A nonsingular predefined time sliding mode control method with continuously variable exponents for nonlinear systems based on emotional neural network

In this paper, we propose a predefined-time nonsingular sliding mode control (SMC) method based on emotional neural networks (ENN) for disturbed nonlinear systems. Firstly, based on the predefined-time stability (PDTS) theory, a new sufficient condition is constructed. By introducing a novel adaptive continuously adjustable exponential term into the Lyapunov stability lemma, this modification abandons the traditional fixed parameterization of exponential terms. Secondly, by introducing ENN to approximate the equivalent control term, a nonsingular SMC is designed, meanwhile solving the issue of unknown dynamics. Using Lyapunov theory, this paper rigorously proves the system's PDTS in both the reaching and sliding phases. Finally, simulations on the inverted pendulum and deep-sea vehicle manipulator show that they can track the target trajectory in 0.22 and 0.23 s respectively, outperforming comparison methods.