LAUSR

This letter reports a novel controller for robot teleoperation systems in the task-space. The local and the remote robots are kinematically and dynamically different and they are modeled as Euler-Lagrange agents. We consider the realistic scenario where the robot actuators are not ideal and thus they are prone to saturation. Moreover, velocity measurements are not available and variable time-delays arise in the communications. The human operator and the remote environment are assumed to be passive. The controller is dynamical and it consists of a gravity cancellation plus a plant-controller interconnection term. The controller dynamics is of second-order and damping is injected to ensure convergence. Unit-quaternions are used to obtain a singularity-free representation of the orientation. When the human and the environment forces are zero, then we prove that the pose of both robots converges to a common pose. Experimental results of the proposed scheme are provided to illustrate the controller performance.