A CAV lane-changing method and system based on macro-micro traffic flow information

Abstract

The application belongs to the field of intelligent transportation systems, and relates to a CAV lane-changing method and system based on macro and micro traffic flow information, comprising the following steps: acquiring state information of a vehicle and road section information on a CAV travel direction in real time, uploading the state information and the road section information to a cloud server; the cloud server pre-processes the uploaded data, calculates a comprehensive macro benefit index according to the pre-processed data; a micro benefit index of the intelligent networked vehicle is calculated according to the vehicle state information and the road section information; a lane-changing decision is generated according to the comprehensive macro and micro benefit indexes; the lane-changing decision is transmitted to a vehicle-mounted unit of the intelligent networked vehicle; the vehicle-mounted unit verifies the lane-changing decision, if the verification fails, no lane-changing operation is performed; if the verification succeeds, the intelligent networked vehicle is controlled to change lanes; the lane-changing decision method and system based on the combination of macro and micro traffic flow information can effectively improve the global optimization capability and local safety of the lane-changing decision.

Description

Technical Field

[0001] This invention belongs to the field of intelligent transportation systems, specifically relating to a CAV lane-changing method and system based on macro and micro traffic flow information. Background Technology

[0002] With the rapid development of Intelligent Transportation Systems (ITS), autonomous driving technology has demonstrated enormous potential in improving traffic efficiency, reducing traffic accidents, and decreasing energy consumption. As a crucial component of autonomous driving technology, lane-changing decisions for connected and autonomous vehicles (CAVs) are a key issue in ensuring vehicle efficiency and safety. Lane-changing behavior involves a comprehensive analysis and decision-making process considering the vehicle's own state, traffic flow conditions, and the dynamics of surrounding vehicles; therefore, its complexity requires vehicles to accurately acquire and process multi-dimensional traffic information.

[0003] Traditional lane-changing decision-making methods are typically based on fixed rules or simple dynamic models, such as making lane-changing decisions solely based on micro-data like the distance and speed difference between a vehicle and those in front and behind. However, these methods often neglect higher-level traffic flow information, such as macro-level indicators like the overall density, flow rate, and comfort of the target lane, resulting in a lack of global and optimal lane-changing decisions. Furthermore, relying solely on micro-information for lane-changing decisions may lead to localized optimization behaviors, such as frequent or unnecessary lane changes, which can negatively impact overall traffic flow.

[0004] In recent years, lane-changing optimization methods based on macroscopic traffic flow characteristics have begun to attract researchers' attention. These methods, through the analysis of macroscopic indicators such as lane density and average vehicle speed, initially screen out target lanes that may bring higher traffic efficiency or better comfort. However, macroscopic optimization cannot accurately assess potential safety hazards and local dynamic constraints that may exist during actual lane-changing processes, such as distances to vehicles in front and behind, and relative speed differences. Therefore, decision-making methods based solely on macroscopic information also have certain limitations. Summary of the Invention

[0005] To address the problems existing in the prior art, this invention proposes a CAV lane-changing method based on macro and micro traffic flow information. This method includes: real-time acquisition of the status information of intelligent connected vehicles and surrounding non-connected vehicles; acquisition of road segment information in the direction of travel of the intelligent connected vehicle, and uploading the acquired vehicle status information and road segment information to a cloud server; at the macro level, the cloud server preprocesses the uploaded data and calculates a comprehensive macro-benefit index based on the preprocessed data; at the micro level, the lane-changing index of the intelligent connected vehicle is calculated based on the vehicle status information and road segment information; a lane-changing decision is generated based on the comprehensive macro-benefit index and the micro-benefit index; the lane-changing decision is transmitted to the on-board unit of the intelligent connected vehicle; the on-board unit verifies the lane-changing decision; if the verification fails, no lane-changing operation is performed; if the verification succeeds, the intelligent connected vehicle is controlled to perform a lane change according to the lane-changing decision, and the lane-changing operation is uploaded to the cloud server.

[0006] A CAV lane-changing system based on macro and micro traffic flow information, comprising: a real-time driving data acquisition unit, a macro benefit analysis module, a micro benefit analysis module, a CAV decision recommendation distribution module, and a cooperative driving decision execution unit;

[0007] The real-time driving data acquisition unit is used to collect status information of intelligent connected vehicles and surrounding non-connected vehicles, as well as road segment information in the direction of travel of intelligent connected vehicles.

[0008] The macro revenue analysis module calculates the macro revenue of each lane under the current conditions based on the collected data.

[0009] The micro-profit analysis module is used to calculate the micro-profit under the current conditions based on the collected data.

[0010] The CAV decision recommendation distribution module constructs lane-changing strategies based on the macro and micro benefits of each lane;

[0011] The cooperative driving decision-making and execution unit controls the vehicle to change lanes according to the lane-changing strategy.

[0012] The beneficial effects of this invention are:

[0013] This invention presents a lane-changing decision-making method and system based on a combination of macroscopic and microscopic traffic flow information, which effectively improves the global optimization capability and local safety of lane-changing decisions. By analyzing macroscopic traffic characteristics such as lane density, average lane speed, and comfort, it achieves global optimization selection of the target lane. Simultaneously, by incorporating microscopic information, such as the space between the target lane and vehicles in front and behind, dynamic weights, and the necessity of lane changing, it ensures the safety and feasibility of lane changes. This invention can quickly respond to real-time traffic environments and intelligently generate lane-changing decisions, not only improving the traffic efficiency and comfort of autonomous vehicles but also significantly enhancing the safety and flexibility of decision-making in complex traffic scenarios. Attached Figure Description

[0014] Figure 1 This is a flowchart of the CAV free lane-changing decision-making method of the present invention;

[0015] Figure 2 This is a schematic diagram of the CAV free lane-changing decision system of the present invention;

[0016] Figure 3 This is a simulation result diagram of the CAV free lane-changing decision system of the present invention;

[0017] Figure 4 This is a graph showing the changes in lane-changing process indicators over time for the CAV free lane-changing decision system of the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] This invention, based on a lane-changing decision-making method that combines macroscopic and microscopic traffic information, represents a significant current research direction. Macroscopic information is used to screen optimal lane candidates, providing global optimization guidance, while microscopic information is used to further evaluate the feasibility and safety of lane changes, ensuring that actual operations meet dynamic constraints. This combined macroscopic and microscopic approach enhances the global optimization capability of lane-changing decisions while effectively mitigating local safety issues.

[0020] This invention proposes a free lane-changing decision generation method for Car Access Vehicles (CAVs) based on macroscopic and microscopic traffic flow information. The method includes: real-time collection of state information (including speed, latitude and longitude, acceleration, heading angle, etc.) of the CAV and surrounding vehicles, as well as road traffic environment data (lane density, average lane speed, etc.); calculating the macroscopic and microscopic benefits of the target lane respectively; and constructing a hierarchical lane-changing decision framework. Finally, the final decision recommendation is sent to the CAV's OBU unit via a 5G communication cloud, integrating macroscopic and microscopic benefits. Upon receiving the decision recommendation, the CAV executes the operation, improving operational efficiency and safety.

[0021] A CAV lane-changing method based on macro and micro traffic flow information, such as Figure 1As shown, the method includes: acquiring real-time status information of the intelligent connected vehicle and surrounding non-connected vehicles; acquiring road segment information in the direction of travel of the intelligent connected vehicle, and uploading the acquired vehicle status information and road segment information to a cloud server; at the macro level, the cloud server preprocesses the uploaded data and calculates a comprehensive macro-benefit index based on the preprocessed data; at the micro level, the cloud server calculates the lane-changing index of the intelligent connected vehicle based on the vehicle status information and road segment information; generates a lane-changing decision based on the comprehensive macro-benefit index and the lane-changing index; transmits the lane-changing decision to the on-board unit of the intelligent connected vehicle; the on-board unit verifies the lane-changing decision, and if the verification fails, no lane-changing operation is performed; if the verification is successful, the cloud server controls the intelligent connected vehicle to perform a lane-changing operation based on the lane-changing decision and uploads the lane-changing operation to the cloud server.

[0022] In this embodiment, a CAV (Carrier Assisted Vehicle) free lane-changing decision-making method based on macro and micro traffic flow information includes: constructing a real-world urban two-lane scenario as a verification platform using the SUMO and CARLA joint simulation platform. The data input source is a simulated roadside RSU (Roadside Unit) in a format suitable for real-time verification. Specific steps include:

[0023] S1: Firstly, at the micro-information level, CAV (Connected Vehicle) uploads its own and surrounding vehicle status information in real time through onboard sensors. At the same time, the Roadside Unit (RSU) senses and acquires the status information of HDV (Unconnected Vehicle) in the range in real time. The data includes vehicle speed, latitude and longitude, heading angle, acceleration, vehicle ID, and other information. At the macro-information level, the road segment information in the direction of CAV travel is acquired through the RSU and advanced mobile communication. Finally, the vehicle data and road segment information are uploaded to the cloud server in real time.

[0024] The simulation data includes vehicle speed, acceleration, latitude and longitude, heading angle, timestamp, and ID. At the macro level, traffic status information related to the road segment is further obtained, including the number of lanes, traffic flow distribution in each lane, average lane speed, and traffic density. The above information is then synchronized and fused in time to ensure real-time consistency between micro and macro information.

[0025] In this embodiment, the information preprocessing includes the following steps: denoising the vehicle status information to remove outliers and invalid data; synchronizing multi-source data from onboard sensors and roadside units (RSUs) according to timestamps to ensure time sequence consistency; fusing real-time latitude, longitude, speed, acceleration, and other information of the vehicle in multiple dimensions to generate a unified data stream; and formatting and standardizing the fused data to convert it into a unified data format for easy subsequent calculations.

[0026] S2: After receiving the data, the cloud performs preprocessing on the acquired vehicle information data, including data filtering and time synchronization fusion of multi-sensor data. First, in the macro-indicator section, based on the acquired lane information and preprocessed vehicle status data, it calculates the evaluation of each lane in the CAV's direction of travel. This is a normalized quantitative evaluation decoupled by lane and index (efficiency-comfort) within the road segment, in order to support micro-level decision-making recommendations.

[0027] In this embodiment, calculating the comprehensive macro-benefit index includes: calculating the driving comfort benefit of the corresponding vehicle in the lane based on the number of vehicles in the lane, the acceleration of the corresponding vehicle in the lane, and the average acceleration of all vehicles in the lane; calculating the density benefit based on the average speed of vehicles in the lane; obtaining the vehicle speed benefit; assigning weights to the driving comfort benefit, density benefit, and speed benefit, and calculating the comprehensive macro-benefit index based on the assigned weights.

[0028] Specifically, the lane-changing recommendation decision generated based on a comprehensive assessment of macro-information includes: firstly, calculating the comfort index benefit, considering that the acceleration and deceleration volatility of each lane can be represented by the standard deviation of the acceleration of all vehicles in the lane. If acceleration and deceleration occur frequently in a lane, it indicates that the traffic flow in that lane is unstable and the comfort level is poor.

[0029]

[0030] Among them, A std,i Let a be the standard deviation of acceleration on lane i. j,i Let μ be the acceleration of the j-th vehicle in lane i. a,i Let N be the average acceleration on lane i. i Let be the number of vehicles in lane i. Next, the macro-efficiency of each lane is calculated, mainly considering two key indicators: traffic flow density and average lane speed. These directly reflect the lane's capacity and efficiency. Traffic flow density is expressed as:

[0031]

[0032] Where, N i L represents the total number of vehicles in lane i. i Let be the length of the area covered by the data collected on the lower side of lane i in this road segment. The average speed of the lane is expressed as:

[0033]

[0034] Among them, v j,i Let be the speed of the j-th vehicle in lane i. Higher speed indicates higher lane efficiency. Since the dimensions of each indicator are different, normalization is required to enable comprehensive calculation of macro-level benefit indicators. A linear normalization method is used:

[0035]

[0036] Among them, X i X is the indicator for the current lane. min X max These are the minimum and maximum values ​​of this indicator across all lanes. Calculate the overall macro-revenue indicator R for each lane. macro,i :

[0037]

[0038]

[0039] in, The normalized density, w is the normalized lane average speed s For comfort index S i The weight, w d For traffic flow density weights, w v S is the lane average speed weight. i For comfort benefits, For density gains, For vehicle speed gains. For negative indicators (D) i A std,i (The smaller the value, the better), after normalization, take This can be transformed into a positive indicator. Overall return R macro,i The higher the value, the higher the macroeconomic benefit of that lane at the current moment.

[0040] In this embodiment, the weight value w s Set to 0.2, w d Set to 0.4, w v Set it to 0.4.

[0041] S3: In the micro-indicators section, preprocessed vehicle status and road segment information are used to evaluate the micro-benefits of CAV. Calculations are performed through two main dimensions: spatial benefit and lane-changing necessity. Spatial benefit assesses whether the target lane has sufficient safe lane-changing space to meet safety requirements. Lane-changing necessity is determined by the distance to the vehicle in front and vehicle speed to determine whether the minimum threshold requirement is met. Finally, the cloud combines macro and micro indicators for comprehensive analysis. If both the micro and macro benefits of the target lane are significantly higher than those of the current lane, a lane-changing decision is generated, preparing for the next step of sending a recommendation decision.

[0042] In this embodiment, a micro-lane-changing recommendation decision is generated based on a comprehensive evaluation of the collected micro-vehicle state information. By using real-time CAV and surrounding vehicle state information (including speed, latitude and longitude, heading angle, lane number, acceleration, etc.), combined with the traffic environment, the necessity and spatial benefit indicators of the lane-changing behavior are dynamically calculated. First, the spatial benefit of the target lane is calculated. Based on lane information, it is determined whether there are vehicles in front and behind in the target lane. If so, their positions are obtained in real-time using latitude and longitude. The actual clearance is calculated using the Haversine formula combined with the azimuth angle.

[0043]

[0044] θ front =arctan2(sin(λ) front -λ cav )·cos(φ front ),cos(φ cav )·sin(φ front )-sin(φ cav )·cos(φ front )·cos(λ front -λ cav ))

[0045] θ rear =arctan2(sin(λ) rear -λ host )·cos(φ rear ),cos(φ host )·sin(φ rear )-sin(φ host )·cos(φ rear )·cos(λ rear -λ host ))

[0046] Where, d front θ is the distance to the vehicle in front. front d represents the relative azimuth angle of the vehicle in front. rear Let θ be the distance to the following vehicle. rear The relative azimuth angle of the following vehicle is r, where r is the Earth's radius (6371 km), and φ is φ. cav φ front and φ rear These are the latitudes of CAV, the front vehicle, and the rear vehicle, respectively, and λ. cav , λ front and λ rear These are the longitudes of the CAV, the preceding vehicle, and the following vehicle, respectively. The relative positions of the preceding and following vehicles are determined as follows:

[0047] Δθ front =(θ front -θ host +360)mod360

[0048] Δθ rear =(θ rear -θ host +360)mod360

[0049] The vehicle ahead in the target lane must satisfy 0 ≤ Δθ front ≤90 or 270<Δθ front A vehicle with a radius of less than 360 degrees can be used as a reference vehicle in front of a CAV, and the vehicle behind in the target lane must meet the condition that 90 degrees < Δθ. rear Vehicles with a radius of ≤270 can be used as rear reference vehicles for a CAV. Then, the dynamic weights of the preceding and following vehicles are calculated separately.

[0050]

[0051] Among them, F front and F rear These represent the weights of the impact of the preceding and following vehicles on the CAV lane change, v front and v rear These represent the speeds of the front and rear vehicles, respectively. In this embodiment, T... human The driver's reaction time is set to 1.5 seconds. maxbrake The maximum braking speed of the vehicle is set to 8 m / s. 2 Therefore, the formula for judging the sufficiency of spatial benefits can be expressed as:

[0052]

[0053] Among them, gap sufficient A value of 1 indicates sufficient space gain, while a value of 0 indicates insufficient gain.

[0054] In this embodiment, Δ safe A safety constant of 5m is set. This method comprehensively describes the dynamic relationship between the CAV and vehicles in front and behind in the target lane, determining whether the space gain from lane changing is sufficient. Next, the necessity of CAV lane changing is calculated, reflecting whether the CAV needs to perform a lane-changing operation, and the safe following distance s. safe Represented as:

[0055] s safe =v pv ·t thw

[0056] Among them, v pv t represents the speed of the vehicle in front in the current lane. thw The time-distance threshold is set to 1.5 seconds. The distance between the CAV and the vehicle in front is compared with the safe following distance (s). safe The necessity of lane changing related to vehicle distance is calculated as follows:

[0057]

[0058] Among them, s pv Let s be the distance between the CAV and the vehicle in front in the current lane. pv <s safe At that time, the necessity is relatively high. The necessity of lane changing related to differences in calculation speed:

[0059] Δv desired =v desired -v pv

[0060] Where, Δv desired This represents the speed of the vehicle in front and the expected speed of CAV, v. desired The difference between them. In this embodiment, v desired The necessity assessment of the speed difference is as follows, assuming a speed of 50 km / h:

[0061]

[0062] According to Δv desired Size determines whether the preceding vehicle limits the CAV's driving efficiency; when v desired >v pv At that time, the necessity of speed difference is relatively high. Finally, based on the current traffic conditions, weights are assigned to the necessity of vehicle distance and the necessity of speed difference:

[0063] w distance +w speed =1

[0064] f necessity =w distance ·distance necessity +w speed ·speed necessity

[0065] When f necessity When the threshold is greater than Threshold, a lane-changing decision is initiated. Threshold is the lane-changing necessity threshold. In this embodiment, it is set to 0.5.

[0066] S4: When comprehensively evaluating macro and micro indicators, the cloud prioritizes the assessment of macro-level traffic conditions, and then combines micro-level benefits and feasibility to make a final decision. When both macro and micro indicators meet the lane-changing conditions, the cloud generates a lane-changing recommendation decision and sends the lane-changing recommendation to the CAV's onboard unit (OBU) in real time via 5G communication to guide the CAV to complete the optimal lane-changing operation.

[0067] S5: After receiving the decision suggestion, the CAV combines its own real-time status (speed, acceleration, lane position) and surrounding environmental information (such as the speed and distance of surrounding vehicles) to verify the rationality and feasibility of the cloud-based recommendation. Once verified, the CAV executes a lane-changing operation based on the cloud-based decision suggestion to maximize driving efficiency at both the macro and micro levels. The actual operation data is then transmitted back to the cloud to help optimize subsequent lane-changing recommendation algorithms, improving the accuracy and response speed of lane-changing strategies.

[0068] In this embodiment, the CAV analyzes the received decision and assesses the vehicle's current driving status, including speed, position, and surrounding traffic environment. Based on the received lane change decision recommendation and the vehicle's own status, after confirming that the decision is correct and the risk is controllable, it executes the corresponding driving behavior and performs a lane change operation.

[0069] In this embodiment, a corresponding feedback mechanism is established. The CAV (OBU) sends feedback to the cloud based on the received recommendation decision information and actual driving conditions, which is used for continuous algorithm optimization.

[0070] A CAV lane-changing system based on macro and micro traffic flow information, whose core modules are as follows: Figure 2 As shown, the system includes: a real-time driving data acquisition unit, a macro-level benefit analysis module, a micro-level benefit analysis module, a CAV decision recommendation distribution module, and a cooperative driving decision execution unit;

[0071] The real-time driving data acquisition unit is used to collect status information of intelligent connected vehicles and surrounding non-connected vehicles, as well as road segment information in the direction of travel of intelligent connected vehicles.

[0072] The macro revenue analysis module calculates the macro revenue of each lane under the current conditions based on the collected data.

[0073] The micro-profit analysis module is used to calculate the micro-profit under the current conditions based on the collected data.

[0074] The CAV decision recommendation distribution module constructs lane-changing strategies based on the macro and micro benefits of each lane;

[0075] The cooperative driving decision-making and execution unit controls the vehicle to change lanes according to the lane-changing strategy.

[0076] In this embodiment, the system includes at least: a roadside sensing device and a vehicle-mounted sensor on the CAV for collecting vehicle status data and macroscopic information; a 5G communication network for realizing high-speed data transmission between the vehicle, road, and cloud; a cloud server equipped with high-performance equipment for processing the transmitted data and meeting the requirements for real-time calculation of macro and micro lane-changing benefits; and an OBU installed on the CAV for receiving cloud-based recommendation decisions and responding accordingly.

[0077] In this embodiment, lane change decisions are divided into three types: left turn (0), keep going straight (1), and right turn (2).

[0078] In this embodiment, Figure 3 The diagram shows the simulation results of the CAV free lane change decision system of the present invention. The cloud calculates the macro and micro benefits, analyzes that the comprehensive index of the right lane is higher than that of the current lane at the current moment, and meets the feasibility. The cloud then sends the decision to the CAV. The CAV verifies the rationality of the decision at the current moment and executes lane change decision 2 (right lane change).

[0079] In this embodiment, Figure 4 The graph shows the changes in macro and micro indicators over time during the lane-changing process. The CAV travels in lane 1, and the overall macro benefit of lane 2 gradually becomes greater than that of lane 1 over time. At this point, the micro benefit and feasibility are verified. While the spatial condition is met (1), the lane-changing necessity also exceeds the threshold of 0.5. The macro and micro benefits are simultaneously satisfied, triggering a lane-changing recommendation decision. It can be seen that after changing lanes, the lane-changing necessity sometimes exceeds the threshold, but at this time, the macro indicator of lane 2 is always greater than that of lane 1, so the lane-changing recommendation decision will not be triggered.

[0080] This embodiment uses the simulation parameters shown in Tables 1 and 2:

[0081] Table 1. Macroscopic parameter weight settings

[0082]

[0083] Table 2. Microparameter weight values ​​settings

[0084]

[0085] In simulation experiments, this invention simulates the real-world information interaction process between the vehicle, road, and cloud, achieving a total calculation and delivery time of less than 60ms for each lane-change decision (including data upload, macro and micro index calculation, comprehensive analysis, and decision generation and delivery), and sending decision information to the CAV every 0.1 seconds. This demonstrates the system's efficiency and real-time performance, enabling it to meet the real-time lane-changing needs of autonomous vehicles in complex and dynamic traffic environments.

[0086] The system implementation method of the present invention is the same as the method implementation method.

[0087] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, which may include ROM, RAM, disk, or optical disk, etc.

[0088] The above-described embodiments further illustrate the purpose, technical solution, and advantages of the present invention. It should be understood that the above-described embodiments are merely preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made to the present invention within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A CAV lane-changing method based on macro and micro traffic flow information, characterized in that, include: Real-time acquisition of status information of intelligent connected vehicles and surrounding non-connected vehicles; Obtain road segment information in the direction of travel of intelligent connected vehicles, and upload the obtained vehicle status information and road segment information to the cloud server; At the macro level, the cloud server preprocesses the uploaded data and calculates the comprehensive macro benefit index based on the preprocessed data. At the micro level, it calculates the micro benefit index of the intelligent connected vehicle based on vehicle status information and road segment information. It generates a lane-changing decision based on the comprehensive macro benefit index and the micro benefit index. The lane-changing decision is transmitted to the on-board unit of the intelligent connected vehicle. The on-board unit verifies the lane-changing decision. If the verification fails, the lane-changing operation is not performed. If the verification is successful, the intelligent connected vehicle will be controlled to change lanes according to the lane change decision, and the lane change operation will be uploaded to the cloud server. The calculation of the comprehensive macro-benefit index includes: calculating the driving comfort benefit of the corresponding vehicle in the lane based on the number of vehicles in the lane, the acceleration of the corresponding vehicle in the lane, and the average acceleration of all vehicles in the lane; calculating the density benefit based on the average speed of vehicles in the lane; obtaining the vehicle speed benefit; assigning weights to the driving comfort benefit, density benefit, and speed benefit, and calculating the comprehensive macro-benefit index based on the assigned weights. The micro-level benefit indicators for intelligent connected vehicles include: spatial benefit indicators and lane-changing necessity indicators; in the spatial benefit indicators, the Haversine formula is used in conjunction with the vehicle's latitude and longitude to calculate the actual gap between vehicles; the relative azimuth angles of the vehicle in front and behind are calculated based on the latitude and longitude of the vehicle in front and behind in the target lane, respectively; the relative positions of the vehicles in front and behind are calculated based on the relative azimuth angles of the vehicles in front and behind; vehicles are filtered based on the relative positions of the vehicles in front and behind, and the dynamic weights of the vehicles in front and behind are calculated based on the filtered vehicles; The spatial benefit of lane changing is calculated based on dynamic weights and the actual gap between vehicles. When the spatial benefit is 0, lane changing is not performed. In the lane changing necessity index, a safe distance between the CAV and the vehicle in front is set. The distance between the CAV and the vehicle in front in the current lane is obtained, and the lane changing necessity related to vehicle distance is calculated based on the distance and the safe distance. The lane changing necessity related to speed difference is calculated and evaluated. Weights are assigned to the lane changing necessity related to vehicle distance and the lane changing necessity related to speed difference. The lane changing index of intelligent connected vehicles is calculated based on the assigned weights.

2. The CAV lane-changing method based on macro and micro traffic flow information according to claim 1, characterized in that, The status information obtained for intelligent connected vehicles and surrounding non-connected vehicles includes: vehicle speed, acceleration, latitude and longitude, heading angle, timestamp, and ID; road segment information includes the number of lanes, road segment traffic distribution, average lane speed, and traffic density.

3. The CAV lane-changing method based on macro and micro traffic flow information according to claim 1, characterized in that, Preprocessing the uploaded data includes: denoising the vehicle status information, deleting abnormal and invalid data from the denoised data; synchronizing the multi-source data after removing invalid data based on the timestamp; fusing the vehicle's real-time latitude, longitude, speed, and acceleration into a unified data stream; and formatting and standardizing the fused data to convert it into a unified data format.

4. The CAV lane-changing method based on macro and micro traffic flow information according to claim 1, characterized in that, Lane change indicators are: f necessity = w distance • distance necessity + w speed • speed necessity In distance +in speed =1 Among them, f necessity In order to obtain the necessary indicators, w distance The weight of lane-changing necessity related to distance. necessity For the necessity of lane changing related to vehicle distance, w speed Assess the necessity of speed difference by assigning weights to speed. necessity The results of the assessment of the necessity of speed differences.

5. The CAV lane-changing method based on macro and micro traffic flow information according to claim 1, characterized in that, The lane-changing decision generated based on comprehensive macro and micro benefit indicators includes: between the current lane and the target lane, determining whether the current lane is the optimal target lane based on macro benefit indicators. If no lane is better than the current lane, the current lane is kept unchanged; otherwise, a target lane is recommended by CAV. Among the recommended target lanes, whether the lane-changing process meets safety conditions is evaluated through micro benefit assessment. If it meets the conditions, the lane-changing decision is executed; otherwise, the current lane is kept and the evaluation is continuously re-evaluated until the conditions are met.

6. The CAV lane-changing method based on macro and micro traffic flow information according to claim 5, characterized in that, Lane-changing decisions are transmitted to the vehicle's onboard unit via a 5G communication network.

7. The CAV lane-changing method based on macro and micro traffic flow information according to claim 1, characterized in that, The onboard unit verifies the lane-changing decision by: after receiving the recommendation, the CAV verifies the rationality and feasibility of the recommended plan by combining its own real-time status and the surrounding vehicle environment information; after the verification is passed, it executes the lane-changing operation according to the recommended plan; after the lane change is completed, the CAV transmits the actual operation data back to the cloud.

8. A CAV lane-changing system based on macro and micro traffic flow information, the system being used to execute the CAV lane-changing method based on macro and micro traffic flow information as described in any one of claims 1 to 7, the system comprising: Real-time driving data acquisition unit, macro-level benefit analysis module, micro-level benefit analysis module, CAV decision recommendation distribution module, and collaborative driving decision execution unit; The real-time driving data acquisition unit is used to collect status information of intelligent connected vehicles and surrounding non-connected vehicles, as well as road segment information in the direction of travel of intelligent connected vehicles. The macro revenue analysis module calculates the macro revenue of each lane under the current conditions based on the collected data. The micro-profit analysis module is used to calculate the micro-profit under the current conditions based on the collected data. The CAV decision recommendation distribution module constructs lane-changing strategies based on the macro and micro benefits of each lane; The cooperative driving decision-making and execution unit controls the vehicle to change lanes according to the lane-changing strategy.

Images