Data-driven sliding mode control of shape memory alloy actuators with prescribed performance

This paper proposes a data-driven sliding mode control (SMC) method with tracking error constraint for a shape memory alloy (SMA) actuated system. More specifically, a new SMC method with prescribed performance is investigated by adopting the equivalent dynamic linearization technique and a novel transformed error algorithm to guarantee that the tracking error can converge into a preassigned region. Meanwhile, the proposed control method can effectively ensure the convergence rate and the steady-state error within a prescribed bound, merely depending on the input/output data without implicit or explicit information of the systems, which is more valuable for the practical application. The presented control method has been implemented on an SMA actuated system successfully and experimental results have shown that the proposed control is easily applicable and highly precise. In addition, the tracking error of the SMA actuated system can be constrained in a preassigned domain by the proposed control method. (paper)