Disturbance observer-based LPV feedback control of a <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" display="inline" id="d1e1738" altimg="si2.svg"><mml:mi>N</mml:mi></mml:math>-DoF robotic manipulator including compliance through gain shifting

This paper proposes a control scheme for N-DoF robotic manipulator in joint-regulation motion problem, dealing with disturbances (as e.g. exogenous forces, unmodelled dynamics) that hinder task fulfilment, and also considering not all the required states are available online. Existing literature tackles this problem through Disturbance Observer (DO) strategies which imply complex analysis design methods or introducing strong assumptions. Conversely, we propose to formulate system as Linear Parameter Varying (LPV) model, allows straightforward application of existing linear structures but without neglecting its non-linear behaviour. We make use Robust Unknown Input (RUIO) obtain (for measurable states) decoupled estimation from unknown disturbance effects, improve noise reduction capabilities new optimal RUIO design. The is controlled state-feedback law that, making LPV paradigm, has been designed seamlessly avoid torque saturation on manipulator’s joints gain shifting strategy modifies compliant Stability performance requirements imposed both synthesis problems stated using LMI framework, applying Polya’s theorems positive forms standard simplex reduce overall conservatism. Experiments, simulated head TIAGo robot testbed various realistic scenarios, show benefits when compared joint-independent PD state-of-art EKF estimation.