Solving for a Class of Linear Matrix Inequalities Using Neural Networks

Abstract

This paper proposes a new approach solving for a class of LMIs, which are commonly encountered in the robust control system analysis and design, using recurrent neural network. The nature of parallel and distributed neural processing renders these networks possessing the computational advantages over the traditional sequential algorithms in real-time applications. The proposed networks are proven to be asymptotically in the large and capable of LMIs solving. Illustrative examples are provided to demonstrate the proposed results.