An integrated and cooperative architecture for multi-intersection traffic signal control

Traffic signal control(TSC)systems are one essential component in intelligent transport systems.However,relevant studies are usually independent of the urban traffic simulation environment,collaborative TSC algorithms and traffic signal communication.In this paper,we propose(1)an integrated and cooperative Internet-of-Things architecture,namely General City Traffic Computing System(GCTCS),which simultaneously leverages an urban traffic simulation environment,TSC algorithms,and traffic signal communication;and(2)a general multi-agent reinforcement learning algorithm,namely General-MARL,considering cooperation and communication between traffic lights for multi-intersection TSC.In experiments,we demonstrate that the integrated and cooperative architecture of GCTCS is much closer to the real-life traffic environment.The General-MARL increases the average movement speed of vehicles in traffic by 23.2%while decreases the network latency by 11.7%.

A Model with Traffic Routers, Dynamically Managing Signal Phases to Address Traffic Congestion in Real Time

On-road Vehicular traffic congestion has detrimental effect on three lifelines: Economy, Productivity and Pollution (EPP). With ever increasing population of vehicles on road, traffic congestion is a major challenge to the economy, productivity and pollution, notwithstanding continuous developments in alternative fuels, alternative sources of energy. The research develops accurate and precise model in real time which computes congestion detection, dynamic signaling algorithm to evenly distribute vehicle densities while ensuring avoidance of starvation and deadlock situation. The model incorporates road segment length and breadth, quality and achievable average speed to compute road capacity. Vehicles installed with GPS enabled devices provide their location, which enables computing road occupancy. Road occupancy is evaluated based on number of vehicles as well as area occupied by vehicles. Ratio of road occupancy and road capacity provides congestion index important to compute signal phases. The algorithm ensures every direction is serviced once during a signaling cycle ensuring no starvation. Secondly, the definition of minimum and maximum signal timings ensures against dead lock situation. A simulator is developed to validate the proposition and proves it can ease congestion by more than 50% which is better than any of the contemporary approaches offering 15% improvement. In case of higher congestion index, alternate routes are suggested based on evaluation of traffic density graphs for shortest route or knowledge database. The algorithm to compute shortest route is optimized drastically, reducing computation cost to 3*√2N vis-à-vis computation cost of N2 by classical algorithms. The proposal brings down the cost of implementation per traffic junction from USD 30,000 to USD 2000.

Assessing the Effects of Indirect Left Turn on a Signalized Intersection Performance: A Case Study for the Tehran Metropolitan Area

Many signalized intersections are characterized with frequent left-turn moves. Vehicles waiting for a protected left turn may form long queues, which will increase the intersection delay and negatively impact the network performance. Researchers and practitioners across various countries underline that access management leads to a smoother traffic flow. One way of access management at intersections is to eliminate the direct left-turn maneuver. This study aims to evaluate how the traffic conditions will be affected from replacing the direct left turn with the right-turn U-turn maneuver at intersections. In case of the right-turn U-turn maneuver, a vehicle turns right instead of making the left turn and travels either to the median opening or to the next intersection to make a U-turn. Two simulation models are built using the Synchro Studio and Aimsun simulation software packages based on the data, collected from one of the busiest intersections in Tehran (Iran), to quantify the effects of replacing the direct left turn with the right-turn U-turn maneuver on the intersection and network performance. Results of a comprehensive simulation analysis indicate that the proposed access management treatment not only significantly reduces the total vehicle queue length and the total delay at the considered intersection, but also decreases the total network delay and the total travel time. Furthermore, elimination of the direct left turn increases the number of vehicles entering the network.

Recent advances in traffic signal performance evaluation

Signal retiming is a prominent way that transportation agencies use to fight congestion and change of traffic pattern.Performance evaluations of traffic conditions at signalized intersections and arterials provide actionable data for agencies to make well-informed and prioritized signal retiming decisions.However,the abundance of data sources,the lack of standardized evaluation methods and oftentimes the shortage of resources make it a difficult endeavor.The review detailed in this paper examines the advances made in traffic signal performance evaluation.We establish the necessity for the evaluations,study the process of continuous improvement of traffic signal performance using the evaluations,and then examine multiple methodologies in a plethora of research endeavors.Particularly,we focus on probe vehicles and sensors data,the two major sources of data.We discuss how sensors are connected to signal controllers to provide relevant in-depth traffic data including speed and occupancy measures.We also review the nature of probe vehicles and the level of penetration.We then define and summarize performance measures derived from both sources,to aid in performance evaluations.For performance evaluation methods,we discuss the research studies and provide summaries including advantages and disadvantages of the methods used,as well as a holistic outlook for future research.This paper is aimed to provide a comprehensive review on the state-of-the-art to benefit researcher,traffic agencies,and commercial entities that thrive to improve safety and efficiency of traffic signals through performance evaluations.

Model predictive control for hybrid vehicle ecological driving using traffic signal and road slope information

This paper presents development of a control system for ecological driving of a hybrid vehicle. Prediction using traffic signal and road slope information is considered to improve the fuel economy. It is assumed that the automobile receives traffic signal information from intelligent transportation systems （ITS）. Model predictive control is used to calculate optimal vehicle control inputs using traffic signal and road slope information. The performance of the proposed method was analyzed through computer simulation results. Both the fuel economy and the driving profile are optimized using the proposed approach. It was observed that fuel economy was improved compared with driving of a typical human driving model.

Adaptive green traffic signal controlling using vehicular communication

The importance of using adaptive traffic signal control for figuring out the unpredictable traffic congestion in today's metropolitan life cannot be overemphasized. The vehicular ad hoc network(VANET), as an integral component of intelligent transportation systems(ITSs), is a new potent technology that has recently gained the attention of academics to replace traditional instruments for providing information for adaptive traffic signal controlling systems(TSCSs). Meanwhile, the suggestions of VANET-based TSCS approaches have some weaknesses:(1) imperfect compatibility of signal timing algorithms with the obtained VANET-based data types, and(2) inefficient process of gathering and transmitting vehicle density information from the perspective of network quality of service(Qo S). This paper proposes an approach that reduces the aforementioned problems and improves the performance of TSCS by decreasing the vehicle waiting time, and subsequently their pollutant emissions at intersections. To achieve these goals, a combination of vehicle-to-vehicle(V2V) and vehicle-to-infrastructure(V2I) communications is used. The V2 V communication scheme incorporates the procedure of density calculation of vehicles in clusters, and V2 I communication is employed to transfer the computed density information and prioritized movements information to the road side traffic controller. The main traffic input for applying traffic assessment in this approach is the queue length of vehicle clusters at the intersections. The proposed approach is compared with one of the popular VANET-based related approaches called MC-DRIVE in addition to the traditional simple adaptive TSCS that uses the Webster method. The evaluation results show the superiority of the proposed approach based on both traffic and network Qo S criteria.

KSUTraffic: A Microscopic Traffic Simulator for Traffic Planning in Smart Cities

Simulation is a powerful tool for improving,evaluating and analyzing the performance of new and existing systems.Traffic simulators provide tools for studying transportation systems in smart cities as they describe the evolution of traffic to the highest level of detail.There are many types of traffic simulators that allow simulating traffic in modern cities.The most popular traffic simulation approach is the microscopic traffic simulation because of its ability to model traffic in a realistic manner.In many cities of Saudi Arabia,traffic management represents a major challenge as a result of expansion in traffic demands and increasing number of incidents.Unfortunately,employing simulation to provide effective traffic management for local scenarios in Saudi Arabia is limited to a number of commercial products in both public and private sectors.Commercial simulators are usually expensive,closed source and inflexible as they allow limited functionalities.In this project,we developed a local traffic simulator“KSUtraffic”for traffic modeling,planning and analysis with respect to different traffic control strategies and considerations.We modeled information specified by GIS and real traffic data.Furthermore,we designed experiments that manipulate simulation parameters and the underlying area.KSUTraffic visualizes traffic and provides statistical results on the simulated traffic which would help to improve traffic management and efficiency.

智能交通环境下的Multi-Agent的交通仿真研究

在总结智能交通Agent技术的基础上,把Multi-Agent仿真应用于无信号交叉口的车流引导中,并以Netlogo语言为开发平台进行了系统的建模、仿真与评价,探讨了具体采用Multi-Agent技术进行交通仿真的方法,为Agent技术的研究奠定了基础。

基于异构多智能体自注意力网络的路网信号协调顺序优化方法

针对路网交通信号控制的复杂性,本文提出基于异构多智能体自注意力网络的路网信号协调顺序优化方法,提升路网范围内多交叉口信号控制策略性能。首先,模型考虑多交叉口交通流的空间相关性,采用基于自注意力机制的价值编码器学习交通观测表征,实现路网级通信;其次,面向多智能体策略更新的非稳态环境,模型在前序智能体的联合动作基础上,基于多智能体优势分解的策略解码器,顺序决策最优反应动作;最后,设计基于有效行驶车辆的动作掩码机制,在时效完备区间自适应调节决策频率,并提出考虑等待公平性的时空压力奖励函数,进一步提高策略性能与实用性。在杭州路网数据集上验证模型有效性,结果表明:所提模型在2个数据集和5个性能指标上均优于基准模型;相比最优基准模型,所提模型平均行程时间降低10.89%,平均排队长度降低18.84%,平均等待时间降低22.21%。此外,所提模型的泛化能力更强,且显著减少车辆等待时间过长的情形。

智能网联环境下单交叉口车辆轨迹优化

为了提高信号灯前车辆的通行效率,改善交通流整体运行水平,本文从减少车辆延误和降低燃油消耗两个角度入手,在智能网联环境下,提出了一种车辆编组识别算法和针对编组头车的多目标线性轨迹优化模型(MOLP-pl)。首先对智能驾驶员跟驰模型(IDM)进行改进,调整车辆状态,减少车辆随机到达状态下车辆速度和车头时距分布的差异,同时为后续MOLP-pl轨迹优化模型的运行提供先决条件。在此基础上,以车辆编组为优化单元,通过车辆编组识别算法识别编组头车和跟随车辆,将编组头车的行驶轨迹作为优化对象并建立相应的数学模型。为了提高车辆轨迹优化模型的求解效率和精度,对其进行线性化重构,采用线性求解器计算编组头车加速度,构建编组头车最佳时空轨迹,然后,利用IDM跟驰模型计算跟随车辆的行驶速度,从而使编组车辆最大效率的通过交叉口。最后,利用SUMO构建的仿真实验表明:本研究提出的车辆轨迹优化算法可显著提高信号灯前车辆的通行效率,在三种不同的交通饱和度条件下,相对于无速度引导场景,车辆延误分别降低了8.56%、12.42%、64.79%,燃油消耗分别降低了17.21%、18.34%、12.64%;相对于逻辑控制场景,延误分别降低了-1.31%、2.63%、60.83%,燃油消耗分别降低了2.47%、7.91%、2.28%。

结合模糊控制的深度强化学习交通灯控制策略

现有交通信号灯控制策略大多针对单一交叉口展开分析,该策略仅考虑车流量的单一因素,难以适应动态的路网状态。对此,提出了一种结合模糊控制的深度强化学习交通灯控制策略,利用SAC(soft actor critic)深度强化学习对两交叉口的交通信号灯相位选择及配时进行联合优化,同时考虑车辆速度、路段车辆排队长度等因素,利用模糊控制对SAC的惩罚函数进行处理。实验结果表明,与固定循环周期策略、SAC控制策略和DDPG(deep deterministic policy gradient)控制策略相比,提出的交通信号灯控制策略能获得更快的车辆通行速度,车辆的油耗和尾气排放情况也得到了改善。

大规模智慧交通信号控制中的强化学习和深度强化学习方法综述

当前在交通信号控制系统中引入智能化检测和控制已是大势所趋,特别是强化学习和深度强化学习方法在可扩展性、稳定性和可推广性等方面展现出巨大的技术优势,已成为该领域的研究热点。针对基于强化学习的交通信号控制任务进行了研究,在广泛调研交通信号控制方法研究成果的基础上,系统地梳理了强化学习和深度强化学习在智慧交通信号控制领域的分类及应用;并归纳了使用多智能体合作的方法解决大规模交通信号控制问题的可行方案,对大规模交通信号控制的交通场景影响因素进行了分类概述;从提高交通信号控制器性能的角度提出了本领域当前所面临的挑战和未来可能极具潜力的研究方向。

基于SAC算法的多交叉口交通信号控制研究

针对深度Q网络(deep Q-learning network,DQN)算法在解决多交叉口交通信号配时方案由于外部环境变化和内部参数波动导致效果不佳的问题,提出了基于柔性“行动器-评判器”(soft actor-critic,SAC)的交叉口交通信号控制方法,并设计了相应的系统采样策略和回报函数.与原采样策略相比,新采样策略将相邻智能体的策略信息加入到系统状态中,使当前智能体能够得到更多的交叉口交通分布和合作策略信息.与原回报函数相比,新回报函数中引入空间折扣因子,缩小了相邻智能体的观察和回报值,使当前智能体更加关注和改善当前交通状况.随后在此基础上分别应用DQN和SAC算法设计交通信号控制方法.Webster配时法是利用相位流量数据开发的一种基于周期的固定相位长度交通信号方法,与DQN和SAC算法相比,其优化目标是降低交叉口延迟时间,不考虑交叉口排队长度.在城市交通模拟软件(simulation of urban mobility,SUMO)中构建一个时变交通流交通网络,并在其中分别对基于DQN、SAC和Webster配时法的信号配时控制方法进行仿真测试.仿真结果表明:基于SAC算法的交通信号控制方法与基于DQN算法和Webster配时法的交通信号控制方法相比,能够显著减少交叉口排队长度和平均延迟时间,具体来说,车辆平均排队长度分别减少了17.8%和28.2%,平均延迟分别减少了26.8%和36.3%,说明所提出的方法具有更好的控制效果.

基于车路协同的智能交通信号灯优化控制策略

交通拥堵、交通事故、公共交通不便及空气污染等已成为制约城市交通发展的主要因素,解决城市交通问题已成为智能城市建设的首要任务。利用车路协同技术搭建车辆和信号灯之间的通信通道,及时收集车辆的实时状态、位置信息、道路的拥堵情况和交通流量等信息,通过车辆和道路设施之间的信息交互系统实现自适应的交通信号控制,达到信号灯的精细化控制和智能化管理。基于车路协同的智能信号灯优化控制策略是一种新型的交通信号控制策略,通过实时采集车辆信息、实时监测交通流和调整信号灯,达到交通信号灯优化控制,对全面提高城市交通效率、安全性和环保性,解决城市交通拥堵等问题具有很好的针对性。

深度强化学习驱动的智能交通信号控制策略综述

随着城市人口快速增加,私家车数量呈指数级增长,使本已不堪重负的交通系统将承受更大的压力,交通拥堵问题愈加凸显。传统交通信号控制技术难以适应复杂多变的交通情况,数据驱动的方法为基于控制的系统带来了新方向。深度强化学习方法与交通控制系统的结合在自适应交通信号控制中扮演着重要角色。首先,文中综述了智能交通信号控制系统应用的最新进展,对智能交通信号控制方法进行了分类讨论,总结了这一领域的现有工作。其次,采用深度强化学习方法能够有效解决智能交通信号控制中状态信息获取不准确、控制算法鲁棒性差以及区域协调控制能力弱等问题,在此基础上,给出了智能交通信号控制的仿真平台和实验设置概述,并通过实例进行了分析和验证。最后,探讨了智能交通信号控制领域面临的挑战和有待解决的问题,并总结了未来的研究方向。

基于智能导航和交通信号灯优化设计的城市智慧交通系统构建研究

本文旨在基于智能导航和交通信号灯优化设计,构建城市智慧交通体系模型。文章首先介绍了基于智能导航和交通信号灯优化设计结合的城市智慧交通体系应用背景,并探讨了智能导航系统及交通信号灯优化设计的研究概述。然后,文章详细讨论了智能导航系统的组成及工作原理,并探讨了交通信号灯优化设计的部件和实现方式。最后,文章提出了智能导航系统和交通信号灯优化设计的方法。此项研究为未来城市智慧交通体系的构建提供了重要的参考和指导。

基于可变相位时长的网联车队交叉口通行信号控制方法

城市信号交叉口是城市交通网络的重要节点、也是交通事件的频发区域之一。为提高网联车队在交叉口处的通行效率,降低车辆平均停车延误,提出一种车联网环境下基于可变相位的车队交叉口信号控制模型。根据车辆初始速度与车辆加入随机性,引导车辆以固定间距车队形式通行;通过信号控制区域栅格化处理,收缩车辆停车等待时间,减少车辆平均停车延误,并对周期内各相位放行时间进行优化,调整绿灯相位时长使车队完整通过交叉口。利用SUMO仿真软件对实际交叉口采集真实数据进行验证,将提出方法与现有信号控制方法进行比对,分析不同放行方式下的延误。结果显示,其效果在平均延误方面减少43.10%和10.86%,平均排队长度减少38.46%和13.80%。构建的方法有较好的可行性与适用性,可以为城市网联信号控制优化提供借鉴与参考。

智能网联环境下管理车道设置策略与影响因素分析

为探索人工驾驶车辆(human driven vehicle,HDV)与自动驾驶车辆(connected and autonomous vehicle,CAV)构成的新型混合交通流的运行规律,研究不同管理车道设置策略对城市快速路新型混合交通流产生的影响。首先,基于不同种类车辆间跟驰与专用道选择概率间的耦合关系,定量描述了不同管理车道设置策略条件下快速路通行能力演变机理。基于此,利用SUMO仿真平台分析了新型混合交通流条件下车辆延误的变化规律。最后,通过对车辆换道类型与换道间隙分析,研究了不同管理车道设置策略对交通流车辆间碰撞风险的影响。结果表明:CAV渗透率低于30%或大于80%,且只限制HDV在普通车道通行时,通行能力最大;CAV渗透率介于30%~80%之间,应考虑设置公交和CAV专用车道;设置公交和CAV专用车道但不限制其通行路权时,路段平均延误最小且几乎不受CAV渗透率的影响;当只为CAV或多乘员车辆(high-occupancy vehicle,HOV)设置管理车道时,会增加车辆碰撞风险。这说明CAV渗透率是建立合理的管理车道设置策略的重要参考因素,CAV渗透率对设置管理车道路段的通行能力有很大影响,而路段平均延误和交通流车辆间碰撞风险则更受管理车道设置策略的影响。

基于大数据技术的城市智慧交通体系建设

建设基于大数据技术的城市智慧交通体系,采集设备感知交通大数据,数据预处理后,通过5G+网络上传云服务处理中心进行标准化数据建模、整合分析,根据不同交通参与者需求进行数据挖掘、分析,构建个性化模型,推送智慧服务方案。通过探讨大数据技术下智慧交通系统架构、实施流程、服务应用场景、仿真应用,助力提升城市交通质量和公众出行体验,推动城市智慧交通健康可持续发展。