Assistive Control System for Upper Limb Rehabilitation Robot

TLDR

The study develops an assistive control system for an upper‑limb rehabilitation robot that incorporates a unique kinematic structure and force/torque sensors. The system, named NTUH‑ARM, uses a dynamic human model with torque sensing to simulate interaction in active, assistive, and passive modes, maps the robot’s 7 degrees of freedom to the human arm’s 4 degrees, and employs Lyapunov theory to ensure controller stability, with preliminary clinical trials involving six patients. Clinical trials with six patients, including a control, yielded positive outcomes, and physical therapists’ STREAM assessments indicated promising results.

Abstract

This paper presents an assistive control system with a special kinematic structure of an upper limb rehabilitation robot embedded with force/torque sensors. A dynamic human model integrated with sensing torque is used to simulate human interaction under three rehabilitation modes: active mode, assistive mode, and passive mode. The hereby proposed rehabilitation robot, called NTUH-ARM, provides 7 degree-of- freedom (DOF) motion and runs subject to an inherent mapping between the 7 DOFs of the robot arm and the 4 DOFs of the human arm. The Lyapunov theory is used to analyze the stability of the proposed controller design. Clinical trials have been conducted with six patients, one of which acts as a control. The results of these experiments are positive and STREAM assessment by physical therapists also reveals promising results.