Learning Feedback Linearization Control Without Torque Measurements

Feedback Linearization (FL) allows the best control performance in executing a desired motion task when an accurate dynamic model of a fully actuated robot is available. However, due to residual parametric uncertainties and unmodeled dynamic effects, a complete cancellation of the nonlinear dynamics by feedback is hardly achieved in practice. In this paper, we summarize a novel learning framework aimed at improving online the torque correction necessary for obtaining perfect cancellation with a FL controller, using only joint position measurements. We extend then this framework to the class of underactuated robots controlled by Partial Feedback Linearization (PFL), where we simultaneously learn a feasible trajectory satisfying the boundary conditions on the desired motion while improving the associated tracking performance.