Cartesian Controllers for Tracking of Robot Manipulators under Parametric Uncertainties

Abstract

The relevance of robot manipulators in different processes has created the need to design efficient controllers with low computational costs. Although several applications for this problem are defined in operational coordinates, a wide variety of controllers reported in the literature are defined in joint coordinates. Then, for a joint robot control the desired joint references are computed from desired Cartesian coordinates using inverse mappings and its derivatives up to second order. However, computing the inverse kinematics mappings is difficult due to the ill-posed nature of these mappings. To circumvent the computation of inverse kinematics, a very old but not less important approach coined as Cartesian control can be used. Cartesian control deals with the problem of designing controllers in terms of desired Cartesian or operational coordinates. This allows saving a significant amount of time in real time applications due to the inherent simplification.