An online trajectory planning method for visually guided assisted reaching through a rehabilitation robot

Several manipulators or exoskeleton are characterized by having a concave workspace in the operational space due to mechanical limits. This article proposes an online trajectory planning method for performing visually guided assisted reaching through a rehabilitation robotic exoskeleton, the L-Exos, in its concave workspace. To evaluate the proposed methodology in a rehabilitation application, we set-up a computer vision based system that can automatically identify target objects in the workspace and generate a robot assisted movement to reach them through the L-Exos.