CAREER: Peer-to-Peer Models and Mechanisms for the Next Generation of Transportation Systems

这项教师早期职业发展（Career）资助将探索连通性和自主性可能塑造下一代交通系统的方式。这些技术突破为旅行者、车辆和交通基础设施之间的通信、互动和自动协商打开了大门，使许多创新应用成为可能，其中价值是通过同行之间的互动创造的。该项目对几个新兴的点对点应用的理论和实践都有贡献，包括：1)共享多式联运系统，按需共享的移动系统可以与传统的交通方式（如过境）集成，提供无缝的门到门移动解决方案；2)队列化（Platooning），即车辆运动的同步化，使一组自动驾驶车辆在距离较近的地方行驶，从而提高燃油效率和道路容量；3)电动汽车之间的车对车无线电力传输，其中电动汽车可以相互配对以交换电力。这笔拨款所产生的模型和算法可以彻底改变（不久）未来智慧城市中人员和货物的移动方式，并提供更高效、更实惠、更可持续和更公平的移动选择。为了应对点对点系统运行中出现的挑战，该项目在算法设计（数学规划和大规模优化）和机制设计（拍卖理论和经济学）领域提供了几个主要的智力贡献。为了充分发挥智慧城市、自动驾驶和互联交通系统的潜力，子系统的规划和优化不能再孤立地进行；相反，它们需要作为一个集成系统同时进行优化。为了解决支持未来点对点系统实际操作的新兴大规模规划和动态匹配问题，该NSF拨款处理图论中的开放问题，并设计了一套新的专门方法（即分解和聚合/分解方案，利用点对点系统的独特结构），以便根据需要高效且快速地获得精确或有界的解决方案，这通常意味着实时。此外，这项拨款为P2P系统中的定价和资源交换设计了新的机制。这些机制将是个别理性的、激励相容的和预算平衡的，并将解决运输系统的响应能力和吞吐量之间的潜在权衡，并缩小在获得运输选择方面的公平差距。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

This Faculty Early Career Development (CAREER) grant will explore the ways in which connectivity and autonomy may shape the next generation of transportation systems. These technological breakthroughs open doors for communication, interaction, and automated negotiation among travelers, vehicles, and the transportation infrastructure, enabling many innovative applications in which value is created from interactions among peers. This project contributes to both theory and practice in several emerging peer-to-peer applications, including: 1) Shared multi-modal transportation systems in which on-demand shared mobility systems can be integrated with traditional transportation modes (such as transit) to provide seamless door-to-door mobility solutions; 2) Platooning, wherein synchronization of vehicle motion allows a group of automated vehicles to travel together at close proximity, thereby increasing fuel efficiency as well as road capacity; and 3) Vehicle-to-vehicle wireless power transfer between electric vehicles, wherein electric vehicles may pair with one another to exchange electric power. The resulting models and algorithms from this grant can revolutionize the way people and goods move in smart cities of the (near) future, and offer more efficient, affordable, sustainable, and equitable mobility options.In response to challenges that arise in operationalizing peer-to-peer systems, this project offers several major intellectual contributions within the realms of algorithm design (mathematical programming and large-scale optimization) and mechanism design (auction theory and economics). To achieve the full potential of what smart cities, autonomy, and connected transportation systems have to offer, planning and optimization of sub-systems can no longer occur in isolation; instead, they need to be optimized simultaneously as one integrated system. To address the emerging massive-scale planning and dynamic matching problems that support the real-life operation of future peer-to-peer systems, this NSF grant tackles open problems in graph theory, and designs a suite of new specialized methods (namely decomposition and aggregation/disaggregation schemes that utilize the distinct structure of peer-to-peer systems) to obtain exact or bounded solutions efficiently and as quickly as needed, which often implies in real-time. In addition, this grant designs novel mechanisms for pricing and resource exchange in P2P systems. These mechanisms will be individually rational, incentive compatible, and budget balanced, and will resolve the potential trade-off between responsiveness and throughput in transportation systems, and to narrow down the equity gap in access to transportation options.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.