Adaptive control of a 7-DOF exoskeleton robot with uncertainties on kinematics and dynamics. (July 2018)

• Record Type: Journal Article
• Authors: Brahmi, Brahim; Saad, Maarouf; Lam, Jacqueline Tu Anh Thu; Luna, Cristobal Ochoa; Archambault, Philippe S.; Rahman, Mohammad H.
• Is Part Of: European journal of control. Volume 42(2018)
• Page Start: 77
• Page End: 87
• Publication Date: 2018-07
• Subjects: Rehabilitation robots -- Time delay control -- Time delay error -- Uncertainties, recursive control; Control theory -- Periodicals; Automatic control -- Periodicals; Automatic control -- Mathematics -- Periodicals; Electronic journals; 629.805
• DOI: 10.1016/j.ejcon.2018.03.002
• Languages: English
• ISSNs: 0947-3580

Abstract

Highlights: New adaptive visual tracking control for an exoskeleton robot (7-DOFs) is introduced. An estimation of the nonlinear kinematics and dynamics uncertainties of an exoskeleton robot is provided. Passive functional therapy treatment. Novel control scheme based on sliding mode Control incorporated with Time Delay Estimation and recursive control. Experiments results and comparison procedure prove the feasibility of the designed strategy. Abstract: In this paper, we propose a new adaptive visual tracking control approach based on sliding mode control in Cartesian space applied to an exoskeleton robot with uncertain kinematics and dynamics, taking into account uncertainties in visual system (camera) parameters. The adaptation of kinematic uncertainties is based on a filtered regressor kinematic matrix, whereas, the adaptation of dynamic uncertainties is based on a Time Delay Estimation approach. This is performed considering the Time Delay Error (TDR) to provide a control action capable of following the designed functional therapy tasks. A new recursive controller is combined with TDE in order to estimate the TDR and limit its effect. The proposed strategy does not need the accurate dynamic and kinematic models of the exoskeleton. The update laws are designed using Lyapunov theory to solve the adaptation problem methodically and to show the stability of the robot system. Experimental results confirm the effectiveness and feasibility of the designed approach.