Network intelligence deals with situations where a group of intelligent agents such as autonomous vehicles (AVs) need to coordinate, cooperatively plan, and execute a task. This concept could be broken down into two main interdependent categories including situational awareness and decision-making. Accurate “Situational Awareness” is a key component for reliable and safe group decision-making. Relying on only onboard sensor suites of individual agents is not sufficient for this purpose, due to its inherent restrictions, like limited detection range and non-line-of-sight limitations. Therefore, to have a safe AV design having a framework that facilitates the collaboration among multiple connected AVs (CAVs) and jointly build up their scene understanding seems to be essential. Our research on this domain is specifically focused on proposing a novel framework by utilizing the capabilities of advanced wireless communication and edge-computing technologies. Additionally, our framework is capable of risk assessment and quantification that enables a clear characterization of the collaboration among CAVs and its impact on collision risk reduction and uncertainty. We further investigate the effectiveness of the proposed approach by utilizing real-world data and postprocessing analysis. On the other hand, for cooperative decision-making, we propose an optimal design solution combined with the social compliance concept to minimize maneuver time and associated energy consumption while aiming to interrupt traffic flow as minimally as possible.
Ehsan Moradi-Pari, Ph.D. is a chief scientist and Network Intelligence lead scientist at Honda Research Institute USA, Inc. in Ann Arbor, Michigan. His research centers on cooperative AI and human-AI teamwork. Dr. Moradi-Pari serves as Honda’s lead and representative for pre-competitive active safety research conducted at Crash Avoidance Metrics Partnership (CAMP) and the 5G Automotive Association (5GAA). He has spearheaded Honda’s efforts and preparations to accelerate connected and automated vehicle deployment in the United States, including the design and deployment of the first-of-its-kind smart intersection concept in Marysville, Ohio and 5G V2X safety and mobility applications for connected and automated vehicles at the University of Michigan Mcity. In addition to his role at Honda, Dr. Moradi-Pari serves on the University of Michigan Center for Connected and Automated Transportation (CCAT) Technology Advisory Board.
His work in the area of designing communication protocols and applications for multi-agent systems including connected and automated vehicles as well as modeling and analysis of intelligent systems has been published in several international journals, conference proceedings, and book chapters.
Dr. Moradi-Pari received his doctorate degree from West Virginia University and Master of Science from Sharif University of Technology.