No label defined (Q4367152)

The authors consider a nonlinear system described as \[ x=A \psi(x)+ B \varphi (u) +C\xi (x,u), \quad x(t_0) =x_0, \] where \(x\in R^n\) is the system state vector, \(u\in R^m\) is the control input vector. The problem is to estimate adaptively the system parameter matrices \(A,B\) and \(C\) without explicit use of the derivatives of the accessible signals. The idea consists in using the modified Lyapunov function candidate \[ V= e^TPe+ \text{tr} (\Phi^T P \Phi+ \Psi^T P\Psi), \] where \(P\) is the solution of the Lyapunov equation. Then \(P\) is excluded from estimation algorithms and it is not necessary to solve the Lyapunov equation. This feature of the algorithm is attractive, especially when the system is of high dimension. Two modifications of the algorithm are presented. A numerical example for a simple nonlinear system is considered.