Control of uncertain non-affine nonlinear systems using neural networks subject to input saturation with unknown control direction

In this paper, based on the implicit function theorem and mean value theorem, a novel neural network controller for trajectory tracking of uncertain non-affine nonlinear systems with input saturation, unknown control direction, and external disturbance is designed. To compensate for actuator saturation, the controller employs an auxiliary system and a modified tracking error. Radial basis function neural networks are employed to approximate uncertainties within the system dynamics. A Nussbaum-type function tackles the challenge of unknown control direction. Adaptive control techniques are implemented to handle actuator saturation and compensate for neural network approximation errors and disturbance. For output feedback control where some states are unavailable, a high-gain observer is utilized for state estimation. Lyapunov analysis guarantees asymptotic convergence of closed-loop error signals. The effectiveness of the proposed approach is validated through simulations.