Construction and initialization of an adaptive neuro-fuzzy control based on the SDRE technique for a two-link manipulator

In modern control theory, one of the open problems is the construction of adaptive control for nonlinear systems with parametric uncertainty and the analysis of the stability of the corresponding closed system. One of the approaches that can take into account the nonlinearity and uncertainty of the control object is fuzzy logic. Affine systems are a class of nonlinear systems whose representatives are often found in various practical problems. For this class, there are a number of developed methods for the synthesis of regulators, in particular, a method based on the Riccati equation with state-dependent coefficients. In this paper, for a given class of nonlinear systems, the adaptation mechanism of a neuro fuzzy controller approximating the control obtained using the SDRE approach is applied for the first time. The main results of the work are the architecture of the neuro fuzzy network, as well as methods of its initialization. The proposed approach is applied to the model of a two-link manipulator with uncertain coefficients. Numerical experiments have shown the effectiveness of the obtained control according to the totality of the quality criteria considered.