As a standard model for a rigid multilink robotic manipulator, we consider a MIMO nonlinear system of uniform vector relative degree {2,.,2}, which has a globally defined normal form with no zero dynamics. We consider a trajectory-following problem for a class of smooth bounded time-varying vector reference signals. We design an output feedback integral controller that ensures ultimately bounded tracking error which could be made as small as required. In addition, if the reference signal has a constant limit, the output approaches this limit. The controller is robust with respect to parameter uncertainties, decentralized, saturated, and simple to implement. Locally, it is a PID regulator with derivatives estimated via a linear high-gain observer. Regional (and semiglobal) practical (and asymptotic) stability is shown with the help of a vector Lyapunov function and a technique, typical for "continuous" sliding mode control. Copyright (C) 2001 IFAC.
Universal integral controllers for robotic manipulators
2002-01-01
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
DDC: | 629 |
Auto-tuning of PID Controllers for Robotic Manipulators Using PSO and MOPSO
Springer Verlag | 2019
|Auto-tuning of PID Controllers for Robotic Manipulators Using PSO and MOPSO
BASE | 2019
|Auto-tuning of PID Controllers for Robotic Manipulators Using PSO and MOPSO
TIBKAT | 2019
|