Learning Koopman Embedding Subspaces for System Identification and Optimal Control of a Wrist Rehabilitation Robot
Goyal T. Hussain S. Martinez-Marroquin E. Brown N.A.T. Jamwal P.K.
1 July 2023Institute of Electrical and Electronics Engineers Inc.
IEEE Transactions on Industrial Electronics
2023#70Issue 77092 - 7101 pp.
Rehabilitation robots have proven their usefulness in assisting with physical therapy. This article presents a trajectory tracking controller for a wrist rehabilitation robot with three degrees of freedom. The nonlinearity of the human-robot interaction dynamics has been defined as the Koopman linear system in terms of nonlinear observable functions of the state variables. Koopman operators are learned using linear regression to encode the states into object-centric embedding space for a linear approximation of a nonlinear dynamical system. The learned Koopman operators ascertain the system dynamics applied to design the wrist robots trajectory tracking task controller. This is a data-driven approach that yields an explicit control-oriented model. The efficiency and feasibility of the controller were evaluated through experiments with three healthy human subjects. The experiments demonstrated the ability of the controller to guide the subjects wrist along the reference trajectory.
Koopman operator , nonlinear control , parallel robot , system identification , wrist rehabilitation robot
Text of the article Перейти на текст статьи
University of Canberra, School of Information Technology and Systems, Bruce, 2617, ACT, Australia
University of Canberra, Faculty of Health, Canberra, 2617, ACT, Australia
Nazarbayev University, Department of Electrical and Computer Engineering, Astana, 010000, Kazakhstan
University of Canberra
University of Canberra
Nazarbayev University
10 лет помогаем публиковать статьи Международный издатель
Книга Публикация научной статьи Волощук 2026 Book Publication of a scientific article 2026