March 12, 2020
Takashi Tanaka, an assistant professor in the Department of Aerospace Engineering and Engineering Mechanics at the Cockrell School of Engineering, has been selected to receive a National Science Foundation (NSF) Faculty Early Career Development Program (CAREER) award for 2020. Over the next five years Tanaka will use the award to help solve the technological challenges that exist within Artificial Intelligence-based Networked Control Systems (NCS).
The prestigious NSF CAREER Program offers awards to junior faculty who exemplify the role of teacher-scholars through outstanding research, excellent education and the integration of education and research within the context of their organization’s mission. Tanaka was awarded for his project “Directed Information Theory for Networked Control Systems in Big Data Regime.”
Safe and efficient operations of large-scale infrastructure that support our society today, such as transportation networks and power systems, require an intelligent coordination of multiple decision-makers over real-time communication networks. Over the decades, the Networked Control Systems (NCSs) theory has been providing a mathematical foundation and design principles for such systems. In recent years, the advancement of Artificial Intelligence (AI) is changing the technological landscape around the NCS theory. While the AI-powered control algorithms are greatly enhancing the capacity of NCSs, the increased demand for AI has brought a set of challenges that are unprecedented in the conventional NCS theory.
Tanaka and his team will use this grant to take on what they consider to be the top three challenges within AI-based NCSs: incorporating extremely large data sets within the existing network; accommodating multiple users within limited network resources; and developing these systems in a way that ensures security and privacy.
Tanaka’s team plans to solve these challenges by applying a novel algorithm for information flow optimization—which Tanaka calls Optimal Information Flow (OIF) synthesis—using the concept known as directed information. The method is able to compute desired information flow over the communication network (what information should be shared with whom, with what accuracy and when) to accomplish given control tasks in real time.
An example of using this new approach to meet the challenge of supporting large amounts of data can be illustrated with connected cars. Due to the growing number of autonomous vehicles and the increasing data size that individual vehicles generate, it is becoming commonplace for the data rate to exceed the shared wireless networks which could eventually cause these networks to crash. Tanaka’s team proposes that their new approach could solve this issue while utilizing the communication resources more efficiently. The proposed method for information flow optimization is also expected to improve state-of-the-art methods to enhance security and privacy issues in NCSs in the big data regime.
Tanaka’s novel technique could result in improved efficiency, security and privacy within AI-based Networked Control Systems. The research project also includes innovative educational outreach activities and the development of a graduate level NCS curriculum.
Tanaka, who joined the department as an assistant professor in 2017, is also a winner of the AFOSR Young Investigator Program Award (2019) and the DARPA Young Faculty Award (2019). He specializes in stochastic and non-stochastic optimal control, robust control, distributed and networked control, optimization and game theory. He is an affiliated member of the Oden Institute. Learn more about Tanaka’s work on his research website.
