Semiglobal exponential control of Euler-Lagrange systems using a sliding-mode disturbance observer

Abstract Sliding-mode disturbance observer (SMDOB) is appealing for its ability in the estimation of system uncertainties and disturbances without introducing serious chattering in the sliding mode technique. However, existing SMDOB-based controllers with stability guarantee were analyzed by only considering disturbances or under a strict assumption that the time derivatives of system uncertainties are bounded prior to control implementation. This paper proposes a SMDOB-based tracking control strategy for Euler–Lagrange systems with modeling uncertainties and external disturbances. The SMDOB is designed by the construction of an extended state observer embedded by an improved super-twisting algorithm. Compared with existing SMDOB-based controllers, the significant advantage of the proposed controller lies in semiglobal exponential stability guarantee for the combined observer–controller system without the assumption that the time derivatives of modeling uncertainties are bounded by constants. The chattering caused by switching signals is alleviated by passing the switching signals through a low-pass filter in the SMDOB and is compressed by the frequency bandwidths of both the SMDOB and the control system. The effectiveness of the designed SMDOB-based controller is illustrated by simulations on a two-link robot manipulator.