A Luenberger-style Observer for Robot Manipulators with Position Measurements

This paper presents a novel Luenberger-style observer for robot manipulators with position measurements. Under the assumption that the state evolutions that are to be observed have bounded velocities, it is shown that the origin of the observation error dynamics is globally exponentially stable and that the corresponding convergence rate can be made arbitrarily high by increasing a gain of the observer. Comparisons and relations between the proposed observer and existing observers are discussed. The effectiveness of the result here presented is illustrated by a simulation of the observer for the Pendubot, an underactuated two-joint manipulator.