A method for ramp merging area vehicle cooperative control in a hybrid intelligent connected environment

Abstract

The present application relates to a kind of ramp merging area vehicle coordination control method under mixed intelligent network connection environment, belong to intelligent transportation technical field.Solve the single scene of prior art adaptation, cannot adapt to the problem of mixed traffic normality of manual driving vehicle and network connection automatic driving vehicle.Adjust technical scheme to obtain traffic state data, dynamically select a small amount of main line network connection automatic driving vehicle, complete lane change by lateral acceleration optimization, generate safe merging gap by combining longitudinal speed optimization, guide ramp vehicle orderly merging.This method adapts to mixed traffic flow, and the degree of intervention is low, through the layered optimization of lateral and longitudinal and adaptive weight control, give consideration to driving safety and traffic efficiency, and the practicality is strong.

Description

Technical Field

[0001] This invention belongs to the field of intelligent transportation technology and relates to a method for collaborative control of vehicles entering the ramp merging area in a hybrid intelligent connected environment. Background Technology

[0002] Ramps are important traffic nodes on highways, urban expressways, and roads with merging main and auxiliary roads. As vehicles on ramps enter the main road, they need to identify the available gaps between vehicles on the main road within a limited merging distance and adjust their speed and merging trajectory according to the speed, position, and movement trends of the main road vehicles.

[0003] Because there are usually speed differences, position differences, and differences in driving intentions between vehicles on the main line and those on the ramps, sudden acceleration, sudden deceleration, forced lane changes, cutting in, and rear-end collisions are prone to occur in the ramp merging area. When the ramp traffic density is high or the ramp queue length is long, it is difficult for ramp vehicles to obtain a merging gap that meets the safety distance requirements in a timely manner, which can easily cause ramp queuing and mainline traffic fluctuations.

[0004] In traditional manual driving environments, whether a vehicle can merge onto the main road from a ramp is primarily determined by the driver based on visual perception. Differences in driver reaction time, driving style, attention levels, and risk preferences lead to significant uncertainty in merging decisions. Furthermore, information asymmetry exists between vehicles, making it difficult for drivers to accurately obtain the acceleration and future operating status of adjacent vehicles, further increasing safety risks during ramp merging.

[0005] With the development of vehicle-road cooperation and autonomous driving technologies, connected autonomous vehicles possess the characteristics of obtainable vehicle status, predictable driving trajectory, and adjustable motion state. By implementing speed control on some connected autonomous vehicles, the longitudinal spacing between vehicles on the main line can be changed, creating merging conditions for vehicles on ramps.

[0006] However, the penetration rate of connected and autonomous vehicles will be difficult to reach 100% for a considerable period of time, and the actual road traffic environment will remain a mixed state of manually driven vehicles, ordinary connected vehicles, and connected and autonomous vehicles for a long time. Existing cooperative control methods designed for fully autonomous vehicle traffic flow are difficult to directly apply to mixed traffic environments.

[0007] Some existing ramp merging control methods directly control connected autonomous vehicles in other lanes to change lanes to the outermost lane of the mainline when there are no controllable vehicles in the outermost lane, and then create a merging gap by decelerating. However, the lateral lane-changing process is not only related to the average speed of the mainline traffic flow, but also to the candidate vehicle's own longitudinal speed, longitudinal acceleration, remaining lane-changing distance, as well as the position, speed, acceleration, and spacing of vehicles in the current lane and the target lane. Calculating the available lane-changing time for a single vehicle using only the average speed of the mainline traffic flow cannot accurately reflect the actual lane-changing conditions of the candidate vehicle.

[0008] Some lateral lane change optimization models directly sum the squared term of lateral acceleration and the reciprocal term of lane change time with weights. However, the squared term of lateral acceleration and the reciprocal term of lane change time have different physical dimensions, and the objective function obtained by directly adding them lacks a clear physical meaning.

[0009] Furthermore, in mixed traffic environments, manually driven vehicles and other vehicles not directly controlled are not directly controlled by the cooperative control system. If the control method assumes that uncontrolled vehicles behind the controlled vehicle must decelerate or remain behind the controlled vehicle, it is difficult to guarantee the following safety of vehicles during the deceleration of the controlled vehicle.

[0010] Under certain traffic conditions, ramp vehicles may be able to merge using the natural gaps in the mainline traffic flow before or during the implementation of coordinated control. If the system directly controls mainline vehicles to change lanes or decelerate without first determining whether ramp vehicles have the conditions for natural merging, it will increase unnecessary lateral maneuvers and longitudinal traffic disturbances.

[0011] Therefore, a vehicle cooperative control method for ramp merging areas suitable for hybrid intelligent connected traffic environments is needed. Before implementing lateral lane changing and longitudinal deceleration control for vehicles on the main line, it is necessary to determine whether the ramp vehicles already meet the conditions for natural merging. When cooperative control is required, connected autonomous vehicles already located in the outermost lane of the main line should be prioritized. When vehicles in other lanes need to change lanes, the motion states of candidate vehicles and vehicles in front of and behind the current lane and the target lane should be comprehensively considered to complete a reproducible lateral lane changing safety judgment. During longitudinal control, uncontrolled vehicles behind should be used as real-time safety constraints, and the control should be updated or withdrawn according to changes in traffic conditions. Summary of the Invention

[0012] In view of this, the purpose of this invention is to provide a method for cooperative vehicle control in the merging area of ​​a ramp under a hybrid intelligent connected environment, in order to solve the following problems existing in the prior art: Existing control methods do not determine whether vehicles on ramps have the conditions for natural merging before implementing coordinated control, which can easily cause vehicles on the main line to perform unnecessary lane changes or deceleration operations. The existing lateral lane change safety assessment does not fully consider the actual motion state of the candidate vehicle, the vehicles in front of and behind in the current lane, and the vehicles in front of and behind in the target lane. The existing objective function for lateral lane change has issues with inconsistent physical dimensions among its different objective terms. The existing longitudinal clearance formation process relies on the cooperation of manually driven vehicles or other uncontrolled vehicles behind, making it difficult to guarantee the safety of following the car during the control process; When traffic conditions change, existing methods lack a rolling exit mechanism to reassess natural merging conditions and to stop unnecessary controls.

[0013] To achieve the above objectives, the present invention provides the following technical solution: A method for cooperative vehicle control in a ramp merging area under a hybrid intelligent connected environment includes the following steps: S1: Obtain traffic status data Traffic status data of mainline vehicles and ramp vehicles are acquired by using roadside sensing devices, roadside units and vehicle communication devices installed in the ramp merging area.

[0014] The traffic status data includes the position, longitudinal speed, longitudinal acceleration, lane and vehicle length of vehicles on the main line, the position, longitudinal speed, longitudinal acceleration and vehicle length of vehicles on the ramp, as well as the average longitudinal speed of the main line traffic flow, the average longitudinal speed of the ramp traffic flow, the ramp traffic density, the real-time queue length of the ramps and the penetration rate of connected and automated vehicles.

[0015] The connected autonomous vehicle is a connected and autonomous vehicle (CAV).

[0016] Establish a longitudinal coordinate axis for the road, define the mainline vehicle travel direction as the positive longitudinal direction, and then... j At any moment t The longitudinal position, longitudinal velocity, and observed longitudinal acceleration are denoted as: , , ; Among them, subscript z Indicates the longitudinal direction of the road; subscript j Indicates the corresponding vehicle; observes longitudinal acceleration. This represents the longitudinal acceleration of a vehicle, which has positive and negative directions. Positive values ​​are taken when the vehicle accelerates in the direction of travel, and negative values ​​are taken when the vehicle decelerates in the direction of travel.

[0017] When the vehicle communication data includes longitudinal acceleration, the longitudinal acceleration is read directly from the vehicle communication data. When the vehicle communication data does not include longitudinal acceleration, the observed longitudinal acceleration is calculated based on the continuously acquired longitudinal velocity.

[0018] in, The sampling period for traffic status data.

[0019] The main road is divided into the Vehicle Preparation Area (VPA), Vehicle Control Area (VCA), and Vehicle Merging Area (VMA) along the direction of vehicle travel.

[0020] The Vehicle Preparation Area (VPA) is located upstream of the Vehicle Control Area (VCA) and is used for selecting candidate controlled Vehicles (CAVs) and determining the safety of lateral lane changes.

[0021] The Vehicle Control Area (VCA) is located between the Vehicle Preparation Area (VPA) and the Vehicle Merging Area (VMA) and is used to control the controlled Vehicle (CAV) to complete longitudinal speed optimization.

[0022] The Vehicle Merging Area (VMA) is located in the actual merging area between the ramp and the mainline, and is used to allow ramp vehicles and mainline vehicles to pass in sequence.

[0023] S2: Determine whether vehicles on the ramp have the conditions for natural merging. When the ramp vehicle detector detects a vehicle about to merge onto the ramp, the detection time is recorded as follows: First, it is predicted whether there is a natural merging gap in the outermost lane of the main line that meets the natural merging conditions when the vehicles waiting to merge arrive at the merging point.

[0024] The outermost lane of the main line is the main line lane adjacent to the ramp and used to receive vehicles merging from the ramp.

[0025] Vehicles waiting to merge onto the ramp are recorded as ramp vehicles. Obtain ramp vehicles At the time of detection Longitudinal distance to the sink point Longitudinal velocity and observation of longitudinal acceleration .

[0026] when and At that time, the predicted travel time for vehicles on the ramp to reach the merging point is calculated according to the following formula:

[0027] when and When it is determined that a valid natural merging prediction result cannot be formed at present, the updated status of ramp vehicles will continue to be acquired according to the preset control cycle; among which, To preset the minimum effective ramp vehicle speed; A preset acceleration threshold is used to determine whether the vehicle's longitudinal acceleration is close to zero.

[0028] When the following conditions are met: At that time, the ramp vehicle speed is calculated according to the following formula. Predicted travel time to the merging point :

[0029] The candidate solutions for predicting the travel time are:

[0030] The smallest positive root satisfying the longitudinal motion equation is determined as the predicted travel time. .

[0031] When the value within the square root is less than zero, the longitudinal motion equation has no positive root, or it is determined from the predicted motion state that the ramp vehicle cannot reach the merging point before stopping, it is determined that a valid natural merging prediction result cannot be formed at present. In the current control cycle, the controlled vehicle selection step is not entered, and the position, longitudinal velocity, and observed longitudinal acceleration of the ramp vehicle continue to be acquired according to the preset control cycle. When the positive predicted travel time of the ramp vehicle to the merging point can be obtained based on the updated vehicle state, the natural merging gap is re-predicted. The controlled vehicle selection step is entered only when a valid natural merging prediction result can be formed and there is no natural merging gap that meets the natural merging conditions.

[0032] The predicted arrival time of vehicles on the ramp is denoted as: : Vehicle calculation using a constant acceleration short-time prediction model j Predicting the inflow time Predicted longitudinal position and predicted longitudinal velocity:

[0033]

[0034] Among them, vehicles j For ramp vehicles Or vehicles in the outermost lane of the main line; superscript This represents the predicted value.

[0035] Based on the predicted merging time of each vehicle The predicted longitudinal position is used to determine the location of vehicles on the ramp within the outermost lane of the mainline. The vehicle in front of you f and vehicles located on ramps The vehicle behind and adjacent to the rear b .

[0036] Calculate ramp vehicles according to the following formula. Adjacent vehicle in front f Predicted longitudinal net spacing between:

[0037] Calculate the following formula for adjacent vehicles behind. b With ramp vehicles Predicted longitudinal net spacing between:

[0038] in, , and The vehicles adjacent to each other in front f Ramp vehicles and adjacent vehicles behind b The length of the vehicle.

[0039] For vehicles following the car and the vehicle located in front of the following vehicle The following formula is used to calculate the following vehicles in a vehicle pair. Relative to the vehicle in front Safe longitudinal spacing:

[0040] in, The minimum stationary distance is preset; To preset a safe headway; This is to preset the safe braking deceleration amplitude.

[0041] The vehicle in front is determined when the following conditions are met simultaneously. f Adjacent vehicles behind b There exist natural inflow gaps that satisfy the natural inflow conditions:

[0042]

[0043]

[0044]

[0045] in, This is the preset maximum allowed import speed difference.

[0046] The first condition is the forward safety distance constraint between the ramp vehicle and the adjacent vehicle in front; the second condition is the backward safety distance constraint between the adjacent vehicle behind and the ramp vehicle; the third and fourth conditions together constitute the merging speed difference constraint.

[0047] When a natural merging gap exists that satisfies the natural merging conditions, the controlled CAV is not selected, and no lateral lane change control or longitudinal deceleration control is implemented for mainline vehicles. Ramp vehicles use the natural merging gap to merge into the outermost lane of the mainline.

[0048] If there is no natural merging gap that meets the natural merging conditions, proceed to the controlled vehicle selection step.

[0049] S3: Determine the number of vehicles to be controlled and the selection range of vehicles to be controlled. Get real-time queue length of ramps The maximum inflow gap that a single controlled CAV on the main line can provide. and the maximum number of vehicles under control The maximum inflow gap The maximum dynamic merging gap that a single controlled CAV can form is defined under the following constraints: longitudinal length of the vehicle control area (VCA), maximum permissible longitudinal deceleration of the controlled CAV, minimum permissible longitudinal speed of the controlled CAV, longitudinal safety distance constraint between the controlled CAV and the vehicle in front, and safety constraint for uncontrolled vehicles following behind the controlled CAV. The maximum inflow gap that a single controlled CAV can provide on the main line Determined through offline calibration or online optimization. When performing online optimization, it is determined according to the following formula:

[0050] Among them, longitudinal deceleration amplitude and the end time of vertical optimization The controllable vehicle (CAV) must simultaneously satisfy the following constraints: acceleration constraint, speed constraint, rate of change of acceleration constraint, longitudinal safety distance constraint between the controlled CAV and the vehicle in front, safety constraint for uncontrolled vehicles following the controlled CAV, and dynamic merging gap constraint.

[0051] When using the offline calibration method, the maximum merging gap is pre-calculated and stored based on the longitudinal length of the vehicle control area (VCA), the road design speed, the allowable longitudinal deceleration range, and the preset safety distance parameters.

[0052] When the following conditions are met: At that time, determine the number of vehicles under control:

[0053] When the following conditions are met: At that time, the number of controlled vehicles is determined according to the following formula: ; in, The number of vehicles under control; This indicates the rounding up operation.

[0054] Obtain the average longitudinal speed of the main line traffic flow Average longitudinal speed of ramp traffic and the longitudinal distance from the ramp vehicle detector to the merging point The range of controlled vehicles is calculated according to the following formula. :

[0055] in, To preset the minimum effective average longitudinal speed of ramp traffic, the time required for vehicles on the ramp to travel to the merging point at the average longitudinal speed of ramp traffic is:

[0056] The distance traveled by vehicles on the main line at the average longitudinal speed of the main line traffic flow within the specified time is used as the range for selecting controlled vehicles. .

[0057] Within the scope of controlled vehicles Within the pool, CAVs are selected in order of distance from the ingress point to the nearest point.

[0058] Prioritize selecting controlled CAVs from those already located in the outermost lane of the main line and meeting the longitudinal control feasibility constraints.

[0059] The longitudinal control feasibility constraints include controlled CAV speed constraints, controlled CAV acceleration constraints, controlled CAV collision risk constraints with the vehicle in front, and uncontrolled vehicles following the controlled CAV for safety constraints.

[0060] When the number of CAVs satisfying the longitudinal control feasibility constraints in the outermost lane of the main line reaches the number of controlled vehicles. At that time, vehicles will no longer be selected from other lanes.

[0061] When the number of CAVs satisfying the longitudinal control feasibility constraints in the outermost lane of the main line is less than the number of controlled vehicles. At that time, candidate controlled lane AVs are selected from the lanes adjacent to the outermost lane of the main line, and lateral lane change safety constraints are determined for the candidate controlled lane AVs.

[0062] Therefore, lateral lane changes are only permitted for candidate controlled CAVs in other lanes when there is no natural merging gap and the number of CAVs meeting the longitudinal control conditions in the outermost lane of the mainline is insufficient.

[0063] When the number of controlled vehicles When the distance is greater than 1, each controlled CAV is numbered in order of proximity to the merging point, and longitudinal speed optimization is performed sequentially according to the numbering order. For a controlled CAV located behind another controlled CAV, the nearest vehicle in front of it is taken as the forward adjacent vehicle. When the forward adjacent vehicle is a controlled CAV whose longitudinal speed optimization has already been completed, longitudinal safety distance constraints are established using the predicted longitudinal position, predicted longitudinal speed, and actual longitudinal acceleration control commands of the forward adjacent vehicle. Each controlled CAV forms a dynamic merging gap, and each dynamic merging gap corresponds to the vehicles waiting to merge onto the ramp in order of their predicted arrival order at the merging point, in order of proximity to the merging point.

[0064] S4: Calculate the available lane-changing time for the candidate controlled CAV. For candidate controlled CAVs that are not located in the outermost lane of the main line, the candidate controlled CAVs will be recorded as vehicles. c .

[0065] Acquire vehicle c At the time of candidate vehicle selection Remaining longitudinal distance to the downstream boundary of the vehicle preparation area VPA Longitudinal velocity and observation of longitudinal acceleration .

[0066] The remaining longitudinal distance Indicates vehicle c The actual longitudinal distance between the current location and the downstream boundary of the Vehicle Preparation Area (VPA) is not equal to the total length of the VPA. This applies only to vehicles... c When located at the upstream boundary of the vehicle preparation area VPA, the remaining longitudinal distance It equals the total length of the vehicle preparation area (VPA).

[0067] When the following conditions are met: and At that time, the available lane-changing time is calculated according to the following formula. :

[0068] when When the candidate controlled CAV does not meet the lane-changing available time constraint, it is determined that the candidate controlled CAV does not meet the lane-changing available time constraint. When the following conditions are met: At that time, the available lane-changing time is calculated according to the following formula. :

[0069] When the value inside the square root is not less than zero, the vehicle will be satisfied. c The minimum positive value of the longitudinal motion equation is determined as the available lane-changing time. .

[0070] When the value inside the square root is less than zero, the result does not have a positive value, or the vehicle is determined based on the prediction result. c When the vehicle has stopped before reaching the downstream boundary of the vehicle preparation area VPA, determine the vehicle. c If the lane change availability time constraint is not met, the vehicle will not be moved. c It has been identified as the accused CAV.

[0071] S5: Constructing the objective function for lateral lane change The lateral lane change process is divided into a lateral acceleration phase and a lateral deceleration phase.

[0072] The lateral acceleration and lateral deceleration phases last for the same duration, but the lateral accelerations are in opposite directions, causing the vehicle to... c The lateral speed returns to zero at the end of the lateral lane change.

[0073] With lateral acceleration amplitude As a decision variable, the vehicle c The lateral distance required to move from the current lane center to the center of the outermost lane of the main line is denoted as . .

[0074] Let the duration of the lateral acceleration phase be... The lateral displacement during the lateral acceleration phase is: ; The lateral displacement during the lateral deceleration phase is: ; The sum of the lateral displacements in the two stages equals Therefore, the lateral lane change execution time is obtained: ; Construct the lateral lane change objective function:

[0075] in, The objective function for lateral lane changing; and Let be the weighting coefficients of the lateral lane change objective function, and satisfy: , g This is the acceleration due to gravity.

[0076] The first objective in the objective function is the square of the ratio of the lateral acceleration amplitude to the maximum permissible lateral acceleration, which is dimensionless. The second objective in the objective function is the ratio of the lateral lane change execution time to the available lane change time, which is also dimensionless. Therefore, the two objective terms have the same physical dimensions and can be weighted and summed.

[0077] The execution time for lateral lane changes should meet the following requirements: ; Substituting the formula for the lateral lane change execution time, we obtain the lower limit of the lateral acceleration amplitude: ; The upper limit of the lateral acceleration amplitude is: Therefore, the lateral acceleration constraint is: ; S6: Calculate the potential collision probability during a lateral lane change. Acquire vehicle c Vehicles in front of and behind in the current lane, as well as vehicles in the outermost lane of the main line. c The vehicles in front and behind at the corresponding positions.

[0078] When there is a vehicle ahead of the candidate controlled CAV in the current lane within the preset effective perception range, the candidate controlled CAV and the vehicle ahead in the current lane are combined to form a first adjacent vehicle pair. When there is a vehicle behind the current lane within the preset effective sensing range, the vehicle behind the current lane and the candidate controlled CAV are combined to form a second adjacent vehicle pair; When there is a vehicle ahead in the outermost lane of the main line within the preset effective sensing range, the candidate controlled CAV and the vehicle ahead in the outermost lane of the main line are combined to form a third adjacent vehicle pair. When there is a vehicle behind the outermost lane of the main line within the preset effective sensing range, the vehicle behind the outermost lane of the main line and the candidate controlled CAV are combined to form a fourth adjacent vehicle pair. The actual adjacent vehicle pairs formed are combined into an adjacent vehicle pair set. When there is no corresponding vehicle in any direction or the corresponding vehicle is outside the preset effective sensing range, no adjacent vehicle pair is established for that direction; when the set of adjacent vehicle pairs... When the set is empty, the potential collision probability is set to 0. For adjacent vehicle pairs Any adjacent vehicle pair , will the vehicle The vehicle was identified as a follower and was put into service. It has been identified as the vehicle ahead.

[0079] According to the preset prediction time step Set the prediction time during the lateral lane change execution and the lateral lane change end time. Set as the last prediction time: in: ; For adjacent vehicle pairs Vehicles in j Calculate the vehicle according to the following formula j At the predicted time Predicted longitudinal velocity:

[0080] The following formula is used to calculate adjacent vehicle pairs. At the predicted time The predicted mean longitudinal net spacing:

[0081] in, For following vehicles Vehicle in front At the time of candidate vehicle selection The initial longitudinal net spacing.

[0082] The following formula is used to calculate the following vehicles. Relative to the vehicle in front At the predicted time Safe longitudinal spacing:

[0083] Considering vehicle state measurement errors and prediction model errors, the error of adjacent vehicle pairs is calculated according to the following formula. Predicted standard deviation of longitudinal net spacing:

[0084] in, For adjacent vehicles The standard deviation of the relative spacing prediction error; The standard deviation of the relative velocity prediction error; This represents the standard deviation of the relative acceleration prediction error.

[0085] The standard deviation of the relative spacing prediction error is determined by the following formula:

[0086] The standard deviation of the relative velocity prediction error is determined by the following formula:

[0087] The standard deviation of the relative acceleration prediction error is determined by the following formula:

[0088] in, , and vehicles Standard deviation of position, velocity, and acceleration measurement errors; , and vehicles The standard deviation of position, velocity, and acceleration measurement errors.

[0089] The standard deviation of each measurement error is determined by the calibration results of the roadside sensing equipment and the vehicle-mounted sensing equipment, or by the root mean square error between the predicted value and the actual measured value within a preset sliding time window.

[0090] When the calculated standard deviation of the longitudinal net spacing prediction is less than the preset minimum standard deviation When, set it to: ,in, This is used to avoid the denominator being zero during the calculation of the probability of collision risk.

[0091] The predicted longitudinal net spacing follows the mean. Standard deviation is Under the condition of normal distribution, the number of adjacent vehicle pairs can be calculated according to the following formula. At the predicted time Collision risk probability:

[0092] in, This is the cumulative distribution function of the standard normal distribution.

[0093] The maximum value among all adjacent vehicle pairs and all predicted time points is determined as the potential collision probability:

[0094] Due to the execution time of lateral lane change With lateral acceleration amplitude Variations in prediction time range and prediction steps with lateral acceleration amplitude The change, therefore the potential collision probability is the lateral acceleration amplitude. The function.

[0095] The potential collision probability constraint is: in, This is a preset safety threshold.

[0096] Solve the objective function for lateral lane change under the following constraints:

[0097]

[0098] Obtain the optimal lateral acceleration amplitude .

[0099] The objective function for the lateral lane change is solved using a sequential quadratic programming algorithm, interior point method, or other numerical optimization algorithms that meet the constraints.

[0100] When there exists an optimal lateral acceleration amplitude that satisfies all constraints. At that time, the vehicle c The controlled CAV was identified, and the optimal lateral acceleration amplitude was determined. The duration of the lateral acceleration phase, the duration of the lateral deceleration phase, and the corresponding lateral acceleration direction are sent to the chassis control system of the controlled CAV.

[0101] The durations of both the lateral acceleration phase and the lateral deceleration phase are: ; When there is no lateral acceleration amplitude that satisfies all constraints, the vehicle will not be... c Once a candidate CAV is identified as the controlled CAV, the next candidate controlled CAV is determined in order of distance from the sink point from closest to furthest.

[0102] When none of the candidate controlled CAVs meet the lateral lane change safety constraints, the system does not force the candidate controlled CAVs to perform lateral lane changes, but continues to monitor the traffic status of the mainline and ramps until a natural merging gap that meets the natural merging conditions appears or a new candidate controlled CAV that meets the control conditions appears.

[0103] For a controlled CAV that is already in the outermost lane of the main line, no lateral lane change control is performed. Instead, the controlled CAV is controlled to maintain its current lane and enter the subsequent longitudinal speed optimization stage.

[0104] S7: Scroll to determine natural import conditions The system operates according to a preset control cycle before the controlled CAV begins a lateral lane change, during the lateral lane change, and after the controlled CAV enters the Vehicle Control Area (VCA). Update the traffic status of mainline vehicles and ramp vehicles, and re-predict whether vehicles waiting to merge on ramps have the conditions for natural merging.

[0105] If a natural merging gap that meets the natural merging conditions is detected before the controlled CAV begins a lateral lane change, the lateral lane change control and longitudinal deceleration control of the controlled CAV are cancelled.

[0106] After the controlled CAV has started a lateral lane change, if a natural merging gap that meets the natural merging conditions is detected, the controlled CAV is controlled to complete the current lateral lane change according to the determined safe lateral lane change trajectory. The controlled CAV is not controlled to reverse the lane change in the middle of the lateral lane change, and the longitudinal deceleration control used to widen the merging gap after the controlled CAV completes the lateral lane change is canceled.

[0107] During longitudinal deceleration control of the controlled CAV, if a natural merging gap satisfying the natural merging condition is detected, the rate of change of the longitudinal acceleration control command of the controlled CAV is limited according to the following formula:

[0108] in, For the accused CAV at the moment t Longitudinal acceleration control commands; This is the preset maximum allowable rate of change of acceleration.

[0109] Under the condition of satisfying the longitudinal acceleration control command change rate constraint, the longitudinal deceleration amplitude of the controlled CAV is gradually reduced so that the controlled CAV can smoothly return to normal driving state.

[0110] S8: Constructing a longitudinal velocity optimization model When there is no natural merging gap that meets the natural merging conditions, the controlled CAV completes longitudinal speed optimization within the vehicle control area (VCA).

[0111] The moment when the controlled CAV enters the Vehicle Control Area (VCA) and begins longitudinal speed optimization is denoted as [missing information]. The moment when the controlled CAV leaves the Vehicle Control Area (VCA) and enters the Vehicle Merging Area (VMA) is recorded as .

[0112] Get Time Corresponding ramp traffic density and the critical density of ramps .

[0113] Ramp Critical Density Pre-calibrated based on historical traffic data of the ramps. Specifically, historical traffic density and historical queue length of the ramps are synchronized in time to determine the ramp traffic density corresponding to the transition from a stable state to a continuously increasing state of the ramp queue length, and this ramp traffic density is determined as the critical ramp density. The critical density of the ramps is recalibrated according to the newly added traffic status data at a preset update cycle.

[0114] When the following conditions are met: When operating condition one is adopted, the controlled CAV is controlled to continuously decelerate throughout the entire process within the vehicle control area (VCA).

[0115] When the following conditions are met: In this case, operating condition two is adopted, and the controlled CAV is controlled to decelerate first and then travel at a constant speed.

[0116] Construct an adaptive multi-objective optimization model:

[0117] in, Optimize the objective function for longitudinal velocity; The proportion of CAVs (Cars Access Vehicles) to the total number of vehicles within the statistical scope of the ramp merging area; To normalize potential conflict indicators; To optimize the baseline time vertically; and These are adaptive weighting coefficients.

[0118] The baseline time for vertical optimization is calculated using the following formula:

[0119] in, This is the longitudinal length of the vehicle control area (VCA). For the accused CAV at the moment longitudinal velocity, The preset minimum effective longitudinal velocity of the controlled CAV.

[0120] The vehicle pairs that require safety judgments during the longitudinal optimization prediction period are grouped into a longitudinal vehicle pair set: ; The longitudinal vehicle pair set includes vehicle pairs consisting of a controlled CAV and the vehicle in front of it, vehicle pairs consisting of an uncontrolled vehicle behind the controlled CAV and the controlled CAV, and adjacent vehicle pairs formed between multiple controlled CAVs.

[0121] For vehicle pairs in the longitudinal vehicle pair set ( p , l ), calculate the predicted longitudinal net spacing based on the predicted longitudinal position of each vehicle:

[0122] in, and Vehicles following each other p and the vehicle in front l The length of the vehicle.

[0123] For vehicles not directly controlled j The longitudinal position and longitudinal velocity are predicted according to the following formula:

[0124]

[0125] With non-negative longitudinal deceleration amplitude The longitudinal velocity optimization variable for the controlled CAV is: The actual longitudinal acceleration corresponding to the controlled CAV is: ; Under operating condition 1, the predicted longitudinal position and predicted longitudinal velocity of the controlled CAV during longitudinal optimization are as follows:

[0126]

[0127] Under operating condition two, the moment when the controlled CAV completes deceleration is recorded as (the time when the controlled CAV completes deceleration is then recorded as ). The duration of deceleration is denoted as:

[0128] when: At that time, the predicted longitudinal position and predicted longitudinal velocity of the controlled CAV are:

[0129]

[0130] when: At that time, the predicted vertical position of the controlled CAV is:

[0131] The predicted longitudinal velocity of the controlled CAV is:

[0132] in:

[0133] For vehicle pairs in the longitudinal vehicle pair set The original number of potential conflict vehicle pairs is calculated according to the following formula:

[0134] in, This is an indicator function; it takes the value 1 when the condition inside the parentheses is true, and takes the value 0 when the condition inside the parentheses is false.

[0135] The normalized potential conflict index is calculated using the following formula:

[0136] in, For longitudinal vehicle pairs The number of vehicles in the system.

[0137] Therefore, we can conclude that:

[0138] The longitudinal optimization time is normalized using the longitudinal optimization baseline time:

[0139] Therefore, the normalized potential conflict index and the normalized longitudinal optimization time in the longitudinal velocity optimization objective function are both dimensionless quantities.

[0140] The adaptive weighting coefficients satisfy:

[0141] Determine the adaptive weighting coefficients in the following order.

[0142] When the following conditions are met: or At that time, it was determined that: At this point, the priority should be to reduce potential traffic conflicts.

[0143] When the aforementioned conditions are not met, and the following conditions are met: or Determine if at least one of the following conditions is met: At this point, priority should be given to shortening the duration of vertical optimization.

[0144] When neither the first set of conditions nor the second set of conditions is met, determine: .

[0145] Under the constraints of controlled CAV acceleration, controlled CAV velocity, rate of change of acceleration, longitudinal safety distance between the controlled CAV and its adjacent vehicle in front, safety constraint for uncontrolled vehicles following the controlled CAV, and dynamic merging gap not less than the preset minimum safe merging gap, the longitudinal velocity optimization objective function is solved to obtain the optimal longitudinal deceleration amplitude: Among them, the optimal longitudinal deceleration amplitude For non-negative values, the corresponding actual longitudinal acceleration control command is: ; The controlled CAV acceleration constraint is: ;in, This is the preset maximum allowable longitudinal deceleration amplitude.

[0146] The controlled CAV velocity constraint is: ;in, and These are the preset minimum longitudinal speed and the preset maximum longitudinal speed, respectively.

[0147] The acceleration rate of change constraint is: ; The accused CAV will be recorded as a vehicle. c The nearest vehicle located in front of the controlled CAV and in the outermost lane of the main line will be recorded as the adjacent vehicle in front. .

[0148] During longitudinal optimization, the predicted longitudinal velocity of the controlled CAV is calculated according to the following formula:

[0149] Calculate the number of adjacent vehicles ahead using the following formula. Predicted longitudinal velocity:

[0150] The controlled CAV and the adjacent vehicle in front are calculated using the following formula. Predicted longitudinal net spacing between:

[0151] The controlled CAV is calculated relative to the adjacent vehicle in front using the following formula. Safe longitudinal spacing: ; The longitudinal safety distance constraint between the controlled CAV and the adjacent vehicle in front is: .

[0152] When there are no vehicles in front of the controlled CAV within the preset effective perception range, the longitudinal safety distance constraint is not set.

[0153] Dynamic ingress gap constraints are used to ensure that the controlled CAV is within a certain time frame. It remains upstream of the inflow point, maintaining a dynamic safety distance between the controlled CAV and the inflow point. The dynamic inflow gap constraint is as follows:

[0154] in, According to the accused CAV at the time At that time Longitudinal driving distance, controlled CAV at any time Longitudinal distance to the sink and time The dynamic inflow gap is obtained by calculating the corresponding dynamic safety distance. This is a preset minimum safe entry gap.

[0155] When the dynamic inflow gap constraint is satisfied, the controlled CAV at time... The remaining longitudinal distance to the merging point is not less than the sum of the dynamic safety distance and the preset minimum safe merging gap.

[0156] S9: Establish safety constraints for uncontrolled vehicles following behind. Manually driven vehicles or other vehicles not directly controlled that are located behind the controlled CAV and in the outermost lane of the same main line as the controlled CAV will be recorded as uncontrolled vehicles behind it. u .

[0157] The system does not warn uncontrolled vehicles behind it. u Sending control commands will not affect uncontrolled vehicles behind. u Deceleration is a prerequisite for longitudinal deceleration control of the controlled CAV.

[0158] Acquire uncontrolled vehicles behind u At any moment The longitudinal position, longitudinal velocity, and observed longitudinal acceleration.

[0159] Calculate the number of uncontrolled vehicles behind using the following formula. u Predicted net longitudinal spacing between the controlled CAV and the controlled CAV during longitudinal optimization:

[0160] in, For uncontrolled vehicles behind u With the accused CAV at the moment The initial longitudinal net spacing; For uncontrolled vehicles behind u At any moment longitudinal velocity; For uncontrolled vehicles behind u At any moment The longitudinal acceleration was observed.

[0161] The following formula is used to calculate uncontrolled vehicles behind. u Relative to the safe longitudinal spacing of the controlled CAV:

[0162] The safety constraints for uncontrolled vehicles following behind are: ;in: ; The system operates according to a preset control cycle. Update the position, longitudinal velocity, and longitudinal acceleration of the uncontrolled vehicle behind, and resolve the longitudinal deceleration amplitude of the controlled CAV.

[0163] When an uncontrolled vehicle behind accelerates, fails to decelerate in time, or the predicted longitudinal clearance tends to be less than the safe longitudinal clearance, the system reduces the longitudinal deceleration amplitude of the controlled CAV.

[0164] When there is no longitudinal deceleration amplitude that satisfies the safety constraint of uncontrolled vehicles following behind, cancel the use of the currently controlled CAV to form a dynamic merging gap, and reselect other controlled CAVs or continue to monitor the natural merging gap.

[0165] Therefore, uncontrolled vehicles behind are only considered as objects of traffic state perception and safety constraints, and the system does not assume that they must obey control commands or necessarily remain behind the controlled CAV.

[0166] S10: Calculate dynamic inflow gap The CAV will be charged at the time The longitudinal distance to the sink point is denoted as: ; The longitudinal distance It represents the remaining distance along the longitudinal direction of the road between the current position of the controlled CAV and the merging point, but does not represent the longitudinal position coordinates of the controlled CAV.

[0167] Under operating condition 1, the controlled CAV is calculated from time 1 according to the following formula. At that time Longitudinal driving distance:

[0168] The dynamic ingress gap under the following operating condition is calculated using the following formula:

[0169] in, This only indicates the operating condition of the controlled CAV from the time point. At that time The longitudinal driving distance.

[0170] Under operating condition two, the moment when the controlled CAV completes deceleration is recorded as (the time when the controlled CAV completes deceleration is then recorded as ). .

[0171] The controlled CAV is calculated from time 1 according to the following formula. At that time Deceleration distance:

[0172] The controlled CAV at time t is calculated according to the following formula. Longitudinal velocity:

[0173] The controlled CAV is calculated from time 1 according to the following formula. At that time The distance traveled at a constant speed:

[0174] The dynamic merging gap under operating condition two is calculated using the following formula:

[0175] in, This only indicates the deceleration distance of the controlled CAV under condition two; This only indicates the constant speed travel distance of the controlled CAV under condition two.

[0176] time The corresponding dynamic safety distance is:

[0177] The aforementioned safety longitudinal spacing The dynamic safety distance is used to determine whether there is a risk of longitudinal collision between a following vehicle and the vehicle in front. Its calculation considers the preset minimum stationary distance, safe headway, and the relative braking distance between the vehicles in front and behind; this step involves the dynamic safety distance. Used to limit the controlled CAV at a certain time. The minimum positional margin that should be maintained between the point of entry and the point of convergence.

[0178] The dynamic safety distance Does not replace the safe longitudinal clearance between vehicle pairs During longitudinal speed optimization, the controlled CAV should simultaneously satisfy the longitudinal safety distance constraint between itself and the vehicle in front, the following safety constraint of uncontrolled vehicles behind it, and the dynamic merging gap constraint determined by the dynamic safety distance.

[0179] The resulting dynamic inflow gap satisfies:

[0180] in, This is a preset minimum safe entry gap.

[0181] The controlled CAV forms the rear boundary of the dynamic merging gap. Mainline vehicles located in front of the controlled CAV continue downstream according to their original driving status, while ramp vehicles enter the outermost lane of the mainline in front of the controlled CAV.

[0182] When the real-time queue length of the ramp exceeds the maximum merging gap that a single controlled CAV can provide, multiple controlled CAVs will form multiple dynamic merging gaps in order of distance from the merging point from near to far, so that the ramp vehicles can merge in groups.

[0183] S11: Vehicle Merging and Control Recovery Control the controlled CAV and the vehicles waiting to merge on the ramp to pass through the vehicle merging area (VMA) in sequence, so that the ramp vehicles merge into the outermost lane of the main line in front of the controlled CAV.

[0184] Once a vehicle on the ramp has completed its merging and left the vehicle merging area (VMA), longitudinal deceleration control of the controlled CAV is terminated.

[0185] The controlled CAV gradually recovers to the desired driving speed on the main line according to the acceleration change rate constraint.

[0186] When vehicles on the ramp have not yet completed merging or the mainline traffic condition changes, the natural merging condition judgment, the following uncontrolled vehicle safety constraint update, and the longitudinal speed optimization rolling solution continue to be executed.

[0187] The beneficial effects of this invention are as follows: (1) Before selecting a controlled CAV and implementing lateral lane change or longitudinal deceleration, the natural merging gap of the ramp vehicle when it arrives at the merging point is predicted. When the ramp vehicle already has the conditions for natural merging, no additional control is implemented on the mainline vehicle, reducing unnecessary lateral lane change and longitudinal deceleration operations. Controlled vehicles are preferentially selected from CAVs already located in the outermost lane of the mainline. Only when the number of CAVs that meet the longitudinal control conditions in the outermost lane of the mainline is insufficient, candidate controlled CAVs are selected from adjacent lanes and lateral lane change safety judgment is performed, reducing the risk of additional lateral operations.

[0188] (2) The available lane-changing time is calculated based on the remaining lane-changing distance, longitudinal speed, and observed longitudinal acceleration of the candidate controlled CAV, instead of using the average speed of the mainline traffic flow to replace the actual motion state of a single vehicle. In the lateral lane-changing safety judgment, the positions, longitudinal speeds, longitudinal accelerations, longitudinal distances, and vehicle state prediction errors of the candidate controlled CAV, the vehicles in front and behind in the current lane, and the vehicles in front and behind in the target lane are considered simultaneously, enabling the calculation of the potential collision probability according to a clear formula. The lateral lane-changing objective function of this invention uses normalized lateral acceleration and normalized lane-changing execution time as objective terms. Both objective terms are dimensionless, eliminating the problem of directly adding different physical dimensions. The longitudinal speed optimization objective function of this invention uses normalized potential conflict index and normalized longitudinal optimization time as objective terms, making both objective terms in the longitudinal optimization objective function dimensionless.

[0189] (3) This invention does not assume that manually driven vehicles or other uncontrolled vehicles behind are subject to system control. Instead, it treats uncontrolled vehicles behind as real-time safety constraints and updates the longitudinal control quantity of the controlled CAV on a rolling basis according to the actual acceleration and deceleration states of the vehicles behind. During the control process, this invention continuously reassesses the natural merging conditions. Once the natural merging conditions are formed, it cancels the control that has not yet started, completes the safe lane change that has already started, or smoothly exits the longitudinal deceleration control according to the control stage of the controlled CAV, avoiding the continued execution of control operations that have lost their necessity. It can adapt to different CAV penetration rates, different ramp traffic densities, and different mainline traffic states, reducing interference with mixed traffic flow while ensuring safety.

[0190] Other advantages, objectives, and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination, or may be learned from practice of the invention. The objectives and other advantages of the invention can be realized and obtained through the following description. Attached Figure Description

[0191] To make the objectives, technical solutions, and advantages of the present invention clearer, the preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings, wherein: Figure 1 A schematic diagram showing the division of the vehicle cooperative control area; Figure 2 This is a flowchart of the overall vehicle cooperative control method. Detailed Implementation

[0192] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. Unless otherwise specified, the following embodiments and features can be combined with each other.

[0193] The accompanying drawings are for illustrative purposes only and are schematic diagrams, not actual pictures. They should not be construed as limiting the invention. To better illustrate the embodiments of the invention, some parts in the drawings may be omitted, enlarged, or reduced, and do not represent the actual product dimensions. It is understandable to those skilled in the art that some well-known structures and their descriptions may be omitted in the drawings.

[0194] In the accompanying drawings of the embodiments of the present invention, the same or similar reference numerals correspond to the same or similar components. In the description of the present invention, it should be understood that if terms such as "upper," "lower," "left," "right," "front," and "rear" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, they are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, the terms used to describe positional relationships in the drawings are only for illustrative purposes and should not be construed as limiting the present invention. For those skilled in the art, the specific meaning of the above terms can be understood according to the specific circumstances.

[0195] The present invention will now be described in further detail with reference to the accompanying drawings. It should be noted that the implementation of the present invention is not limited to the specific parameter settings described below. Other embodiments obtained by those skilled in the art without inventive effort are all within the scope of protection of the present invention.

[0196] like Figure 1 As shown, the present invention provides a method for coordinated vehicle control in the merging area of ​​a ramp under a hybrid intelligent connected environment. Figure 1The diagram illustrates the division of the vehicle cooperative control area, dynamically dividing the main road into three zones. The vehicle preparation zone is located at the upstream end, used for controlled vehicles to prepare for lateral lane changes. The vehicle control zone is located in the middle section, used for controlled vehicles to adjust their longitudinal speed. The vehicle merging zone is located at the downstream end, near the merging nose of the ramp, allowing ramp vehicles and controlled vehicles to pass sequentially.

[0197] like Figure 2 As shown, Figure 2 The overall flowchart of the vehicle cooperative control method is as follows, and the specific implementation steps are as follows: Implementation Method 1: Control Area and Traffic Status Data Collection The main road is divided into the Vehicle Preparation Area (VPA), Vehicle Control Area (VCA), and Vehicle Merging Area (VMA) along the direction of vehicle travel.

[0198] The moment when the ramp vehicle detector detects a vehicle about to merge through the ramp is recorded as . .

[0199] The moment when the controlled CAV enters the Vehicle Control Area (VCA) and begins longitudinal speed optimization is denoted as [missing information]. .

[0200] The moment when the accused CAV leaves the Vehicle Control Area (VCA) and enters the Vehicle Merging Area (VMA) is recorded as .

[0201] The length of the Vehicle Preparation Area (VPA) is set according to the maximum permissible vehicle speed on the road and the preset maximum lane change execution time, so that candidate controlled vehicles (CAVs) that meet the lateral lane change safety constraints can complete the lateral lane change before entering the Vehicle Control Area (VCA).

[0202] Roadside sensing devices include one or more of camera devices, millimeter-wave radar, and lidar. The roadside unit acquires vehicle status data of the CAV via wireless communication and synchronizes the roadside sensing data with the vehicle communication data in time.

[0203] For multiple position, speed and acceleration detection results of the same vehicle, they are merged according to a unified vehicle identification and the merged results are converted to a unified road longitudinal coordinate system.

[0204] Traffic condition data sampling period Set to 0.1 s Control cycle Set to 0.2 s .

[0205] Preset minimum stationary distance Set to 5 m Preset safe headway h Set to 1.5 sPreset safe braking deceleration amplitude Set to 3 Preset maximum allowable import speed difference Set to 5 .

[0206] Preset acceleration determination threshold Set to 0.05 Potential collision probability safety threshold Set to 0.05, lateral lane change prediction time step Set to 0.2 s .

[0207] The above parameters are pre-calibrated based on road grade, road design speed, vehicle composition, and the accuracy of sensing equipment.

[0208] Example 1: Low-permeability, high-density ramp traffic scenario This embodiment applies to the merging area of ​​highway ramps. The CAV penetration rate of the merging area is... The average longitudinal speed of the mainline traffic flow is 30%. 30 Average longitudinal speed of ramp traffic 10 Longitudinal distance from ramp vehicle detector to merging point For 500 m .

[0209] Real-time queue length of ramps For 400 m The maximum inflow gap that a single controlled CAV can provide 150 m Maximum number of vehicles under control The value is 3.

[0210] The system predicts the natural merging gap in the outermost lane of the mainline when vehicles on the ramp arrive at the merging point. Calculations show that the predicted longitudinal clearance between the ramp vehicle and the adjacent vehicle ahead is less than the corresponding safe longitudinal clearance; therefore, there is currently no natural merging gap that meets the natural merging conditions.

[0211] The number of controlled vehicles is determined according to the following formula:

[0212] The range of controlled vehicles can be calculated using the following formula:

[0213] The system is located 1500 meters upstream of the inlet point. m Within the specified range, priority will be given to CAVs that are already located in the outermost lane of the main line.

[0214] Two of the CAVs were already in the outermost lane of the main line and met the longitudinal control feasibility constraints, so they were directly identified as controlled CAVs.

[0215] The third candidate vehicle, a CAV, was located in the lane adjacent to the outermost lane of the main line.

[0216] The third candidate was charged with CAV at the time. Remaining longitudinal distance to the downstream boundary of the vehicle preparation area VPA The longitudinal speed is 600m. 30 Observing longitudinal acceleration 0 .

[0217] therefore:

[0218] In this embodiment, the lateral movement distance W The distance is 3.75m, and the weighting coefficients of the lateral lane change objective function are set as follows: ; ; The lower limit of the lateral acceleration amplitude is:

[0219] The system acquires the status data of vehicles in front of and behind the candidate controlled CAV in the current lane and vehicles in front of and behind the outermost lane of the main line, and calculates the collision risk probability of each adjacent vehicle pair during the lateral lane change execution according to the prediction time step of 0.2s.

[0220] The optimal lateral acceleration amplitude was obtained through optimization. The optimal lateral acceleration amplitude corresponds to a lateral lane change execution time of less than 20 seconds, and satisfies the following: Therefore, the third candidate controlled CAV was identified as the controlled CAV and controlled to complete the lateral lane change.

[0221] because: System settings: , Prioritize vertical optimization to ensure safety objectives.

[0222] Because the traffic density on the ramp was higher than the critical density of the ramp, all three controlled CAVs adopted the first operating condition, continuously decelerating under the safety constraint of the uncontrolled vehicles following behind, thus forming three dynamic merging gaps in sequence.

[0223] The system updates the position, longitudinal velocity, and longitudinal acceleration of the three controlled CAVs behind them within each 0.2s control cycle. If any uncontrolled vehicle behind accelerates and causes the predicted following distance to decrease, the longitudinal deceleration amplitude of the corresponding controlled CAV is recalculated.

[0224] Ramp vehicles are divided into three groups according to their queuing order and enter the main line through three dynamic merging gaps. After all ramp vehicles have left the vehicle merging area (VMA), the three controlled CAVs resume normal driving according to the acceleration rate of change constraint.

[0225] Example 2: Scenario with moderate penetration rate and moderate ramp traffic density This embodiment applies to the merging area of ​​urban expressway ramps. The CAV penetration rate of the merging area is... The average longitudinal speed of the mainline traffic flow is 50%. 25 Average longitudinal speed of ramp traffic 8 Longitudinal distance from ramp vehicle detector to merging point It is 400m.

[0226] Real-time queue length of ramps The maximum inflow gap that a single controlled CAV can provide is 120m. It is 150m.

[0227] because: ;Sure: The scope of vehicles to be controlled is as follows: ; The system predicts that there are currently no natural inflow gaps that meet the natural inflow conditions.

[0228] Within the range of controlled vehicles, the CAV closest to the merging point is already located in the outermost lane of the main line, and the predicted longitudinal clearance between the uncontrolled vehicle behind it and the CAV meets the car-following safety constraint. Therefore, the CAV is directly identified as the controlled CAV, and no lateral lane change is performed.

[0229] Since the traffic density of the ramp is lower than the critical density of the ramp, operating condition two is adopted.

[0230] Since both CAV penetration and ramp traffic density are at a moderate level, the following settings are adopted: , ; The optimal longitudinal deceleration amplitude is obtained by solving the system. and deceleration completion time .

[0231] Accused of CAV from the moment At that time Reduce speed, the corresponding reduction distance is:

[0232] Accused of CAV from the moment At that time Traveling at a constant speed, the corresponding distance traveled at a constant speed is:

[0233] The system calculates the dynamic merging gap based on the remaining longitudinal distance from the controlled CAV to the merging point, and ensures that the dynamic merging gap is not less than the preset minimum safe merging gap.

[0234] Vehicles queuing at the ramp enter the main line through the dynamic merging gap. After merging, the controlled CAV smoothly resumes its normal driving speed.

[0235] Example 3: A scenario with high permeability, low density, and natural inflow gaps. This embodiment is applied to a smart connected demonstration road. The CAV penetration rate in the ramp merging area... The average longitudinal speed of the mainline traffic flow is 80%. 28 Average longitudinal speed of ramp traffic 12 Longitudinal distance from ramp vehicle detector to merging point It is 300m.

[0236] Vehicles on the ramp are in free-flow state; real-time queue length of the ramp. It is 20m.

[0237] The system is at the detection time The system acquires the position, longitudinal velocity, and observed longitudinal acceleration of vehicles on ramps and in the outermost lane of the mainline, and predicts the mainline traffic conditions when vehicles on ramps arrive at the merging point.

[0238] Calculations show that the predicted longitudinal clearance between vehicles on the ramp and adjacent vehicles on the mainline satisfies the following conditions:

[0239] The predicted longitudinal clearance between adjacent vehicles behind the mainline and vehicles on the ramp meets the following requirements:

[0240] The predicted speed difference between vehicles on the ramp and vehicles on the main line before and after it is no greater than the preset maximum allowable merging speed difference.

[0241] Therefore, it is determined that there is a natural merging gap between adjacent vehicles in front and behind on the main line that meets the natural merging conditions.

[0242] The system does not select controlled CAVs, nor does it control any mainline CAVs to perform lateral lane changes or longitudinal deceleration. Ramp vehicles directly utilize the natural merging gap to complete the merging.

[0243] This embodiment demonstrates that when a vehicle on a ramp has met the conditions for safe merging, the present invention does not force coordinated control due to the presence of a controllable CAV in the main line, thereby avoiding additional lane changes and deceleration operations.

[0244] Example 4: Scenario of Re-establishing a Natural Merging Gap During Control In this embodiment, the system detects at the time... No natural merging gap meeting the natural merging condition was detected, and a candidate controlled CAV in the adjacent lane was selected.

[0245] The candidate controlled CAV satisfies the lateral lane change safety constraint, and the system sends a lateral lane change control command to it.

[0246] After the candidate lane-changing vehicle (CAV) began to change lanes laterally, the vehicles ahead on the main line accelerated away, creating a new natural merging gap in the outermost lane of the main line.

[0247] The system re-predicts the forward longitudinal clearance, backward longitudinal clearance, and merging speed difference when vehicles arrive at the merging point on the ramp based on the updated traffic conditions, and determines that the newly formed gap meets the natural merging conditions.

[0248] Since the controlled CAV has already begun to change lanes laterally, the system does not control the vehicle to return to the original lane midway through the lane change. Instead, it controls the vehicle to complete the current lateral lane change according to the established safe lateral lane change trajectory.

[0249] After the controlled CAV completes the lateral lane change, the system no longer controls it to perform the longitudinal deceleration operation used to widen the merging gap.

[0250] Vehicles on the ramps use the newly formed natural merging gaps to enter the main line.

[0251] When a natural merging gap forms before the controlled CAV begins a lateral lane change, the system cancels the lateral lane change command.

[0252] When the natural merging gap forms during longitudinal deceleration of the controlled CAV, the system operates according to the maximum permissible rate of acceleration change. Gradually reduce the longitudinal deceleration amplitude so that the controlled CAV can smoothly return to normal driving status.

[0253] The above embodiments demonstrate that the present invention can continuously determine whether control operations are still necessary based on changes in traffic conditions, and select the corresponding exit method according to the control stage in which the controlled CAV is located.

[0254] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for cooperative vehicle control in a ramp merging area under a hybrid intelligent connected environment, characterized in that: Includes the following steps: S1, acquire traffic state data, which includes the position, longitudinal speed, longitudinal acceleration, lane and vehicle length of vehicles on the main line, the position, longitudinal speed, longitudinal acceleration and vehicle length of vehicles on the ramp, as well as the average longitudinal speed of the main line traffic flow, the average longitudinal speed of the ramp traffic flow, the ramp traffic density, the real-time queue length of the ramp and the penetration rate of connected autonomous vehicles, wherein the connected autonomous vehicles are connected autonomous vehicles (CAVs). S2, when the ramp vehicle detector detects a vehicle waiting to merge onto the ramp, based on the position, longitudinal velocity, and longitudinal acceleration of the vehicle waiting to merge onto the ramp and the vehicles in the outermost lane of the main line adjacent to the ramp and used to receive the merging of the ramp vehicle, predict the natural merging gap between adjacent vehicles in the outermost lane of the main line when the vehicle waiting to merge onto the ramp arrives at the merging point, and determine whether the natural merging gap simultaneously satisfies the forward safety distance constraint, the backward safety distance constraint, and the merging speed difference constraint; When there is a natural merging gap that simultaneously satisfies the forward safety distance constraint, the backward safety distance constraint, and the merging speed difference constraint, the controlled vehicle is not selected, and no lateral lane change control or longitudinal deceleration control is implemented for the mainline vehicle, so that the vehicle waiting to merge on the ramp can use the natural merging gap to merge into the outermost lane of the mainline. S3, when there is no natural merging gap that simultaneously satisfies the forward safety distance constraint, the backward safety distance constraint, and the merging speed difference constraint, determine the number of controlled vehicles and the selection range of controlled vehicles. Within the selection range of controlled vehicles, prioritize selecting controlled CAVs from those already located in the outermost lane of the main line and satisfying the longitudinal control feasibility constraint. The longitudinal control feasibility constraint includes the speed constraint of the controlled CAV, the acceleration constraint of the controlled CAV, the longitudinal safety distance constraint between the controlled CAV and the adjacent vehicle in front, and the following safety constraint of uncontrolled vehicles behind the controlled CAV. When the number of CAVs satisfying the longitudinal control feasibility constraint in the outermost lane of the main line is less than the number of controlled vehicles, candidate controlled CAVs are selected from the lanes adjacent to the outermost lane of the main line. S4. For a candidate controlled CAV not located in the outermost lane of the main line, obtain the position, longitudinal velocity, longitudinal acceleration, and vehicle length of the candidate controlled CAV, the vehicles in front and behind the candidate controlled CAV in its current lane, and the vehicles in front and behind the outermost lane of the main line. Calculate the available lane-changing time based on the remaining longitudinal distance from the candidate controlled CAV to the downstream boundary of the Vehicle Preparation Area (VPA) and the longitudinal velocity and longitudinal acceleration of the candidate controlled CAV. Calculate the potential collision probability based on the predicted longitudinal net distance, predicted safe longitudinal distance, and vehicle state prediction error between the candidate controlled CAV and each of the vehicles in front and behind. Only when the lateral lane change execution time is not greater than the available lane-changing time, the lateral acceleration satisfies the lateral dynamics constraint, and the potential collision probability is not greater than a preset safety threshold, is the candidate controlled CAV identified as a controlled CAV, and the controlled CAV is controlled to change lanes to the outermost lane of the main line within the Vehicle Preparation Area (VPA). If the candidate controlled CAV does not meet any of the lateral lane change execution time constraints, lateral dynamic constraints, and potential collision probability constraints, the candidate controlled CAV will not be determined as a controlled CAV, and the next candidate controlled CAV will be judged for lateral lane change safety constraints in order of distance from the merging point from near to far. For a controlled CAV that is already in the outermost lane of the main line, control the controlled CAV to maintain its current lane; S5. Before the controlled CAV begins to change lanes laterally, during the lateral lane change, and after entering the Vehicle Control Area (VCA), the system re-determines whether the vehicle to be merged on the ramp has the conditions for natural merging according to a preset control cycle. When no natural merging gap satisfying the natural merging condition is continuously present, the controlled CAV is controlled to perform longitudinal speed optimization within the Vehicle Control Area (VCA), and the moment when the controlled CAV enters the VCA and begins longitudinal speed optimization is obtained. Corresponding ramp traffic density It also obtains the critical ramp density pre-calibrated based on the historical ramp traffic density and historical ramp queue length. ,when When the controlled CAV continuously decelerates throughout the entire process, in condition one, when... In case 2, the controlled CAV is first decelerated and then driven at a constant speed. An adaptive multi-objective optimization model is constructed with the objectives of minimizing the normalized potential conflict index and minimizing the normalized longitudinal optimization time. Under the constraints of controlled CAV acceleration, controlled CAV speed, acceleration change rate, longitudinal safety distance between the controlled CAV and the adjacent vehicle in front, safety constraint of uncontrolled vehicles following the controlled CAV, and dynamic merging gap not less than the preset minimum safe merging gap, the non-negative optimal longitudinal deceleration amplitude is solved, and the dynamic merging gap is calculated based on the optimal longitudinal deceleration amplitude. If there is no longitudinal deceleration amplitude that simultaneously satisfies all the constraints, cancel the longitudinal deceleration control of the current controlled CAV and reselect the controlled CAV or continue to monitor the natural merging gap. S6, control the controlled CAV and the vehicle to be merged on the ramp to pass through the vehicle merging area VMA in sequence, so that the vehicle to be merged on the ramp merges into the outermost lane of the main line in front of the corresponding controlled CAV. When there are multiple controlled CAVs, each controlled CAV forms a dynamic merging gap in order of distance from the merging point to the farthest point. Each vehicle waiting to merge into the merging ramp merges into the outermost lane of the main line in the order of predicted arrival at the merging point using the corresponding dynamic merging gap. Once a vehicle on the ramp has completed its merging and left the Vehicle Merging Area (VMA), longitudinal control of the controlled CAV corresponding to that vehicle on the ramp is terminated, and the controlled CAV is restored to its normal driving state.

2. The method for coordinated vehicle control in the merging area of ​​a ramp under a hybrid intelligent connected environment according to claim 1, characterized in that, The process of predicting the natural inflow gap and determining whether the natural inflow gap meets the natural inflow conditions specifically includes: The vehicles waiting to merge onto the ramp are referred to as ramp vehicles. The ramp vehicle detector detects ramp vehicles. The moment of passing is recorded as the detection moment. Obtain ramp vehicles At the time of detection Longitudinal distance to the sink point Longitudinal velocity and observation of longitudinal acceleration ; when and At that time, the predicted travel time for vehicles on the ramp to reach the merging point is calculated according to the following formula. : when and When it is determined that a valid natural merging prediction result cannot be formed at present, the updated status of ramp vehicles will continue to be acquired according to the preset control cycle; among which, To preset the minimum effective ramp vehicle speed; when Then, the predicted travel time is calculated using the following formula. : The smallest positive root satisfying the above formula is determined as the predicted driving time. ,in, The preset acceleration determination threshold is used; When the value within the square root of the longitudinal motion equation is less than zero, the longitudinal motion equation does not have a positive root, or it is determined from the predicted motion state that the ramp vehicle r cannot reach the merging point before stopping, it is determined that a valid natural merging prediction result cannot be formed at present. Therefore, the controlled vehicle selection step is not entered within the current control cycle, and the acquisition of ramp vehicles continues according to the preset control cycle. r Position, longitudinal velocity, and observed longitudinal acceleration; when the position of the ramp vehicle can be determined based on the updated vehicle state. r When the predicted travel time to the merging point is reached, the natural merging gap is re-predicted; only when a valid natural merging prediction result can be formed and there is no natural merging gap that meets the natural merging conditions, the controlled vehicle selection step is entered. Will The predicted merging time is determined, and the vehicle is calculated according to the following formula. j Predicting the inflow time Predicted longitudinal position and predicting longitudinal velocity : Among them, vehicles j For ramp vehicles Or vehicles in the outermost lane of the main line, , and vehicles j At the time of detection The longitudinal position, longitudinal velocity, and observed longitudinal acceleration; Determine the predicted merging time from the outermost lane of the main line. Vehicles located on the ramp The vehicle in front of you f and vehicles located on ramps The vehicle behind and adjacent to the rear b Calculate the number of vehicles on the ramp according to the following formula. Adjacent vehicle in front f Predicted longitudinal net spacing between and adjacent vehicles behind b With ramp vehicles Predicted longitudinal net spacing between : in, , and The vehicles adjacent to each other in front f Ramp vehicles and adjacent vehicles b The length of the vehicle; The following formula is used to calculate the following vehicles. Relative to the vehicle in front Safety longitudinal spacing : in, Indicates following vehicles. Indicates the location of the following vehicle The vehicles ahead, To preset the minimum stationary distance, To preset a safe headway, To preset the safe braking deceleration amplitude; When both conditions are met , as well as At that time, determine the adjacent vehicles ahead. f and adjacent vehicles b There exist natural inflow gaps that satisfy the aforementioned natural inflow conditions, wherein, This is the preset maximum allowed import speed difference.

3. The method for coordinated vehicle control in the merging area of ​​a ramp under a hybrid intelligent connected environment according to claim 1, characterized in that, The determination of the number of controlled vehicles and the selection range of controlled vehicles specifically includes: Get real-time queue length of ramps The maximum inflow gap that a single controlled CAV on the main line can provide. and the maximum number of vehicles under control The maximum inflow gap The maximum dynamic merging gap that a single controlled CAV can form is defined under the following constraints: longitudinal length of the vehicle control area (VCA), maximum permissible longitudinal deceleration of the controlled CAV, minimum permissible longitudinal speed of the controlled CAV, longitudinal safety distance constraint between the controlled CAV and the vehicle in front, and safety constraint for uncontrolled vehicles following behind the controlled CAV. when At that time, determine the number of vehicles under control. for: ; when At that time, the number of controlled vehicles shall be determined according to the following formula. : ; in, This indicates the rounding up operation; The range of controlled vehicles is calculated using the following formula. : in, To preset the minimum effective average longitudinal speed of ramp traffic flow, The average longitudinal speed of the main line traffic flow The average longitudinal speed of the traffic flow on the ramp. The longitudinal distance from the ramp vehicle detector to the merging point, and within the selected range of the controlled vehicles. The inner parts are selected in order of distance from the ingress point to the nearest point. A CAV charged with a crime or a candidate for a crime; When the number of controlled vehicles When the value is greater than 1, each controlled CAV is recorded as a vehicle in order of distance from the merging point from closest to furthest. ,vehicle ,……,vehicle The longitudinal deceleration amplitude of each controlled CAV is solved sequentially according to the order stated therein. For any controlled CAV located behind another controlled CAV, the predicted driving state of the nearest vehicle in front of it is taken as the state of the vehicle in front, so that the longitudinal safety distance constraint is satisfied between adjacent controlled CAVs.

4. The method for coordinated vehicle control in the merging area of ​​a ramp under a hybrid intelligent connected environment according to claim 1, characterized in that, The determination of lateral lane change safety constraints for candidate controlled CAVs not located in the outermost lane of the mainline specifically includes: The candidate CAV will be designated as a vehicle. c Obtain vehicle c At the time of candidate vehicle selection Remaining longitudinal distance to the downstream boundary of the vehicle preparation area VPA Longitudinal velocity and observation of longitudinal acceleration ; when and At that time, the available lane-changing time is calculated according to the following formula. : in, The preset acceleration determination threshold; when When the candidate controlled CAV does not meet the lane-changing available time constraint, it is determined that the candidate controlled CAV does not meet the lane-changing available time constraint. when At that time, the available lane-changing time is calculated according to the following formula. : When the value inside the square root is not less than zero, the vehicle condition will be satisfied. c The minimum positive value of the longitudinal motion equation is determined as the available lane-changing time. When the value within the square root is less than zero, the result does not contain a positive value, or the prediction result indicates that vehicle c has stopped before reaching the downstream boundary of the vehicle preparation area VPA, the vehicle is determined to be... c If the available time constraint for lane changing is not met, vehicle c will not be identified as a controlled CAV, and the lateral lane change safety constraint will be determined for the next candidate controlled CAV in order of distance from the merging point from closest to furthest. The lateral lane change process is divided into a lateral acceleration phase and a lateral deceleration phase. The duration of the lateral acceleration phase and the lateral deceleration phase are equal and the directions of lateral acceleration are opposite. The lateral acceleration amplitude is used as the metric. As the decision variable, the lateral lane change execution time is calculated according to the following formula. : in, For vehicles c The lateral movement distance required to move from the current lane center to the outermost lane center of the main line; Construct the lateral lane change objective function: in, Let be the objective function for lateral lane changing. and Let be the weight coefficients of the lateral lane change objective function, and satisfy . , g It is the acceleration due to gravity; Solve for the optimal lateral acceleration magnitude under the following constraints. : in, For potential collision probability, Preset safety threshold; When there exists an optimal lateral acceleration amplitude that satisfies the aforementioned constraints. At that time, the optimal lateral acceleration amplitude will be... The duration of the lateral acceleration phase, the duration of the lateral deceleration phase, and the corresponding lateral acceleration direction are sent to the chassis control system of the controlled CAV.

5. The method for coordinated vehicle control in the merging area of ​​a ramp under a hybrid intelligent connected environment according to claim 4, characterized in that, The potential collision probability The calculation specifically includes: When there is a vehicle ahead of the candidate controlled CAV in the current lane within the preset effective perception range, the candidate controlled CAV and the vehicle ahead in the current lane are combined to form a first adjacent vehicle pair. When there is a vehicle behind the current lane within the preset effective sensing range, the vehicle behind the current lane and the candidate controlled CAV are combined to form a second adjacent vehicle pair; When there is a vehicle ahead in the outermost lane of the main line within the preset effective sensing range, the candidate controlled CAV and the vehicle ahead in the outermost lane of the main line are combined to form a third adjacent vehicle pair. When there is a vehicle behind the outermost lane of the main line within the preset effective sensing range, the vehicle behind the outermost lane of the main line and the candidate controlled CAV are combined to form a fourth adjacent vehicle pair. The actual adjacent vehicle pairs formed are combined into an adjacent vehicle pair set. When there is no corresponding vehicle in any direction or the corresponding vehicle is outside the preset effective sensing range, no adjacent vehicle pair is established for that direction; when the set of adjacent vehicle pairs... When the set is empty, the potential collision probability is set to 0. For the set of adjacent vehicle pairs Adjacent vehicle pairs , will the vehicle The vehicle was identified as a follower and was put into service. It has been identified as the vehicle ahead. According to the preset prediction time step Set the prediction time during the lateral lane change execution and the lateral lane change end time. Set as the last prediction time: in: ; The following formula is used to calculate adjacent vehicle pairs. China Vehicle j At the predicted time Predicted longitudinal velocity: Among them, vehicles j For following vehicles or the vehicle ahead ; The following formula is used to calculate adjacent vehicle pairs. At the predicted time The predicted mean longitudinal net spacing : in, For following vehicles Vehicle in front At the time of candidate vehicle selection The initial longitudinal net spacing; The following formula is used to calculate the following vehicles. Relative to the vehicle in front At the predicted time Safety longitudinal spacing : The following formula is used to calculate adjacent vehicle pairs. Longitudinal net spacing prediction standard deviation : in, , and Adjacent vehicle pairs The standard deviation of the relative spacing prediction error, the standard deviation of the relative velocity prediction error, and the standard deviation of the relative acceleration prediction error; The following formula is used to calculate adjacent vehicle pairs. At the predicted time Collision risk probability : in, The cumulative distribution function of the standard normal distribution; The potential collision probability is determined by the maximum value among the collision risk probabilities of all adjacent vehicle pairs and all predicted times, according to the following formula. : 。 6. The method for coordinated vehicle control in the merging area of ​​a ramp under a hybrid intelligent connected environment according to claim 1, characterized in that, The objective function of the adaptive multi-objective optimization model is: in, To optimize the objective function for longitudinal velocity, This represents the proportion of CAVs (Caravans and Aerial Vehicles) within the statistical scope of the ramp merging area to the total number of vehicles. The moment when the controlled CAV enters the vehicle control area (VCA) and begins longitudinal speed optimization. The moment when the accused CAV leaves the Vehicle Control Area (VCA) and enters the Vehicle Merging Area (VMA). To normalize the potential conflict indicator, To optimize the baseline time vertically; The vertical optimization reference time Calculate according to the following formula: in, This is the longitudinal length of the vehicle control area (VCA). For the accused CAV at the moment longitudinal velocity, The minimum effective longitudinal velocity of the controlled CAV is preset; The vehicle pairs during the longitudinal optimization prediction period will be grouped into a longitudinal vehicle pair set. The original number of potential conflict vehicle pairs is calculated according to the following formula. : in, This is an indicator function; it takes the value 1 when the condition inside the parentheses is true, and takes the value 0 when the condition inside the parentheses is false. The normalized potential conflict index is calculated according to the following formula. : in, For longitudinal vehicle pairs The number of vehicles in the middle; and The weights are adaptive weights and satisfy the following conditions: The adaptive weighting coefficients are determined in the following order: First set of conditions: When or At that time, it was determined that: ; The second set of conditions: When the first set of conditions is not met, but the following conditions are met. or Determine if at least one of the following conditions is met: ; When neither the first set of conditions nor the second set of conditions mentioned above are met, it is determined that: ; The accused CAV will be recorded as a vehicle. c The nearest vehicle located in front of the controlled CAV and in the outermost lane of the main line is recorded as the adjacent vehicle in front. The CAV being charged will be in contact with the vehicle adjacent to it. Form a target longitudinal vehicle pair; Based on the longitudinal deceleration amplitude of the controlled CAV, and the relationship between the controlled CAV and the adjacent vehicle in front. At any moment Based on the longitudinal velocity, observed longitudinal acceleration, and initial longitudinal clearance, the target longitudinal vehicle pair is calculated at the predicted time. Predicted longitudinal net spacing ; The controlled CAV is calculated relative to the adjacent vehicle in front using the following formula. Safe longitudinal spacing: And ensure that the longitudinal deceleration magnitude satisfies the following during longitudinal optimization: When there are no vehicles in front of the controlled CAV within the preset effective perception range, no longitudinal safety distance constraint is set between the controlled CAV and the adjacent vehicle in front. By solving the longitudinal velocity optimization objective function, the non-negative optimal longitudinal deceleration amplitude is obtained. The actual longitudinal acceleration control command corresponding to the controlled CAV is: .

7. The method for coordinated vehicle control in the merging area of ​​a ramp under a hybrid intelligent connected environment according to claim 6, characterized in that, The calculation of the following uncontrolled vehicle safety constraint and the dynamic merging gap specifically includes: Manually driven vehicles or vehicles not directly controlled that are located behind the controlled CAV and in the outermost lane of the main road are recorded as uncontrolled vehicles behind it. u With longitudinal deceleration amplitude For the longitudinal velocity optimization variable, calculate the uncontrolled vehicles behind according to the following formula. u Predicted net longitudinal spacing between the controlled CAV and the controlled CAV during longitudinal optimization : in, For uncontrolled vehicles behind u With the accused CAV at the moment The initial longitudinal net spacing, and Uncontrolled vehicles behind u At any moment The longitudinal velocity and observed longitudinal acceleration; The following formula is used to calculate uncontrolled vehicles behind. u The safe longitudinal spacing relative to the controlled CAV : and make the longitudinal deceleration amplitude During vertical optimization, the following conditions must be met: ; When the following safety constraint for uncontrolled vehicles is not met, the following is based on the uncontrolled vehicles behind. u The longitudinal deceleration amplitude is recalculated based on the updated position, longitudinal velocity, and longitudinal acceleration. If there is no longitudinal deceleration amplitude that satisfies the following safety constraint of the uncontrolled vehicle behind, the dynamic merging gap formed by the current controlled CAV is cancelled. The CAV will be charged at the time The longitudinal distance to the sink point is denoted as ; Under operating condition 1, the controlled CAV is calculated from time 1 according to the following formula. At that time Longitudinal driving distance : The dynamic inflow gap is calculated according to the following formula. : Under operating condition two, the moment when the controlled CAV completes deceleration is recorded as (the time when the controlled CAV completes deceleration is then recorded as ). The controlled CAV is calculated from time t according to the following formula. At that time Deceleration distance The accused CAV at the moment longitudinal velocity And the accused CAV from the moment At that time constant speed travel distance : The dynamic inflow gap is calculated according to the following formula. : in, For a moment The corresponding dynamic safety distance, and satisfies: The resulting dynamic inflow gap satisfy: in, To preset the minimum safe entry gap; The dynamic safety distance is used to limit the remaining positional margin between the controlled CAV and the merging point, and the safety longitudinal spacing is used to limit the following safety between the following vehicle and the vehicle in front. The two are applicable to different safety constraint objects.

8. The method for coordinated vehicle control in the merging area of ​​a ramp under a hybrid intelligent connected environment according to claim 6, characterized in that, The step of re-determining whether the vehicle to be merged on the ramp meets the conditions for natural merging according to a preset control cycle specifically includes: When a natural merging gap that meets the natural merging conditions is detected before the controlled CAV begins a lateral lane change, the lateral lane change control and longitudinal deceleration control of the controlled CAV are cancelled. When a natural merging gap that meets the natural merging conditions is detected after the controlled CAV has started a lateral lane change, the controlled CAV is controlled to complete the current lateral lane change according to the determined lateral lane change trajectory, and the longitudinal deceleration control used to widen the merging gap after the controlled CAV completes the lateral lane change is canceled. When a natural merging gap satisfying the natural merging condition is detected during longitudinal deceleration control of the controlled CAV, the rate of change of the longitudinal acceleration control command of the controlled CAV is limited according to the following formula: And under the condition of satisfying the longitudinal acceleration control command rate of change constraint, gradually reduce the longitudinal deceleration amplitude of the controlled CAV, so that the controlled CAV returns to normal driving state, wherein, For the accused CAV at the moment t Longitudinal acceleration control commands, To preset the control cycle, This is the preset maximum allowable rate of change of acceleration.

Images