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

Abstract:

As autonomous systems technologies penetrate into aerial, road, underwater, and sea vehicles, it becomes possible to coordinate their actions to substantially increase their performance while their enhancing safety, by embedding them into an intelligent network of vehicles. The potential opportunities were not gone unnoticed. For example, the competition to robust build drone delivery systems that has already attracted tech giants like Amazon and Google. Autonomous vehicles and vehicle sharing technologies may offer efficient and safe transportation infrastructure in the future. However, a deep understanding of the fundamental limits of and practical algorithm for autonomy-enabled transportation networks is essential for the large-scale deployment of these networks.

Intellectual Merit: The proposed research 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. The proposed research effort aims to advance the science of cyber-physical systems by following a unique and radical approach, drawing inspiration and techniques from nonequilibrium statistical mechanics and self-organizing systems, and blending this inspiration with the foundational tools of queueing theory, control theory, and optimization. The proposed research addresses the Science of Cyber-Physical Systems target area indicated in the solicitation.

Broader Impacts: 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.