate, and vehicle motion status over time. As new data enters the time window, historical data exceeding the time window can be removed, thus forming continuously updated time-series data.

[0044] Based on time-series data within a time window, the project predicts the trend of traffic flow changes at entrances and exits within the next 5 to 30 seconds. This prediction time range primarily covers the short-term process required for vehicles to travel from upstream of the coordinated control area to the merging point at the entrance, the diverging point at the exit, or the weaving zone. The prediction comprehensively considers changes in vehicle arrival rate, existing queue length, average lane speed, expected vehicle passage time, and vehicle trajectory to obtain the changing trends of the number of vehicles arriving at the entrance, the number of vehicles diverting at the exit, the mainline vehicle throughput capacity, and the vehicle concentration in the weaving zone within a short period of time.

[0045] Preferably, the traffic condition indicators also include an entrance / exit queue overflow risk value; this value characterizes the likelihood that vehicle queues at entrance or exit ramps will continue to grow and extend into adjacent main roads or upstream road areas. In practice, this risk value can be determined based on the current queue length, the trend of queue length changes, the relationship between vehicle arrival and departure rates, and available road space for queuing. The queue overflow risk value increases accordingly when arriving vehicles consistently outnumber departing vehicles, the queue length continues to increase, and it approaches available queuing space.

[0046] The risk of conflict in weaving areas is determined based on the movement trends of vehicles waiting to merge, vehicles waiting to divert, and vehicles traveling straight on the main line over the next 5–30 seconds. Based on real-time vehicle location, speed, estimated travel time, and trajectory, it is determined whether different vehicles are likely to enter weaving areas with intersecting paths at similar times; the more conflicting vehicles that are likely to arrive at the same or similar times, the higher the risk of conflict in the weaving area. Road surface conditions, weather conditions, and lighting conditions reflected in road environment data can also be used to adjust risk assessments, ensuring that risk assessments in conditions such as slippery surfaces and low visibility maintain the necessary safety margin.

[0047] In one implementation, traffic conditions can be categorized into smooth, slow, congested, and severely congested levels based on the relationship between mainline traffic saturation, inlet merging pressure, outlet diversion load, weaving zone conflict risk, and inlet / outlet queue overflow risk values ​​and corresponding thresholds. This categorization visually represents the current road condition, while the prediction results represent the future trend of the road condition in the short term. Therefore, subsequent control measures do not need to wait for congestion or queue backflow to occur before taking action.

[0048] In this implementation, traffic flow changes over the next 5 to 30 seconds are predicted through continuous time windows. The predicted trend is then combined with the current queuing status, vehicle trajectory, and expected passage time. This allows for the identification of risks before queue overflow or interleaving conflicts actually occur, transforming passive control based solely on the current state into proactive control with short-term predictive capabilities, thereby improving the timeliness of entrance and exit management.

[0049] In the above implementation, preferably, the target merging flow of the entrance ramp, the target diversion flow of the exit ramp, and the mainline lane control parameters are jointly determined based on the current traffic condition assessment and short-term risk prediction.

[0050] Within each control cycle, with the goal of maintaining overall traffic efficiency at entrances and exits and reducing traffic flow conflicts, a comprehensive analysis is conducted on mainline capacity, the number of vehicles waiting to merge at entrances, the number of vehicles waiting to be diverted at exits, conflict risks in weaving zones, and queue overflow risks. Target merging flow, target diversion flow, and mainline lane control parameters are determined using the same set of traffic state indicators, ensuring coordination between entrance release, exit diversion, and mainline lane organization.

[0051] When the mainline traffic saturation is below a preset low saturation threshold, it indicates that the mainline still has sufficient capacity to accommodate incoming vehicles. In this case, the target merging flow rate at the entrance ramps is increased, allowing more vehicles to enter the mainline within the current control cycle. This utilizes the remaining mainline capacity and reduces unnecessary queuing at the entrance ramps. The degree to which the target merging flow rate is increased can be determined by considering both the entrance merging pressure and the risk of conflict in the weaving area, avoiding new conflicts in the weaving area caused by adding too many merging vehicles at once.

[0052] When the mainline traffic saturation reaches the preset congestion threshold, it indicates that the mainline may continue to accept incoming vehicles, which may cause the mainline speed to decrease or the congestion to worsen. At this time, the target merging flow of the entrance ramp is reduced, the number of vehicles entering the mainline in the current control cycle is limited, the existing vehicles on the mainline are given priority to pass through the entrance and exit areas, and the mainline traffic flow gap is reserved for subsequent vehicles to merge smoothly.

[0053] When the exit diversion load or the risk value of queue overflow at entrances and exits reaches a corresponding preset threshold, the target diversion flow of the exit ramp is reduced. This target diversion flow represents the vehicle flow allowed to enter the exit diversion process from the mainline within the current control cycle, rather than the physical capacity of the exit ramp itself. By limiting the number of vehicles concentrated on the exit ramp in a short period of time, the vehicle diversion rate can be matched with the real-time carrying capacity of the exit ramp, preventing the exit queue from flowing back into the mainline.

[0054] The functional configuration of mainline lanes is determined based on the traffic load of the mainline, entrance ramps, and exit ramps. This configuration can be achieved by specifying the passage requirements or priority relationships for through traffic, merging traffic, and exit traffic on each lane within the current control cycle. For example, when merging pressure is high, the mainline lane adjacent to the entrance ramp can handle the merging connection, while other lanes can divert through traffic. When exit traffic load is high, the outer lanes near the exit ramps can maintain their diversion connection conditions, reducing the need for vehicles to continuously cross lanes near the exit. This configuration changes the organization of traffic flow among the mainline lanes without requiring changes to the road's physical structure.

[0055] Simultaneously, speed constraints for mainline lanes are determined based on traffic load, average speed, road conditions, and merging / diverting connection requirements for each lane. For lanes requiring merging or diverting connections, speed constraints adapted to the weaving zone's traffic sequence can be set; for lanes handling straight-through traffic on the mainline, corresponding speed constraints can be set based on their current saturation level. When subsequently generating target speeds for straight-through vehicles on the mainline and related merging / diverting vehicles, the target speeds are ensured not to exceed the speed constraints of the corresponding lanes.

[0056] The determined target inflow, target diversion flow, mainline lane function configuration, and mainline lane speed constraints together form the global control result and serve as individual vehicle scheduling constraints. This ensures that the order of individual vehicle passage, passage time, target speed, and lane-changing trajectory are all constrained by the overall traffic flow organization target, preventing local optimal operations at the individual vehicle level from disrupting the overall traffic flow.

[0057] In this implementation, by increasing or decreasing the target inflow at the entrance according to the saturation level of the main line, and controlling the target diversion flow according to the exit load and queuing overflow risk, the problems of excessive flow restriction at the entrance when the main line is smooth and continuous inflow at the entrance when the main line is congested can be reduced. By simultaneously determining the lane function configuration and speed constraints, the lane distribution and driving status of vehicles on the main line can be further organized in addition to flow control, thereby improving the connection between the main line, entrances and exits.

[0058] In the above embodiments, preferably, determining the passage order and passage time of each vehicle entering the entrance / exit weaving area includes: Based on the road geometry of the pre-defined coordinated control area, areas where the merging paths of vehicles to be merged and the diverging paths of vehicles to be diverted are spatially intersected or overlapped are identified as conflict zones. One or more conflict zones may exist within a single entrance / exit weaving zone. The travel trajectories of each vehicle are used to determine the conflict zones they will pass through, and the estimated passage time is used to determine the estimated time when a vehicle will enter the corresponding conflict zone.

[0059] Based on the estimated travel time and trajectory of each vehicle, vehicles with conflicting relationships entering the same conflict area within the entrance / exit weaving zone are identified. When the trajectories of two or more vehicles pass through the same conflict area, and their estimated entry times into that conflict area may overlap, these vehicles are identified as having conflicting relationships. Conflict relationships are defined by both spatial path overlap and temporal proximity, preventing vehicles without actual path conflicts from being included in mutually exclusive scheduling simply because they are located in adjacent road areas.

[0060] For vehicles with conflicting relationships, the passage order is first determined. In one implementation, an initial passage order can be formed according to the expected passage time of each vehicle, and then the order can be dynamically adjusted by taking into account the merging demand, diversion demand, current speed adjustment range, and the traffic status of existing vehicles in the weaving area. After the passage order is determined, dedicated passage slots are allocated according to the passage order. Each dedicated passage slot constitutes the passage time period for the corresponding vehicle, and the dedicated passage slots corresponding to vehicles with conflicting relationships do not overlap.

[0061] A dedicated passage slot represents the time range within which a vehicle can enter and occupy a corresponding conflict zone. When the same vehicle passes through multiple conflict zones in sequence, multiple passage slots can be arranged consecutively based on the vehicle's direction of travel and estimated travel time, allowing the vehicle to pass through the relevant conflict zones sequentially after completing a speed adjustment. For vehicles that do not conflict with each other, their passage slots do not need to be mutually exclusive, thus avoiding unnecessary reduction in the utilization rate of the weaving area.

[0062] The target speed is determined based on the remaining distance from the vehicle's current location to the conflict zone, its current speed, and the designated passage time slot. If the vehicle will arrive earlier than the designated passage time slot at its current speed, the target speed is appropriately reduced; if the vehicle cannot arrive within the designated passage time slot at its current speed, the target speed is increased while meeting the speed constraints of the mainline lanes and the conditions for safe vehicle operation. Determining the target speed allows vehicles to reach the designated passage time slot through continuous speed adjustments, reducing temporary stops or sudden accelerations and decelerations at the entrance to the weaving zone.

[0063] For vehicles merging into the main line, the merging lane-changing trajectory is determined based on their current entrance ramp lane, the main line lane they need to enter, and their corresponding dedicated traffic slot. For vehicles diverting from the main line, the diversion lane is determined based on their current main line lane, the target outer diversion lane, and their corresponding dedicated traffic slot. The lane-changing trajectory is coordinated with the target driving speed to ensure that the time it takes for a vehicle to complete a lateral lane change is consistent with its dedicated traffic slot.

[0064] In this implementation, by simultaneously identifying vehicles with genuine spatiotemporal conflicts based on their driving trajectories and expected passage times, and allocating exclusive passage slots that do not overlap to the conflicting vehicles, it is possible to avoid merging vehicles and diverting vehicles occupying the same conflict area at the same time. By reverse-determining the target driving speed and lane-changing trajectory based on the exclusive passage slots, vehicles can enter the weaving area in a continuous and executable motion process, reducing phenomena such as cutting in, rapid acceleration, and rapid deceleration.

[0065] In the above implementation, preferably, regulating vehicle merging at the entrance ramp according to the target merging flow includes: Based on vehicle operation data of vehicles waiting to merge within the entrance ramps, vehicles that are adjacent in location, travel in the same direction, and have similar expected merging times are grouped into vehicle platoons. Each vehicle platoon contains one or more vehicles waiting to merge, and the vehicles within the platoon operate in a predetermined order. The division of vehicle platoons is not fixed; when new vehicles enter the coordinated control area or the mainline traffic flow changes, the platoon composition can be redefined in subsequent control cycles.

[0066] The total number of vehicles allowed to merge into the mainline within the current control cycle is determined based on the target inflow rate, and the platoon size is determined under this total constraint. Mainline traffic gaps are determined based on the time and spatial intervals between available entry points for vehicles on the mainline, and vehicle platoons of different sizes are matched with their corresponding mainline traffic gaps. When the mainline traffic gap is large, platoons containing more vehicles can be arranged to merge; when the mainline traffic gap is small, the platoon size can be reduced or waiting for subsequent available gaps.

[0067] The platoon's speed and merging timing are determined based on its current position, arrival time within the mainline traffic gap, and the operational relationships between vehicles within the platoon. The platoon speed ensures that the platoon's arrival time at the merging point corresponds to the available mainline traffic gap, while the merging timing determines when the platoon begins to enter the mainline. Each vehicle within the platoon merges according to its target speed, travel time, and lane-changing trajectory, ensuring the entire platoon continuously utilizes the mainline traffic gap.

[0068] When the mainline traffic saturation reaches a preset congestion threshold, flow control is implemented at the entrance ramps. This involves adjusting vehicle platoon size and merging timing to restrict vehicles from entering the mainline. Flow control is achieved by reducing the number of vehicles that can enter the mainline within the current control cycle, rather than allowing all entrance vehicles to simultaneously head towards the merging point. Specifically, this can be achieved by reducing vehicle platoon size, delaying the merging timing of some platoons, or extending the merging interval between adjacent platoons, ensuring that the actual merging flow matches the reduced target merging flow, and prioritizing the passage of existing vehicles on the mainline.

[0069] In this implementation, by organizing scattered vehicles waiting to merge into vehicle platoons and matching the platoon size, platoon speed, and merging timing with the target merging flow and the gap in the mainline traffic flow, the utilization of available mainline gaps can be improved, and the continuous disturbance to the mainline traffic flow caused by random merging of single vehicles can be reduced. When the mainline is congested, entrance flow restriction can be implemented by adjusting the platooning parameters to avoid the continuous occupation of the mainline passage space by the entrance traffic flow.

[0070] In the above implementation, preferably, regulating the vehicle diversion of the exit ramp according to the target diversion flow includes: The diversion demand for mainline vehicles is predicted based on traffic flow demand. Traffic flow demand may include the vehicle's target exit or subsequent travel path. Based on the vehicle's current location, current lane, speed, and target exit, it is determined whether the vehicle needs to move from the mainline to the exit ramp within the current preset cooperative control area, and the corresponding vehicle is identified as a vehicle to be diverted.

[0071] A pre-set lane-changing area is set up upstream of the exit ramp to allow vehicles waiting to be diverted to complete lane adjustments to the outer diversion lanes before reaching the exit diversion point. In one specific implementation, the pre-changing area can be set up approximately 500m upstream of the exit; the specific distance can be determined based on the number of mainline lanes, road design speed, exit ramp structure, and the distance required for vehicles to complete a smooth lane change.

[0072] Once a vehicle waiting to be diverted enters the pre-designated lane-changing area, a lane-changing guidance instruction is issued. This instruction corresponds to the lane-changing trajectory and target speed, guiding the vehicle to gradually change lanes from its current main lane to the outer diversion lane. For vehicles needing to cross multiple main lanes, the lane-changing process can be arranged as continuous, multi-segment lane adjustments to avoid abrupt, continuous lateral movements as the vehicle approaches the exit.

[0073] The vehicle diversion rate is determined based on the target diversion flow and the real-time capacity of the exit ramp. The real-time capacity of the exit ramp reflects the number of vehicles that the exit ramp can receive and discharge within the current control cycle. The vehicle diversion rate does not exceed the flow limit set by the target diversion flow and is adapted to the real-time capacity of the exit ramp. When the capacity of the exit ramp decreases or the queue of vehicles at the exit increases, the speed of vehicles entering the outer diversion lane and heading towards the exit ramp is reduced; when the exit ramp has remaining capacity, the vehicle diversion rate can be increased within the target diversion flow range.

[0074] By adjusting the target speed, lane-changing time, and diversion interval between adjacent vehicles after they enter the pre-diversion area, vehicles awaiting diversion are directed to enter the exit ramps sequentially at a predetermined diversion rate, thereby reducing the risk of queues at the exit ramps backing onto the mainline. Vehicles not yet in the current diversion sequence continue to travel in lanes that conform to the mainline lane function configuration, awaiting subsequent diversion control instructions, thus avoiding a large number of vehicles concentrating simultaneously at the exit diversion location.

[0075] In this implementation, by identifying diversion demand in advance and guiding vehicles to be diverted upstream of the exit to gradually switch to the outer diversion lane, emergency lane changes and mainline traffic disturbances near the exit can be reduced. By making the vehicle diversion rate constrained by both the target diversion flow and the real-time carrying capacity of the exit ramp, the speed at which the exit ramp receives vehicles can be prevented from exceeding its discharge capacity, thus reducing the risk of exit queues flowing back into the mainline.

[0076] In the above embodiments, preferably, after issuing the passage control command, the method further includes: During the execution of traffic control instructions or at the end of a control cycle, the traffic efficiency, number of vehicle conflicts, queue length, and vehicle passage time after control are acquired to form instruction execution feedback data. Traffic efficiency reflects the number of vehicles passing through the preset collaborative control area or the vehicle capacity per unit time; the number of vehicle conflicts reflects the number of times path competition occurs or conflict avoidance measures are required in the weaving area; queue length reflects the vehicle accumulation at entrance ramps, exit ramps, or adjacent road areas; and vehicle passage time reflects the time required for vehicles to pass through the preset collaborative control area.

[0077] The system compares traffic flow efficiency with preset traffic efficiency thresholds, the number of vehicle conflicts with preset vehicle conflict thresholds, queue length with preset queue length thresholds, and vehicle transit time with preset vehicle transit time thresholds. These thresholds correspond to four evaluation dimensions: traffic capacity, weaving safety, queue status, and transit time, ensuring that the control effect is not judged based on a single traffic flow indicator.

[0078] When the traffic flow efficiency is not lower than the preset traffic efficiency threshold, the number of vehicle conflicts is lower than the corresponding threshold, the queue length is lower than the corresponding threshold, and the vehicle passage time is lower than the corresponding threshold, the current control parameters can continue to be used in the next control cycle, or only a regular update can be performed based on the newly acquired traffic status data.

[0079] When traffic flow efficiency falls below a preset efficiency threshold, the number of vehicle conflicts reaches a preset conflict threshold, the queue length reaches a preset queue length threshold, or the vehicle transit time reaches a preset transit time threshold, it indicates that the current global control result or individual vehicle scheduling result fails to adequately adapt to the actual traffic conditions. In this case, the control parameters used to determine the global control result and individual vehicle scheduling result are corrected. The corrected control parameters may include weighting parameters for different traffic state indicators during the global decision-making process, flow allocation parameters, and timing adjustment parameters during the individual vehicle scheduling process.

[0080] The target merging flow of the entrance ramps, the target diversion flow of the exit ramps, and the mainline lane control parameters are redefined using the revised control parameters. The passage order, passage time, and target speed of each vehicle, as well as the lane-changing trajectories of vehicles to be merged and diverted, are also redefined. During this redefined process, vehicle operation data currently within the preset cooperative control area are used, ensuring that the updated results reflect the latest vehicle operating status rather than continuing to use vehicle positions from the previous control cycle.

[0081] Based on the redefined global control results and vehicle scheduling results, updated traffic control commands are issued to the relevant intelligent connected vehicles. Subsequently, command execution feedback data is acquired and threshold comparisons are performed, thus forming a closed-loop process of control command execution – effect acquisition – threshold judgment – ​​parameter correction – control result update – new command execution.

[0082] In this implementation, by evaluating the actual control effect from multiple dimensions such as traffic efficiency, vehicle conflict, queuing status and traffic time, and re-determining the global control result and individual vehicle scheduling result when any indicator reaches an abnormal condition, the control strategy can continuously adapt to traffic flow fluctuations and changes in traffic conditions. This closed-loop process reduces the reliance on fixed signal timing and manual intervention, and improves the adaptability and continuous operation capability of coordinated control.

[0083] This invention also proposes a future intelligent control system for highway entrance and exit coordination, including a data acquisition module, a traffic condition assessment module, a global decision-making module, a vehicle scheduling module, a control execution module, and a feedback optimization module. The above modules establish data connections according to traffic data flow, control result flow, and feedback data flow.

[0084] The data acquisition module connects to roadside sensing devices and vehicle-to-infrastructure (V2I) communication devices within a pre-defined collaborative control area. Roadside sensing devices collect lane-level traffic data and transmit it to the data acquisition module. V2I communication devices receive vehicle operation data and traffic demand from vehicle terminals and transmit them to the data acquisition module. The data acquisition module organizes the data according to road segment, lane, and vehicle, and outputs lane-level traffic data and vehicle operation data to the traffic condition assessment module. Simultaneously, it outputs vehicle operation data and traffic demand to the vehicle dispatching module.

[0085] The traffic condition assessment module receives lane-level traffic data and vehicle operation data output from the data acquisition module, and determines traffic condition indicators, including mainline traffic saturation, entrance merging pressure, exit diversion load, and weaving zone conflict risk. The traffic condition assessment module outputs the traffic condition indicators to the global decision-making module, thereby enabling the global decision-making module to obtain a unified evaluation result reflecting the operating status of the mainline, entrances, exits, and weaving zones.

[0086] The global decision-making module determines the target inbound traffic flow of the entrance ramps, the target diversion traffic flow of the exit ramps, and the mainline lane control parameters based on traffic state indicators. The mainline lane control parameters include mainline lane function configuration and mainline lane speed constraints. These parameters collectively constitute the global control result. The global decision-making module outputs the global control result to both the vehicle scheduling module and the control execution module. This enables the vehicle scheduling module to allocate vehicles under overall traffic flow and lane organization constraints, and allows the control execution module to convert the overall control requirements into executable vehicle instructions.

[0087] The vehicle dispatching module identifies vehicles to merge, vehicles to be diverted, and vehicles traveling straight on the main line based on vehicle operation data and traffic demand output by the data acquisition module. Using the global control results output by the global decision-making module as constraints, the vehicle dispatching module determines the passage order, passage time period, and target speed for each vehicle, and generates lane-changing trajectories for vehicles to merge and vehicles to be diverted. These results constitute the vehicle dispatching outcome and are output to the control execution module.

[0088] The control execution module simultaneously receives global control results and individual vehicle scheduling results, generates corresponding traffic control instructions according to vehicle identity, and issues them to intelligent connected vehicles via the vehicle-to-infrastructure communication link. Instructions issued to vehicles traveling straight on the mainline include the corresponding lane passage requirements based on the mainline lane function configuration and the target driving speed determined based on the mainline lane speed constraints. Instructions issued to vehicles needing to merge or diverge may also include corresponding lane-changing trajectories, enabling different types of vehicles to execute their respective individual vehicle scheduling results according to the global traffic flow organization requirements.

[0089] The feedback optimization module acquires the traffic flow efficiency, number of vehicle conflicts, queue length, and vehicle passage time after control measures are implemented, and compares the above instruction execution feedback data with the corresponding preset thresholds. When any feedback indicator reaches the trigger condition, the feedback optimization module corrects the control parameters used to determine the global control result and the single-vehicle scheduling result, and sends recalculation trigger information to the global decision module and the single-vehicle scheduling module, respectively.

[0090] The global decision-making module responds to the trigger information and regenerates the global control result. The vehicle scheduling module then regenerates the vehicle scheduling result based on the regenerated global control result and the current vehicle data. The feedback optimization module further triggers the control execution module, causing it to issue updated traffic control commands based on the new global control result and vehicle scheduling result. These modules form a closed-loop data connection relationship, from data acquisition, status assessment, global decision-making, vehicle scheduling, command execution to feedback optimization.

[0091] In this implementation, by using the traffic condition assessment results simultaneously for global flow and lane decisions, and using the global decision results as constraints for the vehicle scheduling module, it can be ensured that the global decision and vehicle control in the modular system remain consistent. By using the feedback optimization module to trigger the global decision module, vehicle scheduling module, and control execution module in a coordinated manner, continuous updates of the entrance and exit coordination control results can be achieved under modular deployment conditions.

[0092] The present invention also proposes a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements any of the intelligent control methods for coordinated access to and from future highways disclosed in the above embodiments.

[0093] By transforming data acquisition, status assessment, global decision-making, vehicle scheduling, instruction generation, and optional data processing, risk prediction, formation control, diversion control, and feedback optimization processes into program instructions that can be read and executed by a processor, the corresponding coordination control methods can be repeatedly deployed and executed on different computing devices, and the data transfer relationships during program operation can be kept consistent with the method steps defined in the above implementation.

[0094] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A future intelligent control method for coordinated access to highway entrances and exits, characterized in that, include: Acquire lane-level traffic data, vehicle operation data, and vehicle traffic demand for the main line, entrance ramps, exit ramps, and entrance / exit weaving areas of the expressway within the preset collaborative control area; Based on the vehicle operation data and the vehicle traffic demand, identify vehicles to merge, vehicles to be diverted, and mainline straight-through vehicles within the preset collaborative control area, wherein the vehicles to merge, vehicles to be diverted, and mainline straight-through vehicles are all intelligent connected vehicles. Based on the lane-level traffic data and the vehicle operation data, traffic status indicators including mainline traffic saturation, inlet merging pressure, outlet diversion load and weaving zone conflict risk are determined. Based on the traffic state indicators, the target merging flow of the entrance ramps and the target diversion flow of the exit ramps are jointly determined, along with the mainline lane control parameters, to form a global control result. The mainline lane control parameters include the mainline lane function configuration and the mainline lane speed constraint. Using the global control result as a single-vehicle scheduling constraint, based on the vehicle operation data of the vehicles to be merged, the vehicles to be diverted, and the mainline straight-through vehicles, the passage order and passage time of each vehicle entering the entrance and exit weaving area are determined, the target driving speed corresponding to each vehicle is generated, and the corresponding lane-changing trajectory is generated for the vehicles to be merged and the vehicles to be diverted, forming a single-vehicle scheduling result. Among them, the passage time of vehicles to be merged and vehicles to be diverted that have conflicting relationships do not overlap. Based on the global control results and the single-vehicle scheduling results, traffic control instructions containing corresponding travel time periods and target speeds are issued to the vehicles to be merged, the vehicles to be diverted, and the vehicles traveling straight on the main line. The traffic control instructions issued to the vehicles to be merged and the vehicles to be diverted also include corresponding lane-changing trajectories, and the traffic control instructions issued to the vehicles traveling straight on the main line also include corresponding lane passage requirements based on the main line lane function configuration. The traffic control commands regulate the merging of vehicles at the entrance ramps according to the target merging flow, regulate the diversion of vehicles at the exit ramps according to the target diversion flow, and allow vehicles with conflicting relationships to pass through the entrance / exit weaving area according to the passage order and passage time. In addition, the commands regulate the lane passage and driving speed of the mainline straight-through vehicles according to the mainline lane function configuration and the mainline lane speed constraints.

2. The intelligent control method for coordinated access to and from future highways as described in claim 1, characterized in that, The lane-level traffic data is acquired through roadside sensing devices deployed on the main line of the expressway, entrance ramps, exit ramps, and intersections of entrances and exits. The vehicle operation data and vehicle traffic demand are obtained by receiving data sent by vehicle terminals through vehicle-road communication equipment. The lane-level traffic data includes traffic density, average speed of each lane, lane occupancy, queue length, and traffic arrival rate; the vehicle operation data includes real-time vehicle location, speed, acceleration, trajectory, vehicle type, and estimated travel time. Before determining traffic status indicators based on lane-level traffic data and vehicle operation data, the process further includes acquiring road environment data and performing data fusion, cleaning, and spatiotemporal calibration on the lane-level traffic data, vehicle operation data, and road environment data to obtain processed lane-level traffic data and vehicle operation data. The traffic status indicators are determined based on the processed lane-level traffic data and vehicle operation data, wherein the road environment data includes road surface condition, weather conditions, light intensity, and temporary road control status.

3. The intelligent control method for coordinated access to and from future highways according to claim 2, characterized in that, The traffic status indicators also include the risk value of queue overflow at entrances and exits; The determination of traffic state indicators includes predicting the trend of entrance and exit traffic flow changes within the next 5 to 30 seconds based on processed lane-level traffic data and vehicle operation data continuously acquired within a time window, and determining the conflict risk of the weaving area and the queuing overflow risk value of the entrance and exit based on the trend of entrance and exit traffic flow changes and the processed lane-level traffic data and vehicle operation data.

4. The intelligent control method for coordinated access to and from future highways as described in claim 3, characterized in that, The joint determination of the target merging flow of the entrance ramp and the target diversion flow of the exit ramp, as well as the mainline lane control parameters, includes: When the mainline traffic saturation is lower than the preset low saturation threshold, increase the target inflow of the entrance ramps; When the main line traffic saturation reaches the preset congestion threshold, the target inflow of the entrance ramps will be reduced. When the exit diversion load or the risk value of queuing overflow at the entrance or exit reaches a preset threshold, the target diversion flow of the exit ramp is reduced. Based on the traffic load of the mainline, entrance ramps, and exit ramps, the mainline lane function configuration and mainline lane speed constraints are determined. The mainline lane control parameters include the mainline lane function configuration and the mainline lane speed constraints, and the mainline lane function configuration and the mainline lane speed constraints serve as single-vehicle scheduling constraints for the mainline through traffic.

5. The intelligent control method for coordinated access to and from future highways according to claim 2, characterized in that, The determination of the passage order and passage time of each vehicle entering the entrance / exit weaving area includes: Based on the expected passage time and travel trajectory of each vehicle, identify the vehicles with conflicting relationships that enter the same conflict area within the entrance / exit weaving zone; Determine the passage order for the vehicles with conflicting relationships, and allocate non-overlapping dedicated passage time slots as the passage time periods according to the passage order; The target driving speed is determined based on the dedicated passage time slot corresponding to each vehicle, and the lane-changing trajectory of the vehicle to be merged and the vehicle to be diverted is determined.

6. The intelligent control method for coordinated access to and from future highways according to claim 1, characterized in that, The regulation of vehicle merging at the entrance ramp according to the target merging flow includes: Vehicles waiting to merge at the entrance ramp are divided into vehicle platoons; Based on the target merging flow and the gap in the main traffic flow, the formation size, formation speed, and merging timing of the vehicle formation are determined; When the mainline traffic saturation reaches a preset congestion threshold, flow control is implemented on the entrance ramps, and the entry of vehicles is restricted from entering the mainline by adjusting the platoon size and merging timing of the vehicle platoons.

7. The intelligent control method for coordinated access to and from future highways according to claim 1, characterized in that, The method of adjusting the vehicle diversion at the exit ramp according to the target diversion flow includes: The diversion demand of the vehicles to be diverted is predicted based on the vehicle traffic demand. In the pre-designated lane-changing area upstream of the exit ramp, a lane-changing guidance instruction is issued to the vehicles to be diverted, guiding them to change lanes to the outer diversion lane. The vehicle diversion rate is determined based on the target diversion flow and the real-time capacity of the exit ramps to reduce the risk of queues at the exit ramps flowing back onto the main line.

8. The intelligent control method for coordinated access to and from future highways according to claim 1, characterized in that, After issuing the access control command, the following is also included: The efficiency of traffic flow after control, the number of vehicle conflicts, the queue length and the time it takes for vehicles to pass through are obtained as feedback data for instruction execution. The instruction execution feedback data is compared with preset traffic efficiency threshold, preset vehicle conflict number threshold, preset queue length threshold, and preset vehicle passage time threshold. When the traffic flow efficiency is lower than the preset traffic flow efficiency threshold, the number of vehicle conflicts reaches the preset number of vehicle conflicts threshold, the queue length reaches the preset queue length threshold, or the vehicle passage time reaches the preset vehicle passage time threshold, the control parameters used to determine the global control result and the single-vehicle scheduling result are corrected, and the target merging flow of the entrance ramp, the target diversion flow of the exit ramp, the mainline lane control parameters, the passage order of each vehicle, the passage time period, the target driving speed, and the lane-changing trajectory of the vehicles to be merged and the vehicles to be diverted are re-determined. Based on the re-determined global control result and single-vehicle scheduling result, an updated traffic control command is issued.

9. A future intelligent control system for coordinating highway entrances and exits, characterized in that, It includes a data acquisition module, a traffic condition assessment module, a global decision-making module, a vehicle dispatching module, a control execution module, and a feedback optimization module, among which: The data acquisition module is used to acquire lane-level traffic data, vehicle operation data and vehicle traffic demand within a preset cooperative control area through roadside sensing devices and vehicle-road communication devices. The traffic condition assessment module is used to determine traffic condition indicators based on the lane-level traffic data and vehicle operation data output by the data acquisition module. The global decision-making module is used to determine the target merging flow of the entrance ramp, the target diversion flow of the exit ramp, and the mainline lane control parameters based on the traffic state indicators output by the traffic state assessment module, and to form a global control result. The mainline lane control parameters include the mainline lane function configuration and the mainline lane speed constraint. The single-vehicle scheduling module is used to identify vehicles to merge, vehicles to be diverted, and mainline straight-through vehicles based on the vehicle operation data and vehicle traffic demand output by the data acquisition module, and to determine the passage order, passage time period, and target driving speed of each vehicle using the global control result output by the global decision module as constraints, as well as to generate the lane-changing trajectory of vehicles to merge and vehicles to be diverted, thus forming a single-vehicle scheduling result. The control execution module is used to issue passage control instructions to intelligent connected vehicles based on the global control results and the single vehicle scheduling results. The passage control instructions issued to vehicles traveling straight on the main line include the corresponding lane passage requirements based on the main line lane function configuration and the target driving speed based on the main line lane speed constraints. The feedback optimization module is used to acquire traffic flow efficiency, vehicle conflict frequency, queue length, and vehicle passage time as instruction execution feedback data, and compare the instruction execution feedback data with preset traffic flow efficiency thresholds, preset vehicle conflict frequency thresholds, preset queue length thresholds, and preset vehicle passage time thresholds. When the traffic flow efficiency is lower than the preset traffic flow efficiency threshold, the vehicle conflict frequency reaches the preset vehicle conflict frequency threshold, the queue length reaches the preset queue length threshold, or the vehicle passage time reaches the preset vehicle passage time threshold, the control parameters used to determine the global control result and the single-vehicle scheduling result are corrected, and the global decision module and the single-vehicle scheduling module are triggered to regenerate the global control result and the single-vehicle scheduling result, and the control execution module is triggered to issue updated traffic control instructions based on the regenerated global control result and single-vehicle scheduling result.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the future highway entrance and exit coordination intelligent control method as described in any one of claims 1 to 8.