Robot for Rehabilitation /Robot pour la réhabilitation

Dear Members,
I would like to inform you that our laboratory Lata is actually a partner of a new project concerning the application of robotics for rehabilitation. So, thos e who are interested and those who want to cooperate into such topics have to contact me.
Here is the description of the project:

RObotique pour la Rehabilitation, les mouvements Articulaires Simul\'es et l'ASsistance, RObots for Rehabilitation, humAn joint Simulation and ASsistance



Partners


CECCARELLI M., Laboratory di Robotica e Meccatronica, LARM (CU) Email : ##ERROR##.unicas.it
ROMDHANE L., Sousse Engineer School (SE), Email : lotfi.##ERROR##.rnu.tn
ZAATRI A., Mentouri University (MU), Email: ##ERROR##.com
MERLET J-P., INRIA (C), Email: Jean-Pierre.##ERROR##.fr

Scientific objectives
RORAS II is the natural follow-up of RORAS I that will focus on the robotics part with an extension of the application toward assistance robotics. The main objectives here are to develop and experiment robotized rehabilitation and assistance systems with the following objectives:


low intrusivity
low cost
allowing passive monitoring and/or force control
high modularity
multi-organ
Our motivations to work on this type of platforms in RORAS was that commercially available systems are dedicated to specific human joints (e.g. the knee) for adults, are very costly and intrusive. Such facts are still valid in 2008 and we believe that parallel wire-driven robots may be an efficient solution to address the previous issues. In such robotized system the moving part of the robot (the one that will be attached to the limb) is connected to the ground by several wires whose lengths may be controlled. The position and orientation of the moving part are controlled by adjusting the wire lengths (e.g. by coiling the wire on a drum). Furthermore the forces and torques exerted on the limb may be measured/controlled by measuring/constraining the tension in the wires. Such system has several advantages:


low intrusivity and low cost: instead of being confronted to an impressive mechanical system the patient may see only a small set of low dimension wires. The mechanical hardware part of the system is based on low-cost rotary motors and only a dedicated robot controller has to be used
position monitoring/force control: pose monitoring may be obtained just by measuring the lengths of the wires. In that case it is not even necessary to have a motor in the coiling system as only a pre-tensioning spring is needed (such systems are already commercially available). On the other hand, force control will require an actuated coiling system and tension measurement sensors. Indeed, it is easy to calculate what should be the tension in the wire to exert a given force/torque on the moving part
high modularity, multi-organ: clearly a rehabilitation robot has to provide very different performances when dealing with a finger or with a knee, with a child or with an adult. Indeed the amount of exertable force/torque, the amplitude of motion of the moving part are quite different according to the patient and the pathology. The advantage of parallel robots is that the performances they are provided may be drastically modified just by changing the geometry of the system (the location of the coiling systems on the ground). The system may be mechanically designed in such a way that changing this geometry is easy but the hard part is to determine what should be the location of the coiling systems so that the resulting robot will provide in a guaranteed way the required performances
safety: the wires may be dimensioned in such a way that they will break down if submitted to a large tension while the robotwill immediately stop in its current pose in case of a power outrage.

But such systems may also be a solution for other applications such as assistive robots. Current assistance robots suffer from the same drawbacks as rehabilitation robots: they are bulky, impressive, costly and not easy to use. A typical (and crucial) task for such a robot is to allow elderly people to sit down and stand up in their toilet and bath room. We believe that parallel wire-driven robots may also be used for this type of tasks and that their modularity is appropriate for them. In classical assistance robot the end-user and its surrounding have to be adapted to the robot (e.g. the robot cannot even enter a narrow bath room) while our robot may be designed according to the room dimensions and the end-user mass. They are less universal than the classical concept (they can perform only within the room) but we are believing that they provide a good compromise in terms of cost, flexibility and facility of interface.

[b]