(Jan 17) Control and Estimation with Limited Information: A Game-Theoretic Approach

Modern control systems can be viewed as interconnections of spatially distributed multiple subsystems, where the individual subsystems interact and share their information with each other through an underlying network. The network inherently introduces limitations on information flow, which are induced by structural (topological) and/or communication constraints in the network. Hence, in order to design optimal control and estimation mechanisms for modern control systems, we must answer the following two practical but important questions:

1) What are the fundamental communication limits to achieve the desired control performance and stability?

2) What are the efficient approaches one has to adopt to design a distributed optimal controller for large-scale complex systems to deal with structural constraints?

This talk consists of two main parts. The first part will cover the fundamental communication limits in worst-case estimation, and considers minimax estimation with intermittent observations, where the estimator receives the sensor measurements only sporadically, with the availability governed by a random switching. In the second part of the talk, we study risk-sensitive mean field games under stochastic uncertainty to address structural constraints in large-scale complex systems. Conclusions and suggestions for future work will be drawn accordingly.