Distributed Vehicular Traffic Management via DSRC-Enabled Vehicles

在这项研究中，我们利用车辆组成的具有健壮的无线通信设备、存储、处理和感知能力的自组织网络来执行健壮的交通状态估计和分布式交通管理。首先，我们将利用车辆的感知和计算能力以及自组织的网格计算引擎来开发稳健的估计和控制算法，以畅通高速公路上的车辆交通。通过仿真和分析，我们将考察这些方案的有效性，目标是减少事故，最大限度地减少拥塞延迟和最大化吞吐量。我们还将调查使这一制度有效所需的渗透程度。其次，作为这项研究的一部分，我们将开发软件体系结构、网络协议和资源管理算法，以创建网格计算引擎VGrid，并将其与路边传感器基础设施集成。由于自组织网格计算机的动态特性，新的挑战出现了，因为拓扑和节点成员都随着时间而改变。第三，我们将开发一个集成的仿真工具，该工具既具有逼真的车辆机动性模型，又具有捕获无线信道动态的通信/网络层。利用该仿真工具，我们将研究感知-计算-控制混合系统的性能特征，并确定设计和建模问题，以提高此类系统的性能。我们提出的研究为管理交通流量和提高交通安全提供了一种替代的基础设施和新的方法。这里开发的分布式动态传感和控制体系结构将使在任何存在VGrid类型的车辆的地方无处不在地部署交通安全、安保和管理措施成为现实，这减轻了对昂贵的固定基础设施的依赖，并且与传统的集中式智能交通系统相比，具有加快响应时间的潜力。此外，了解车载自组织网络和覆盖网格计算平台的特点将有助于为其他应用程序(如车辆碰撞避免、紧急疏散和灾难恢复)开发通用框架。这项拟议的研究有重大的更广泛的影响。首先，通过提供替代基础设施和管理交通流量和加强交通安全的新方法，可以预期通过增加机动性和拯救人类生命的形式带来重大的社会效益。其次，由于这个项目研究的是车辆和信息交通交叉口的问题，它为参与其中的学生提供了跨学科的培训。最后，通过一个研究车辆自组织网络的科学项目，提高了当地高中生的教育体验。

In this research, we exploit the ad hoc networks formed by vehicles equipped with robust wireless communication devices, storage, processing, and sensing capability to perform robust traffic state estimation and distributed traffic management. First, we will utilize the sensing and computation capabilities of vehicles and the self-organized grid computing engine to develop robust estimation and control algorithms to smooth vehicular traffic flow on freeways. Through simulation and analysis we will investigate the effectiveness of these schemes with the goal to reduce accidents, minimize congestion delays and maximize throughput. We will also investigate the required degree of penetration to make such a system effective. Second, as part of this research, we will develop the software architecture, the networking protocols, and the resource management algorithms to create the grid computing engine, VGrid, and integrate it with the roadside sensor infrastructure. New challenges arise due to the dynamic nature of the ad hoc grid computer as both the topology and the node membership change with time. Third, we will develop an integrated simulation tool that has both a realistic vehicular mobility model and communication/networking layers that capture the dynamics of wireless channels. Using this simulation tool, we will investigate the performance characteristics of a hybrid sensing-computing-control system, and identify design and modeling issues to improve the performance of such a system.Our proposed research provides both an alternative infrastructure and new ways to manage traffic flow and enhance traffic safety. The distributed dynamic sensing and control architecture developed here would make ubiquitous deployment of traffic safety, security and management measures a reality wherever there are VGrid types of vehicles, which alleviates the reliance on expensive fixed infrastructure and has the potential to speed up the response time compared to traditional centralized intelligent transportation systems. Moreover, understanding the characteristics of vehicular ad hoc networks and the overlay grid computing platform will aid the development of a general framework for other applications such as vehicular collision avoidance, emergency evacuation, and disaster recovery. There are significant broader impacts of this proposed research. First, by providing both an alternative infrastructure and new ways to manage traffic flow and enhance traffic safety, significant societal benefits can be expected in the form of increased mobility and saved human lives. Second, since this project examines problems at the intersection of vehicular and information traffic, it provides interdisciplinary training to the students involved. Lastly, it enhances the educational experience of local high school students through a science project studying vehicular ad hoc networks.