Experimental validation of a time scaling algorithm for robotics systems

Saturation phenomena, that characterize physical systems, worsen the performances of control loops and, in the worst cases, could also lead to instability. Possible problems can be avoided by devising appropriate feedback control systems or by means of trajectory scaling algorithms. The main advantage of these latter is that they do not require any controller adaptation. The paper describes an experimental validation, obtained by means of a two link planar manipulator, of a recently proposed scaling algorithm, which performances have been verified by considering modeling errors and an uncertain knowledge of the plant parameters.