Gravitational Base Parameters Identification for a Knee Rehabilitation Parallel Robot

New robotic technology is emerging nowadays to tackle lower limb rehabilitation issues. However, the commercial robots available for are usually oversize and expensive. Knee generally aided by a professional therapist, making this clinical procedure an interesting scope robotics. Parallel suitable candidates knee due their high load capacity, stiffness, accuracy compared serial ones. In contrast, robot has singular configurations inside its workspace, dynamic model complex. For these reasons, parallel needs advanced control unit solve complex mathematical problems that ensure patient security. This study proposes base parameters identification of compact gravitational linear 3UPS+RPU using value decomposition. paper recommends adding statistical method focused on condition number minimization decomposition process. reduces computational resources taken searching best inertial combination at beginning parameter identification. The identified have physical meaning low complexity. fact makes results research basis adaptative applied robot. shows term most influential tasks, with inertial, Coriolis, centrifugal components, regarding behavior