CPS: Synergy: Collaborative Research: Design and Control of High-performance Provably-safe Autonomy-enabled Dynamic Transportation Networks

CPS：协同：协作研究：高性能、可证明安全、支持自主的动态运输网络的设计和控制

• 批准号: 1544578
• 负责人: Zhi-Hong Mao
• 金额: $ 21.31万
• 依托单位: University of Pittsburgh
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2015
• 资助国家: 美国
• 起止时间: 2015-09-15 至 2019-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1544578&HistoricalAwards=false
• 关键词: CPS; Synergy; Collaborative; Research; Design

During the last decade, we have witnessed a rapid penetration of autonomous systems technology into aerial, road, underwater, and sea vehicles. The autonomy assumed by these vehicles holds the potential to increase performance significantly, for instance, by reducing delays and increasing capacity, while enhancing safety, in a number of transportation systems. However, to exploit the full potential of these autonomy-enabled transportation systems, we must rethink transportation networks and control algorithms that coordinate autonomous vehicles operating on such networks. This project focuses on the design and operation of autonomy-enabled transportation networks that provide provable guarantees on achieving high performance and maintaining safety at all times. The foundational problems arising in this domain involve taking into account the physics governing the vehicles in order to coordinate them using cyber means. This research effort aims to advance the science of cyber-physical systems by following a unique and radical approach, drawing inspiration and techniques from non-equilibrium statistical mechanics and self-organizing systems, and blending this inspiration with the foundational tools of queueing theory, control theory, and optimization. This approach may allow orders of magnitude improvement in the servicing capabilities of various transportation networks for moving goods or people. The applications include the automation of warehouses, factory floors, sea ports, aircraft carrier decks, transportation networks involving driverless cars, drone-enabled delivery networks, air traffic management, and military logistics networks. The project also aims to start a new wave of classes and tutorials that will create trained engineers and a research community in the area of safe and efficient transportation networks enabled by autonomous cyber-physical systems.

在过去的十年中，我们见证了自主系统技术迅速渗透到空中，公路，水下和海上车辆中。这些车辆所具有的自主性具有显著提高性能的潜力，例如，在许多运输系统中，通过减少延误和增加容量，同时提高安全性。然而，为了充分利用这些自动驾驶交通系统的潜力，我们必须重新思考交通网络和控制算法，以协调在这些网络上运行的自动驾驶汽车。该项目的重点是设计和运营安全的交通网络，为实现高性能和始终保持安全提供可证明的保证。在这一领域出现的基本问题涉及到考虑到管理车辆的物理学，以便使用网络手段协调它们。这项研究工作的目的是通过以下独特和激进的方法，从非平衡统计力学和自组织系统中汲取灵感和技术，并将这种灵感与控制理论，控制理论和优化的基础工具相结合，来推进网络物理系统科学。这种方法可以允许用于移动货物或人员的各种运输网络的服务能力的数量级改进。这些应用包括仓库、工厂车间、海港、航空母舰甲板、涉及无人驾驶汽车的运输网络、无人机支持的交付网络、空中交通管理和军事物流网络的自动化。该项目还旨在启动新一轮课程和教程，培养训练有素的工程师和研究社区，以实现自主网络物理系统所支持的安全高效的交通网络。