Adaptive neural control of MIMO stochastic systems with unknown high-frequency gains

This paper addresses the control problem of MIMO stochastic nonlinear systems with unknown high-frequency gains. In the existing works, the prior knowledge of the high-frequency gain or its control direction is assumed known for the control design. This paper removes such assumptions, and proposes an adaptive neural network algorithm that allows the high-frequency gains to be time-varying and their control directions to be unknown. Nussbaum gain based approach and adaptive neural network mechanism are brought together such that all the signals in the closed-loop system are ensured bounded. A simulation study is carried out to confirm the validity of the proposed algorithm.