BLF-based neuroadaptive control for full state constrained nonlinear systems with input saturation: A feasibility-condition-free approach

For nonlinear systems under full state constraints, current barrier Lyapunov function (BLF)-based control methods depend on feasibility conditions. In this paper, we propose a novel BLF-based adaptive neural network control method to deal with full state constrained nonlinear systems without feasibility conditions, both unknown external disturbances and input saturation are considered in the design process. Firstly, BLFs are introduced to guarantee the tracking errors obviate the symmetric constraints. Then, in order to obtain the nominal control signals within the predefined asymmetric constraints, hyperbolic tangent functions are employed to limit the virtual control signals, which is the key to completely circumvent feasibility conditions. Furthermore, neural networks are used to approximate the unknown functions and the control signal errors. Finally, Both Lyapunov stability theory-based analysis and simulation verify the effectiveness of the proposed control method. The proposed BLF-based neuroadaptive control strategy facilitates more extensive application to practical engineering.