Improved Haptic Linear Lines for Better Movement Accuracy in Upper Limb Rehabilitation

Abstract

Force feedback has proven to be beneficial in the domain of robot-assisted rehabilitation. According to the patients' personal needs, the generated forces may either be used to assist, support, or oppose their movements. In our current research project, we focus onto the upper limb training for MS (multiple sclerosis) and CVA (cerebrovascular accident) patients, in which a basic building block to implement many rehabilitation exercises was found. This building block is a haptic linear path: a second-order continuous path, defined by a list of points in space. Earlier, different attempts have been investigated to realize haptic linear paths. In order to have a good training quality, it is important that the haptic simulation is continuous up to the second derivative while the patient is enforced to follow the path tightly, even when low or no guiding forces are provided. In this paper, we describe our best solution to these haptic linear paths, discuss the weaknesses found in practice, and propose and validate an improvement.