Fixed-Time Control of a Fractional-Order Brain-Computer Interfaces System with AI-Approximated Nonlinearities and Mobile Neuro-Sensing Units

This paper is concerned with the fixed-time control of a fractional-order neurodynamic model for brain-computer interfaces. The model features unknown nonlinearities and bounded perturbations. A distributed control framework is proposed using mobile neuro-sensing and stimulation units to enhance control efficiency. Radial basis function neural networks (RBFNNs) are employed to approximate unknown nonlinearities. By constructing operator-dependent Lyapunov functional, a sufficient condition which contains the velocity law of each mobile unit can be derived to ensure that it is uniformly ultimately bounded in fixed time. Numerical simulations verified the improved performance of the controlled system.