CN102592028A

CN102592028A - Method for optimizing setting position and number of horizontal passageway on lane side

Info

Publication number: CN102592028A
Application number: CN2012100435273A
Authority: CN
Inventors: 王晓华; 白子建; 刘治国; 马红伟; 钟石泉; 郑利; 赵巍; 王海燕; 张磊; 严西华
Original assignee: Tianjin Municipal Engineering Design and Research Institute
Current assignee: Tianjin Municipal Engineering Design and Research Institute
Priority date: 2012-02-24
Filing date: 2012-02-24
Publication date: 2012-07-18

Abstract

The invention relates to the technical field of road design. In order to provide a clear method to guide the setting of the location and quantity of the lateral passages on the side of the lane of the large traffic hub. In order to achieve the above object, the technical solution adopted by the present invention is that the position and quantity optimization method of the lateral passages on the side of the roadway are adopted as a platform for the simulation of the two-dimensional cellular automata model, and the traffic flow of the roadside design form including the side passageway is included. Analysis of the characteristics, detailed description of the multiple influences of the docking model objective function, docking rules, cells of the transverse passage, and turning trajectories, and integrated simulation of the target to obtain the optimization data of the passage setting position and quantity. The invention is mainly applied to road design.

Description

Optimizing method for setting position and quantity of transverse passages on lane side

技术领域 technical field

本发明属于道路设计技术领域，尤其是涉及车道边横向通道设置位置及数量优化方法。The invention belongs to the technical field of road design, and in particular relates to a method for optimizing the position and quantity of transverse passages at the side of a lane.

背景技术 Background technique

为了满足日益增长的现代出行需求，各个城市相继修建大型交通枢纽。作为交通枢纽特有的交通设施，车道边相关问题的研究日益受到国内外学者的重视。国外早在20世纪60年代就开始了对大型枢纽系统车道边设施的规划布局及相关研究。近年来，随着基于交通行为理论的一些微观动态交通分析模型研究的深入，开发了一些城市交通微观模拟系统。这些研究重心的主要放在枢纽内旅客和行李的流动仿真上，对枢纽外路边和道路单元的仿真研究并不多。国内方面，在对于车道边的设置形式上缺乏系统的理论支持。尤其对于横向通道设置的位置及数量往往是根据以往的设计资料结合实际的使用经验进行计算估计，缺乏使得横向通道设置位置及数量达到最优化的方法。In order to meet the growing demand for modern travel, cities have successively built large-scale transportation hubs. As a unique traffic facility in a transportation hub, the research on the related problems of the lane side has been paid more and more attention by scholars at home and abroad. As early as the 1960s, foreign countries began to plan the layout and related research on the laneside facilities of large-scale hub systems. In recent years, with the in-depth study of some microscopic dynamic traffic analysis models based on traffic behavior theory, some urban traffic microscopic simulation systems have been developed. The focus of these studies is mainly on the flow simulation of passengers and luggage in the hub, and there are not many simulation studies on the roadside and road units outside the hub. Domestically, there is a lack of systematic theoretical support for the setting of lane edges. In particular, the position and quantity of transverse passages are often calculated and estimated based on previous design data combined with actual use experience, and there is a lack of methods to optimize the position and quantity of transverse passages.

从已有相关文献阅读分析可知，针对大型枢纽车道边的研究的已有技术具有以下特点：From the reading and analysis of the existing relevant literature, it can be seen that the existing technologies for the research on the lanes of large hubs have the following characteristics:

(1)以往针对大型枢纽车道边的研究方向主要是对枢纽各个交通设施之间的动态分析研究，针对车道边的车道设置的方式方法的研究尚且不足，其中对有横向通道车道边的设置亦是通过经验得出的估计值，缺乏确切的理论方法及数据的支撑。(1) In the past, the research direction of the lanes of large-scale hubs was mainly the dynamic analysis of the various traffic facilities in the hub. The research on the way of setting up the lanes of the lanes is still insufficient, and the setting of the lanes with transverse passages is also not enough. It is an estimated value obtained through experience and lacks the support of exact theoretical methods and data.

(2)在已开发的软件中，主要的仿真技术大多基于交通行为理论的一些微观动态交通分析模型。应用基于一体化的集成仿真环境的元胞自动机进行研究尚处于初级发展阶段。(2) Among the developed software, most of the main simulation technologies are based on some microscopic dynamic traffic analysis models of traffic behavior theory. The application of cellular automata based on an integrated simulation environment is still in its infancy.

发明内容 Contents of the invention

本发明旨在解决克服现有技术的不足，提供明确的方法对大型交通枢纽车道边横向通道位置及数量的设置进行指导。为达到上述目的，本发明采取的技术方案是，车道边横向通道设置位置及数量优化方法，采用二维元胞自动机模型模拟仿真作为平台，通过包括带有横向通道的车道边设计形式交通流特性的分析，停靠模型目标函数、停靠规则、横向通道的元胞、转弯轨迹的多种影响的详细描述，对目标进行集成仿真，得出通道设置位置及数量优化数据。The invention aims to overcome the deficiencies of the prior art, and provides a clear method to guide the setting of the position and quantity of the lateral passages on the side of the lane of the large-scale traffic hub. In order to achieve the above object, the technical solution adopted by the present invention is that the position and quantity optimization method of the lateral passages on the side of the roadway are adopted as a platform for the simulation of the two-dimensional cellular automata model, and the traffic flow of the roadside design form including the side passageway is included. Analysis of the characteristics, detailed description of the multiple influences of the docking model objective function, docking rules, cells of the transverse passage, and turning trajectories, and integrated simulation of the target to obtain the optimization data of the passage setting position and quantity.

横向通道的车道边设计形式交通流特性的分析是指道边设置有双排停车道的情况。The analysis of the traffic flow characteristics of the design form of the lane side of the transverse aisle refers to the situation that there are double-row parking lanes on the side of the road.

停靠模型目标函数是指：令x_e表示车辆当前位置，x_e表示下车客人要求的入站口行车方向的起点位置，D_e表示下车客人要求的入站口宽度，驾驶员的换道意愿函数可表示为：The objective function of the docking model refers to: let x _e represent the current position of the vehicle, x _e represent the starting position of the driving direction of the entrance required by the passengers getting off the bus, D _e represents the width of the entrance gate required by the passengers getting off the bus, and the driver's lane change The desirability function can be expressed as:

${v v}_{n no}^{d d} ((k k,, {x x}_{c c},, {x x}_{e e})) = = \{\begin{matrix} random random ((00,, {δ δ}_{44,, k k})) & if if {x x}_{c c} > > {d d}_{s the s}^{n no} ((k k)) + + {x x}_{e e} \\ 11 - - {[[(({x x}_{c c} - - {x x}_{e e} - - {d d}_{s the s}^{n no} ((k k)) + + {D D.}_{e e})) / / (({x x}_{e e} + + {d d}_{s the s}^{n no} ((k k)) - - {D D.}_{e e}))]]}^{{λ λ}_{44}} & if if {x x}_{e e} - - {D D.}_{e e} + + {d d}_{s the s}^{n no} ((k k)) < < {x x}_{c c} \leq \leq {d d}_{s the s}^{n no} ((k k)) + + {x x}_{e e} \\ 11 & if if {x x}_{c c} \leq \leq {x x}_{e e} - - {D D.}_{e e} + + {d d}_{s the s}^{n no} ((k k)) \end{matrix}$

$v_{w}^{d} (k, x_{c}, x_{e}) = \{\begin{matrix} random (0, δ_{5, k}) & if x_{c} > d_{s}^{w} (k) + x_{e} \\ 1 - {[(x_{c} - x_{e} - d_{s}^{w} (k) + D_{e}) / (x_{e} + d_{s}^{w} (k) - D_{e})]}^{λ_{5}} & if x_{e} - D_{e} + d_{s}^{w} (k) < x_{c} \leq d_{s}^{w} (k) + x_{e} \\ 1 & if x_{c} \leq x_{e} - D_{e} + d_{s}^{w} (k) \end{matrix}$ 上式中，δ_4，k表示第k类驾驶员在带有横向通道的车道边内侧非换道区域的换道意愿上限，与驾驶员的风格和当前交通状况有关；λ₄表示在带有横向通道的车道边内侧目标换道区域的换道意愿系数；换入内侧车道的换道意愿

是驾驶员类型k、车辆当前位置x_c、下车客人要求的入站口起点位置x_e的函数；random(0，δ_4，k)表示在区间(0，δ_4，k)随机产生的意愿；

表示内侧经验换道停靠距离；

表示的位置为内侧非换道区域；

表示的位置为内侧目标换道区域；

表示的位置为内侧惩罚换道区域，δ_5，k表示第k类驾驶员在带有横向通道的车道边外侧非换道区域的换道意愿上限，与驾驶员的风格和当前交通状况有关；λ₅表示在带有横向通道的车道边外侧目标换道区域的换道意愿系数；换入外侧车道的换道意愿

是驾驶员类型k、车辆当前位置x_c、下车客人要求的入站口起点位置x_e的函数；random(0，δ_5，k)表示在区间(0，δ_5，k)随机产生的意愿；

表示外侧经验换道停靠距离；

表示的位置为外侧非换道区域；表示的位置为外侧目标换道区域；

表示的位置为外侧惩罚换道区域；

v_{w}^{d} (k, x_{c}, x_{e}) = \{\begin{matrix} random (0, δ_{5, k}) & if x_{c} > d_{the s}^{w} (k) + x_{e} \\ 1 - {[(x_{c} - x_{e} - d_{the s}^{w} (k) + {D.}_{e}) / (x_{e} + d_{the s}^{w} (k) - {D.}_{e})]}^{λ_{5}} & if x_{e} - {D.}_{e} + d_{the s}^{w} (k) < x_{c} \leq d_{the s}^{w} (k) + x_{e} \\ 1 & if x_{c} \leq x_{e} - {D.}_{e} + d_{the s}^{w} (k) \end{matrix}

In the above formula, δ _{4, k} represents the upper limit of the willingness to change lanes of drivers of type k in the non-lane-changing area on the inner side of the lane with a transverse passage, which is related to the driver’s style and current traffic conditions; λ ₄ represents Lane-changing willingness coefficient of the target lane-changing area inside the lane edge of the transverse passage; lane-changing willingness to change into the inner lane

is a function of the driver type k, the current position of the vehicle x _c , and the starting position x _e of the entrance required by the passengers getting off the bus; random(0, δ _{4, k} ) means that it is randomly generated in the interval (0, δ _{4, k} ) will;

Indicates the inside experience lane change stop distance;

The position indicated is the inner non-lane changing area;

The indicated position is the inner target lane change area;

The position represented by δ5 is the inner penalty lane-changing area, δ _{5, and k} represents the upper limit of the willingness to change lanes of drivers of type k in the non-lane-changing area on the outer side of the lane with a transverse passage, which is related to the driver's style and current traffic conditions; λ ₅ represents the lane-changing willingness coefficient in the target lane-changing area on the outer side of the lane with a transverse passage; the lane-changing willingness to change into the outer lane

is a function of the driver type k, the current position of the vehicle x _c , and the starting position x _e of the entry gate requested by the passengers getting off the bus; random(0, δ _{5, k} ) means randomly generated in the interval (0, δ _{5, k} ) will;

Indicates the outside experience lane change stop distance;

The position indicated is the outer non-lane changing area; The indicated position is the outer target lane change area;

The indicated position is the outer penalty lane change area;

下车乘客的换道意愿函数可表示为：The lane-changing willingness function of passengers getting off the bus can be expressed as:

${v v}_{n no}^{p p} ((h h,, {x x}_{c c},, {x x}_{e e})) = = \{\begin{matrix} random random ((00,, {δ δ}_{66,, k k})) & if if {x x}_{c c} > > {d d}_{s the s}^{n no} ((k k)) + + {x x}_{e e} \\ 11 - - {[[(({x x}_{c c} - - {x x}_{e e} - - {d d}_{s the s}^{n no} ((k k)) + + {D D.}_{e e})) / / (({x x}_{e e} + + {d d}_{s the s}^{n no} ((k k)) - - {D D.}_{e e}))]]}^{{λ λ}_{66}} & if if {x x}_{e e} - - {D D.}_{e e} + + {d d}_{s the s}^{n no} ((k k)) < < {x x}_{c c} \leq \leq {d d}_{s the s}^{n no} ((k k)) + + {x x}_{e e} \\ 11 & if if {x x}_{c c} \leq \leq {x x}_{e e} - - {D D.}_{e e} + + {d d}_{s the s}^{n no} ((k k)) \end{matrix}$

${v v}_{w w}^{p p} ((h h,, {x x}_{c c},, {x x}_{e e})) = = \{\begin{matrix} random random ((00,, {δ δ}_{77,, k k})) & if if {x x}_{c c} > > {d d}_{s the s}^{w w} ((k k)) + + {x x}_{e e} \\ 11 - - {[[(({x x}_{c c} - - {x x}_{e e} - - {d d}_{s the s}^{w w} ((k k)) + + {D D.}_{e e})) / / (({x x}_{e e} + + {d d}_{s the s}^{w w} ((k k)) - - {D D.}_{e e}))]]}^{{λ λ}_{77}} & if if {x x}_{e e} - - {D D.}_{e e} + + {d d}_{s the s}^{w w} ((k k)) < < {x x}_{c c} \leq \leq {d d}_{s the s}^{w w} ((k k)) + + {x x}_{e e} \\ 11 & if if {x x}_{c c} \leq \leq {x x}_{e e} - - {D D.}_{e e} + + {d d}_{s the s}^{w w} ((k k)) \end{matrix}$

上式中，δ_6，k表示第h类下车乘客在带有横向通道的车道边内侧非换道区域的换道意愿上限，与下车乘客的风格和当前交通状况有关；λ₆表示在带有横向通道的车道边内侧目标换道区域的换道意愿系数；random(0，δ_6，k)表示在区间(0，δ_6，k)随机产生的意愿；δ_7，k表示第h类下车乘客在带有横向通道的车道边外侧非换道区域的换道意愿上限，与下车乘客的风格和当前交通状况有关；λ₇表示在带有横向通道的车道边外侧目标换道区域的换道意愿系数；random(0，δ_7，k)表示在区间(0，δ_7，k)随机产生的意愿；In the above formula, δ _{6, k} represents the upper limit of the willingness to change lanes of passengers of the h category in the non-lane-changing area on the inner side of the lane with a transverse passage, which is related to the style of the passengers getting off and the current traffic conditions; λ ₆ represents the Lane-changing willingness coefficient of the target lane-changing area on the inner side of the lane with a transverse channel; random(0, δ _{6, k} ) represents the willingness randomly generated in the interval (0, δ _{6, k} ); δ _{7, k} represents the The upper limit of lane-changing willingness of passengers getting off in the non-lane-changing area on the outer side of the lane with a transverse passage is related to the style of the disembarking passengers and the current traffic conditions; λ ₇ represents the target lane change on the outer side of the lane with a transverse passage Lane-changing willingness coefficient of the area; random(0, δ _{7, k} ) represents the willingness generated randomly in the interval (0, δ _{7, k} );

驾驶员内侧停靠目标函数、驾驶员外侧停靠目标函数、乘客内侧停靠目标函数和乘客外侧停靠目标函数，将这四个目标函数对应的换道意愿值用v_i，i＝1，2，3，4表示，则对应驾驶员内侧停靠区域、驾驶员外侧停靠区域、乘客内侧停靠区域和乘客外侧停靠区域的概率用p_i(v_i)，i＝1，2，3，4表示，其博弈后的选择概率：The driver’s inner parking objective function, the driver’s outer parking objective function, the passenger’s inner parking objective function and the passenger’s outer parking objective function, the lane change willingness values corresponding to these four objective functions are used as v _i , i=1, 2, 3, 4, then the probability corresponding to the parking area inside the driver, the parking area outside the driver, the parking area inside the passenger, and the parking area outside the passenger is represented by p _i (v _i ), i=1, 2, 3, 4, after the game The selection probability of :

${p p}_{i i} (({v v}_{i i})) = = \frac{{v v}_{i i}}{{Σ Σ}_{j j = = 11}^{44} {v v}_{j j}},, i i = = 1,2,3,4 1,2,3,4$

上式中v_i是当前位置、驾驶员类型、乘客要求的入站口起点位置和乘客类型的函数，实际上是随着车辆运行而发生变化的，而最终选择是这四个意愿值之间的博弈结果。In the above formula, v _i is a function of the current position, driver type, the starting position of the entrance required by passengers, and the type of passengers. It actually changes with the operation of the vehicle, and the final choice is between these four willingness values. result of the game.

横向通道的元胞是指：对车道边横向通道及其左、右一个大型客车长范围内采用二维元胞进行划分，其它道路区域采用一维元胞，区域被二维元胞离散后，仿真中需要采用元胞数量及元胞位置来描述车辆的运行轨迹。The cells of the transverse aisle refer to two-dimensional cells used to divide the lateral aisle on the side of the lane and the long range of a large passenger car on the left and right, and one-dimensional cells used in other road areas. After the area is separated by two-dimensional cells, In the simulation, the number of cells and the position of cells need to be used to describe the running track of the vehicle.

转弯轨迹的仿真规则为：The simulation rules for the turning trajectory are:

1)仿真时刻t驶出横向通道车辆车轮与车道线夹角为β(t)，车身偏转角为车辆转向轮的速度为V_q(t)，最前方车身角点位置为l(x(t)，y(t))；1) At the simulation time t, the angle between the vehicle wheel and the lane line is β(t), and the deflection angle of the vehicle body is The speed of the steering wheel of the vehicle is V _q (t), and the corner position of the frontmost body is l(x(t), y(t));

2)t：＝t+1；2) t:=t+1;

3)如果l(x(t)，y(t))已经进入行车道，则β(t)＝max(β(t)-δ_d，0)，δ_d表示前轮回转步长，转5；3) If l(x(t), y(t)) has entered the traffic lane, then β(t)=max(β(t)-δ _d , 0), δ _d represents the front wheel rotation step size, turn 5 ;

4)如果l(x(t)，y(t))没有进入行车道，则β(t)＝β(t)+δ_a，δ_a表示前轮转向步长；4) If l(x(t), y(t)) does not enter the carriageway, then β(t)=β(t)+ _δa , _{where δa} represents the front wheel steering step;

5)如果

则v_B(t)＝v_B(t)+1，为横向通道行驶的最高速度；5) if

Then v _B (t) = v _B (t) + 1, is the maximum speed for traveling in the transverse aisle;

6)延续当前状态，估计t+2时刻，当前车辆和相邻车辆的元胞占用情况；6) Continuing the current state and estimating the cell occupancy of the current vehicle and adjacent vehicles at time t+2;

7)如果t+2时刻当前车辆与其它车辆均无占用元胞重叠的情况，转2；否则继续；7) If there is no cell overlap between the current vehicle and other vehicles at time t+2, go to 2; otherwise, continue;

8)如果t+2时刻当前车辆仅与相邻车辆1有占用元胞重叠，则v_B(t)：＝v_B(t)-1，转6；8) If the current vehicle only overlaps with the adjacent vehicle 1 in occupied cells at time t+2, then v _B (t):=v _B (t)-1, go to 6;

9)如果t+2时刻当前车辆仅仅与相邻车辆2、5有占用元胞重叠，则β(t)＝β(t)+δ_d，转6；9) If the current vehicle only overlaps the occupied cells of adjacent vehicles 2 and 5 at time t+2, then β(t)=β(t)+ _δd , go to 6;

10)如果t+2时刻当前车辆仅与相邻车辆3、4有占用元胞重叠，则β(t)＝β(t)-δ_a，转6；10) If the current vehicle only overlaps the occupied cells of adjacent vehicles 3 and 4 at time t+2, then β(t)=β(t)-δ _a , turn to 6;

11)如果t+2时刻出现多处占用元胞重叠，则v_B(t)＝0，转6；11) If multiple occupied cells overlap at time t+2, then v _B (t)=0, go to 6;

12)转弯仿真结束。12) Turning simulation ends.

本发明的技术特点及效果：Technical characteristics and effects of the present invention:

1)带有横向通道的车道边长度至今没有量化的确定方法，只能通过经验进行估计。本发明通过细致刻画车辆停靠目标函数、停靠规则、转弯轨迹等多种影响，采用元胞自动机智能体进行集成，具有首创性。1) There is no quantified determination method for the length of the side of the lane with a transverse passage, and it can only be estimated empirically. In the present invention, the cellular automaton intelligent body is used to integrate various influences such as vehicle parking objective functions, parking rules, and turning trajectories in detail, which is original.

2)在实际应用中，本本发明的方法数据采集较为简单，只要预测车道边流量及车型比例，而这些数据在航站楼设计时已经具备，因此本发明仿真方法的实际使用价值较大。2) In practical application, the method data collection of the present invention is comparatively simple, only need to predict lane side flow rate and vehicle type ratio, and these data have been possessed when terminal building design, so the actual use value of simulation method of the present invention is bigger.

附图说明 Description of drawings

图1带有横向通道的车道边仿真场景示意图。Figure 1. Schematic diagram of a lane-side simulation scene with a transverse passage.

图2个横向通道时交通流时空图。Fig. 2 Spatio-temporal diagram of traffic flow during transverse passage.

图3个横向通道时交通流时空图。Fig. 3 Spatio-temporal diagram of traffic flow during transverse passage.

图4个横向通道时交通流时空图。Fig. 4 Spatio-temporal diagram of traffic flow during transverse passage.

图5个横向通道时交通流时空图。Fig. 5 Spatio-temporal diagram of traffic flow in transverse corridor.

图6个横向通道时交通流时空图。Fig. 6 Spatio-temporal diagram of traffic flow during horizontal passage.

图7个横向通道时交通流时空图。Fig. 7 Spatio-temporal diagram of traffic flow during transverse passage.

图8不同横向通道数下的延误时间。Figure 8 Delay time under different number of transverse channels.

图9交通冲突示意图。Fig. 9 Schematic diagram of traffic conflict.

图10将带有横向通道的车道边停靠模型目标区域进行划分。Figure 10 divides the target area of the lane-side parking model with transverse passages.

图11横向通道及上游区域二维元胞划分。Figure 11 Two-dimensional cell division of the transverse channel and upstream region.

图12大型车辆中车辆位置及元胞表示图。Fig. 12 Vehicle position and cell representation in a large vehicle.

图13大型客车横向通道转弯元胞示意。Figure 13 is a schematic diagram of the turning cell of the transverse aisle of a large passenger car.

图14车辆在横向通道转弯位置示意图Figure 14 Schematic diagram of the turning position of the vehicle in the transverse passage

具体实施方式 Detailed ways

1.1带有横向通道的车道边设计形式交通流特性分析1.1 Analysis of Traffic Flow Characteristics of Lane Side Design Form with Transverse Passage

带有横向通道的车道边设计形式下，除了正常换道外，车侧车道车辆还可以通过横向通道进入行车道，减少了由于外车道拥堵而使得内车道车辆不能进入行车道的概率，其交通冲突示意图如图9所示。In the lane side design form with transverse passages, in addition to changing lanes normally, vehicles in the side lanes can also enter the carriageway through the cross passage, which reduces the probability that vehicles in the inner lane cannot enter the carriageway due to congestion in the outer lane, and its traffic conflicts The schematic diagram is shown in Figure 9.

注：m+1，m，m-1为行车道行驶中车辆。Note: m+1, m, m-1 are vehicles running on the carriageway.

n+4，n-1，n-3为外侧停车道正在停车落客车辆。n+4, n-1, and n-3 are vehicles that are parking and dropping off passengers in the outer parking lane.

n+3，n+1，n为内侧停车道正在停车落客车辆。n+3, n+1, n are vehicles that are parking and dropping off passengers in the inner parking lane.

n+2为内侧停车道车辆正在通过横向通道驶入行车道。n+2 means that the vehicle in the inner parking lane is entering the driving lane through the transverse passage.

n-2为内侧停车道车辆准备进入横向通道。n-2 is for vehicles in the inner parking lane to prepare to enter the transverse passage.

n-1为外侧停车道车辆驶入行车道准备交织车。n-1 prepares weaving vehicles for the vehicles in the outer parking lane to enter the driving lane.

如图9所示，当设置了横向通道时，由于在横向通道位置不能停车，使得内侧车道车辆不会完全被外侧车道车辆阻挡而只能进行跟驰行车，能提高内侧车道的行车速度，但是这种通道会占用一定的空间，特别是内侧车道进出行车道便利性越好，所需要的通道数量也越多，因此该形式需要在一定的交通条件下进行适应性设计和优化。As shown in Figure 9, when the transverse passage is set, since the vehicle in the transverse passage cannot be parked, the vehicle in the inner lane will not be completely blocked by the vehicle in the outer lane and can only follow the car, which can increase the driving speed of the inner lane, but This kind of passage will occupy a certain amount of space, especially the better the convenience of the inner lane entering and exiting the traffic lane, the more passages will be needed. Therefore, this form needs to be adaptively designed and optimized under certain traffic conditions.

1.2带有横向通道的车道边停靠模型目标函数1.2 The objective function of the lane-side parking model with transverse passages

带有横向通道的车道边设计形式中，内外侧停靠车道对于驾驶员和乘客的影响不同。在外侧停靠车道，车辆更容易换入停靠和换出进入行车道，而且在外侧停靠车道车辆不会由于相邻车辆阻碍而不能进入行车道，而如果车辆要进入内侧车道和从内车道启动进行行车道，则需要进行两次换道，特别是当外侧车道拥堵，内侧车辆换入行车道受到前方车辆和是否存在横向通道极大影响，因此对驾驶员而言，其理想选择是外侧车道。对于下车客人而言，外侧车道下车后需要穿过内侧车道才能进入航站楼，由于该车道边设计形式下一般没有控制信号，因此在行人在穿越过程中还存在人车冲突，因此对下车客人而言，其理想选择是内侧车道。由于存在上述矛盾，因此需要在驾驶员和下车客人之间形成博弈均衡，由于大型车辆一般为公共服务型车辆，私家车一般接送熟悉人群，本发明将其设置为客人优先级别，即驾驶员也优先选择内侧车道进行停靠，仅考虑出租车驾驶员和下车客人之间的博弈关系。本发明模型中将带有横向通道的车道边停靠模型目标区域进行划分，如图10所示。In lane-side designs with transverse passages, the inner and outer parking lanes affect drivers and passengers differently. In the outer parking lane, it is easier for vehicles to change into and out of the driving lane, and the vehicle in the outer parking lane will not be unable to enter the driving lane due to the obstruction of adjacent vehicles, and if the vehicle wants to enter the inner lane and start from the inner lane In the driving lane, it is necessary to change lanes twice, especially when the outer lane is congested, the inner vehicle changing into the driving lane is greatly affected by the vehicle in front and whether there is a transverse passage, so for the driver, the ideal choice is the outer lane. For passengers who get off the bus, they need to cross the inner lane to enter the terminal building after getting off in the outer lane. Because there is generally no control signal in the design form of the lane, there are conflicts between pedestrians and vehicles during the crossing process. Ideal for drop-off passengers is the inside lane. Due to the above-mentioned contradictions, it is necessary to form a game balance between the driver and the guests who get off the bus. Since large vehicles are generally public service vehicles, and private cars generally pick up and drop off familiar people, the present invention sets it as the guest priority level, that is, the driver The inner lane is also given priority for parking, and only the game relationship between the taxi driver and the passengers getting off is considered. In the model of the present invention, the target area of the parking model on the side of the lane with the transverse passage is divided, as shown in FIG. 10 .

图10中的参数含义与带有中间快速通道的车道边停靠模型目标区域划分参数基本一致，本发明不进行赘述，只对其停车目标函数不一致部分进行说明，主要表现在出租车停靠带有横向通道的车道边模型中。The meaning of the parameters in Fig. 10 is basically consistent with the target area division parameters of the lane-side parking model with the middle fast lane. In the lane edge model of the channel.

出租车停靠带有横向通道的车道边模型中，目标函数分为驾驶员目标函数和乘客目标函数，最终停靠车道和位置为二者之间博弈结果。令x_c表示车辆当前位置，x_e表示下车客人要求的入站口行车方向的起点位置，D_e表示下车客人要求的入站口宽度。In the lane-side model of taxis with transverse passages, the objective function is divided into driver objective function and passenger objective function, and the final parking lane and position are the results of the game between the two. Let x _c represent the current position of the vehicle, x _e represent the starting position of the driving direction of the entrance required by the passengers getting off the bus, and D _e represent the width of the entrance required by the passengers getting off the bus.

驾驶员的换道意愿函数可表示为：The driver's willingness to change lanes can be expressed as:

${v v}_{w w}^{d d} ((k k,, {x x}_{c c},, {x x}_{e e})) = = \{\begin{matrix} random random ((00,, {δ δ}_{55,, k k})) & if if {x x}_{c c} > > {d d}_{s the s}^{w w} ((k k)) + + {x x}_{e e} \\ 11 - - {[[(({x x}_{c c} - - {x x}_{e e} - - {d d}_{s the s}^{w w} ((k k)) + + {D D.}_{e e})) / / (({x x}_{e e} + + {d d}_{s the s}^{w w} ((k k)) - - {D D.}_{e e}))]]}^{{λ λ}_{55}} & if if {x x}_{e e} - - {D D.}_{e e} + + {d d}_{s the s}^{w w} ((k k)) < < {x x}_{c c} \leq \leq {d d}_{s the s}^{w w} ((k k)) + + {x x}_{e e} \\ 11 & if if {x x}_{c c} \leq \leq {x x}_{e e} - - {D D.}_{e e} + + {d d}_{s the s}^{w w} ((k k)) \end{matrix}$

上式中，δ_4，k表示第k类驾驶员在带有横向通道的车道边内侧非换道区域的换道意愿上限，与驾驶员的风格和当前交通状况有关；λ₄表示在带有横向通道的车道边内侧目标换道区域的换道意愿系数；换入内侧车道的换道意愿是驾驶员类型k、车辆当前位置x_c、下车客人要求的入站口起点位置x_e的函数；random(0，δ_4，k)表示在区间(0，δ_4，k)随机产生的意愿；表示的位置为内侧非换道区域；

表示的位置为内侧目标换道区域；

表示的位置为内侧惩罚换道区域。δ_5，k表示第k类驾驶员在带有横向通道的车道边外侧非换道区域的换道意愿上限，与驾驶员的风格和当前交通状况有关；λ₅表示在带有横向通道的车道边外侧目标换道区域的换道意愿系数；换入外侧车道的换道意愿是驾驶员类型k、车辆当前位置x_c、下车客人要求的入站口起点位置x_e的函数；random(0，δ_5，k)表示在区间(0，δ_5，k)随机产生的意愿；

表示的位置为外侧惩罚换道区域。In the above formula, δ _{4, k} represents the upper limit of the willingness to change lanes of drivers of type k in the non-lane-changing area on the inner side of the lane with a transverse passage, which is related to the driver’s style and current traffic conditions; λ ₄ represents Lane-changing willingness coefficient of the target lane-changing area inside the lane edge of the transverse passage; lane-changing willingness to change into the inner lane is a function of the driver type k, the current position of the vehicle x _c , and the starting position x _e of the entrance required by the passengers getting off the bus; random(0, δ _{4, k} ) means that it is randomly generated in the interval (0, δ _{4, k} ) will; The position indicated is the inner non-lane changing area;

The indicated position is the inner target lane change area;

The position indicated is the inner penalty lane change area. δ _{5, k} represents the upper limit of the willingness to change lanes of drivers of type k in the non-lane-changing area on the outer side of the lane with a transverse passage, which is related to the driver's style and current traffic _conditions ; Lane-changing willingness coefficient in the outer target lane-changing area; lane-changing willingness to change into the outer lane is a function of the driver type k, the current position of the vehicle x _c , and the starting position x _e of the entry gate requested by the passengers getting off the bus; random(0, δ _{5, k} ) means randomly generated in the interval (0, δ _{5, k} ) will;

The indicated position is the outer penalty lane change area.

$v_{w}^{p} (h, x_{c}, x_{e}) = \{\begin{matrix} random (0, δ_{7, k}) & if x_{c} > d_{s}^{w} (k) + x_{e} \\ 1 - {[(x_{c} - x_{e} - d_{s}^{w} (k) + D_{e}) / (x_{e} + d_{s}^{w} (k) - D_{e})]}^{λ_{7}} & if x_{e} - D_{e} + d_{s}^{w} (k) < x_{c} \leq d_{s}^{w} (k) + x_{e} \\ 1 & if x_{c} \leq x_{e} - D_{e} + d_{s}^{w} (k) \end{matrix}$ 上式中，δ_6，k表示第h类下车乘客在带有横向通道的车道边内侧非换道区域的换道意愿上限，与下车乘客的风格和当前交通状况有关；λ₆表示在带有横向通道的车道边内侧目标换道区域的换道意愿系数；换入内侧车道的换道意愿

是下车乘客类型k、车辆当前位置x_e、下车客人要求的入站口起点位置x_e的函数；random(0，δ_6，k)表示在区间(0，δ_6，k)随机产生的意愿；

表示的位置为内侧非换道区域；

表示的位置为内侧目标换道区域；

表示的位置为内侧惩罚换道区域。δ_7，k表示第h类下车乘客在带有横向通道的车道边外侧非换道区域的换道意愿上限，与下车乘客的风格和当前交通状况有关；λ₇表示在带有横向通道的车道边外侧目标换道区域的换道意愿系数；换入外侧车道的换道意愿

是下车乘客类型k、车辆当前位置x_c、下车客人要求的入站口起点位置x_e的函数；random(0，δ_7，k)表示在区间(0，δ_7，k)随机产生的意愿；

表示的位置为外侧非换道区域；

表示的位置为外侧目标换道区域；

表示的位置为外侧惩罚换道区域。

v_{w}^{p} (h, x_{c}, x_{e}) = \{\begin{matrix} random (0, δ_{7, k}) & if x_{c} > d_{the s}^{w} (k) + x_{e} \\ 1 - {[(x_{c} - x_{e} - d_{the s}^{w} (k) + {D.}_{e}) / (x_{e} + d_{the s}^{w} (k) - {D.}_{e})]}^{λ_{7}} & if x_{e} - {D.}_{e} + d_{the s}^{w} (k) < x_{c} \leq d_{the s}^{w} (k) + x_{e} \\ 1 & if x_{c} \leq x_{e} - {D.}_{e} + d_{the s}^{w} (k) \end{matrix}

In the above formula, δ _{6, k} represents the upper limit of the willingness to change lanes of passengers of the h category in the non-lane-changing area on the inner side of the lane with a transverse passage, which is related to the style of the passengers getting off and the current traffic conditions; λ ₆ represents the Lane-change willingness coefficient for the inner target lane-changing area of the lane edge with a transverse passage; lane-changing willingness to change into the inner lane

is a function of the type k of passengers getting off the bus, the current position of the vehicle x _e , and the starting position x _e of the entrance required by the passengers getting off the bus; random(0, δ _{6, k} ) means that it is randomly generated in the interval (0, δ _{6, k} ) will;

The position indicated is the inner non-lane changing area;

The indicated position is the inner target lane change area;

The position indicated is the inner penalty lane change area. δ _{7, k} represents the upper limit of the willingness to change lanes in the non-lane-changing area on the outer side of the lane with cross passages for passengers getting off in category h, which is related to the style of the passengers getting off and the current traffic conditions; λ ₇ represents The lane-changing willingness coefficient of the outer target lane-changing area on the edge of the lane; the lane-changing willingness to change into the outer lane

is a function of the type k of passengers getting off the bus, the current position of the vehicle x _c , and the starting position x _e of the entrance required by the passengers getting off the bus; random(0, δ _{7, k} ) means that it is randomly generated in the interval (0, δ _{7, k} ) will;

The position indicated is the outer non-lane changing area;

The indicated position is the outer target lane change area;

The indicated position is the outer penalty lane change area.

1.3带有横向通道的车道边停靠选择规则1.3 Selection rules for lane-side stops with transverse passages

出租车停靠带有横向通道的车道边模型中，存在四个换道区域目标函数，分别为驾驶员内侧停靠目标函数、驾驶员外侧停靠目标函数、乘客内侧停靠目标函数和乘客外侧停靠目标函数，即前文下车乘客的换道意愿函数和驾驶员的换道意愿函数，为方便表述，本发明将这四个目标函数对应的换道意愿值用v_i，i＝1，2，3，4表示。则对应驾驶员内侧停靠区域、驾驶员外侧停靠区域、乘客内侧停靠区域和乘客外侧停靠区域的概率用p_i(v_i)，i＝1，2，3，4表示，其博弈后的选择概率：In the lane-side model of taxis with transverse passages, there are four lane-changing area objective functions, which are the driver’s inner parking objective function, the driver’s outer parking objective function, the passenger’s inner parking objective function and the passenger’s outer parking objective function. That is, the lane-changing willingness function of the passengers getting off the bus and the lane-changing willingness function of the driver. For the convenience of expression, the present invention uses v _i as the lane-changing willingness values corresponding to these four objective functions, i=1, 2, 3, 4 express. Then the probability corresponding to the driver’s inner parking area, the driver’s outer parking area, the passenger’s inner parking area and the passenger’s outer parking area is represented by p _i (v _i ), i=1, 2, 3, 4, and the selection probability after the game :

1.4横向通道的元胞描述1.4 Cellular description of the transverse channel

对车道边横向通道及其左、右一个大型客车长范围内采用二维元胞进行划分，其它道路区域采用一维元胞。由于二维元胞的存在，能在横向通道附近对大型客车转弯等行为进行细致描述，而其它部分的一维元胞能使仿真保持较快速度。横向通道及上游区域二维元胞划分如图11所示。Two-dimensional cells are used to divide the transverse passage on the side of the lane and the long range of a large passenger car on the left and right, and one-dimensional cells are used for other road areas. Due to the existence of two-dimensional cells, it is possible to describe the behavior of large passenger cars such as turning in detail near the transverse passage, while the one-dimensional cells in other parts can keep the simulation at a relatively fast speed. The two-dimensional cell division of the transverse channel and the upstream area is shown in Figure 11.

图11中，L_h表示横向通道长度，L_k表示靠近横向通道的上游大型客车转弯区域长度，本发明中设为1个大型客车长度。每个二维元胞的长宽均为1.25米，一维元胞长度为1.25米。由于大型客车宽度一般不超过2.5米，所以车辆宽度占有的元胞数最多不超过3个，一个车道3.75米刚好分成3歌元胞。区域被二维元胞离散后，仿真中需要采用元胞数量及元胞位置来描述车辆的运行轨迹，图12为大型车辆中车辆位置及元胞表示图。In Fig. 11, L _h represents the length of the transverse passage, and L _k represents the length of the turning area of the upstream large passenger car close to the transverse passage, which is set as the length of one large passenger car in the present invention. The length and width of each two-dimensional cell are 1.25 meters, and the length of one-dimensional cell is 1.25 meters. Since the width of a large bus generally does not exceed 2.5 meters, the number of cells occupied by the vehicle width does not exceed 3 at most, and a lane of 3.75 meters is just divided into 3 cells. After the area is discretized by two-dimensional cells, the number of cells and cell positions need to be used in the simulation to describe the running track of the vehicle. Figure 12 shows the vehicle position and cell representation in a large vehicle.

1.5车辆横向通道转弯规则1.5 Turning rules for vehicles in lateral passages

图13为大型客车横向通道转弯元胞示意，车辆在横向通道转弯时，相邻车辆位置和信号会影响当前车辆的转向、制动、加速等操作，因此需要对相邻车辆位置和信号进行感知并判断操作空间和风险。本发明采用如图14中的1-5位置车辆作为相邻车辆。Figure 13 is a schematic diagram of the turning cell of a large passenger car in a transverse aisle. When a vehicle is turning in a transverse aisle, the position and signal of adjacent vehicles will affect the steering, braking, acceleration and other operations of the current vehicle. Therefore, it is necessary to sense the position and signal of adjacent vehicles. And judge the operating space and risks. The present invention adopts the vehicles at positions 1-5 as shown in Figure 14 as adjacent vehicles.

车辆驶在横向通道转弯的仿真规则为：The simulation rules for a vehicle turning in a transverse channel are:

1)仿真时刻t驶出横向通道车辆车轮与车道线夹角为β(t)，车身偏转角为

车辆转向轮的速度为V_q(t)，最前方车身角点位置为l(x(t)，y(t))。1) At the simulation time t, the angle between the vehicle wheel and the lane line is β(t), and the deflection angle of the vehicle body is

The speed of the steering wheel of the vehicle is V _q (t), and the position of the frontmost corner of the vehicle body is l(x(t), y(t)).

2)t：＝t+1；2) t:=t+1;

4)如果l(x(t)，y(t))没有进入行车道，则β(t)＝β(t)+δ_a，δ_a表示前轮转向步长，；4) If l(x(t), y(t)) does not enter the carriageway, then β(t)=β(t)+ _δa , _{where δa} represents the front wheel steering step;

5)如果

则v_B(t)＝v_B(t)+1，为横向通道行驶的最高速度；5) if

12)转弯仿真结束。12) Turning simulation ends.

1.6带有横向通道的车道边通行能力交通仿真1.6 Lane-side capacity traffic simulation with transverse passages

1.6.1仿真环境设置1.6.1 Simulation environment settings

仿真场景中停车道和行车道各设置为两条，车道边长度150米，三个入口位置如下图所示，其中L_e＝L_b/2＝25米，行车道限速为20公里/小时，停靠时间限定为大型车2秒/人，出租车30秒，私家车：18秒。大型客车、私家车和出租车比例分别为1∶1∶1。横向通道及上游区域采用二维元胞进行划分，长宽各为1.25米，其它区域采用一维元胞对车道进行划分，每个元胞长度为1.25米。本节设置的航空枢纽带有横向通道的车道边场景示意图如图1所示，其中横向通道设置位置、宽度、数量根据不同实验要求进行不同设置。In the simulation scene, there are two parking lanes and two driving lanes, and the length of the side of the lane is 150 meters. The positions of the three entrances are shown in the figure below, where L _e = L _b /2 = 25 meters, and the speed limit of the driving lane is 20 km/h , The parking time is limited to 2 seconds per person for large vehicles, 30 seconds for taxis, and 18 seconds for private cars. The ratio of large passenger cars, private cars and taxis is 1:1:1. The horizontal passage and the upstream area are divided by two-dimensional cells, each with a length and width of 1.25 meters, and other areas are divided by one-dimensional cells, each with a length of 1.25 meters. The schematic diagram of the laneside scene with transverse passages in the aviation hub set up in this section is shown in Figure 1, where the location, width, and number of transverse passages are set differently according to different experimental requirements.

1.6.2仿真结果及分析1.6.2 Simulation results and analysis

将进入车道边的发车频率设为1000辆/小时，得到车道边交通流时空图。Set the departure frequency at the side of the lane as 1000 vehicles/hour, and obtain the space-time map of the traffic flow at the side of the lane.

由上图交通流时空图可以看出，随着横向通道数量由2个增加到7个，在相同的发车频率下交通拥堵呈现先降低后增加的趋势，当横向通道数为5，分布在出入口和出入口之间时，车道边拥堵程度最低，停止延误和行车延误为13.48和8.23秒，是本发明推荐的横向通道位置和数量。实际上，由于人行道区域不能停车，因此可将正对出口的人行道部分适当加宽作为同时满足行人穿越和车辆横向通过功能，同时在出入口之间各加入1个横向通道，使车道边分段数增多，车道边车辆进入行车道更加便捷。From the time-space diagram of traffic flow in the above figure, it can be seen that as the number of horizontal passages increases from 2 to 7, the traffic congestion presents a trend of first decreasing and then increasing under the same departure frequency. Between the entrance and exit, the lane side congestion is the lowest, and the stop delay and driving delay are 13.48 and 8.23 seconds, which is the recommended position and quantity of the transverse passage in the present invention. In fact, since parking is not allowed in the sidewalk area, the part of the sidewalk facing the exit can be appropriately widened to meet the functions of pedestrian crossing and vehicle lateral passage at the same time. It is more convenient for vehicles on the side of the lane to enter the lane.

Claims

1. limit, track interconnection is provided with position and quantity optimization method; It is characterized in that; Adopt the emulation of two dimensional cellular automaton modeling as platform,, stop the detailed description of the multiple influence of simulated target function, the cellular of stopping rule, interconnection, turning path through comprising the analysis of limit, the track design form traffic stream characteristics that has interconnection; Target is carried out integrated emulation, draw passage position and quantity optimization data are set.

2. the method for claim 1 is characterized in that, the analysis of limit, the track design form traffic stream characteristics of interconnection is meant that the limit is provided with the situation of double stopway.

3. the method for claim 1 is characterized in that, stops the simulated target function and is meant: make x _eExpression vehicle current location, x _eThe get off start position of the station entrance direction of traffic that the guest requires of expression, D _eThe expression station entrance width that the guest requires of getting off, the driver changes the wish function and can be expressed as:

In the following formula, δ _{4, k}Represent k class driver on the limit, track that has interconnection inboard non-change the zone change the wish upper limit, relevant with driver's style and current traffic condition; λ ₄What be illustrated in that the inboard target in the limit, track that has interconnection changes the zone changes the wish coefficient; Change to the wish of changing of fast lane

Be driver's type k, vehicle current location x _c, the station entrance start position x that requires of the guest that gets off _eFunction; Random (0, δ _{4, k}) be illustrated in the interval (0, δ _{4, k}) wish that produces at random;

Represent that inboard experience is changed and stop distance;

The position of expression is the inboard non-zone of changing;

The position of expression is that inboard target is changed the zone;

Zone, δ are changed for inboard punishment in the position of expression _{5, k}Represent k class driver in limit, the track outside that has interconnection non-change the zone change the wish upper limit, relevant with driver's style and current traffic condition; λ ₅What be illustrated in that limit, the track outside target that has interconnection changes the zone changes the wish coefficient; Change to the wish of changing of kerb lane

Be driver's type k, vehicle current location x _c, the station entrance start position x that requires of the guest that gets off _eFunction; Random (0, δ _{5, k}) be illustrated in the interval (0, δ _{5, k}) wish that produces at random;

Expression outside experience is changed and is stopped distance;

The position of expression is the non-zone of changing, the outside; The zone is changed for outside target in the position of expression;

The position of expression is that the zone is changed in outside punishment;

Get off the changing the wish function and can be expressed as of passenger:

In the following formula, δ _{6, k}Represent the h class get off the passenger on the limit, track that has interconnection inboard non-change the zone change the wish upper limit, relevant with the passenger's that gets off style and current traffic condition; λ ₆What be illustrated in that the inboard target in the limit, track that has interconnection changes the zone changes the wish coefficient; Random (0, δ _{6, k}) be illustrated in the interval (0, δ _{6, k}) wish that produces at random; δ _{7, k}Represent the h class get off the passenger in limit, the track outside that has interconnection non-change the zone change the wish upper limit, relevant with the passenger's that gets off style and current traffic condition; λ ₇What be illustrated in that limit, the track outside target that has interconnection changes the zone changes the wish coefficient; Random (0, δ _{7, k}) be illustrated in the interval (0, δ _{7, k}) wish that produces at random.

4. method as claimed in claim 3 is characterized in that, objective function is stopped in the inboard stop of driver objective function, the driver outside, objective function is stopped outside stopping objective function and passenger in the passenger inboard, and the wish value of changing that these four objective functions are corresponding is used v _i, i=1,2,3,4 represent, the inboard probability of stopping the zone in zone, stop zone, the driver outside, passenger inboard stop zone and the passenger outside of stopping of then corresponding driver is used p _i(v _i), i=1,2,3,4 expressions, the selection probability after its game:

V in the following formula _iBe current location, driver's type, the station entrance start position of passenger's requirement and the function of passenger type, be actually along with vehicle operating changes, and final the selection be the payoff between these four wish values.

5. the method for claim 1; It is characterized in that; The cellular of interconnection is meant: to adopting two-dimentional cellular to divide in limit, track interconnection and the long scope of a left and right motorbus thereof; Other road area adopts the one dimension cellular, after being dispersed by two-dimentional cellular in the zone, needs in the emulation to adopt cellular quantity and cellular position to describe the running orbit of vehicle.

6. the method for claim 1 is characterized in that, the emulation rule of turning path is:

1) emulation constantly t roll the interconnection wheel of vehicle away from and the lane line angle is β (t), the vehicle body deflection angle does

The speed of pivoted wheels on vehicle is V _q(t), forefront vehicle body corner location is l (x (t), y (t));

2)t：＝t+1；

3) if l (x (t), y (t)) has got into runway, then β (t)=max (β (t)-δ _d, 0), δ _dExpression front-wheel revolution step-length changes 5;

4) if l (x (t), y (t)) does not get into runway, then β (t)=β (t)+δ _a, δ _aExpression front-wheel steer step-length;

5) if

V then _B(t)=v _B(t)+1,

The top speed of going for interconnection;

6) continuity current state is estimated t+2 constantly, and the cellular of current vehicle and adjacent vehicle takies situation;

7) if t+2 moment current vehicle and other vehicle all do not have the overlapping situation of cellular that takies, change 2; Otherwise continue;

8) if t+2 constantly current vehicle only to take cellular overlapping with adjacent vehicle 1, v then _B(t) :=v _B(t)-1, change 6;

9) if t+2 constantly current vehicle only to take cellular overlapping with adjacent vehicle 2,5, β (t)=β (t)+δ then _d, change 6;

10) if t+2 constantly current vehicle only to take cellular overlapping with adjacent vehicle 3,4, β (t)=β (t)-δ then _a, change 6;

11) to take cellular overlapping if many places appear in t+2 constantly, then v _B(t)=0, change 6;

12) turning emulation finishes.