Chattering-free adaptive sliding-mode control of nonlinear systems with unknown disturbances

Chattering occurs in many sliding-mode control systems and consequently constrains their applications in practice. Though filtering technique has been proposed to suppress the chattering, it comes with the limitation that the information such as the upper bound and/or lower bound of the disturbances/uncertainties must be known a priori. In reality, this information may not be acquirable and thus the control system has to presume a sufficiently large upper bound of the unknown disturbances, which often results in unnecessarily large control gains that reversely amplify the measurement noise and deteriorate the control performance. To overcome this limitation, this paper proposes an adaptive sliding-mode control scheme that incorporates super-twisting algorithm, low-pass filter, and a leakage-type adaptive law to achieve the benefits of free chattering as well as removal of the need for any disturbance information. Lyapunov analysis and simulation results are presented to demonstrate the superiority of the proposed method.