A five degree-of-freedom body-machine interface for children with severe motor impairments

Abstract

Children with severe motor impairments require the use of assistive devices to perform activities of daily living. Brain-machine interfaces are not suited for children due to various factors such as surgical risks. In this paper, we present a non-invasive body-machine interface where upper body movements are recorded by wireless inertial measurement units (IMUs) and used to control a robot arm. We develop a novel approach, called the virtual body model (VBM), which allows for control of high number of degrees-of-freedom (DOF). Our results show that participants could use the VBM to control up to five DOFs of a robot arm, and perform a pick-and-place task. Even with minimal training, trajectories of the end effector were smooth and positioning was accurate. These results show the potential of this safe, non-invasive approach to control high DOFs in children.