CN110570049B - A low-level control method for synergistic optimization of highway mixed traffic flow convergence - Google Patents

A low-level control method for synergistic optimization of highway mixed traffic flow convergence Download PDF

Info

Publication number
CN110570049B
CN110570049B CN201910886980.2A CN201910886980A CN110570049B CN 110570049 B CN110570049 B CN 110570049B CN 201910886980 A CN201910886980 A CN 201910886980A CN 110570049 B CN110570049 B CN 110570049B
Authority
CN
China
Prior art keywords
vehicle
time
confluence
vehicles
control
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201910886980.2A
Other languages
Chinese (zh)
Other versions
CN110570049A (en
Inventor
孙湛博
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Southwest Jiaotong University
Original Assignee
Southwest Jiaotong University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Southwest Jiaotong University filed Critical Southwest Jiaotong University
Priority to CN201910886980.2A priority Critical patent/CN110570049B/en
Publication of CN110570049A publication Critical patent/CN110570049A/en
Application granted granted Critical
Publication of CN110570049B publication Critical patent/CN110570049B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/40—Business processes related to the transportation industry
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G08G1/052—Detecting movement of traffic to be counted or controlled with provision for determining speed or overspeed
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/123—Traffic control systems for road vehicles indicating the position of vehicles, e.g. scheduled vehicles; Managing passenger vehicles circulating according to a fixed timetable, e.g. buses, trains, trams

Landscapes

  • Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Strategic Management (AREA)
  • Human Resources & Organizations (AREA)
  • Economics (AREA)
  • General Business, Economics & Management (AREA)
  • Marketing (AREA)
  • Theoretical Computer Science (AREA)
  • Tourism & Hospitality (AREA)
  • Quality & Reliability (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Chemical & Material Sciences (AREA)
  • Operations Research (AREA)
  • Development Economics (AREA)
  • Game Theory and Decision Science (AREA)
  • Remote Sensing (AREA)
  • Analytical Chemistry (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Primary Health Care (AREA)
  • Control Of Driving Devices And Active Controlling Of Vehicle (AREA)
  • Traffic Control Systems (AREA)

Abstract

The invention relates to a conflux cooperative optimization bottom layer control method for mixed traffic flow of a highway, belonging to the field of traffic engineering. The method comprises the following steps: determining a microscopic following model; predicting an initial track of the vehicle; establishing a confluence model; simulating a cooperative control strategy set; judging whether the vehicles can smoothly complete confluence; if the judgment result is that the vehicle can smoothly complete the confluence, the vehicle continues to run according to the speed of the microscopic follow-up model; if the judgment result is that the vehicles can not smoothly complete the confluence, the specific situation of the vehicles in the confluence process needs to be further judged, and a corresponding cooperative control strategy is made according to the specific situation; and optimizing the running track of the target vehicle according to the made cooperative control strategy to obtain the cooperative optimization control strategy related to the target vehicle, and acting the cooperative optimization control strategy on the target vehicle to control the running of the target vehicle. The method can ensure that the vehicles run stably at a higher speed, thereby improving the vehicle traffic capacity of the expressway ramps in the confluence area.

Description

一种高速公路混合交通流汇流协同优化底层控制方法A low-level control method for synergistic optimization of highway mixed traffic flow convergence

技术领域technical field

本发明涉及一种高速公路混合交通流汇流协同优化底层控制方法，属于交通工程领域。The invention relates to a bottom layer control method for synergistic optimization of mixed traffic flow confluence of expressways, and belongs to the field of traffic engineering.

背景技术Background technique

高速公路入口匝道作为整个高速公路系统的一个交通需求输入环节，也是拥堵容易产生的一个环节，对于整个高速公路系统的流畅平稳运行有十分重要的意义。随着智能网联汽车的出现与发展，未来的高速公路将会面临智能网联汽车与传统人类驾驶车辆混行的交通状况。智能网联汽车与传统人类驾驶车辆混合交通流环境下的决策控制是未来交通出行中需要长期面临的现实问题。因此，研究混合交通流环境下高速公路匝道汇流优化控制具有重要意义。As a traffic demand input link of the entire highway system, the highway on-ramp is also a link that is prone to congestion, and is of great significance to the smooth and stable operation of the entire highway system. With the emergence and development of intelligent networked vehicles, future highways will face a traffic situation where intelligent networked vehicles and traditional human-driven vehicles are mixed together. Decision control in the mixed traffic flow environment of intelligent networked vehicles and traditional human-driven vehicles is a long-term realistic problem that needs to be faced in future traffic travel. Therefore, it is of great significance to study the optimal control of expressway ramp convergence under mixed traffic flow environment.

对于高速公路匝道协同汇流优化控制问题，已有一些模型方法被提出，但现阶段所研究的决策控制方法多为假设自动驾驶车辆渗透率为100％的交通环境设计的决策方法，且仅都是在特定场景下讨论，未能全面地描述自动驾驶车辆与人类驾驶车辆组合排序场景且给出对应的轨迹优化方案。For the optimal control problem of synergistic convergence of expressway ramps, some model methods have been proposed, but most of the decision-making control methods studied at this stage are decision-making methods for the design of the traffic environment assuming that the penetration rate of autonomous vehicles is 100%, and only all of them are Discussed in specific scenarios, it fails to comprehensively describe the combination and sorting scenarios of autonomous vehicles and human-driven vehicles and give the corresponding trajectory optimization scheme.

发明内容SUMMARY OF THE INVENTION

将车辆轨迹优化控制问题定义为底层问题。该底层问题的求解需要根据场景以及优化目标来考虑具体的约束条件，包括道路几何约束、安全约束、车辆类型约束。本发明为解决在人类驾驶车辆(即传统驾驶车辆)和自动驾驶车辆(即智能网联车辆)混行交通状态下高速公路匝道车流汇入主干道时存在的该底层问题，提出一种高速公路混合交通流汇流协同优化底层控制方法。人类驾驶车辆为不可优化控制的车辆，自动驾驶车辆为可优化控制的车辆。The vehicle trajectory optimization control problem is defined as a low-level problem. The solution of the underlying problem needs to consider specific constraints according to the scene and optimization objectives, including road geometry constraints, safety constraints, and vehicle type constraints. In order to solve the underlying problem when the traffic flow of the expressway ramp merges into the main road in the mixed traffic state of human-driven vehicles (that is, traditional driving vehicles) and automatic-driving vehicles (that is, intelligent networked vehicles), the present invention proposes an expressway Synergistic optimization of the underlying control method for the convergence of mixed traffic flows. A human-driven vehicle is a vehicle that cannot be optimally controlled, and an autonomous vehicle is a vehicle that can be optimally controlled.

本发明为实现上述发明目的所采取的技术方案如下：The technical scheme adopted by the present invention for realizing the above-mentioned purpose of the invention is as follows:

一种高速公路混合交通流汇流协同优化底层控制方法，包括步骤：A low-level control method for synergistic optimization of highway mixed traffic flow convergence, comprising the steps of:

S1、确定微观跟驰模型，用所述微观跟驰模型描述车辆的跟驰状态，所述车辆的跟驰状态包括车辆的速度、加速度和位置；S1. Determine a microscopic car-following model, and use the microscopic car-following model to describe the car-following state of the vehicle, and the car-following state of the vehicle includes the speed, acceleration and position of the vehicle;

S2、获取混合交通流中车辆在通过汇流区域前的上游监测点的时刻和速度，并用所述微观跟驰模型预测车辆在上游监测点与汇流终点之间的车辆初始轨迹；所述上游监测点与汇流起点之间有一定距离的路段；所述汇流起点位于所述上游监测点和所述汇流终点之间；所述汇流起点和所述汇流终点之间的路段构成所述汇流区域；S2. Acquire the time and speed of the upstream monitoring point of the vehicle before passing through the confluence area in the mixed traffic flow, and use the microscopic car-following model to predict the initial vehicle trajectory of the vehicle between the upstream monitoring point and the end point of the confluence; the upstream monitoring point A road section with a certain distance from the starting point of the confluence; the starting point of the confluence is located between the upstream monitoring point and the end point of the confluence; the road section between the starting point of the confluence and the end point of the confluence constitutes the confluence area;

S3、基于所述微观跟驰模型，加入加速度约束、距离约束与安全约束，建立汇流模型；S3, based on the microscopic car-following model, adding acceleration constraints, distance constraints and safety constraints to establish a confluence model;

S4、针对混合交通流场景下的汇流过程中可能出现的各种无法顺利完成汇流的情况，拟制协同控制策略集；S4. According to various situations that may occur in the convergence process in the mixed traffic flow scenario, the convergence cannot be successfully completed, formulate a collaborative control strategy set;

S5、基于所述车辆初始轨迹，由所述汇流模型判断车辆是否可以顺利完成汇流；若判断结果为车辆可以顺利完成汇流，则车辆继续遵从所述微观跟驰模型的速度行驶；若判断结果为车辆无法顺利完成汇流，则需进一步判断车辆在汇流过程中出现的具体情况，并依据所述车辆在汇流过程中出现的具体情况对车辆做出所述步骤S4拟制的协同策略集中对应的协同控制策略，进而执行步骤S6；S5. Based on the initial trajectory of the vehicle, it is judged by the confluence model whether the vehicle can successfully complete the confluence; if the judgment result is that the vehicle can successfully complete the confluence, the vehicle continues to run at the speed of the microscopic car-following model; if the judgment result is If the vehicle cannot successfully complete the confluence, it is necessary to further judge the specific situation of the vehicle during the confluence process, and according to the specific situation of the vehicle during the confluence process, make a centralized coordination corresponding to the coordination strategy prepared in step S4 for the vehicle. Control strategy, and then execute step S6;

S6、将参与汇流过程的可优化控制的车辆确定为目标车辆，由步骤S5所做出的协同控制策略对所述目标车辆的行驶轨迹进行优化，并将该优化问题归结为离散时间状态约束的最优控制问题，用动态规划的思想求解得到关于所述目标车辆的协同优化控制策略，并将关于所述目标车辆的协同优化控制策略作用于所述目标车辆，控制所述目标车辆的运行。S6. Determine the optimally controllable vehicle participating in the confluence process as the target vehicle, optimize the driving trajectory of the target vehicle by the collaborative control strategy made in step S5, and attribute the optimization problem to a discrete-time state-constrained The optimal control problem is solved by using the idea of dynamic programming to obtain a cooperative optimal control strategy for the target vehicle, and the cooperative optimal control strategy for the target vehicle is applied to the target vehicle to control the operation of the target vehicle.

进一步地，所述步骤S4具体包括：Further, the step S4 specifically includes:

假定匝道和主干道均为单向单行车道，匝道上的车辆k将汇入主干道的连续车流的两车辆之间的间隔，所述主干道的连续车流的两车辆分别用车辆

Figure BDA0002207615000000021
和车辆
Figure BDA0002207615000000022
表示，其中车辆
Figure BDA0002207615000000023
表示前车，车辆
Figure BDA0002207615000000024
表示后车；Assuming that both the ramp and the main road are one-way one-way lanes, the vehicle k on the ramp will merge into the interval between the two vehicles of the continuous traffic flow of the main road, and the two vehicles of the continuous traffic flow of the main road use the vehicle
Figure BDA0002207615000000021
and vehicles
Figure BDA0002207615000000022
means that the vehicle
Figure BDA0002207615000000023
Indicates the preceding vehicle, the vehicle
Figure BDA0002207615000000024
Indicates the rear car;

针对混合交通流场景下的汇流过程中可能出现的各种情况，基于所述微观跟驰模型将车辆

Figure BDA0002207615000000025
车辆k、车辆
Figure BDA0002207615000000026
之间的关系分为可以顺利完成汇流与无法顺利完成汇流；所述无法顺利完成汇流又分为四种情况，第一种情况记为R1，表示车辆k与车辆
Figure BDA0002207615000000027
之间距离太近，不满足可顺利完成汇流的约束条件；第二种情况记为R2，表示车辆k与车辆
Figure BDA0002207615000000028
之间距离太近，不满足可顺利完成汇流的约束条件；第三种情况记为R3，表示车辆k与车辆
Figure BDA0002207615000000029
且与车辆
Figure BDA00022076150000000210
之间满足基本间隔要求但汇流过程不舒适；第四种情况记为R4，表示车辆k与车辆
Figure BDA00022076150000000211
且与车辆
Figure BDA00022076150000000212
之间的距离均太近，不满足可顺利汇流的约束条件；In view of various situations that may occur in the convergence process in the mixed traffic flow scenario, the vehicle
Figure BDA0002207615000000025
vehicle k, vehicle
Figure BDA0002207615000000026
The relationship between the two can be divided into those that can successfully complete the confluence and those that cannot successfully complete the confluence; the inability to successfully complete the confluence is further divided into four cases, the first case is recorded as R1, indicating that the vehicle k and the vehicle
Figure BDA0002207615000000027
The distance between them is too close to meet the constraints that the confluence can be successfully completed; the second case is recorded as R2, indicating that the vehicle k and the vehicle
Figure BDA0002207615000000028
The distance between them is too close to meet the constraints that the confluence can be successfully completed; the third case is recorded as R3, indicating that the vehicle k and the vehicle
Figure BDA0002207615000000029
and with the vehicle
Figure BDA00022076150000000210
The basic interval requirements are met, but the confluence process is uncomfortable; the fourth case is recorded as R4, indicating that the vehicle k and the vehicle
Figure BDA00022076150000000211
and with the vehicle
Figure BDA00022076150000000212
The distances between them are too close to meet the constraints of smooth confluence;

针对无法顺利完成汇流的四种情况，需要对当中的可优化控制的车辆进行协同控制；用H表示人类驾驶车辆，为不可优化控制的车辆；用A表示自动驾驶车辆，为可优化控制的车辆；用N表示无前车参与或无后车参与汇流；并规定车辆的组合顺序依次为车辆

Figure BDA00022076150000000213
车辆k、车辆
Figure BDA00022076150000000214
(例如：用HAN表示车辆
Figure BDA00022076150000000215
为人类驾驶车辆、车辆k为自动驾驶车辆、车辆
Figure BDA00022076150000000216
为无后车参与汇流。)For the four situations where the confluence cannot be successfully completed, it is necessary to coordinately control the vehicles that can be optimally controlled; use H to represent a human-driven vehicle, which is a vehicle that cannot be optimally controlled; use A to represent an autonomous vehicle, which can be optimally controlled. ;Use N to indicate that no preceding vehicle participates or no rear vehicle participates in the confluence;
Figure BDA00022076150000000213
vehicle k, vehicle
Figure BDA00022076150000000214
(Example: use HAN for vehicles
Figure BDA00022076150000000215
driving vehicles for humans, vehicle k for autonomous vehicles, vehicles
Figure BDA00022076150000000216
Participate in the confluence for no vehicles behind. )

基于不同的车辆组合、不同的车型组合，以及所述无法顺利完成汇流的四种情况，拟制协同控制策略集，如下表所示：Based on different vehicle combinations, different vehicle types, and the four situations in which the convergence cannot be successfully completed, a collaborative control strategy set is drawn up, as shown in the following table:

Figure BDA00022076150000000217
Figure BDA00022076150000000217

Figure BDA0002207615000000031
Figure BDA0002207615000000031

Figure BDA0002207615000000041
Figure BDA0002207615000000041

上表中，所述无优化是指车辆在对应的无法顺利完成汇流的情况下没有相应的控制策略，此时，匝道上的车辆k将以匝道尽头作为一个停止的虚拟前车，遵循所述微观跟驰模型持续减速甚至停车等待，直到主干道上出现满足可汇流的车辆间隔，车辆k才汇入主干道；In the above table, the non-optimization means that the vehicle does not have a corresponding control strategy when the corresponding confluence cannot be successfully completed. The microscopic car-following model continuously decelerates or even stops and waits, and vehicle k does not merge into the main road until there is a vehicle interval that satisfies the merging on the main road;

所述控制车辆k加速，是使t时刻车辆k的决策变量uk(t)满足：The acceleration of the control vehicle k is to make the decision variable u k (t) of the vehicle k at time t satisfy:

Figure BDA0002207615000000042
Figure BDA0002207615000000042

且vk(t)+uk(t)τ≤ve；and v k (t)+u k (t) τ≤ve ;

所述控制车辆k减速，是使t时刻车辆k的决策变量uk(t)满足：The control of vehicle k to decelerate is to make the decision variable u k (t) of vehicle k at time t satisfy:

Figure BDA0002207615000000043
Figure BDA0002207615000000043

且vk(t)+uk(t)τ≥0；and v k (t)+u k (t)τ≥0;

所述车辆k控制状态未知是对车辆k的控制方式可能是控制车辆k减速，也可能是控制车辆k加速，还可能是不控制车辆k，此时t时刻车辆k的决策变量uk(t)满足：The unknown control state of the vehicle k means that the control method for the vehicle k may be to control the vehicle k to decelerate, it may also be to control the vehicle k to accelerate, or it may not control the vehicle k. At this time, the decision variable u k (t )Satisfy:

vk(t)+uk(t)τ≤ve，v k (t)+u k (t) τ≤ve ,

且vk(t)+uk(t)τ≥0；and v k (t)+u k (t)τ≥0;

所述不控制车辆k，是使t时刻车辆k的决策变量uk(t)满足：The non-controlling vehicle k is to make the decision variable u k (t) of the vehicle k at time t satisfy:

Figure BDA0002207615000000044
即uk(t)＝0；
Figure BDA0002207615000000044
That is, u k (t) = 0;

所述控制车辆

Figure BDA0002207615000000045
加速，是使t时刻车辆
Figure BDA0002207615000000046
的决策变量
Figure BDA0002207615000000047
满足：the control vehicle
Figure BDA0002207615000000045
Acceleration is to make the vehicle at time t
Figure BDA0002207615000000046
decision variables
Figure BDA0002207615000000047
Satisfy:

Figure BDA0002207615000000048
Figure BDA0002207615000000048

且

Figure BDA0002207615000000049
and
Figure BDA0002207615000000049

所述不控制车辆

Figure BDA00022076150000000410
是使t时刻车辆
Figure BDA00022076150000000411
的决策变量
Figure BDA00022076150000000412
满足：the uncontrolled vehicle
Figure BDA00022076150000000410
is the vehicle at time t
Figure BDA00022076150000000411
decision variables
Figure BDA00022076150000000412
Satisfy:

Figure BDA00022076150000000413
即
Figure BDA00022076150000000414
Figure BDA00022076150000000413
which is
Figure BDA00022076150000000414

所述控制车辆

Figure BDA00022076150000000415
减速，是使t时刻车辆
Figure BDA00022076150000000416
的决策变量
Figure BDA00022076150000000417
满足：the control vehicle
Figure BDA00022076150000000415
Deceleration is to make the vehicle at time t
Figure BDA00022076150000000416
decision variables
Figure BDA00022076150000000417
Satisfy:

Figure BDA00022076150000000418
Figure BDA00022076150000000418

且

Figure BDA0002207615000000051
and
Figure BDA0002207615000000051

所述不控制车辆

Figure BDA0002207615000000052
是使t时刻车辆
Figure BDA0002207615000000053
的决策变量
Figure BDA0002207615000000054
满足：the uncontrolled vehicle
Figure BDA0002207615000000052
is the vehicle at time t
Figure BDA0002207615000000053
decision variables
Figure BDA0002207615000000054
Satisfy:

Figure BDA0002207615000000055
即
Figure BDA0002207615000000056
Figure BDA0002207615000000055
which is
Figure BDA0002207615000000056

其中，uk(t)作为t时刻车辆k的决策变量，表示在t时刻车辆k的加速度；vk(t)是车辆k在t时刻的速度；

Figure BDA0002207615000000057
是根据所述微观跟驰模型所预测的车辆k在t+τ时刻的安全跟驰速度(其中Lk(t)表示在t时刻车辆k与其跟驰前车之间的相对距离，vk(t)表示车辆k在t时刻的速度，
Figure BDA0002207615000000058
表示车辆k的跟驰前车在t时刻的速度)；
Figure BDA0002207615000000059
作为t时刻车辆
Figure BDA00022076150000000510
的决策变量，表示在t时刻车辆
Figure BDA00022076150000000511
的加速度；
Figure BDA00022076150000000512
是车辆
Figure BDA00022076150000000513
在t时刻的速度；
Figure BDA00022076150000000514
是根据所述微观跟驰模型所预测的车辆
Figure BDA00022076150000000515
在t+τ时刻的安全跟驰速度(其中
Figure BDA00022076150000000516
表示在t时刻车辆
Figure BDA00022076150000000517
与其跟驰前车之间的相对距离，
Figure BDA00022076150000000518
表示车辆
Figure BDA00022076150000000519
在t时刻的速度，
Figure BDA00022076150000000520
表示车辆
Figure BDA00022076150000000521
的跟驰前车在t时刻的速度)；
Figure BDA00022076150000000522
作为t时刻车辆
Figure BDA00022076150000000523
的决策变量，表示在t时刻车辆
Figure BDA00022076150000000524
的加速度；
Figure BDA00022076150000000525
是车辆
Figure BDA00022076150000000526
在t时刻的速度；
Figure BDA00022076150000000527
是根据所述微观跟驰模型所预测的车辆
Figure BDA00022076150000000528
在t+τ时刻的安全跟驰速度(其中
Figure BDA00022076150000000529
表示在t时刻车辆
Figure BDA00022076150000000530
与其跟驰前车之间的相对距离，
Figure BDA00022076150000000531
表示车辆
Figure BDA00022076150000000532
在t时刻的速度，
Figure BDA00022076150000000533
表示车辆
Figure BDA00022076150000000534
的跟驰前车在t时刻的速度)；τ是车辆驾驶的反应时间；ve是期望速度。Among them, u k (t), as the decision variable of vehicle k at time t, represents the acceleration of vehicle k at time t; v k (t) is the speed of vehicle k at time t;
Figure BDA0002207615000000057
is the safe car-following speed of vehicle k at time t+τ predicted according to the microscopic car-following model (wherein L k (t) represents the relative distance between vehicle k and the car in front of it at time t, v k ( t) represents the speed of vehicle k at time t,
Figure BDA0002207615000000058
represents the speed of the vehicle in front of vehicle k at time t);
Figure BDA0002207615000000059
as the vehicle at time t
Figure BDA00022076150000000510
is the decision variable, representing the vehicle at time t
Figure BDA00022076150000000511
acceleration;
Figure BDA00022076150000000512
is a vehicle
Figure BDA00022076150000000513
velocity at time t;
Figure BDA00022076150000000514
is the vehicle predicted by the micro-following model
Figure BDA00022076150000000515
Safe following speed at time t+τ (where
Figure BDA00022076150000000516
represents the vehicle at time t
Figure BDA00022076150000000517
The relative distance between it and the car in front of it,
Figure BDA00022076150000000518
Indicates the vehicle
Figure BDA00022076150000000519
The velocity at time t,
Figure BDA00022076150000000520
Indicates the vehicle
Figure BDA00022076150000000521
the speed of the following car at time t);
Figure BDA00022076150000000522
as the vehicle at time t
Figure BDA00022076150000000523
is the decision variable, representing the vehicle at time t
Figure BDA00022076150000000524
acceleration;
Figure BDA00022076150000000525
is a vehicle
Figure BDA00022076150000000526
velocity at time t;
Figure BDA00022076150000000527
is the vehicle predicted by the micro-following model
Figure BDA00022076150000000528
Safe following speed at time t+τ (where
Figure BDA00022076150000000529
represents the vehicle at time t
Figure BDA00022076150000000530
The relative distance between it and the car in front of it,
Figure BDA00022076150000000531
Indicates the vehicle
Figure BDA00022076150000000532
The velocity at time t,
Figure BDA00022076150000000533
Indicates the vehicle
Figure BDA00022076150000000534
is the speed of the following vehicle at time t); τ is the driving reaction time of the vehicle; ve is the expected speed.

进一步地，所述步骤S6具体包括：Further, the step S6 specifically includes:

S6-1、基于所述微观跟驰模型预测目标车辆进入控制区域的时刻为t0，离开控制区域的时刻为tf；并将t0到tf的时间按离散时间间隔τ′平均分为N段，即N＝(tf-t0)/τ′，定义控制决策时刻为t0+τ′,t0+2τ′,t0+3τ′,t0+4τ′,…,t0+(N-1)τ′,tf；所述控制区域为控制起点和所述汇流终点之间的路段；所述控制起点位于所述上游监测点和所述汇流起点之间；S6-1. Based on the microscopic car-following model, it is predicted that the time when the target vehicle enters the control area is t 0 , and the time when it leaves the control area is t f ; and the time from t 0 to t f is divided into discrete time intervals τ′ equally divided into N segments, namely N=(t f -t 0 )/τ′, define the control decision time as t 0 +τ′,t 0 +2τ′,t 0 +3τ′,t 0 +4τ′,…,t 0 +(N-1)τ′,t f ; the control area is the road section between the control starting point and the convergence end point; the control starting point is located between the upstream monitoring point and the convergence starting point;

S6-2、根据目标车辆在t0时刻的状态，计算目标车辆在t0+τ′时刻的容许状态集，以及目标车辆从t0时刻的状态到t0+τ′时刻的容许状态集中各个容许状态的转移成本；所述目标车辆在t0时刻的状态包括目标车辆在t0时刻的速度和位置；S6-2. According to the state of the target vehicle at time t 0 , calculate the allowable state set of the target vehicle at time t 0 +τ', and the allowable state set of the target vehicle from the state at time t 0 to time t 0 +τ'. The transition cost of the allowable state; the state of the target vehicle at time t 0 includes the speed and position of the target vehicle at time t 0 ;

S6-3、根据目标车辆在t0+τ′时刻的容许状态集，计算目标车辆在t0+2τ′时刻的容许状态集，以及目标车辆从t0+τ′时刻的状态到t0+2τ′时刻的容许状态集中各个容许状态的转移成本和累计成本；S6-3. According to the allowable state set of the target vehicle at time t 0 +τ', calculate the allowable state set of the target vehicle at time t 0 +2τ', and the state of the target vehicle from time t 0 +τ' to t 0 + The transition cost and cumulative cost of each allowable state in the allowable state set at time 2τ′;

S6-4、根据步骤S6-3的方法依次计算目标车辆在t0+3τ′,t0+4τ′,…,t0+(N-1)τ′,tf时刻的容许状态集，并计算得到tf时刻的容许状态集中的各个容许状态的累计成本；S6-4, according to the method of step S6-3, sequentially calculate the allowable state set of the target vehicle at time t 0 +3τ', t 0 +4τ',..., t 0 +(N-1)τ', t f , and Calculate the cumulative cost of each allowable state in the allowable state set at time t f ;

S6-5、判断目标车辆在tf时刻的容许状态集中的各个容许状态是否满足可以顺利完成汇流的条件，并将满足条件的容许状态纳入最终容许状态集中；S6-5, judging whether each allowable state in the allowable state set of the target vehicle at time tf satisfies the conditions for successfully completing the confluence, and incorporating the allowable states that meet the conditions into the final allowable state set;

S6-6、选择最终容许状态集中累计成本最小的容许状态作为tf时刻的最优状态；S6-6, select the allowable state with the smallest cumulative cost in the final allowable state set as the optimal state at time t f ;

S6-7、根据tf时刻的最优状态逆推得到t0+τ′,t0+2τ′,t0+3τ′,t0+4τ′,…,t0+(N-1)τ′,tf各时刻的最优状态，并将每个最优状态对应的控制决策纳入协同优化控制策略；S6-7. According to the optimal state at time t f , t 0 +τ′,t 0 +2τ′,t 0 +3τ′,t 0 +4τ′,…,t 0 +(N-1)τ are obtained by inverse inference ′, t f the optimal state at each moment, and incorporate the control decision corresponding to each optimal state into the collaborative optimization control strategy;

S6-8、根据所述协同优化控制策略对目标车辆在所述控制区域间的运行进行控制。S6-8. Control the operation of the target vehicle between the control areas according to the cooperative optimization control strategy.

进一步地，构建微观交通流仿真环境，对比不同交通情景下优化前后的仿真结果。Furthermore, a microscopic traffic flow simulation environment is constructed to compare the simulation results before and after optimization under different traffic scenarios.

与现有技术相比，本发明方法的有益效果是：Compared with the prior art, the beneficial effects of the method of the present invention are:

本发明提供的一种高速公路混合交通流汇流协同优化底层控制方法是对混合交通流状态下的高度公路匝道汇流过程进行建模，并针对自动驾驶车辆与人类驾驶车辆的各种混合排列组合场景分别给出了相应的协同汇流轨迹优化策略，本方法可用于分析不同交通状态、不同自动驾驶车辆渗透率等对交通的影响。本方法以微观跟驰模型描述微观的车辆跟驰状态，考虑高速公路的交通特征、几何约束、安全约束，将协同汇流轨迹优化问题归结为离散时间状态约束最优控制问题，并提出一种基于动态规划的求解方法来有效地解决这一问题，通过此方法可以使车辆以较高的速度平稳运行，进而提高了高速公路匝道在汇流区域的车辆通行能力，同时提高了车辆的平均行驶时间和路段交通流的稳定性。The low-level control method for collaborative optimization of highway mixed traffic flow convergence provided by the present invention is to model the high-level highway ramp convergence process under the mixed traffic flow state, and aims at various mixed arrangement and combination scenarios of automatic driving vehicles and human driving vehicles. Corresponding co-convergence trajectory optimization strategies are given respectively. This method can be used to analyze the impact of different traffic states and different penetration rates of autonomous vehicles on traffic. This method describes the micro-vehicle following state with a micro-following model, considers the traffic characteristics, geometric constraints, and safety constraints of the expressway, and reduces the coordinated convergence trajectory optimization problem to a discrete-time state-constrained optimal control problem, and proposes a method based on The dynamic programming solution method can effectively solve this problem. Through this method, the vehicle can run smoothly at a higher speed, thereby improving the vehicle capacity of the expressway ramp in the confluence area, and at the same time improving the average travel time and Stability of road traffic flow.

通过大量仿真实验证明，在引入协同汇流轨迹优化控制策略时，可以有效减少汇流行为下的车辆冲突次数，并且能有效提高汇流效率：单个匝道通行能力提高8％～10％。Through a large number of simulation experiments, it is proved that the introduction of the cooperative confluence trajectory optimization control strategy can effectively reduce the number of vehicle conflicts under the confluence behavior, and can effectively improve the confluence efficiency: the capacity of a single ramp is increased by 8% to 10%.

下面通过具体实施方式及附图对本发明作进一步详细说明，但并不意味着对本发明保护范围的限制。The present invention will be further described in detail below through specific embodiments and accompanying drawings, but it is not intended to limit the protection scope of the present invention.

附图说明Description of drawings

图1是本发明实施例中高速公路匝道在混合交通流场景下的汇流示意图。FIG. 1 is a schematic diagram of the convergence of expressway ramps in a mixed traffic flow scenario in an embodiment of the present invention.

图2是本发明实施例中车辆无法顺利完成汇流的第一种情况(R1)的示意图。FIG. 2 is a schematic diagram of the first situation ( R1 ) in which the vehicle cannot successfully complete the confluence in the embodiment of the present invention.

图3是本发明实施例中车辆无法顺利完成汇流的第二种情况(R2)的示意图。FIG. 3 is a schematic diagram of the second situation (R2) in which the vehicle cannot successfully complete the confluence in the embodiment of the present invention.

图4是本发明实施例中车辆无法顺利完成汇流的第三种情况(R3)的示意图。FIG. 4 is a schematic diagram of a third situation ( R3 ) in which the vehicle cannot successfully complete the confluence in the embodiment of the present invention.

图5是本发明实施例中车辆无法顺利完成汇流的第四种情况(R4)的示意图。FIG. 5 is a schematic diagram of the fourth situation ( R4 ) in which the vehicle cannot successfully complete the confluence in the embodiment of the present invention.

图6是本发明实施例中参与汇流过程的车辆组合和车型组合为HHA的三辆车，在无法顺利完成汇流的第二种情况(R2)下，且在无协同优化控制情况下的汇流轨迹图。Fig. 6 shows the combination of vehicles participating in the merging process and the three vehicles whose vehicle type combination is HHA in the embodiment of the present invention, in the second case (R2) where the merging cannot be successfully completed, and the trajectories of the merging in the case of no collaborative optimization control picture.

图7是本发明实施例中参与汇流过程的车辆组合和车型组合为HHA的三辆车，在无法顺利完成汇流的第二种情况(R2)下，且在有协同优化控制情况下的汇流轨迹图。Fig. 7 shows the combination of vehicles participating in the merging process and the three vehicles whose vehicle type combination is HHA in the embodiment of the present invention, in the second case (R2) where the merging cannot be successfully completed, and the trajectories of the merging under the condition of collaborative optimization control picture.

图8是本发明实施例中参与汇流过程的车辆组合和车型组合为HAH的三辆车，在无法顺利完成汇流的第一种情况(R1)下，且在无协同优化控制情况下的汇流轨迹图。Fig. 8 shows the combination of vehicles participating in the merging process and the three vehicles whose vehicle type combination is HAH in the embodiment of the present invention, in the first case (R1) where the merging cannot be successfully completed, and the trajectories of the merging in the case of no collaborative optimal control picture.

图9是本发明实施例中参与汇流过程的车辆组合和车型组合为HAH的三辆车，在无法顺利完成汇流的第一种情况(R1)下，且在有协同优化控制情况下的汇流轨迹图。Fig. 9 shows the combination of vehicles participating in the merging process and three vehicles whose vehicle type combination is HAH in the embodiment of the present invention, in the first case (R1) where the merging cannot be successfully completed, and the trajectories of the merging under the condition of collaborative optimization control picture.

具体实施方式Detailed ways

下面结合附图，通过对实施例的描述，对本发明的具体实施方式作进一步的说明。The specific embodiments of the present invention will be further described below through the description of the embodiments with reference to the accompanying drawings.

一种高速公路混合交通流汇流协同优化底层控制方法，包括步骤：A low-level control method for synergistic optimization of highway mixed traffic flow convergence, comprising the steps of:

S1、确定微观跟驰模型，用所述微观跟驰模型描述车辆的跟驰状态，所述车辆的跟驰状态包括车辆的速度、加速度和位置；S1. Determine a microscopic car-following model, and use the microscopic car-following model to describe the car-following state of the vehicle, and the car-following state of the vehicle includes the speed, acceleration and position of the vehicle;

S2、获取混合交通流中车辆在通过汇流区域前的上游监测点的时刻和速度，并用所述微观跟驰模型预测车辆在上游监测点与汇流终点之间的车辆初始轨迹；所述上游监测点与汇流起点之间有一定距离的路段；所述汇流起点位于所述上游监测点和所述汇流终点之间；所述汇流起点和所述汇流终点之间的路段构成所述汇流区域；S2. Acquire the time and speed of the upstream monitoring point of the vehicle before passing through the confluence area in the mixed traffic flow, and use the microscopic car-following model to predict the initial vehicle trajectory of the vehicle between the upstream monitoring point and the end point of the confluence; the upstream monitoring point A road section with a certain distance from the starting point of the confluence; the starting point of the confluence is located between the upstream monitoring point and the end point of the confluence; the road section between the starting point of the confluence and the end point of the confluence constitutes the confluence area;

S3、基于所述微观跟驰模型，加入加速度约束、距离约束与安全约束，建立汇流模型；S3, based on the microscopic car-following model, adding acceleration constraints, distance constraints and safety constraints to establish a confluence model;

S4、针对混合交通流场景下的汇流过程中可能出现的各种无法顺利完成汇流的情况，拟制协同控制策略集；S4. According to various situations that may occur in the convergence process in the mixed traffic flow scenario, the convergence cannot be successfully completed, formulate a collaborative control strategy set;

S5、基于所述车辆初始轨迹，由所述汇流模型判断车辆是否可以顺利完成汇流；若判断结果为车辆可以顺利完成汇流，则车辆继续遵从所述微观跟驰模型的速度行驶；若判断结果为车辆无法顺利完成汇流，则需进一步判断车辆在汇流过程中出现的具体情况，并依据所述车辆在汇流过程中出现的具体情况对车辆做出所述步骤S4拟制的协同策略集中对应的协同控制策略，进而执行步骤S6；S5. Based on the initial trajectory of the vehicle, it is judged by the confluence model whether the vehicle can successfully complete the confluence; if the judgment result is that the vehicle can successfully complete the confluence, the vehicle continues to run at the speed of the microscopic car-following model; if the judgment result is If the vehicle cannot successfully complete the confluence, it is necessary to further judge the specific situation of the vehicle during the confluence process, and according to the specific situation of the vehicle during the confluence process, make a centralized coordination corresponding to the coordination strategy prepared in step S4 for the vehicle. Control strategy, and then execute step S6;

S6、将参与汇流过程的可优化控制的车辆确定为目标车辆，由步骤S5所做出的协同控制策略对所述目标车辆的行驶轨迹进行优化，并将该优化问题归结为离散时间状态约束的最优控制问题，用动态规划的思想求解得到关于所述目标车辆的协同优化控制策略，并将关于所述目标车辆的协同优化控制策略作用于所述目标车辆，控制所述目标车辆的运行。S6. Determine the optimally controllable vehicle participating in the confluence process as the target vehicle, optimize the driving trajectory of the target vehicle by the collaborative control strategy made in step S5, and attribute the optimization problem to a discrete-time state-constrained The optimal control problem is solved by using the idea of dynamic programming to obtain a cooperative optimal control strategy for the target vehicle, and the cooperative optimal control strategy for the target vehicle is applied to the target vehicle to control the operation of the target vehicle.

实施例Example

如图1所示，为高速公路匝道在混合交通流场景下的汇流示意图，其中主干道和匝道均为单向单行车道，主干道上有若干自动驾驶车辆A和人类驾驶车辆H无规则排列的车流，匝道上有若干自动驾驶车辆A和人类驾驶车辆H无规则排列的车流需要通过汇流区域汇入主干道上的车流之间。假定主干道和匝道上的自动驾驶车辆A和人类驾驶车辆H均遵循微观跟驰模型。As shown in Figure 1, it is a schematic diagram of the confluence of expressway ramps in a mixed traffic flow scenario, in which both the main road and the ramp are one-way one-way lanes, and there are several autonomous vehicles A and human-driven vehicles H on the main road in an irregular arrangement. Traffic flow, there are a number of autonomous vehicles A and human-driven vehicles H on the ramp. The random arrangement of traffic needs to be merged into the traffic flow on the main road through the convergence area. It is assumed that both the autonomous vehicle A and the human-driven vehicle H on the main road and ramp follow the micro-following model.

S1、确定微观跟驰模型，用微观跟驰模型描述车辆的跟驰状态，车辆的跟驰状态包括车辆的速度、加速度和位置。S1. Determine a microscopic car-following model, and use the microscopic car-following model to describe the car-following state of the vehicle. The car-following state of the vehicle includes the speed, acceleration and position of the vehicle.

其中，微观跟驰模型如下：Among them, the micro car following model is as follows:

v(t+τ)＝vmic(L(t),v(t),vlead(t))， (1)v(t+τ)=v mic (L(t), v(t), v lead (t)), (1)

u(t)＝(v(t+τ)-v(t))/τ， (2)u(t)=(v(t+τ)-v(t))/τ, (2)

x(t+τ)＝x(t)-v(t)τ-0.5u(t)τ2， (3)x(t+τ)=x(t)-v(t)τ-0.5u(t)τ 2 , (3)

方程(1)为一般化的车辆跟驰模型，描述车辆在t+τ时刻的速度，其中，L(t)为t时刻车辆与其跟驰前车的车间距；v(t)为t时刻车辆的速度，vlead(t)为t时刻车辆所跟驰的前车的速度。u(t)表示t时刻车辆的加速度。车辆在t时刻的位置用x(t)表示，在t+τ时刻的位置用x(t+τ)表示。Equation (1) is a generalized vehicle following model, which describes the speed of the vehicle at time t+τ, where L(t) is the distance between the vehicle and the preceding vehicle at time t; v(t) is the vehicle at time t The speed of , v lead (t) is the speed of the preceding vehicle followed by the vehicle at time t. u(t) represents the acceleration of the vehicle at time t. The position of the vehicle at time t is denoted by x(t), and the position at time t+τ is denoted by x(t+τ).

S2、获取混合交通流中车辆在通过汇流区域前的上游监测点的时刻和速度，并用微观跟驰模型预测车辆在上游监测点与汇流终点之间的车辆初始轨迹；上游监测点与汇流起点之间有一定距离的路段；汇流起点位于上游监测点和汇流终点之间；汇流起点和汇流终点之间的路段构成汇流区域。S2. Obtain the time and speed of the vehicle in the mixed traffic flow at the upstream monitoring point before passing through the confluence area, and use the micro-car following model to predict the initial trajectory of the vehicle between the upstream monitoring point and the end point of the confluence; the difference between the upstream monitoring point and the starting point of the confluence The starting point of the confluence is located between the upstream monitoring point and the end point of the confluence; the road segment between the starting point and the end point of the confluence constitutes the confluence area.

S3、基于微观跟驰模型，加入加速度约束、距离约束与安全约束，建立汇流模型。S3. Based on the microscopic car-following model, acceleration constraints, distance constraints and safety constraints are added to establish a confluence model.

其中，汇流模型如下：Among them, the confluence model is as follows:

假设匝道上的车辆k处于汇流区域，它将汇入主干道的连续车流的两车辆

Figure BDA0002207615000000081
和
Figure BDA0002207615000000082
之间，其中车辆
Figure BDA0002207615000000083
为前车，车辆
Figure BDA0002207615000000084
为后车。Assuming that vehicle k on the ramp is in the merge area, it will merge into the two vehicles of the continuous traffic flow of the main road
Figure BDA0002207615000000081
and
Figure BDA0002207615000000082
between which vehicles
Figure BDA0002207615000000083
for the preceding vehicle, the vehicle
Figure BDA0002207615000000084
for the rear car.

Figure BDA0002207615000000085
Figure BDA0002207615000000085

Figure BDA0002207615000000086
Figure BDA0002207615000000086

Figure BDA0002207615000000087
Figure BDA0002207615000000087

方程(4)用来表示汇流效用，反映汇流时的舒适度，是以汇流时的车间距以及汇流时匝道上的汇流车辆k和主干道后车

Figure BDA0002207615000000088
的加速度来标定。其中，
Figure BDA0002207615000000089
表示的是汇流行为在不受约束条件限制时的汇流效用。la为车辆车身长，
Figure BDA00022076150000000810
为自动驾驶车辆与前车的最小安全车头距离，
Figure BDA00022076150000000811
为人类驾驶车辆与前车的最小安全车头距离。汇流时，汇流车辆k实际跟驰主干道前车
Figure BDA00022076150000000812
运行，而主干道后车
Figure BDA00022076150000000813
实际跟驰汇流车辆k运行，它们的加速度都可以根据车辆跟驰模型计算得到。
Figure BDA0002207615000000091
表示车辆k的加速度的绝对值；
Figure BDA0002207615000000092
表示主干道后车
Figure BDA0002207615000000093
的加速度的绝对值；bsafe表示最大允许减速度。ΦA为自动驾驶车辆集合，ΦH为人类驾驶车辆集合。η1和η2分别表示安全系数与礼貌系数，安全系数η1为常数，礼貌系数η2采用分段连续形式，如方程(5)所示，Vth是给定的速度阈值，ve为期望速度，β1和β2为常数。方程(6)中lk(t+τ)表示汇流决策，值为0则表示车辆k在t+τ时刻无法顺利完成汇流，值为1则表示车辆k在t+τ时刻可以顺利完成汇流。Equation (4) is used to express the confluence utility, which reflects the comfort level at the time of confluence.
Figure BDA0002207615000000088
to calibrate the acceleration. in,
Figure BDA0002207615000000089
Represents the sinking utility of sinking behavior when it is not restricted by constraints. l a is the length of the vehicle body,
Figure BDA00022076150000000810
is the minimum safe frontal distance between the autonomous vehicle and the preceding vehicle,
Figure BDA00022076150000000811
The minimum safe frontal distance between a human-driven vehicle and the vehicle in front. When merging, the merging vehicle k actually follows the preceding vehicle on the main road
Figure BDA00022076150000000812
run while the main road behind the car
Figure BDA00022076150000000813
The actual car-following and convergent vehicles k run, and their accelerations can be calculated according to the car-following model.
Figure BDA0002207615000000091
represents the absolute value of the acceleration of vehicle k;
Figure BDA0002207615000000092
Indicates the car behind the main road
Figure BDA0002207615000000093
The absolute value of the acceleration; b safe represents the maximum allowable deceleration. Φ A is the set of autonomous vehicles, and Φ H is the set of human-driven vehicles. η 1 and η 2 represent the safety factor and the politeness factor, respectively, the safety factor η 1 is a constant, and the politeness factor η 2 adopts a piecewise continuous form, as shown in equation (5), V th is the given speed threshold, and ve is The desired velocity, β 1 and β 2 are constants. In equation (6), l k (t+τ) represents the convergence decision. A value of 0 means that vehicle k cannot successfully complete the convergence at time t+τ, and a value of 1 means that vehicle k can successfully complete the convergence at time t+τ.

S4、针对混合交通流场景下的汇流过程中可能出现的各种无法顺利完成汇流的情况，拟制协同控制策略集。具体如下：S4. In view of various situations that may occur in the convergence process in the mixed traffic flow scenario, the convergence cannot be successfully completed, formulate a collaborative control strategy set. details as follows:

假定匝道和主干道均为单向单行车道，匝道上的车辆k将汇入主干道的连续车流的两车辆之间的间隔，主干道的连续车流的两车辆分别用车辆

Figure BDA0002207615000000094
和车辆
Figure BDA0002207615000000095
表示，其中车辆
Figure BDA0002207615000000096
表示前车，车辆
Figure BDA0002207615000000097
表示后车。Assuming that both the ramp and the main road are one-way one-way lanes, the vehicle k on the ramp will merge into the interval between the two vehicles of the continuous traffic flow of the main road, and the two vehicles of the continuous traffic flow of the main road will use the vehicle
Figure BDA0002207615000000094
and vehicles
Figure BDA0002207615000000095
means that the vehicle
Figure BDA0002207615000000096
Indicates the preceding vehicle, the vehicle
Figure BDA0002207615000000097
Indicates the car behind.

针对混合交通流场景下的汇流过程中可能出现的各种情况，基于微观跟驰模型将车辆

Figure BDA0002207615000000098
车辆k、车辆
Figure BDA0002207615000000099
之间的关系分为可以顺利完成汇流与无法顺利完成汇流；无法顺利完成汇流分为四种情况：第一种情况记为R1，如图2所示，表示车辆k与车辆
Figure BDA00022076150000000910
之间距离太近，不满足可顺利完成汇流的约束条件；第二种情况记为R2，如图3所示，表示车辆k与车辆
Figure BDA00022076150000000911
之间距离太近，不满足可顺利完成汇流的约束条件；第三种情况记为R3，如图4所示，表示车辆k与车辆
Figure BDA00022076150000000912
且与车辆
Figure BDA00022076150000000913
之间满足基本间隔要求但汇流过程不舒适；第四种情况记为R4，如图5所示，表示车辆k与车辆
Figure BDA00022076150000000914
且与车辆
Figure BDA00022076150000000915
之间的距离均太近，不满足可顺利汇流的约束条件。Aiming at various situations that may occur in the convergence process in the mixed traffic flow scenario, the vehicle is divided into
Figure BDA0002207615000000098
vehicle k, vehicle
Figure BDA0002207615000000099
The relationship between the two can be divided into that the confluence can be successfully completed and the confluence cannot be successfully completed; the failure to complete the confluence is divided into four cases: the first case is recorded as R1, as shown in Figure 2, indicating that the vehicle k and the vehicle
Figure BDA00022076150000000910
The distance between them is too close to meet the constraints that the confluence can be successfully completed; the second case is recorded as R2, as shown in Figure 3, indicating that the vehicle k and the vehicle
Figure BDA00022076150000000911
The distance between them is too close to meet the constraints that the confluence can be successfully completed; the third case is recorded as R3, as shown in Figure 4, indicating that the vehicle k and the vehicle
Figure BDA00022076150000000912
and with the vehicle
Figure BDA00022076150000000913
The basic interval requirements are met, but the confluence process is uncomfortable; the fourth case is recorded as R4, as shown in Figure 5, indicating that the vehicle k and the vehicle
Figure BDA00022076150000000914
and with the vehicle
Figure BDA00022076150000000915
The distances between them are too close to meet the constraints of smooth confluence.

针对无法顺利完成汇流的四种情况，需要对当中的可优化控制的车辆进行协同控制；用H表示人类驾驶车辆，为不可优化控制的车辆；用A表示自动驾驶车辆，为可优化控制的车辆；用N表示无前车参与或无后车参与汇流；并规定车辆的组合顺序依次为车辆

Figure BDA00022076150000000916
车辆k、车辆
Figure BDA00022076150000000917
例如：用HAN表示车辆
Figure BDA00022076150000000918
为人类驾驶车辆、车辆k为自动驾驶车辆、车辆
Figure BDA00022076150000000919
为无后车参与汇流。For the four situations where the confluence cannot be successfully completed, it is necessary to coordinately control the vehicles that can be optimally controlled; use H to represent a human-driven vehicle, which is a vehicle that cannot be optimally controlled; use A to represent an autonomous vehicle, which can be optimally controlled. ;Use N to indicate that no preceding vehicle participates or no rear vehicle participates in the confluence;
Figure BDA00022076150000000916
vehicle k, vehicle
Figure BDA00022076150000000917
For example: use HAN for vehicles
Figure BDA00022076150000000918
driving vehicles for humans, vehicle k for autonomous vehicles, vehicles
Figure BDA00022076150000000919
Participate in the confluence for no vehicles behind.

基于不同的车辆组合、不同的车型组合，以及所述无法顺利完成汇流的四种情况，拟制协同控制策略集，如下表所示：Based on different vehicle combinations, different vehicle types, and the four situations in which the convergence cannot be successfully completed, a collaborative control strategy set is drawn up, as shown in the following table:

Figure BDA00022076150000000920
Figure BDA00022076150000000920

Figure BDA0002207615000000101
Figure BDA0002207615000000101

Figure BDA0002207615000000111
Figure BDA0002207615000000111

上表中，无优化是指车辆在对应的无法顺利完成汇流的情况下没有相应的控制策略，此时，匝道上的车辆k将以匝道尽头作为一个停止的虚拟前车，遵循微观跟驰模型持续减速甚至停车等待，直到主干道上出现满足可汇流的车辆间隔，车辆k才汇入主干道；In the above table, no optimization means that the vehicle does not have a corresponding control strategy when the corresponding confluence cannot be successfully completed. At this time, the vehicle k on the ramp will take the end of the ramp as a stopped virtual vehicle in front and follow the microscopic car following model. Continue to decelerate or even stop and wait until vehicle k merges into the main road until there is a vehicle interval on the main road that can meet the merging;

控制车辆k加速，是使t时刻车辆k的决策变量uk(t)满足：To control the acceleration of vehicle k is to make the decision variable u k (t) of vehicle k at time t satisfy:

Figure BDA0002207615000000112
Figure BDA0002207615000000112

且vk(t)+uk(t)τ≤ve； (7-1)And v k (t)+u k (t) τ≤ve ; (7-1)

控制车辆k减速，是使t时刻车辆k的决策变量uk(t)满足：Controlling vehicle k to decelerate is to make the decision variable u k (t) of vehicle k at time t satisfy:

Figure BDA0002207615000000113
Figure BDA0002207615000000113

且vk(t)+uk(t)τ≥0； (7-2)and v k (t)+u k (t)τ≥0; (7-2)

车辆k控制状态未知是对车辆k的控制方式可能是控制车辆k减速，也可能是控制车辆k加速，还可能是不控制车辆k，此时t时刻车辆k的决策变量uk(t)满足：The control state of vehicle k is unknown. The control method for vehicle k may be to control vehicle k to decelerate, or to control vehicle k to accelerate, or not to control vehicle k. At this time, the decision variable u k (t) of vehicle k at time t satisfies :

vk(t)+uk(t)τ≤ve，v k (t)+u k (t) τ≤ve ,

且vk(t)+uk(t)τ≥0； (7-3)and v k (t)+u k (t)τ≥0; (7-3)

不控制车辆k，是使t时刻车辆k的决策变量uk(t)满足：Not controlling the vehicle k is to make the decision variable u k (t) of the vehicle k at time t satisfy:

Figure BDA0002207615000000114
Figure BDA0002207615000000114

控制车辆

Figure BDA0002207615000000115
加速，是使t时刻车辆
Figure BDA0002207615000000116
的决策变量
Figure BDA0002207615000000117
满足：control the vehicle
Figure BDA0002207615000000115
Acceleration is to make the vehicle at time t
Figure BDA0002207615000000116
decision variables
Figure BDA0002207615000000117
Satisfy:

Figure BDA0002207615000000118
Figure BDA0002207615000000118

且

Figure BDA0002207615000000119
and
Figure BDA0002207615000000119

不控制车辆

Figure BDA00022076150000001110
是使t时刻车辆
Figure BDA00022076150000001111
的决策变量
Figure BDA00022076150000001112
满足：do not control the vehicle
Figure BDA00022076150000001110
is the vehicle at time t
Figure BDA00022076150000001111
decision variables
Figure BDA00022076150000001112
Satisfy:

Figure BDA00022076150000001113
Figure BDA00022076150000001113

控制车辆

Figure BDA00022076150000001114
减速，是使t时刻车辆
Figure BDA00022076150000001115
的决策变量
Figure BDA00022076150000001116
满足：control the vehicle
Figure BDA00022076150000001114
Deceleration is to make the vehicle at time t
Figure BDA00022076150000001115
decision variables
Figure BDA00022076150000001116
Satisfy:

Figure BDA00022076150000001117
Figure BDA00022076150000001117

且

Figure BDA00022076150000001118
and
Figure BDA00022076150000001118

不控制车辆

Figure BDA00022076150000001119
是使t时刻车辆
Figure BDA00022076150000001120
的决策变量
Figure BDA00022076150000001121
满足：do not control the vehicle
Figure BDA00022076150000001119
is the vehicle at time t
Figure BDA00022076150000001120
decision variables
Figure BDA00022076150000001121
Satisfy:

Figure BDA0002207615000000121
Figure BDA0002207615000000121

其中，uk(t)作为t时刻车辆k的决策变量，表示在t时刻车辆k的加速度；vk(t)是车辆k在t时刻的速度；

Figure BDA0002207615000000122
是根据微观跟驰模型所预测的车辆k在t+τ时刻的安全跟驰速度；
Figure BDA0002207615000000123
作为t时刻车辆
Figure BDA0002207615000000124
的决策变量，表示在t时刻车辆
Figure BDA0002207615000000125
的加速度；
Figure BDA0002207615000000126
是车辆
Figure BDA0002207615000000127
在t时刻的速度；
Figure BDA0002207615000000128
是根据微观跟驰模型所预测的车辆
Figure BDA0002207615000000129
在t+τ时刻的安全跟驰速度；
Figure BDA00022076150000001210
作为t时刻车辆
Figure BDA00022076150000001211
的决策变量，表示在t时刻车辆
Figure BDA00022076150000001212
的加速度；
Figure BDA00022076150000001213
是车辆
Figure BDA00022076150000001214
在t时刻的速度；
Figure BDA00022076150000001215
是根据微观跟驰模型所预测的车辆
Figure BDA00022076150000001216
在t+τ时刻的安全跟驰速度；τ是车辆驾驶的反应时间；ve是期望速度。Among them, u k (t), as the decision variable of vehicle k at time t, represents the acceleration of vehicle k at time t; v k (t) is the speed of vehicle k at time t;
Figure BDA0002207615000000122
is the safe car-following speed of vehicle k at time t+τ predicted by the micro-car-following model;
Figure BDA0002207615000000123
as the vehicle at time t
Figure BDA0002207615000000124
is the decision variable, representing the vehicle at time t
Figure BDA0002207615000000125
acceleration;
Figure BDA0002207615000000126
is a vehicle
Figure BDA0002207615000000127
velocity at time t;
Figure BDA0002207615000000128
is the vehicle predicted by the micro-following model
Figure BDA0002207615000000129
Safe following speed at time t+τ;
Figure BDA00022076150000001210
as the vehicle at time t
Figure BDA00022076150000001211
is the decision variable, representing the vehicle at time t
Figure BDA00022076150000001212
acceleration;
Figure BDA00022076150000001213
is a vehicle
Figure BDA00022076150000001214
velocity at time t;
Figure BDA00022076150000001215
is the vehicle predicted by the micro-following model
Figure BDA00022076150000001216
Safe following speed at time t+τ; τ is the reaction time of vehicle driving; ve is the desired speed.

S5、基于车辆初始轨迹，由汇流模型判断车辆是否可以顺利完成汇流；若判断结果为车辆可以顺利完成汇流，则车辆继续遵从微观跟驰模型的速度行驶；若判断结果为车辆无法顺利完成汇流，则需进一步判断车辆在汇流过程中出现的具体情况，并依据车辆在汇流过程中出现的具体情况对车辆做出步骤S4拟制的协同策略集中对应的协同控制策略，进而执行步骤S6。S5. Based on the initial trajectory of the vehicle, the confluence model judges whether the vehicle can successfully complete the confluence; if the judgment result is that the vehicle can successfully complete the confluence, the vehicle continues to follow the speed of the microscopic car following model; if the judgment result is that the vehicle cannot successfully complete the confluence, Then, it is necessary to further judge the specific situation of the vehicle during the merging process, and according to the specific situation of the vehicle during the merging process, make a coordinated control strategy corresponding to the coordinated strategy set prepared in step S4 for the vehicle, and then execute step S6.

S6、将参与汇流过程的可优化控制的车辆确定为目标车辆，由步骤S5所做出的协同控制策略对目标车辆的行驶轨迹进行优化，并将该优化问题归结为离散时间状态约束的最优控制问题，用动态规划的思想求解得到关于目标车辆的协同优化控制策略，并将关于目标车辆的协同优化控制策略作用于目标车辆，控制目标车辆的运行。具体如下：S6. Determine the optimally controllable vehicle participating in the confluence process as the target vehicle, optimize the driving trajectory of the target vehicle by the collaborative control strategy made in step S5, and reduce the optimization problem to the optimal discrete-time state constraint The control problem is solved by using the idea of dynamic programming to obtain the cooperative optimal control strategy of the target vehicle, and the cooperative optimal control strategy of the target vehicle is applied to the target vehicle to control the operation of the target vehicle. details as follows:

S6-1、基于微观跟驰模型预测目标车辆进入控制区域的时刻为t0，离开控制区域的时刻为tf；并将t0到tf的时间按离散时间间隔τ′平均分为N段，即N＝(tf-t0)/τ′，定义控制决策时刻t为t0+τ′,t0+2τ′,t0+3τ′,t0+4τ′,…,t0+(N-1)τ′,tf；控制区域为控制起点和汇流终点之间的路段；控制起点位于上游监测点和汇流起点之间；S6-1. Based on the microscopic car-following model, it is predicted that the time when the target vehicle enters the control area is t 0 , and the time when it leaves the control area is t f ; and the time from t 0 to t f is divided into N segments equally according to the discrete time interval τ′ , namely N=(t f -t 0 )/τ′, the control decision time t is defined as t 0 +τ′,t 0 +2τ′,t 0 +3τ′,t 0 +4τ′,…,t 0 + (N-1)τ′,t f ; the control area is the road section between the control starting point and the confluence end point; the control starting point is located between the upstream monitoring point and the confluence starting point;

S6-2、根据目标车辆在t0时刻的状态，计算目标车辆在t0+τ′时刻的容许状态集，以及目标车辆从t0时刻的状态到t0+τ′时刻的容许状态集中各个容许状态的转移成本；目标车辆在t0时刻的状态包括目标车辆在t0时刻的速度和位置；S6-2. According to the state of the target vehicle at time t 0 , calculate the allowable state set of the target vehicle at time t 0 +τ', and the allowable state set of the target vehicle from the state at time t 0 to time t 0 +τ'. The transition cost of the allowable state; the state of the target vehicle at time t 0 includes the speed and position of the target vehicle at time t 0 ;

S6-3、根据目标车辆在t0+τ′时刻的容许状态集，计算目标车辆在t0+2τ′时刻的容许状态集，以及目标车辆从t0+τ′时刻的状态到t0+2τ′时刻的容许状态集中各个容许状态的转移成本和累计成本；S6-3. According to the allowable state set of the target vehicle at time t 0 +τ', calculate the allowable state set of the target vehicle at time t 0 +2τ', and the state of the target vehicle from time t 0 +τ' to t 0 + The transition cost and cumulative cost of each allowable state in the allowable state set at time 2τ′;

S6-4、根据步骤S6-3的方法依次计算目标车辆在t0+3τ′,t0+4τ′,…,t0+(N-1)τ′,tf时刻的容许状态集，并计算得到tf时刻的容许状态集中的各个容许状态的累计成本；S6-4, according to the method of step S6-3, sequentially calculate the allowable state set of the target vehicle at time t 0 +3τ', t 0 +4τ',..., t 0 +(N-1)τ', t f , and Calculate the cumulative cost of each allowable state in the allowable state set at time t f ;

S6-5、判断目标车辆在tf时刻的容许状态集中的各个容许状态是否满足可以顺利完成汇流的条件，并将满足条件的容许状态纳入最终容许状态集中；S6-5, judging whether each allowable state in the allowable state set of the target vehicle at time tf satisfies the conditions for successfully completing the confluence, and incorporating the allowable states that meet the conditions into the final allowable state set;

S6-6、选择最终容许状态集中累计成本最小的容许状态作为tf时刻的最优状态；S6-6, select the allowable state with the smallest cumulative cost in the final allowable state set as the optimal state at time t f ;

S6-7、根据tf时刻的最优状态逆推得到t0+τ′,t0+2τ′,t0+3τ′,t0+4τ′,…,t0+(N-1)τ′,tf各时刻的最优状态，并将每个最优状态对应的控制决策纳入协同优化控制策略；S6-7. According to the optimal state at time t f , t 0 +τ′,t 0 +2τ′,t 0 +3τ′,t 0 +4τ′,…,t 0 +(N-1)τ are obtained by inverse inference ′, t f the optimal state at each moment, and incorporate the control decision corresponding to each optimal state into the collaborative optimization control strategy;

S6-8、根据协同优化控制策略对目标车辆在控制区域间的运行进行控制。S6-8. Control the operation of the target vehicle between the control areas according to the collaborative optimization control strategy.

步骤S6用动态规划的思想求解得到关于目标车辆的协同优化控制策略，模型如下：Step S6 uses the idea of dynamic programming to obtain the collaborative optimization control strategy for the target vehicle. The model is as follows:

Figure BDA0002207615000000131
Figure BDA0002207615000000131

Figure BDA0002207615000000132
Figure BDA0002207615000000132

Figure BDA0002207615000000133
Figure BDA0002207615000000133

Figure BDA0002207615000000134
Figure BDA0002207615000000134

Figure BDA0002207615000000135
Figure BDA0002207615000000135

Figure BDA0002207615000000136
Figure BDA0002207615000000136

Figure BDA0002207615000000137
Figure BDA0002207615000000137

Figure BDA0002207615000000138
Figure BDA0002207615000000138

将目标车辆记为目标车辆i。目标函数方程(8)表示目标车辆i在控制区间内，行驶足够平缓且行驶速度接近期望速度，ve为期望速度。方程(9)表示目标车辆i的初始状态，即目标车辆i在t0时刻的状态。方程(10)与(11)表示目标车辆的过渡状态，方程(12)则是对汇流车辆k最终状态的汇流效用约束。K为车辆组合，表示匝道和主干道上直接参与汇流的车辆组合，且K∈{K1,K2,K3}，车辆组合

Figure BDA0002207615000000139
表示直接参与汇流的车辆有车辆k、车辆
Figure BDA00022076150000001310
车辆
Figure BDA00022076150000001311
车辆组合
Figure BDA00022076150000001312
表示直接参与汇流的车辆有车辆k、车辆
Figure BDA00022076150000001313
车辆组合
Figure BDA00022076150000001314
表示直接参与汇流的车辆有车辆
Figure BDA00022076150000001315
车辆k；而目标车辆i是车辆组合K中可优化控制的车辆。定义目标车辆i在每个阶段n的一组容许状态(即每个决策时刻的容许状态集)，表示为
Figure BDA00022076150000001316
Belman递推公式如方程(13)和方程(14)所示，用于求解子问题。方程(13)表示目标车辆i从初始状态(t0时刻的状态)到第1阶段(t0+τ时刻)的转移成本；第1阶段(t0+τ时刻)的容许状态是
Figure BDA00022076150000001317
Figure BDA00022076150000001318
为第1阶段(t0+τ时刻)的容许状态集。方程(14)中，
Figure BDA00022076150000001319
是指每个子问题的目标值，即目标车辆i从初始状态(t0时刻的状态)到第n阶段(t0+nτ时刻)的累计成本；第n-1阶段(t0+(n-1)τ时刻)的容许状态是
Figure BDA00022076150000001320
第n阶段(t0+nτ时刻)的容许状态是
Figure BDA00022076150000001321
Figure BDA00022076150000001322
为第n阶段(t0+nτ时刻)的容许状态集；
Figure BDA00022076150000001323
是从第n-1阶段(t0+(n-1)τ时刻)的容许状态
Figure BDA00022076150000001324
到第n阶段(t0+nτ时刻)的容许状态
Figure BDA00022076150000001325
的转移成本，用方程(15)表示。Denote the target vehicle as target vehicle i. The objective function equation (8) indicates that the target vehicle i is in the control interval, travels smoothly enough and the travel speed is close to the desired speed, and ve is the desired speed. Equation (9) represents the initial state of the target vehicle i, that is, the state of the target vehicle i at time t 0 . Equations (10) and (11) represent the transition state of the target vehicle, and equation (12) is the merging utility constraint on the final state of the merging vehicle k. K is the vehicle combination, which represents the vehicle combination directly participating in the convergence on the ramp and the main road, and K∈{K 1 ,K 2 ,K 3 }, the vehicle combination
Figure BDA0002207615000000139
Indicates that the vehicles directly participating in the confluence include vehicle k, vehicle
Figure BDA00022076150000001310
vehicle
Figure BDA00022076150000001311
vehicle combination
Figure BDA00022076150000001312
Indicates that the vehicles directly participating in the confluence include vehicle k, vehicle
Figure BDA00022076150000001313
vehicle combination
Figure BDA00022076150000001314
Indicates that vehicles directly participating in the confluence have vehicles
Figure BDA00022076150000001315
vehicle k; and the target vehicle i is the vehicle that can be optimally controlled in the vehicle combination K. Define a set of permissible states of the target vehicle i at each stage n (that is, the set of permissible states at each decision moment), expressed as
Figure BDA00022076150000001316
The Belman recursion formulations are shown in Equation (13) and Equation (14) for solving the subproblems. Equation (13) represents the transition cost of the target vehicle i from the initial state (state at time t 0 ) to the first stage (time t 0 +τ ); the allowable state in the first stage (time t 0 +τ ) is
Figure BDA00022076150000001317
Figure BDA00022076150000001318
is the allowable state set of the first stage (time t 0 +τ). In equation (14),
Figure BDA00022076150000001319
refers to the target value of each sub-problem, that is, the cumulative cost of the target vehicle i from the initial state (state at time t 0 ) to the nth stage (time t 0 +nτ); the n-1th stage (t 0 +(n- 1) The allowable state at time τ) is
Figure BDA00022076150000001320
The allowable state of the nth stage (time t 0 +nτ) is
Figure BDA00022076150000001321
Figure BDA00022076150000001322
is the allowable state set of the nth stage (time t 0 +nτ);
Figure BDA00022076150000001323
is the allowable state from the n-1th stage (time t 0 +(n-1)τ)
Figure BDA00022076150000001324
Admissible state to the nth stage (time t 0 +nτ)
Figure BDA00022076150000001325
The transfer cost of , expressed by equation (15).

本发明方法通过MATLAB编程建立微观交通流仿真环境(包括车辆跟驰与汇流)，计算机编程实现高速公路混合交通流的仿真实验环境的参数及取值如下表所示：The method of the present invention establishes a microscopic traffic flow simulation environment (including vehicle following and merging) through MATLAB programming, and the parameters and values of the simulation experiment environment for realizing the mixed traffic flow of the expressway by computer programming are shown in the following table:

Figure BDA0002207615000000141
Figure BDA0002207615000000141

图6给出了参与汇流过程的车辆组合和车型组合为HHA的三辆车，在无法顺利完成汇流的第二种情况(R2)下，且在无协同优化控制情况下的汇流轨迹图，其中图(a)为位置-时间关系图，图(b)为速度-时间关系图，图(c)为加速度-时间关系图。Figure 6 shows the combination of vehicles participating in the convergence process and the three vehicles whose vehicle type combination is HHA. In the second case (R2) where the convergence cannot be successfully completed, and without collaborative optimal control, the convergence trajectory diagram, where Figure (a) is a position-time relationship diagram, Figure (b) is a velocity-time relationship diagram, and Figure (c) is an acceleration-time relationship diagram.

图7给出了参与汇流过程的车辆组合和车型组合为HHA的三辆车，在无法顺利完成汇流的第二种情况(R2)下，且在有协同优化控制情况下的汇流轨迹图，其中图(a)为位置-时间关系图，图(b)为速度-时间关系图，图(c)为加速度-时间关系图。Figure 7 shows the combination of vehicles participating in the convergence process and the three vehicles whose vehicle type combination is HHA. In the second case (R2) where the convergence cannot be successfully completed, and the convergence trajectory diagram under the condition of collaborative optimization control, where Figure (a) is a position-time relationship diagram, Figure (b) is a velocity-time relationship diagram, and Figure (c) is an acceleration-time relationship diagram.

对比优化前、后的汇流轨迹图，从速度-时间关系图(图6中(b)和图7中(b))以及加速度-时间关系图(图6中(c)和图7中(c))中可以看出，参与汇流的车辆在有协同优化控制情况下要比在无协同优化控制情况下的运行速度高且平稳，汇流车辆也能以较为平滑的速度尽早的完成汇流。Comparing the confluence trajectories before and after optimization, from the velocity-time relationship diagram (Fig. 6(b) and Fig. 7(b)) and the acceleration-time relationship diagram (Fig. 6(c) and Fig. 7(c) ))), it can be seen that the running speed of the vehicles participating in the merging is higher and smoother than that without the cooperative optimal control, and the merging vehicles can also complete the merging as soon as possible at a relatively smooth speed.

图8给出了参与汇流过程的车辆组合和车型组合为HAH的三辆车，在无法顺利完成汇流的第一种情况(R1)下，且在无协同优化控制情况下的汇流轨迹图，其中图(a)为位置-时间关系图，图(b)为速度-时间关系图，图(c)为加速度-时间关系图。Figure 8 shows the combination of vehicles participating in the convergence process and the three vehicles whose vehicle type combination is HAH. In the first case (R1) where the convergence cannot be successfully completed, and without collaborative optimal control, the convergence trajectory diagram, where Figure (a) is a position-time relationship diagram, Figure (b) is a velocity-time relationship diagram, and Figure (c) is an acceleration-time relationship diagram.

图9给出了参与汇流过程的车辆组合和车型组合为HAH的三辆车，在无法顺利完成汇流的第一种情况(R1)下，且在有协同优化控制情况下的汇流轨迹图，其中图(a)为位置-时间关系图，图(b)为速度-时间关系图，图(c)为加速度-时间关系图。Figure 9 shows the combination of vehicles participating in the convergence process and the three vehicles whose vehicle type combination is HAH. In the first case (R1) where the convergence cannot be successfully completed, and with the collaborative optimization control, the convergence trajectory diagram, where Figure (a) is a position-time relationship diagram, Figure (b) is a velocity-time relationship diagram, and Figure (c) is an acceleration-time relationship diagram.

对比优化前、后的汇流轨迹图，从速度-时间关系图(图8中(b)和图9中(b))以及加速度-时间关系图(图8中(c)和图9中(c))中可以看出，参与汇流的车辆在有协同优化控制情况下要比在无协同优化控制情况下的运行速度高且平稳，主干道后车受到汇流行为影响也较小。Comparing the confluence trajectories before and after optimization, from the velocity-time relationship diagram (Fig. 8(b) and Fig. 9(b)) and the acceleration-time relationship diagram (Fig. 8(c) and Fig. 9(c) ))), it can be seen that the running speed of the vehicles participating in the confluence is higher and smoother than that without the cooperative optimal control, and the vehicles behind the main road are less affected by the confluence behavior.

上述结合附图对本发明进行了示例性描述，显然本发明的具体实现并不受本文所示的实施例限制。The present invention has been exemplarily described above with reference to the accompanying drawings, and it is obvious that the specific implementation of the present invention is not limited by the embodiments shown herein.

Claims (3)

1. A highway mixed traffic flow convergence collaborative optimization bottom layer control method is characterized by comprising the following steps:
s1, determining a micro-following model, and describing the following state of the vehicle by using the micro-following model, wherein the following state of the vehicle comprises the speed, the acceleration and the position of the vehicle;
s2, acquiring the time and the speed of an upstream monitoring point of the vehicle in the mixed traffic flow before the vehicle passes through the confluence area, and predicting the initial trajectory of the vehicle between the upstream monitoring point and the confluence terminal point by using the micro-following model; a road section with a certain distance between the upstream monitoring point and the confluence starting point; the confluence starting point is positioned between the upstream monitoring point and the confluence terminal point; a section between the confluence start point and the confluence end point constitutes the confluence area;
s3, adding acceleration constraint, distance constraint and safety constraint based on the microcosmic car-following model, and establishing a convergence model;
s4, aiming at various situations which can not smoothly complete the convergence in the convergence process under the mixed traffic flow scene, a cooperative control strategy set is prepared;
s5, judging whether the vehicle can smoothly complete the confluence through the confluence model based on the initial track of the vehicle; if the judgment result is that the vehicle can smoothly complete the confluence, the vehicle continues to run according to the speed of the microscopic follow-up model; if the judgment result is that the vehicles cannot smoothly complete the confluence, further judging the specific situation of the vehicles in the confluence process, and making a cooperative control strategy corresponding to the cooperative strategy set simulated in the step S4 for the vehicles according to the specific situation of the vehicles in the confluence process, so as to execute a step S6;
s6, determining the vehicles which participate in the confluence process and can be controlled in an optimized mode as target vehicles, optimizing the running tracks of the target vehicles through the cooperative control strategy made in the step S5, solving the optimization problem into an optimal control problem of discrete time state constraint, solving the cooperative optimization control strategy related to the target vehicles through a dynamic programming idea, acting the cooperative optimization control strategy related to the target vehicles on the target vehicles, and controlling the running of the target vehicles; the optimization problem is the problem of optimizing the driving track of the target vehicle by the cooperative control strategy made in step S5;
the step S4 specifically includes:
assuming that the ramp and the main road are both one-way lanes, the vehicle k on the ramp is converged into the interval between two vehicles of the continuous traffic flow of the main road, and the two vehicles of the continuous traffic flow of the main road are respectively used as vehicles
Figure FDA0003548221780000011
And a vehicle
Figure FDA0003548221780000012
Show, wherein the vehicle
Figure FDA0003548221780000013
Indicating the front vehicle, vehicle
Figure FDA0003548221780000014
Representing a rear vehicle;
aiming at various conditions possibly occurring in the confluence process in the mixed traffic flow scene, vehicles are driven based on the microcosmic following model
Figure FDA0003548221780000015
Vehicle k and vehicle
Figure FDA0003548221780000016
The relationship between them is divided into that the confluence can be smoothly completed and that the confluence cannot be smoothly completed; the failure to smoothly complete the confluence is divided into four cases, the first case is denoted as R1, which indicates that the vehicle k and the vehicle
Figure FDA0003548221780000017
The distance between the two parts is too close, and the constraint condition that the confluence can be smoothly completed is not satisfied; the second case is denoted as R2 and represents vehicle k and vehicle
Figure FDA0003548221780000018
The distance between the two parts is too close, and the constraint condition that the confluence can be smoothly completed is not satisfied; the third case is denoted as R3 and represents vehicle k and vehicle
Figure FDA0003548221780000019
And with vehicles
Figure FDA00035482217800000110
Meets the basic spacing requirement but the confluence process is not comfortable; the fourth case is denoted as R4 and represents vehicle k and vehicle
Figure FDA00035482217800000111
And with vehicles
Figure FDA00035482217800000112
The distances between the two are too close to meet the requirement of smooth confluenceThe constraint of (2);
aiming at four conditions that confluence cannot be smoothly completed, vehicles which can be optimally controlled need to be cooperatively controlled; h represents a human driving vehicle, which is a vehicle which cannot be optimally controlled; an automatic driving vehicle is represented by A and is an optimally controllable vehicle; n represents that no front vehicle participates in the confluence or no rear vehicle participates in the confluence; and prescribes the order of combination of the vehicles as vehicles in turn
Figure FDA0003548221780000021
Vehicle k and vehicle
Figure FDA0003548221780000022
Based on different vehicle combinations, different vehicle type combinations and the four conditions that confluence cannot be smoothly completed, a collaborative control strategy set is prepared, and is shown in the following table:
Figure FDA0003548221780000023
Figure FDA0003548221780000031
in the above table, the non-optimization means that the vehicle has no corresponding control strategy under the condition that the convergence cannot be smoothly completed correspondingly, at this time, the vehicle k on the ramp takes the end of the ramp as a stopped virtual front vehicle, and continuously decelerates or even stops to wait by following the microscopic following model until the vehicle interval meeting the convergence appears on the main road, and the vehicle k is converged into the main road;
the acceleration of the vehicle k is controlled by a decision variable u of the vehicle k at the time tk(t) satisfies:
Figure FDA0003548221780000032
and v isk(t)+uk(t)τ≤ve；
The control vehicle k decelerates, and a decision variable u of the vehicle k at the time t is usedk(t) satisfies:
Figure FDA0003548221780000033
and v isk(t)+uk(t)τ≥0；
The unknown control state of the vehicle k is that the control mode of the vehicle k can be to control the vehicle k to decelerate, also can control the vehicle k to accelerate or not control the vehicle k, and at the moment t, the decision variable u of the vehicle k is determinedk(t) satisfies:
vk(t)+uk(t)τ≤ve，
and v isk(t)+uk(t)τ≥0；
The uncontrolled vehicle k is a decision variable u for the vehicle k at time tk(t) satisfies:
Figure FDA0003548221780000041
i.e. uk(t)＝0；
The control vehicle
Figure FDA0003548221780000042
Acceleration of the vehicle at time t
Figure FDA0003548221780000043
Decision variables of
Figure FDA0003548221780000044
Satisfies the following conditions:
Figure FDA0003548221780000045
and is
Figure FDA0003548221780000046
The uncontrolled vehicle
Figure FDA0003548221780000047
Make the vehicle at the time t
Figure FDA0003548221780000048
Decision variables of
Figure FDA0003548221780000049
Satisfies the following conditions:
Figure FDA00035482217800000410
namely, it is
Figure FDA00035482217800000411
The control vehicle
Figure FDA00035482217800000412
Decelerating the vehicle at time t
Figure FDA00035482217800000413
Decision variables of
Figure FDA00035482217800000414
Satisfies the following conditions:
Figure FDA00035482217800000415
and is
Figure FDA00035482217800000416
The uncontrolled vehicle
Figure FDA00035482217800000417
Make the vehicle at the time t
Figure FDA00035482217800000418
Decision variables of
Figure FDA00035482217800000419
Satisfies the following conditions:
Figure FDA00035482217800000420
namely, it is
Figure FDA00035482217800000421
Wherein u isk(t) as a decision variable for vehicle k at time t, representing the acceleration of vehicle k at time t; v. ofk(t) is the speed of vehicle k at time t;
Figure FDA00035482217800000422
is the safe following speed of the vehicle k at the moment of t + tau predicted according to the microcosmic following model;
Figure FDA00035482217800000423
as the vehicle at time t
Figure FDA00035482217800000424
Represents the vehicle at time t
Figure FDA00035482217800000425
Acceleration of (2);
Figure FDA00035482217800000426
is a vehicle
Figure FDA00035482217800000427
Velocity at time t;
Figure FDA00035482217800000428
is a vehicle predicted according to the micro-following model
Figure FDA00035482217800000429
Safe following speed at time t + τ;
Figure FDA00035482217800000430
as the vehicle at time t
Figure FDA00035482217800000431
Represents the vehicle at time t
Figure FDA00035482217800000432
Acceleration of (2);
Figure FDA00035482217800000433
is a vehicle
Figure FDA00035482217800000434
Velocity at time t;
Figure FDA00035482217800000435
is a vehicle predicted according to the micro-following model
Figure FDA00035482217800000436
Safe following speed at time t + τ; τ is the reaction time of the vehicle driving; v. ofeIs the desired speed; l isk(t) represents the relative distance between the vehicle k and its following preceding vehicle at time t;
Figure FDA00035482217800000437
representing the speed of the car k before the car k follows at the time t;
Figure FDA00035482217800000438
indicating vehicle at time t
Figure FDA00035482217800000439
Relative distance to the car before it is driven;
Figure FDA00035482217800000440
indicating vehicles
Figure FDA00035482217800000441
The speed of the car before the car is followed at the time t;
Figure FDA00035482217800000442
indicating vehicle at time t
Figure FDA00035482217800000443
Relative distance to the car before it is driven;
Figure FDA00035482217800000444
indicating vehicles
Figure FDA00035482217800000445
The speed of the car ahead at time t.
2. The highway mixed traffic flow convergence collaborative optimization floor control method according to claim 1, wherein the step S6 specifically comprises:
s6-1, predicting the moment t when the target vehicle enters the control area based on the microcosmic car-following model0The time when the user leaves the control area is tf(ii) a And will t0To tfIs divided into N segments on average in discrete time intervals τ', i.e., N ═ tf-t0) T', defining the control decision time as t0+τ′,t0+2τ′,t0+3τ′,t0+4τ′,…,t0+(N-1)τ′,tf(ii) a The control area is a road section between a control starting point and the convergence end point; the control starting point is positioned between the upstream monitoring point and the bus starting pointA (c) is added;
s6-2, according to the target vehicle at t0The state of the time, the target vehicle is calculated at t0The set of allowable states at time + τ', and the target vehicle from t0State of time t0The allowable states at time + τ' aggregate the transition costs for each allowable state; said target vehicle is at t0The state at time t includes the target vehicle being at t0Speed and position of the moment;
s6-3, according to the target vehicle at t0The allowable state set at the time + tau' is calculated for the target vehicle at t0The allowable state set at time +2 τ', and the target vehicle from t0State at time + τ' to t0The allowable states at time +2 τ' are grouped into transition costs and cumulative costs for each allowable state;
s6-4, sequentially calculating the target vehicle at t according to the method of the step S6-30+3τ′,t0+4τ′,…,t0+(N-1)τ′,tfThe allowable state set of the time is calculated to obtain tfThe cumulative cost of each allowable state in the time-wise allowable state set;
s6-5, judging that the target vehicle is at tfWhether each allowable state in the allowable state set at the moment meets the condition that the confluence can be smoothly completed or not is judged, and the allowable state meeting the condition is brought into the final allowable state set;
s6-6, selecting the allowable state with the minimum accumulated cost in the final allowable state set as tfThe optimal state of the moment;
s6-7, according to tfThe optimal state of the moment is reversely deduced to obtain t0+τ′,t0+2τ′,t0+3τ′,t0+4τ′,…,t0+(N-1)τ′,tfThe optimal state at each moment, and the control decision corresponding to each optimal state is brought into a cooperative optimization control strategy;
and S6-8, controlling the operation of the target vehicle among the control areas according to the cooperative optimization control strategy.
3. The expressway mixed traffic flow convergence collaborative optimization floor control method according to claim 1 or 2, wherein a microscopic traffic flow simulation environment is constructed, and simulation results before and after optimization under different traffic situations are compared.
CN201910886980.2A 2019-09-19 2019-09-19 A low-level control method for synergistic optimization of highway mixed traffic flow convergence Active CN110570049B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910886980.2A CN110570049B (en) 2019-09-19 2019-09-19 A low-level control method for synergistic optimization of highway mixed traffic flow convergence

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910886980.2A CN110570049B (en) 2019-09-19 2019-09-19 A low-level control method for synergistic optimization of highway mixed traffic flow convergence

Publications (2)

Publication Number Publication Date
CN110570049A CN110570049A (en) 2019-12-13
CN110570049B true CN110570049B (en) 2022-04-29

Family

ID=68781088

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910886980.2A Active CN110570049B (en) 2019-09-19 2019-09-19 A low-level control method for synergistic optimization of highway mixed traffic flow convergence

Country Status (1)

Country Link
CN (1) CN110570049B (en)

Families Citing this family (20)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109978260B (en) * 2019-03-26 2023-02-21 重庆邮电大学 Prediction method for car-following behavior of networked vehicles under mixed traffic flow
JP7141421B2 (en) * 2020-03-16 2022-09-22 本田技研工業株式会社 VEHICLE CONTROL DEVICE, VEHICLE CONTROL METHOD, AND PROGRAM
CN111583640B (en) * 2020-04-30 2022-04-22 重庆大学 On-ramp cooperative merging control method based on vehicle-road coordination
CN111583644B (en) * 2020-05-08 2021-11-26 重庆大学 Control method for network connection automatic vehicle of ramp convergence area on hybrid traffic express way
CN111599194B (en) * 2020-05-27 2024-06-18 中汽研汽车检验中心（天津）有限公司 Highway entrance ramp heterogeneous traffic flow speed guiding system and guiding method
CN111785088B (en) * 2020-06-23 2021-08-20 大连理工大学 A two-layer collaborative optimization method for ramp merging of connected vehicles
CN111754774B (en) * 2020-07-02 2022-04-19 清华大学 Safe self-organizing traffic control method for intelligent network-connected automobile at expressway ramp port
CN114201847B (en) * 2020-10-12 2024-06-21 西南交通大学 Simplifying method for multi-vehicle track control under mixed traffic flow
CN112735126B (en) * 2020-12-24 2023-02-03 成都格林希尔德交通科技有限公司 Mixed traffic flow cooperative optimization control method based on model predictive control
CN113055474B (en) * 2021-03-12 2022-11-11 成都格林希尔德交通科技有限公司 Micro road right transaction system
CN113033883A (en) * 2021-03-12 2021-06-25 西南交通大学 Optimization control and profit allocation method in mixed traffic flow environment
CN113071481B (en) * 2021-04-09 2022-11-29 长沙智能驾驶研究院有限公司 Vehicle control method and device, computer readable storage medium and vehicle
CN113781770B (en) * 2021-08-13 2023-03-31 同济大学 Simulation double-clutch type confluence area vehicle-road cooperative control system
CN114655214B (en) * 2022-04-24 2024-04-26 重庆长安汽车股份有限公司 Sectional control method and system for driving through converging port
CN114707364B (en) * 2022-06-02 2022-09-02 西南交通大学 Ramp vehicle convergence simulation method, device, equipment and readable storage medium
CN115188184B (en) * 2022-06-20 2024-03-19 海信集团控股股份有限公司 Vehicle speed limit processing method, device and apparatus
CN115131965B (en) * 2022-06-23 2023-07-07 重庆长安汽车股份有限公司 Vehicle control method, device, system, electronic equipment and storage medium
CN115620541B (en) * 2022-09-09 2023-06-02 交通运输部公路科学研究所 Intelligent vehicle import method and device and intelligent highway information physical system
CN116153137B (en) * 2022-12-28 2024-05-31 清华大学 Multi-lane ramp collaborative import optimization scheduling method based on vehicle conflict relation graph
CN118038702B (en) * 2024-03-11 2024-08-13 北京工业大学 Method and system for calculating and analyzing upper and lower limits of traffic capacity of lanes in mixed traffic scene

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102368351A (en) * 2011-10-19 2012-03-07 北京航空航天大学 Method for eliminating traffic conflict of two vehicles at intersection without signal
CN102662320A (en) * 2012-03-05 2012-09-12 吴建平 Car-following simulation method based on fuzzy mathematics
CN105631793A (en) * 2015-12-18 2016-06-01 华南理工大学 Intelligent traffic flow congestion dispersal method through vehicle group autonomous cooperative scheduling
CN108806252A (en) * 2018-06-19 2018-11-13 西南交通大学 A kind of Mixed Freeway Traffic Flows collaboration optimal control method
CN108986488A (en) * 2018-07-23 2018-12-11 东南大学 Ring road imports collaboration track and determines method and apparatus under a kind of truck traffic environment

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20180276986A1 (en) * 2017-03-22 2018-09-27 Toyota Research Institute, Inc. Vehicle-to-human communication in an autonomous vehicle operation

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102368351A (en) * 2011-10-19 2012-03-07 北京航空航天大学 Method for eliminating traffic conflict of two vehicles at intersection without signal
CN102662320A (en) * 2012-03-05 2012-09-12 吴建平 Car-following simulation method based on fuzzy mathematics
CN105631793A (en) * 2015-12-18 2016-06-01 华南理工大学 Intelligent traffic flow congestion dispersal method through vehicle group autonomous cooperative scheduling
CN108806252A (en) * 2018-06-19 2018-11-13 西南交通大学 A kind of Mixed Freeway Traffic Flows collaboration optimal control method
CN108986488A (en) * 2018-07-23 2018-12-11 东南大学 Ring road imports collaboration track and determines method and apparatus under a kind of truck traffic environment

Also Published As

Publication number Publication date
CN110570049A (en) 2019-12-13

Similar Documents

Publication Publication Date Title
CN110570049B (en) A low-level control method for synergistic optimization of highway mixed traffic flow convergence
CN110473404B (en) A bottom control method for coordinated optimization of mixed traffic flow at urban intersections
CN108806252B (en) A kind of Mixed Freeway Traffic Flows collaboration optimal control method
CN110599772B (en) Mixed traffic flow cooperative optimization control method based on double-layer planning
CN110298131A (en) Automatic Pilot lane-change decision model method for building up under a kind of mixing driving environment
CN112614340B (en) Method and device for enabling branch vehicles to converge into main road, electronic equipment and storage medium
CN111746539B (en) A method for strict and safe lane change and queuing control for intelligent networked vehicles
CN110992695B (en) Vehicle urban intersection traffic decision multi-objective optimization method based on conflict resolution
CN110362910A (en) Automatic driving vehicle lane-change conflict coordination method for establishing model based on game theory
CN111081065A (en) Intelligent vehicle cooperative lane change decision-making model under mixed traffic conditions
CN109410561B (en) Uniform and heterogeneous formation driving control method for vehicles on highway
CN114707364B (en) Ramp vehicle convergence simulation method, device, equipment and readable storage medium
CN110930697A (en) A rule-based intelligent networked vehicle collaborative entry control method
CN114023108B (en) A mixed traffic flow lane change model and lane change simulation method
CN114842644A (en) Traffic capacity calculation method for mixed traffic flow intersection area
CN115641717A (en) Method, device and storage medium for cooperative merging control of expressway mainline-ramp vehicles based on mixed traffic flow
Liu et al. Eco-Friendly On-Ramp Merging Strategy for Connected and Automated Vehicles in Heterogeneous Traffic
CN116229715A (en) Method and system for generating continuous flow in road weaving area
CN105788238B (en) Class spring vehicle following-model method for building up based on quantum door and self adaptive control
CN113823076B (en) Instant-stop and instant-walking road section blockage relieving method based on networked vehicle coordination control
CN113593225A (en) Single-point intersection vehicle control method oriented to pure internet environment
CN114201847B (en) Simplifying method for multi-vehicle track control under mixed traffic flow
CN116811882A (en) Vehicle following speed control method based on deep reinforcement learning
Zhao et al. Coordinated throttle and brake fuzzy controller design for vehicle following
CN113781770B (en) Simulation double-clutch type confluence area vehicle-road cooperative control system

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant