Robust Kalman Filter and Robust Regulator for Discrete-Time Markovian Jump Linear Systems: Control of Series Elastic Actuators

Abstract

In this paper we deal with control systems for human-robot interaction where force interaction between them is taken into account. Due to the nonlinear and uncertain nature of this interaction, we propose a unified framework to deal with estimation and force control based on discrete-time Markovian jump linear systems. Robust Kalman filter and robust regulator for this class of systems are considered. A Series Elastic Actuator (SEA)-based robotic platform for ankle rehabilitation is used in the evaluation of the proposed control methodology. Also, levels of participation of the user are considered in the interaction with the rehabilitation robotic platform. Preliminary results show that the proposed approach guarantees robust stability and robust performance in presence of parametric uncertainties and disturbances.