A general control theory for uncertain systems with various objectives

A study conducted by Yuji Ito et al. was published in the IEEE Transactions on Cybernetics.

It is challenging to control systems, such as nanoscale devices, that are prone to uncertainty owing to individual differences and environmental changes. Designing control laws accounting for the uncertainty improves performance of the systems for desired control indices. In this study, we proposed a theory for designing optimal control laws for various performance indices; however, limited indices could be handled using existing methods. By deriving a generalized relationship between the parameters of a control law and a performance index, the parameters could be optimized for various performance indices, improving the versatility of conventional optimal control. This proposed concept serves as a fundamental theory for solving problems caused by uncertainty in complex systems, such as control applications for nanoscale devices.

Title: Stochastic Optimal Linear Control for Generalized Cost Functions With Time-Invariant Stochastic Parameters
Authors: Ito, Y., Fujimoto, K., Tadokoro, Y.
Journal Name: IEEE Transactions on Cybernetics
Published: March 21, 2023
https://doi.org/10.1109/TCYB.2023.3252673