Implementation of adaptive control algorithms in robot manipulators using parallel computing

Abstract

The dynamics equation of robot manipulators is non linear and coupled. An inverse dynamic control algorithm that requires a full knowledge of the dynamics of the system is one way to solve the control movement. Adaptive control is used to identify the unknown parameters (inertial parameters, mass, etc). The adaptive control algorithms are based on the linear relationship of inertial parameters in the dynamic equation. A formulation to generalize this relationship is applied to the Johansson adaptive algorithm. The objective of this paper is to present the implementation of this relationship using parallel computing and apply it to an on-line identification problem in real-time. © Springer-Verlag 2003.