Model Predictive Control Applied to Different Time-Scale Dynamics of Flexible Joint Robots

Modern Lightweight robots are constructed to be collaborative, which often results in a low structural stiffness compared conventional rigid robots. Therefore, the controller must able handle dynamic oscillatory effect mainly due intrinsic joint elasticity. Singular perturbation theory makes it possible decompose flexible dynamics into fast and slow subsystems. This model separation provides additional features incorporate future knowledge of joint-level dynamical behavior within design using Model Predictive Control (MPC) technique. In this study, different architectures considered that combine method Perturbation MPC. For Perturbation, parameters influence validity technique control flexible-joint robot investigated. Furthermore, limits on input constraints for trajectory with The position performance robustness against external forces each architecture validated experimentally robot. experimental validation shows superior practice presented MPC framework, especially respecting actuator torque limits.