High-gain PI-like control of Euler–Lagrange mechanical systems: Simulations and experiments in 2DoF robotic manipulators

Abstract

Euler–Lagrange mechanical systems usually possess nonlinear models, which are subject to uncertainties and unknown disturbances. In this article, a robust but smooth controller is proposed, relying on a high-gain proportional–integral structure that renders an accurate tracking, even in the case of uncertain effects. The tracking performance can be improved by adjusting a single parameter, deriving into a prescribed stability regime, where the tracking errors are locally uniformly ultimately bounded, and where the controller depends on a change of coordinates that integrates the proportional, integral, and derivative information of the tracking error. The advantages of this proposal become evident without considering an intricate implementation procedure. Numerical results are provided to highlight the reliability of the proposed method.