技术领域technical field
本发明涉及智能交通领域，更具体地，涉及丢包环境下迭代学习的快速路入口匝道控制方法及系统。The present invention relates to the field of intelligent transportation, and more specifically, relates to an expressway entrance ramp control method and system for iterative learning in a packet loss environment.
背景技术Background technique
快速路入口匝道控制是智能交通领域的重要组成部分。交通流模型每天具有重复性，但由于地区社会活动的人员流动、天气、路况等因素变化影响，在实际中，快速路模型系统参数随之发生变化，快速路系统具有切换系统特性，因此现阶段快速路入口匝道控制很多是基于切换系统特性下的研究。On-ramp control of expressways is an important part of the field of intelligent transportation. The traffic flow model is repetitive every day, but due to the influence of factors such as the flow of people in regional social activities, weather, and road conditions, in practice, the system parameters of the expressway model change accordingly, and the expressway system has the characteristics of a switching system. Therefore, at this stage Many studies on the on-ramp control of expressway are based on the characteristics of switching systems.
快速路交通控制属于远程控制场景，实际上快速路与控制中心之间物理距离遥远，且快速路的测量设备经常缺乏维护并暴露野外，故使得测量数据输出与传输可能出现丢包或延时的情况。现阶段很多研究成果都是基于没有传输丢包发生的理想情况下的控制场景，不能更好地切合实际需要。故基于迭代学习的快速路入口匝道控制实用性降低。Expressway traffic control is a remote control scenario. In fact, the physical distance between the expressway and the control center is far away, and the measurement equipment of the expressway is often lack of maintenance and exposed to the wild, so the output and transmission of measurement data may be lost or delayed. Happening. At this stage, many research results are based on control scenarios under ideal conditions without transmission packet loss, which cannot better meet actual needs. Therefore, the practicability of expressway on-ramp control based on iterative learning is reduced.
发明内容Contents of the invention
本发明首要目的是提供一种丢包环境下迭代学习的快速路入口匝道控制方法及系统，以解决丢包环境下具有切换系统特性的带有匝道路口的快速路的控制问题。The primary purpose of the present invention is to provide an iterative learning expressway entrance ramp control method and system in a packet loss environment to solve the control problem of an expressway with ramp intersections with switching system characteristics in a packet loss environment.
本发明还提供一种具有单边传输丢包情况迭代学习的快速路入口匝道远程控制系统。The invention also provides a remote control system for an expressway entrance ramp with iterative learning of unilateral transmission packet loss.
为解决上述技术问题，本发明的技术方案如下：In order to solve the problems of the technologies described above, the technical solution of the present invention is as follows:
丢包环境下迭代学习的快速路入口匝道控制方法，包括以下步骤：An expressway entrance ramp control method for iterative learning in a packet loss environment, comprising the following steps:
S1：控制中心通过通信模块获取当前快速路交通流密度值；S1: The control center obtains the current traffic flow density value of the expressway through the communication module;
S2：控制中心获取本地存储中快速路的期望交通流密度值与匝道入口的交通流控制量；S2: The control center obtains the expected traffic flow density value of the expressway and the traffic flow control amount of the ramp entrance in the local storage;
S3：根据接收到的实际交通流密度值和期望交通流密度值而获得当前的误差函数S3: Obtain the current error function according to the received actual traffic flow density value and expected traffic flow density value
ek(t)＝αk(t)(yd(t)-yk(t))ek (t)=αk (t)(yd (t)-yk (t))
其中yd(t)为期望交通流密度，yk(t)为第k次迭代时控制中心接收到的实际交通流密度，αk(t)为第k次迭代的yk(t)不发生随机丢失的事件序列，快速路端至控制中心端的传输丢包概率为Where yd (t) is the expected traffic flow density, yk (t) is the actual traffic flow density received by the control center at the k-th iteration, αk (t) is the yk (t) of the k-th iteration In the event sequence of random loss, the transmission packet loss probability from the express road end to the control center end is
S4：基于误差函数，计算遗忘因子值，计算公式如下S4: Based on the error function, calculate the value of the forgetting factor, the calculation formula is as follows
其中γ(k)为第k次迭代误差函数序列中不为0的元素个数，θmax与θmin为θ(k)函数的上下界值，φ(k)为最大值选择函数，用于选择从第1次迭代至第k次迭代以来最大的误差函数均方值；Among them, γ(k) is the number of elements that are not 0 in the error function sequence of the kth iteration, θmax and θmin are the upper and lower bounds of the θ(k) function, and φ(k) is the maximum value selection function for Select the largest mean square value of the error function from the 1st iteration to the kth iteration;
S5：根据所述误差函数与遗忘因子，设置基于传输丢包情况下具有切换特性的迭代学习控制率以及迭代学习增益。其中迭代学习控制率表示如下：S5: According to the error function and the forgetting factor, set an iterative learning control rate and an iterative learning gain with switching characteristics based on transmission packet loss. The iterative learning control rate is expressed as follows:
uk+1(t)＝(1-αk(t+1)θ(k+1))sat[uk(t)]+Γiek(t+1)uk+1 (t)＝(1-αk (t+1)θ(k+1))sat[uk (t)]+Γi ek (t+1)
式子中，uk+1(t)表示第k+1次迭代的匝道入口的交通流，uk(t)表示第k次迭代的匝道入口的交通流，θ(k)为自适应遗忘因子函数，误差函数ek(t+1)表示第k次迭代t+1时刻的输出误差，而Γi为第i个切换系统子系统的控制增益，饱和函数sat[·]表达式如下：In the formula, uk+1 (t) represents the traffic flow at the ramp entrance of the k+1 iteration, uk (t) represents the traffic flow at the ramp entrance of the k iteration, and θ(k) is the adaptive forgetting The factor function, the error function ek (t+1) represents the output error at the time of the kth iteration t+1, andΓi is the control gain of the ith switching system subsystem, and the expression of the saturation function sat[·] is as follows:
其中in
其中j∈{1,2,...,K}，K为一段快速路中的匝道数，为第k次迭代时匝道j的交通流控制量，u(j),max(t)与u(j),min(t)分别为匝道j交通流控制量的饱和上下界；Where j∈{1,2,...,K}, K is the number of ramps in a section of expressway, is the traffic flow control quantity of ramp j at the kth iteration, u(j), max (t) and u(j), min (t) are the saturated upper and lower bounds of the traffic flow control quantity of ramp j respectively;
S6：控制中心结合所述迭代学习控制律以及迭代学习增益，生成快速路匝道入口的交通流控制信号，并通过可靠通信方式将信号发送至快速路端；S6: The control center combines the iterative learning control law and the iterative learning gain to generate a traffic flow control signal at the ramp entrance of the expressway, and send the signal to the end of the expressway through a reliable communication method;
S7：快速路端通过可靠的通信方式接收匝道入口的交通流控制信号；S7: The expressway end receives the traffic flow control signal at the entrance of the ramp through a reliable communication method;
S8：将交通流控制信号应用到带有匝道入口的快速路系统控制，使得能够在一定迭代次数内使快速路能够达到期望交通流密度；S8: Apply the traffic flow control signal to the expressway system control with the ramp entrance, so that the expressway can reach the desired traffic flow density within a certain number of iterations;
S9：快速路端将快速路交通密度流信号发送至控制中心。S9: The expressway end sends the expressway traffic density flow signal to the control center.
一种丢包环境下迭代学习的快速路入口匝道远程控制系统，包括：A remote control system for an expressway entrance ramp with iterative learning in a packet loss environment, comprising:
当前交通流获取模块：用于控制中心以通信方式获取当前快速路实际交通流密度；Current traffic flow acquisition module: used for the control center to obtain the actual traffic flow density of the current expressway through communication;
交通信息获取模块：用于控制中心获取本地存储设备中期望交通流、匝道入口的交通流控制量；Traffic information acquisition module: used for the control center to obtain the expected traffic flow in the local storage device and the traffic flow control amount at the entrance of the ramp;
误差函数计算模块：用于计算当前误差函数，其根据当前实际接收到的交通流密度值与期望交通流密度计算获得当前误差函数Error function calculation module: used to calculate the current error function, which calculates and obtains the current error function according to the current actually received traffic flow density value and the expected traffic flow density
ek(t)＝αk(t)(yd(t)-yk(t))ek (t)=αk (t)(yd (t)-yk (t))
其中yd(t)为期望交通流密度，yk(t)为第k次迭代时控制中心接收到的实际交通流密度，αk(t)为第k次迭代的yk(t)不发生随机丢失的事件序列，快速路端至控制中心端的传输丢包概率为Where yd (t) is the expected traffic flow density, yk (t) is the actual traffic flow density received by the control center at the k-th iteration, αk (t) is the yk (t) of the k-th iteration In the event sequence of random loss, the transmission packet loss probability from the express road end to the control center end is
遗忘因子计算模块：用于根据误差函数计算出遗忘因子值，计算公式如下：Forgetting factor calculation module: used to calculate the value of the forgetting factor according to the error function, the calculation formula is as follows:
其中γ(k)为第k次迭代误差函数序列中不为0的元素个数，θmax与θmin为θ(k)函数的上下界值，φ(k)为最大值选择函数，用于选择从第1次迭代至第k次迭代以来最大的误差函数均方值；Among them, γ(k) is the number of elements that are not 0 in the error function sequence of the kth iteration, θmax and θmin are the upper and lower bounds of the θ(k) function, and φ(k) is the maximum value selection function for Select the largest mean square value of the error function from the 1st iteration to the kth iteration;
迭代学习律与学习增益设置模块：用于根据误差函数与自适应遗忘因子设置传输丢包情况下基于切换系统特征的迭代学习控制律以及迭代学习增益，其中，迭代学习控制律表示如下：Iterative learning law and learning gain setting module: used to set the iterative learning control law and iterative learning gain based on the characteristics of the switching system in the case of transmission packet loss according to the error function and adaptive forgetting factor. The iterative learning control law is expressed as follows:
uk+1(t)＝(1-αk(t+1)θ(k+1))sat[uk(t)]+Γiek(t+1)uk+1 (t)＝(1-αk (t+1)θ(k+1))sat[uk (t)]+Γi ek (t+1)
式子中，uk+1(t)表示第k+1次迭代的匝道入口的交通流，uk(t)表示第k次迭代的匝道入口的交通流，θ(k)为自适应遗忘因子函数，误差函数ek(t+1)表示第k次迭代t+1时刻的输出误差，而Γi为第i个切换系统子系统的控制增益，饱和函数sat[·]表达式如下：In the formula, uk+1 (t) represents the traffic flow at the ramp entrance of the k+1 iteration, uk (t) represents the traffic flow at the ramp entrance of the k iteration, and θ(k) is the adaptive forgetting The factor function, the error function ek (t+1) represents the output error at the time of the kth iteration t+1, andΓi is the control gain of the ith switching system subsystem, and the expression of the saturation function sat[·] is as follows:
其中in
其中j∈{1,2,...,K}，K为一段快速路中的匝道数，为第k次迭代时匝道j的交通流控制量，u(j),max(t)与u(j),min(t)分别为匝道j交通流控制量的饱和上下界；Where j∈{1,2,...,K}, K is the number of ramps in a section of expressway, is the traffic flow control quantity of ramp j at the kth iteration, u(j), max (t) and u(j), min (t) are the saturated upper and lower bounds of the traffic flow control quantity of ramp j respectively;
控制信号输出模块：用于控制中心生成快速路匝道入口的交通流控制信号，并以可靠通信方式发送至快速路端；Control signal output module: used for the control center to generate the traffic flow control signal at the ramp entrance of the expressway, and send it to the end of the expressway through reliable communication;
控制信号获取模块：用于快速路端通过可靠通信方式获取交通流控制信号；Control signal acquisition module: used for obtaining traffic flow control signals at the express road end through reliable communication;
交通控制模块：用于将交通流控制信号应用到带有匝道入口的快速路系统控制，使得能够在一定迭代次数内使快速路能够达到期望交通流密度；Traffic control module: used to apply the traffic flow control signal to the expressway system control with ramp entrance, so that the expressway can achieve the desired traffic flow density within a certain number of iterations;
交通流密度输出模块：用于快速路端通过通信方式将当前快速路交通流密度信号发送至控制中心。Traffic flow density output module: used for the expressway end to send the current expressway traffic flow density signal to the control center through communication.
与现有技术相比，本发明技术方案的有益效果是：本发明提供一种丢包环境下的快速路入口匝道控制方法，基于快速路入口匝道的切换系统特性，根据实际上未传输丢失的交通流密度值和期望交通流密度值获得当前误差函数；根据误差函数与基于误差的自适应遗忘因子设置传输丢包环境下基于切换系统特征的迭代学习控制律以及迭代学习增益；将传输丢包环境下基于切换系统特征的迭代学习控制律生成的控制信号发送至带有匝道入口的快速路系统中，能够在一定迭代次数内使得快速路能够达到期望交通流密度。本发明方法不仅能够解决丢包环境下具有切换系统特性的带有匝道路口的快速路控制问题，同时能够改善由于数据丢包导致的交通控制收敛速度变慢的问题，更好地切合实际需要。Compared with the prior art, the beneficial effect of the technical solution of the present invention is: the present invention provides a method for controlling the on-ramp of the expressway under the environment of packet loss, based on the characteristics of the switching system of the on-ramp of the expressway, according to the fact that no transmission is lost The current error function is obtained from the traffic flow density value and the expected traffic flow density value; according to the error function and the error-based adaptive forgetting factor, the iterative learning control law and iterative learning gain based on the characteristics of the switching system in the transmission packet loss environment are set; the transmission packet loss The control signal generated by the iterative learning control law based on the characteristics of the switching system in the environment is sent to the expressway system with ramp entrances, which can make the expressway reach the desired traffic flow density within a certain number of iterations. The method of the present invention can not only solve the control problem of the expressway with the ramp intersection with switching system characteristics in the packet loss environment, but also can improve the problem of slow convergence speed of traffic control caused by the data packet loss, and better meet the actual needs.
附图说明Description of drawings
图1为实施例1丢包环境下迭代学习的快速路入口匝道控制方法的流程图。。FIG. 1 is a flow chart of an iterative learning method for controlling an expressway on-ramp in a packet loss environment in Embodiment 1. .
图2为实施例1中快速路模型图。FIG. 2 is a model diagram of the expressway in Embodiment 1.
图3为实施例1中一个应用场景下所述丢包环境下迭代学习的控制律方框图。FIG. 3 is a block diagram of the iterative learning control law in the packet loss environment in an application scenario in Embodiment 1. FIG.
图4为实施例1中一个应用场景下所述丢包环境下迭代学习的快速路系统的一种切换规则。FIG. 4 is a switching rule of the iterative learning expressway system in the packet loss environment in an application scenario in Embodiment 1. FIG.
图5为实施例1中一个应用场景下快速路系统中各路段的匝道入口车辆需求量。Fig. 5 shows the vehicle demand at the ramp entrance of each road section in the expressway system in an application scenario in Embodiment 1.
图6为实施例1中一个应用场景下快速路系统中各路段的匝道出口车流量。FIG. 6 shows the off-ramp traffic flow of each road section in the expressway system in an application scenario in Embodiment 1.
图7为本发明一个应用场景下各路段的交通流输出误差指标分析图。Fig. 7 is an analysis diagram of traffic flow output error indicators of each road section in an application scenario of the present invention.
图8为本发明一个应用场景下两种控制方法的快速路交通的输出误差指标与及理想无丢包环境下控制的快速路交通的输出误差指标比较图。Fig. 8 is a comparison diagram of the output error indicators of expressway traffic of two control methods in an application scenario of the present invention and the output error indicators of expressway traffic controlled in an ideal environment without packet loss.
图9为本发明实施例1具有单边传输丢包情况的基于切换系统特性迭代学习的快速路入口匝道远程控制系统的示意图。9 is a schematic diagram of an expressway on-ramp remote control system based on iterative learning of switching system characteristics with unilateral transmission packet loss in Embodiment 1 of the present invention.
具体实施方式Detailed ways
附图仅用于示例性说明，不能理解为对本专利的限制；The accompanying drawings are for illustrative purposes only and cannot be construed as limiting the patent;
为了更好说明本实施例，附图某些部件会有省略、放大或缩小，并不代表实际产品的尺寸；In order to better illustrate this embodiment, some parts in the drawings will be omitted, enlarged or reduced, and do not represent the size of the actual product;
对于本领域技术人员来说，附图中某些公知结构及其说明可能省略是可以理解的。For those skilled in the art, it is understandable that some well-known structures and descriptions thereof may be omitted in the drawings.
下面结合附图和实施例对本发明的技术方案做进一步的说明。The technical solutions of the present invention will be further described below in conjunction with the accompanying drawings and embodiments.
实施例1Example 1
如图1所示，一种丢包环境下迭代学习的快速路入口匝道控制方法，包括以下步骤：As shown in Figure 1, an expressway entrance ramp control method for iterative learning in a packet loss environment includes the following steps:
101、获取当前快速路的实际交通流密度；首先，控制中心通过通信方式接收当前迭代次数的实际输出交通流密度；101. Obtain the actual traffic flow density of the current expressway; first, the control center receives the actual output traffic flow density of the current iteration times through communication;
102、获取本地存储的快速路期望交通流密度值与当前迭代的匝道入口交通流控制量；102. Obtain the locally stored expected traffic flow density value of the expressway and the current iterative ramp entrance traffic flow control amount;
103、根据所述实际接收到的交通流密度和所述期望交通流密度值而获得所述方法的当前误差函数ek(t)；103. Obtain the current error function ek (t) of the method according to the actually received traffic flow density and the expected traffic flow density value;
ek(t)＝αk(t)(yd(t)-yk(t))ek (t)=αk (t)(yd (t)-yk (t))
其中yd(t)为期望交通流密度，yk(t)为第k次迭代时控制中心接收到的实际交通流密度，αk(t)为第k次迭代的yk(t)不发生随机丢失的事件序列，快速路端至控制中心端的传输丢包概率为Where yd (t) is the expected traffic flow density, yk (t) is the actual traffic flow density received by the control center at the k-th iteration, αk (t) is the yk (t) of the k-th iteration In the event sequence of random loss, the transmission packet loss probability from the express road end to the control center end is
在获取当前快速路的实际交通流密度值和快速路的期望交通流密度值后，根据快速路的期望交通流密度值与当前快速路的实际交通流密度值之差，结果乘上αk(t)，可以得到当前次迭代的误差函数ek(t)；After obtaining the actual traffic flow density value of the current expressway and the expected traffic flow density value of the expressway, according to the difference between the expected traffic flow density value of the expressway and the actual traffic flow density value of the current expressway, multiply the result by αk ( t), the error function ek (t) of the current iteration can be obtained;
104、根据当前迭代的误差函数ek(t)计算得到遗忘因子θ(k+1)，计算公式如下104. Calculate the forgetting factor θ(k+1) according to the error function ek (t) of the current iteration, and the calculation formula is as follows
其中γ(k)为第k次迭代误差函数序列中不为0的元素个数，θmax与θmin为θ(k)函数的上下界值。φ(k)为最大值选择函数，用于选择从第1次迭代至第k次迭代以来最大的误差函数均方值；Where γ(k) is the number of elements that are not 0 in the error function sequence of the kth iteration, and θmax and θmin are the upper and lower bounds of the θ(k) function. φ(k) is the maximum value selection function, which is used to select the largest mean square value of the error function from the first iteration to the kth iteration;
105、根据所述误差函数与遗忘因子设置迭代学习律以及迭代学习增益，使得所述方法的误差指标在一定迭代次数内收敛，其中，在获得当前次的迭代误差函数后，根据该误差函数给出传输丢包环境下基于切换系统的迭代学习控制律并设置迭代学习增益105. Set the iterative learning law and the iterative learning gain according to the error function and the forgetting factor, so that the error index of the method converges within a certain number of iterations, wherein, after obtaining the current iterative error function, give Iterative learning control law based on switching system and setting iterative learning gain in the environment of outbound packet loss
uk+1(t)＝(1-αk(t+1)θ(k+1))sat[uk(t)]+Γiek(t+1)uk+1 (t)＝(1-αk (t+1)θ(k+1))sat[uk (t)]+Γi ek (t+1)
式子中，uk+1(t)表示第k+1次迭代的匝道入口的交通流，uk(t)表示第k次迭代的匝道入口的交通流，θ(k)为自适应遗忘因子函数。误差函数ek(t+1)表示第k次迭代t+1时刻的输出误差，而Γi为第i个切换系统子系统的控制增益，饱和函数sat[·]表达式如下：In the formula, uk+1 (t) represents the traffic flow at the ramp entrance of the k+1 iteration, uk (t) represents the traffic flow at the ramp entrance of the k iteration, and θ(k) is the adaptive forgetting factor function. The error function ek (t+1) represents the output error at the time of the kth iteration t+1, andΓi is the control gain of the ith switching system subsystem. The expression of the saturation function sat[·] is as follows:
其中in
其中j∈{1,2,...,K}，K为一段快速路中的匝道数。为第k次迭代时匝道j的交通流控制量，u(j),max(t)与u(j),min(t)分别为匝道j交通流控制量的饱和上下界；Where j∈{1,2,...,K}, K is the number of ramps in a section of expressway. is the traffic flow control quantity of ramp j at the kth iteration, u(j), max (t) and u(j), min (t) are the saturated upper and lower bounds of the traffic flow control quantity of ramp j respectively;
106、根据所述学习律与迭代学习增益生成快速路匝道入口的交通流控制信号，通过可靠通信传输到快速路交通；106. Generate a traffic flow control signal at the entrance of the expressway ramp according to the learning law and iterative learning gain, and transmit it to the expressway traffic through reliable communication;
107、获取控制中心发送的控制信号，快速路端通过可靠的通信方式接收匝道入口的交通流控制信号；107. Obtain the control signal sent by the control center, and the express road end receives the traffic flow control signal at the entrance of the ramp through a reliable communication method;
108、将接收到的交通流控制信号应用到带有匝道入口的快速路系统控制，使得能够在一定迭代次数内使快速路能够达到期望交通流密度；108. Apply the received traffic flow control signal to the control of the expressway system with ramp entrances, so that the expressway can reach the desired traffic flow density within a certain number of iterations;
109、快速路端将快速路交通密度流信号发送至控制中心。109. The expressway end sends the expressway traffic density flow signal to the control center.
本实施例中，首先，获取当前快速路的实际交通流密度；其次，控制中心获取快速路的期望交通流密度值与当前迭代次数的匝道入口的交通流控制量；然后，根据所述当前实际接收到的交通流密度和所述期望交通流密度值获得所述方法的误差函数；接着，根据所述误差函数与基于误差的自适应遗忘因子设置迭代学习律以及迭代学习增益；再接着，根据所述学习律与迭代学习增益生成的控制信号传输到快速路端，应用于基于切换系统特性的快速路入口匝道控制，使得所述方法的误差指标在一定迭代次数内收敛；最后，将快速路交通密度流信号发送至控制中心。本实施例在现有的快速路系统基于迭代学习控制方法研究中进行进一步深入研究，使得快速路控制系统考虑了实际因素，即受天气、路况等因素，快速路模型系统参数随之发生变化。此外快速路与控制中心之间物理距离遥远，且快速路的测量设备经常缺乏维护并暴露野外。故快速路系统符合切换系统特性同时又可能有测量输出丢包的实际问题，更符合实际应用的需要。In this embodiment, first, the actual traffic flow density of the current expressway is obtained; secondly, the control center obtains the expected traffic flow density value of the expressway and the traffic flow control amount of the ramp entrance of the current iteration number; then, according to the current actual The received traffic flow density and the expected traffic flow density value obtain the error function of the method; then, set the iterative learning law and the iterative learning gain according to the error function and the adaptive forgetting factor based on the error; then, according to The control signal generated by the learning law and iterative learning gain is transmitted to the end of the expressway, and applied to the on-ramp control of the expressway based on the characteristics of the switching system, so that the error index of the method converges within a certain number of iterations; finally, the expressway The traffic density flow signal is sent to the control center. This embodiment conducts further in-depth research on the existing expressway system based on the iterative learning control method, so that the expressway control system takes into account actual factors, that is, the parameters of the expressway model system change accordingly due to factors such as weather and road conditions. In addition, the physical distance between the expressway and the control center is long, and the measurement equipment of the expressway is often poorly maintained and exposed to the wild. Therefore, the expressway system conforms to the characteristics of the switching system and may have the practical problem of measurement output packet loss, which is more in line with the needs of practical applications.
为便于理解，根据图1的实施例，下面以一个实际应用场景对本发明实施例中的一种丢包环境下基于切换系统特性的迭代学习快速路入口匝道控制方法进行描述：For ease of understanding, according to the embodiment of FIG. 1 , a practical application scenario is used to describe an iterative learning method for controlling an expressway on-ramp based on switching system characteristics in a packet loss environment in the embodiment of the present invention:
图2为本发明用到的交通流模型，该时空离散模型将所描述的一条快速路分为多个路段，每个路段均有一个入口匝道和一个出口匝道，如图所示，交通流模型如下所示：Fig. 2 is the traffic flow model used in the present invention, and this space-time discrete model divides a described expressway into a plurality of sections, and each section has an entrance ramp and an exit ramp, as shown in the figure, the traffic flow model As follows:
q(j)(t)＝ρ(j)(t)v(j)(t)；q(j) (t) = ρ(j) (t) v(j) (t);
其中，T是采样周期(小时)，t表示采样间隔，K表示该路段被分为K个子路段，j∈{1,2,...,K}表示每个子段路的标号。其他模型变量的含义如下：ρ(j)(t)(veh/lane/km)表示第j段的平均密度；v(j)(t)(km/h)表示第j段的平均速度；q(j)(t)(veh/h)表示从第j段进入第j+1段的车流量；r(j)(t)(veh/h)表示第j段的匝道入口进入的车流量；s(j)(t)(veh/h)表示第j段匝道出口流出的车流量；L(j)(km)表示第j段路的长度；vfree表示自由流速度；ρjam单个车道的最大可能密度；τ,υ,κ,l,m是常参数，反映特定交通系统的道路几何特点、驾驶员行为与行驶车辆特征等。Among them, T is the sampling period (hour), t represents the sampling interval, K represents that the road segment is divided into K sub-segments, and j∈{1,2,...,K} represents the label of each sub-segment road. The meanings of other model variables are as follows: ρ(j) (t) (veh/lane/km) represents the average density of segment j; v(j) (t) (km/h) represents the average speed of segment j; q(j) (t)(veh/h) represents the traffic flow entering the j+1 segment from segment j; r(j) (t)(veh/h) represents the traffic flow entering the ramp entrance of segment j; s(j) (t) (veh/h) represents the traffic flow out of the j-th ramp exit; L(j) (km) represents the length of the j-th road; vfree represents the free-flow velocity; ρjam the single lane The maximum possible density; τ, υ, κ, l, m are constant parameters, which reflect the road geometry characteristics, driver behavior and driving vehicle characteristics of a specific traffic system.
在本实际应用场景中，每一段路段均都包含一个匝道入口与一个匝道出口。且交通流模型中的变量定义为In this actual application scenario, each road section includes a ramp entrance and a ramp exit. And the variables in the traffic flow model are defined as
x(t)＝[v(1)(t) v(2)(t) ... v(K)(t)]T,y(t)＝[ρ(1)(t) ρ(2)(t) ... ρ(K)(t)]T,x(t)＝[v(1) (t) v(2) (t) ... v(K) (t)]T ,y(t)＝[ρ(1) (t) ρ(2) (t) ... ρ(K) (t)]T ,
u(t)＝[r(1)(t) r(2)(t) ... r(K)(t)]T,s(t)＝[s(1)(t) s(2)(t) ... s(K)(t)]T,u(t)＝[r(1) (t) r(2) (t) ... r(K) (t)]T , s(t)＝[s(1) (t) s(2) (t) ... s(K) (t)]T ,
其中，j∈{1,2,...,K}。快速路模型在t∈{0,1,2,...,N}上具有重复特性，可以表示为以下状态空间表达式：in, j∈{1,2,...,K}. The expressway model has repetitive properties on t∈{0,1,2,...,N}, which can be expressed as the following state space expression:
yk(t+1)＝A(xk(t),t)yk(t)+B(sat[uk(t)])+η(xk(t))-Bs(t)yk (t+1)＝A(xk (t),t)yk (t)+B(sat[uk (t)])+η(xk (t))-Bs(t)
xk(t+1)＝C(xk(t),t)+D(yk(t),t)xk (t+1)＝C(xk (t),t)+D(yk (t),t)
其中，k是迭代次数，C(x(t),t)，D(x(t),t)为两个非线性函数。由于有限时间段内进入的车流量是有限的且不能为负，故uk(t)受到匝道实际条件的饱和限制。在实际系统中，快速路系统的参数会随环境因素发生变化，在这些参数中，受到影响最大的是vfree，ρjam。这两个参数随时间轴发生变化，在迭代轴上具有重复特性，故符合切换系统特性，表达式可表示以下：Among them, k is the number of iterations, C(x(t),t), D(x(t),t) are two nonlinear functions. Since the incoming traffic flow in a limited time period is limited and cannot be negative, uk (t) is limited by the saturation of the actual conditions of the ramp. In the actual system, the parameters of the expressway system will change with environmental factors. Among these parameters, vfree and ρjam are most affected. These two parameters change with the time axis and have repetition characteristics on the iteration axis, so they conform to the characteristics of the switching system. The expression can be expressed as follows:
其中i＝i(t)是在t∈{0,1,2,...,N}切换规则，取值于有限序列P＝{1,2,...,m}，m是子系统的个数。在t∈{0,1,2,...,N}，尽管因为匝道车流量流出会有一些模型不确定性和干扰，系统输入使得系统输出也能够达到期望输出yd(t)＝[yd,(1)(t) yd,(2)(t) ... yd,(K)(t)]T。Where i=i(t) is the switching rule at t ∈ {0,1,2,...,N}, the value is in the finite sequence P={1,2,...,m}, m is the subsystem the number of . At t ∈ {0,1,2,...,N}, although there are some model uncertainties and disturbances due to outflow of on-ramp traffic, the system input make the system output The desired output yd (t)=[yd,(1) (t) yd,(2) (t) ... yd,(K) (t)]T can also be achieved.
而对于输出的交通流密度信号yk(t)，其从快速路端传输至控制中心端的丢包概率为则控制中心端根据接收到的实际交通流密度信号与本地存储的期望交通流密度值所获得的误差函定义为For the output traffic flow density signal yk (t), the packet loss probability transmitted from the expressway end to the control center end is Then the error function obtained by the control center terminal according to the received actual traffic flow density signal and the expected traffic flow density value stored locally is defined as
ek(t)＝ak(t)(yd(t)-ak(t)yk(t))＝ak(t)(yd(t)-yk(t))ek (t)=ak (t)(yd (t)-ak (t)yk (t))=ak (t)(yd (t)-yk (t))
定义αk(t)为第k次迭代的yk(t)不发生随机丢失的事件序列。Define αk (t) as the sequence of events in which random loss does not occur in yk (t) of the kth iteration.
图3是在获得当前次的迭代误差函数后，根据该误差函数给出传输丢包环境下基于切换系统的迭代学习控制律。Fig. 3 shows the iterative learning control law based on the switching system in the transmission packet loss environment after obtaining the current iterative error function according to the error function.
uk+1(t)＝(1-αk(t+1)θ(k+1))sat[uk(t)]+Γiek(t+1)uk+1 (t)＝(1-αk (t+1)θ(k+1))sat[uk (t)]+Γi ek (t+1)
K＝12表示第k+1次迭代的匝道入口的交通流，uk(t)表示第k次迭代的匝道入口的交通流，θ(k)为自适应遗忘因子函数。αk(t)为第k次迭代的yk(t)不发生随机丢失的事件序列，误差函数ek(t+1)表示第k次迭代t+1时刻的输出误差，而Γi为第i个切换系统子系统的控制增益，饱和函数sat[·]表达式如下：K=12 represents the traffic flow at the ramp entrance of the k+1 iteration, uk (t) represents the traffic flow at the ramp entrance of the k iteration, and θ(k) is an adaptive forgetting factor function. αk (t) is the event sequence in which random loss does not occur in yk (t) of the k-th iteration, the error function ek (t+1) represents the output error at the time of the k-th iteration t+1, and Γi is The control gain of the i-th switching system subsystem, the saturation function sat[·] is expressed as follows:
其中in
其中j∈{1,2,...,K}，K为一段快速路中的匝道数。为第k次迭代时匝道j的交通流控制量，u(j),max(t)与u(j),min(t)分别为匝道j交通流控制量的饱和上下界。Where j∈{1,2,...,K}, K is the number of ramps in a section of expressway. is the traffic flow control volume of ramp j at the kth iteration, u(j), max (t) and u(j), min (t) are the saturation upper and lower bounds of the traffic flow control volume of ramp j respectively.
在本应用场景下，我们有K＝5and L(j)＝500m，j∈{1,2,...,K}，yk(t)丢失概率为0.3，每个路段均有一个匝道入口与一个匝道出口。设置遗忘因子的上下限为θmin＝0.0，θmax＝0.03，收敛指标的容忍误差为0.3。In this application scenario, we have K=5and L(j) =500m, j∈{1,2,...,K}, yk (t) loss probability is 0.3, each section has a ramp entrance with a ramp exit. The upper and lower limits of the forgetting factor are set as θmin =0.0, θmax =0.03, and the tolerance error of the convergence index is 0.3.
图4是本应用场景下的切换规则。因为在实际应用中，受天气、路况等因素影响，快速路模型系统参数随之发生变化，在这些参数中，受到影响最大的是vfree，ρjam。在本应用场景下，快速路系统在时间轴上两个系统之间进行切换，切换规则如图4。Figure 4 shows the switching rules in this application scenario. Because in practical applications, affected by factors such as weather and road conditions, the parameters of the expressway model system will change accordingly. Among these parameters, vfree and ρjam are most affected. In this application scenario, the expressway system switches between the two systems on the time axis, and the switching rules are shown in Figure 4.
图5为本应用场景下，快速路系统中各路段的匝道入口车辆需求量。Figure 5 shows the vehicle demand at the ramp entrance of each road section in the expressway system in this application scenario.
图6为本应用场景下，快速路系统中各路段的匝道出口的车流量。Figure 6 shows the traffic flow at the ramp exit of each road section in the expressway system in this application scenario.
图7为本应用场景下各路段的输出误差指标图。在本应用场景中，可以看到误差指标在有限次内快速收敛到容忍误差之内，误差指标是，j＝1,2,3,4,5，即，误差指标是各路段的输出误差在固定时间区间上的均方误差。Fig. 7 is an output error indicator diagram of each road section in this application scenario. In this application scenario, it can be seen that the error index quickly converges to within the tolerance error within a limited number of times. The error index is, j=1, 2, 3, 4, 5, that is, the error index is the mean square error of the output error of each road section in a fixed time interval.
从图7中可以看出，第3段、第4段与第5段快速路的误差指标在第30次迭代收敛至容忍误差之内，即这两段车流量密度近似达到期望车流量密度。接着，第1段与第2段快速路的误差指标依第40次迭代收敛至容忍误差之内，即这两段车流量密度近似达到期望车流量密度。It can be seen from Figure 7 that the error indicators of the third, fourth, and fifth expressways converged to within the tolerance error at the 30th iteration, that is, the traffic flow density of these two sections approximately reached the expected traffic flow density. Then, the error indicators of the first section and the second section of the expressway converge to within the tolerance error according to the 40th iteration, that is, the traffic flow density of the two sections approximately reaches the expected traffic flow density.
图8为本应用场景下两种控制方法的快速路交通输出误差指标与无传输丢包环境下控制的快速路交通输出误差指标比较图。在本应用场景中，误差指标是，j＝1,2,3,4,5。可看出，使用所述传输丢包环境下基于切换系统特性的迭代学习快速路入口匝道控制方法，快速路的误差指标在第40次迭代收敛至容忍误差之内，即快速路交通车流量密度近似达到期望车流量密度。其比传输丢包环境下使用常规迭代学习控制方法的快速路交通输出误差指标收敛需要更少的迭代次数，而且与无传输丢包环境下使用常规迭代学习控制方法的迭代次数相近。可以看出所述方法不仅在可一定迭代次数内使快速路能够达到期望交通流密度，同时能够改善由于数据丢包导致的交通控制收敛速度变慢的问题。Figure 8 is a comparison diagram of the output error indicators of the expressway traffic of the two control methods in this application scenario and the output error indicators of the expressway traffic controlled in the environment of no transmission packet loss. In this application scenario, the error index is, j=1,2,3,4,5. It can be seen that the error index of the expressway converges to within the tolerance error at the 40th iteration, that is, the traffic flow density of the expressway is Approximately achieves the desired traffic flow density. It requires fewer iterations than the conventional iterative learning control method for the convergence of the expressway traffic output error index in the environment of transmission packet loss, and is similar to the number of iterations of the conventional iterative learning control method in the environment of no transmission packet loss. It can be seen that the method not only enables the expressway to achieve the desired traffic flow density within a certain number of iterations, but also improves the problem of slow convergence of traffic control caused by data packet loss.
实施例2Example 2
如图9所示，一种具有单边传输丢包情况的基于切换系统特性迭代学习的快速路入口匝道远程控制系统，包括：As shown in Figure 9, an expressway on-ramp remote control system based on iterative learning of switching system characteristics with unilateral transmission packet loss, including:
当前交通流获取模块901：用于控制中心以通信方式获取当前快速路实际交通流密度；The current traffic flow acquisition module 901: used for the control center to obtain the actual traffic flow density of the current expressway by means of communication;
交通信息获取模块902：用于控制中心获取本地存储设备中期望交通流、匝道入口的交通流控制量；Traffic information acquisition module 902: used for the control center to acquire the expected traffic flow in the local storage device and the traffic flow control amount at the entrance of the ramp;
误差函数计算模块903：用于计算当前误差函数。其根据当前实际接收到的交通流密度值与期望交通流密度计算获得当前误差函数；Error function calculation module 903: for calculating the current error function. It calculates and obtains the current error function according to the currently received traffic flow density value and the expected traffic flow density;
ek(t)＝αk(t)(yd(t)-yk(t))ek (t)=αk (t)(yd (t)-yk (t))
αk(t)为第k次迭代的yk(t)不发生随机丢失的事件序列，快速路端至控制中心端的传输丢包概率为αk (t) is the event sequence in which random loss does not occur in yk (t) of the k-th iteration, and the transmission packet loss probability from the express road end to the control center end is
遗忘因子计算模块904：用于根据误差函数计算遗忘因子值，计算公式如下Forgetting factor calculation module 904: used to calculate the value of the forgetting factor according to the error function, the calculation formula is as follows
其中γ(k)不为第k次迭代误差函数序列中不为0的元素个数。Where γ(k) is not the number of elements that are not 0 in the error function sequence of the kth iteration.
迭代学习律和学习增益设置模块905：用于根据误差函数与自适应遗忘因子设置传输丢包情况下基于切换系统特征的迭代学习控制律以及迭代学习增益，其中，迭代学习控制律表示如下：Iterative learning law and learning gain setting module 905: used to set the iterative learning control law and iterative learning gain based on the characteristics of the switching system in the case of transmission packet loss according to the error function and the adaptive forgetting factor, wherein the iterative learning control law is expressed as follows:
uk+1(t)＝(1-αk(t+1)θ(k+1))sat[uk(t)]+Γiek(t+1)uk+1 (t)＝(1-αk (t+1)θ(k+1))sat[uk (t)]+Γi ek (t+1)
式子中，uk+1(t)表示第k+1次迭代的匝道入口的交通流，uk(t)表示第k次迭代的匝道入口的交通流，θ(k)为自适应遗忘因子函数，误差函数ek(t+1)表示第k次迭代t+1时刻的输出误差，而Γi为第i个切换系统子系统的控制增益，饱和函数sat[·]表达式如下：In the formula, uk+1 (t) represents the traffic flow at the ramp entrance of the k+1 iteration, uk (t) represents the traffic flow at the ramp entrance of the k iteration, and θ(k) is the adaptive forgetting The factor function, the error function ek (t+1) represents the output error at the time of the kth iteration t+1, andΓi is the control gain of the ith switching system subsystem, and the expression of the saturation function sat[·] is as follows:
其中in
其中j∈{1,2,...,K}，u(j),max(t)与u(j),min(t)分别为闸道j的饱和函数上下界；Where j∈{1,2,...,K}, u(j), max (t) and u(j), min (t) are the upper and lower bounds of the saturation function of gateway j respectively;
控制信号输出模块906：用于控制中心生成快速路匝道入口的交通流控制信号，并以可靠通信方式发送至快速路端；Control signal output module 906: used for the control center to generate the traffic flow control signal at the ramp entrance of the expressway, and send it to the end of the expressway through reliable communication;
控制信号获取模块907：用于快速路端通过可靠通信方式获取交通流控制信号；Control signal acquisition module 907: used for obtaining traffic flow control signals at the expressway end through reliable communication;
交通控制模块908：用于将交通流控制信号应用到带有匝道入口的快速路系统控制，使得能够在一定迭代次数内使快速路能够达到期望交通流密度；Traffic control module 908: used to apply the traffic flow control signal to the expressway system control with ramp entrance, so that the expressway can achieve the desired traffic flow density within a certain number of iterations;
交通流密度输出模块909：用于快速路端通过通信方式将当前快速路交通流密度信号发送至控制中心。Traffic flow density output module 909: used for the expressway end to send the current expressway traffic flow density signal to the control center through communication.
本实施例中，首先，交通流密度实际输出的当前交通流获取模块901，用于获取快速路交通发送的当前主路的实际交通流密度；其次，交通信息获取模块902，用于获取期望交通流密度值与以及本次迭代的匝道入口交通流控制量；然后，误差函数获取模块903，用于根据所述实际接收到的交通流密度和所述期望交通流密度值获得所述方法的当前误差函数；接着，遗忘因子计算模块904，用于根据误差函数计算出遗忘因子值；再接着，迭代学习律和学习增益设置模块905，用于根据所述误差函数与自适应遗忘因子设置迭代学习律的迭代学习增益以及传输丢包环境下基于切换系统的迭代学习控制律；再接着，控制信号输出模块906：将生成的快速路匝道入口交通流控制信号通过可靠通信方式发送至快速路交通；再接着，控制信号获取模块907：快速路端通过可靠通信方式获取交通流控制信号；最终，交通控制模块908：将交通流控制信号输入到到带有匝道入口的快速路系统中，使得能够在一定迭代次数内使快速路能够达到期望交通流密度；此外，交通流密度输出模块909：用于发送当前快速路交通流密度信号至控制中心。本实施例在现有的快速路系统基于迭代学习控制方法研究中进行进一步深入研究，使得快速路控制系统考虑了实际因素，即受天气、路况等因素。此外快速路与控制中心之间物理距离遥远，且快速路的测量设备经常缺乏维护并暴露野外。故快速路系统符合切换系统特性同时又可能有测量输出丢包的实际问题，在这种情形下进行控制更符合实际应用的需要。In this embodiment, first, the current traffic flow acquisition module 901 actually output by the traffic flow density is used to acquire the actual traffic flow density of the current main road sent by the expressway traffic; secondly, the traffic information acquisition module 902 is used to acquire the expected traffic flow The flow density value and the ramp entrance traffic flow control amount of this iteration; then, the error function acquisition module 903 is used to obtain the current value of the method according to the actually received traffic flow density and the expected traffic flow density value. Error function; then, the forgetting factor calculation module 904 is used to calculate the forgetting factor value according to the error function; then, the iterative learning law and the learning gain setting module 905 are used to set iterative learning according to the error function and the adaptive forgetting factor The iterative learning gain of the law and the iterative learning control law based on the switching system in the transmission packet loss environment; then, the control signal output module 906: send the generated traffic flow control signal at the entrance of the expressway ramp to the expressway traffic through reliable communication; Next, the control signal acquisition module 907: the express road end obtains the traffic flow control signal through reliable communication; finally, the traffic control module 908: inputs the traffic flow control signal into the expressway system with the ramp entrance, so that the traffic flow control signal can be The expressway can achieve the desired traffic flow density within a certain number of iterations; in addition, the traffic flow density output module 909 is used to send the current expressway traffic density signal to the control center. In this embodiment, further in-depth research is carried out in the research of the existing expressway system based on iterative learning control method, so that the expressway control system takes into account actual factors, namely factors such as weather and road conditions. In addition, the physical distance between the expressway and the control center is long, and the measurement equipment of the expressway is often poorly maintained and exposed to the wild. Therefore, the expressway system conforms to the characteristics of the switching system and may have the practical problem of measurement output packet loss. In this case, the control is more in line with the needs of practical applications.
相同或相似的标号对应相同或相似的部件；The same or similar reference numerals correspond to the same or similar components;
附图中描述位置关系的用语仅用于示例性说明，不能理解为对本专利的限制；显然，本发明的上述实施例仅仅是为清楚地说明本发明所作的举例，而并非是对本发明的实施方式的限定。对于所属领域的普通技术人员来说，在上述说明的基础上还可以做出其它不同形式的变化或变动。这里无需也无法对所有的实施方式予以穷举。凡在本发明的精神和原则之内所作的任何修改、等同替换和改进等，均应包含在本发明权利要求的保护范围之内。The terms describing the positional relationship in the accompanying drawings are only for illustrative purposes, and cannot be interpreted as limitations on the patent; obviously, the above-mentioned embodiments of the present invention are only examples for clearly illustrating the present invention, rather than implementation of the present invention limitation of the method. For those of ordinary skill in the art, other changes or changes in different forms can be made on the basis of the above description. It is not necessary and impossible to exhaustively list all the implementation manners here. All modifications, equivalent replacements and improvements made within the spirit and principles of the present invention shall be included within the protection scope of the claims of the present invention.
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711336908.XACN108154688B (en) | 2017-12-14 | 2017-12-14 | Expressway on-ramp control method and system for iterative learning in packet loss environment |
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711336908.XACN108154688B (en) | 2017-12-14 | 2017-12-14 | Expressway on-ramp control method and system for iterative learning in packet loss environment |
| Publication Number | Publication Date |
|---|---|
| CN108154688Atrue CN108154688A (en) | 2018-06-12 |
| CN108154688B CN108154688B (en) | 2019-07-02 |
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711336908.XAExpired - Fee RelatedCN108154688B (en) | 2017-12-14 | 2017-12-14 | Expressway on-ramp control method and system for iterative learning in packet loss environment |
| Country | Link |
|---|---|
| CN (1) | CN108154688B (en) |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101045863B1 (en)* | 2011-02-16 | 2011-07-01 | 공주대학교 산학협력단 | Highway ramp metering system to eliminate non-recurring congestion |
| CN103606284A (en)* | 2013-11-14 | 2014-02-26 | 北京交通大学 | Intelligent ramp entrance traffic demand combined control system and method |
| CN104240521A (en)* | 2014-09-30 | 2014-12-24 | 天津通翔智能交通系统有限公司 | Intelligent ramp traffic control machine, system and method |
| CN106355905A (en)* | 2016-10-28 | 2017-01-25 | 银江股份有限公司 | Control method for overhead signal based on checkpoint data |
| CN106600991A (en)* | 2017-01-17 | 2017-04-26 | 河北工业大学 | City expressway multi-ramp coordination control method based on chaos |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101045863B1 (en)* | 2011-02-16 | 2011-07-01 | 공주대학교 산학협력단 | Highway ramp metering system to eliminate non-recurring congestion |
| CN103606284A (en)* | 2013-11-14 | 2014-02-26 | 北京交通大学 | Intelligent ramp entrance traffic demand combined control system and method |
| CN104240521A (en)* | 2014-09-30 | 2014-12-24 | 天津通翔智能交通系统有限公司 | Intelligent ramp traffic control machine, system and method |
| CN106355905A (en)* | 2016-10-28 | 2017-01-25 | 银江股份有限公司 | Control method for overhead signal based on checkpoint data |
| CN106600991A (en)* | 2017-01-17 | 2017-04-26 | 河北工业大学 | City expressway multi-ramp coordination control method based on chaos |
| Title |
|---|
| CHI RONGHU ET AL.: "A new PI controller for freeway ramp metering based on fuzzy logic", 《2013 25TH CHINESE CONTROL AND DECISION CONFERENCE (CCDC)》* |
| HOU ZHONGSHENG ET AL.: "An iterative learning approach for density control of freeway traffic flow via ramp metering", 《TRANSPORTATION RESEARCH PART C: EMERGING TECHNOLOGIES》* |
| 侯忠生等: "带有迭代学习前馈的快速路无模型自适应入口匝道控制", 《自动化学报》* |
| 刘祥鹏: "快速路交通入口匝道的智能学习控制方法", 《中国优秀硕士学位论文全文数据库工程科技Ⅱ辑》* |
| 赵明等: "受限迭代学习入口匝道控制收敛性分析", 《交通运输系统工程与信息》* |
| Publication number | Publication date |
|---|---|
| CN108154688B (en) | 2019-07-02 |
| Publication | Publication Date | Title |
|---|---|---|
| James et al. | Characterizing the impact of production adaptive cruise control on traffic flow: An investigation | |
| Gao et al. | Adaptive traffic signal control: Deep reinforcement learning algorithm with experience replay and target network | |
| Li et al. | Backpressure control with estimated queue lengths for urban network traffic | |
| US10847029B2 (en) | Method, apparatus, and system for automatic road closure detection | |
| US10210765B2 (en) | Inference and interpolation of continuous 4D trajectories | |
| CN106846836B (en) | A kind of Single Intersection signal timing control method and system | |
| CN101706888B (en) | Method for predicting travel time | |
| CN108986512A (en) | A kind of bus arrival time querying method and device | |
| CN110288170A (en) | Prediction model construction method, traffic flow forecasting method, device and electronic equipment | |
| US20220353728A1 (en) | Method and apparatus for managing roadside device in vehicle road cooperation, and cloud control platform system | |
| CN106023629A (en) | Path recommendation method and path recommendation device | |
| Roopa et al. | Dynamic management of traffic signals through social IoT | |
| Mishra et al. | Adaptive traffic signal control for developing countries using fused parameters derived from crowd-source data | |
| CN114200835A (en) | A Distributed Control Method for Heterogeneous Fleet Stability Based on Tube MPC | |
| CN113269339B (en) | Method and system for automatically creating and distributing network appointment tasks | |
| CN106846830A (en) | Through street On-ramp Control method and system based on switching system characteristic | |
| Fadila et al. | Comprehensive Review of Smart Urban Traffic Management in the Context of the Fourth Industrial Revolution | |
| Dong et al. | Analysis and control of intelligent traffic signal system based on adaptive fuzzy neural network | |
| CN108154688A (en) | The through street On-ramp Control method and system of iterative learning under packet loss environment | |
| EP3985636B1 (en) | Computer system and method for determining reliable vehicle control instructions | |
| CN108594799A (en) | The real-time acquisition device of pilotless automobile traffic information and system | |
| Li et al. | Elevating adaptive traffic signal control in semi‐autonomous traffic dynamics by using connected and automated vehicles as probes | |
| CN112258856A (en) | Method for establishing regional traffic signal data drive control model | |
| Wu et al. | Development of platoon‐based actuated signal control systems to coordinated intersections: Application in corridors in Houston | |
| KR20210096403A (en) | Artificial intellignece-based optimal distribution system and method thereof |
| 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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee | Granted publication date:20190702 Termination date:20201214 |