Functional Recovery of the Upper Limb With an Active Personal Device

A new personal device to assist the upper limb capable to be mounted on a wheelchair is being tested. The robot is equipped with three brushless motors powered by four electronic boards appositely designed and communicating via I²C protocol; one board works as master for the other three, which have simpler tasks. Most of the driving software has been developed with Matlab and mainly translated into C++ for memory space and boards’ efficiency matters. The system’s end-effector is connected to the subject’s forearm and can cooperate to the arm motion in several different ways. In order to avoid the overstress of the natural joints no further connections are made to the upper limb. The working volume of the limb connected to the device allows the execution of the trajectories required for most of daily living activities. It is addressed to post stroke rehabilitation and to the self-treatment of other patients with serious deficiency of arm forces, like individuals affected by muscular dystrophy. Some working modes exploit the mechanical impedance control to gently interfere with the residual natural motion capability of the subjects. The very preliminary tests of the prototype fitted on a power wheelchair are encouraging: it is light, not too noisy and easy to move for the subject. The first working mode, with full arm gravity compensation, and the second working mode, with partial compensation, have been implemented and are currently being tested with patients, as well as the mode in which the subject is helped to repeat a stereotyped exercise for self-physiotherapy. Beside the fact the control system must be calibrated on patient characteristics, for these initial tasks the system reveals to be user-friendly. Other working modes require to interpret the patient intention to move the upper limb. For a natural approach it is sufficient to decode the movement intention of the patient and the final position he wants the hand to reach, whilst the rest of the limb can be automatically positioned by the system. Different approaches to solve the problem have been experimentally tested, including the use of a headgear with a brain interface. To present knowledge the best results have been obtained by monitoring the movement of another body segment such as the head. The device designed for the functional recovery of upper limb can furthermore be used to monitor and easily certify the evolution of the patient conditions.