Project Type:
Contract No.:
Project Acronym:
Project Full Title:
Starting date:
Duration:

 UE STREP
 FP6-NMP4-CT-2006-0033423
 SmartHand
 The Smart Bio-adaptive Hand Prosthesis
 November 1, 2006
 36 months

 

 

Brief Presentation

The overall scientific objective of the SmartHand project is to develop an intelligent artificial hand that looks and feels like a real hand . This is a challenging and visionary goal. However, recent development in the field and several converging scientific areas makes it possible to state that the perfect artificial hand is no longer a fantasy. SmartHand aims to integrate recent advances in nanobioscience, cognitive neuroscience and information technologies in order to develop such an intelligent artificial prosthetic hand with all basic features displayed by a real one.

The SmartHand project is divided into two major approaches, i.e. long- and short-term research efforts, respectively. In long-term, a highly advanced neural interface will be developed and different studies will be conducted to achieve a better knowledge in cognitive neuroscience, and thus also contribute to the field of nerve injuries in general. In short-term research efforts will target the development of a hand prosthesis system, namely the SmartHand , that directly could be used by hand amputees. This system will consist of a highly dexterous prosthetic hand controlled by the remaining muscles in the forearm stump. It will be equipped with artificial sensibility put into effect by artificial skin relaying sensation to a tactile display. A cognitive software training system will be developed to tune the SmartHand to the user and to train the subject. Furthermore, numerous qualitative and quantitative studies will be performed with amputees. Since the SmartHand project will take advantage from the outcomes of the previous major projects ( The Artificial Hand, CyberHand, and FreeHand ), it will substantially enhance functionality, controllability, and acceptability of hand prostheses and nerve injuries in general.

 
Participant List
 


No

1
2
3
4
5
6
7


Leader Name

Thomas Laurell, Coordinator
Maria Chiara Carrozza
Dejan Popovic
Yosi Shacham
John Alderman
Freygardur Thorsteinsson
Elisabet Csöregi

 


Participant Name

Lund University, Lund
ARTS Lab, Scuola Superiore Sant'Anna, Pisa
Aalborg University, Aalborg
Tel Aviv University, Tel Aviv
Tyndall Institute, Cork
Össur
SciTech Link HB, Lund


Country

Sweden
Italy
Denmark
Israel
Ireland
Iceland
Sweden

ARTS Lab Contribution
 
ARTS Lab is responsible for the development of WP5.

WP 5.1 First implementation of multi-degrees of freedom prosthesis and of artificial skin. Cosmetics is one of the most important requirement for a prosthetic hand, and an appropriate cosmetic skin will be developed to provide external appearance as the human hand. In addition, the robotic hand must be perceived by the amputee as the natural hand. Concerning functionality, it is important that the hand is able to perform grasping, manipulation and exploratory tasks that are fundamental for Activities of Daily Living. This will be achieved by a dextrous robotic hand endowed with 16 degrees of freedom and able to flex/extend each finger independently, and to oppose thumb to different fingers tip for providing different hand shapes and grasps adaptive to objects to be grasped and manipulated. The hand will be actuated by motors integrated in the palm or in the socket, and the transmission will be based on sensorized tendons and underactuated mechanisms. The robotic hand will be designed with a biomechatronic approach in order to mimic the functional anatomy of the musco-skeletal system and the natural motor control loop. Particular attention will be devoted to develop an artificial sensory system integrated in the hand that is aimed at providing the same functionality of natural mechanoreceptors for exteroception and proprioception, a purposely designed sensory skin will be designed and fabricated. The targeted hand performance is that of the natural hand, for example the design goal for the maximum force value is more than 100 N. This value should be compatible with the available energy stored in the battery. In order to perform stable grasp against disturbance is fundamental to develop slippage sensing, and global force measurement with pressure sensors distributed in the hand surface. WP5.1 is dedicated to the development of a first prototype of 16 DoF robotic hand. The hand will be covered by a soft and compliant material in order to increase the grip during grasping and manipulation tasks. This will provide compliance to the hand in order to protect from interaction forces generated during hand-object contacts. In addition, an early prototype of artificial skin integrating available micro sensors will be developed.

 

WP 5.2 Development of the final SmartHand robotic hand and artificial skin.

The results of the experimental characterization of the prototypes developed in WP 5.1 will be exploited by P2 and P6 to co-design and develop the final SmartHand prosthetic hand and its sensorized artificial skin. The development of the last prototype will take into account the possible engineering plan for the industrial exploitation in agreement with P6.

 
April 2007: System Definition
 


A five finger anthropomorphic hand prosthesis, employing robust sensors and electronics will be developed. The main requirement will be the hand assessment with amputees during the last year of the project. Since the majority of amputations are distal (quite near to wrist level), the prosthetic hand will employ a low number of motors fitted in a small volume package inside or nearby the palm. Characteristics such as grasping capabilities, robustness, cosmetics, small weight and human size will be preferred in respect to high dexterity, high power and manipulation capabilities. The prosthetic hand will be designed with a biomechatronic approach in order to mimic the functional anatomy of the musco-skeletal system and the natural motor control loop. The finger flexion/extension actuation will be based on underactuated and differential mechanisms and on torsion springs in order to reduce both encumbrance and weight. Moreover, the hand will have an opposable thumb in order to perform an high number of prehensile patterns such as lateral, bi-digital, tri-digital and cylindrical grasps, and will be actuated with a sensorized tendon driven transmission in order to decrease weight.

 


The SmartHand mechanical design: Five underactuated fingers are driven by four non-back-drivable actuation units based on DC-Motors. All fingers are made of 3 phalanxes and endowed with angular and grasping sensors. Thumb opposition axis positionin is also provided.

Particular attention will be devoted to the development of an artificial sensory system integrated in the hand in order to try to replicate some functionality of natural mechanoreceptors for exteroception and proprioception. Position sensors, force sensors and touch sensors will be developed and embedded in the mechanical structure of the hand. Sensory information will be used both for automatic closed-loop control implementation and for sensory feedback to the amputees.
A dedicated controller will be developed for proper hand operation and external world dealing. This controller will be able to directly drive all degrees of motion, both using force control algorithms and position control ones. Moreover, on this controller some low level control strategies for functional tasks will be designed, implemented and tested in order to exploit in the best way the features of the novel neural interfaces developed by the consortium partners. Communication protocols will be developed in collaboration with the partners in order to exchange information among the developed stand-alone prosthetic hand and the stimulation and recording electronic circuitries.
Cosmetics will be one of the main requirements: an appropriate cosmetic glove will be used to provide external appearance as the human hand.
From a functional point of view, the hand will be able to perform at least the 80% of the grips necessary in activities of daily living (ADLs) , i.e. palm opposition grasps (power grasps), a small set of pad opposition grasps (precision grasps) and lateral grasps. For this reason, the developed hand will be a fine device for grasping tasks but not for manipulation. In fact, the hand transmission and actuator system based on underactuated mechanisms do not permit the direct control of all degrees of freedom.

 
May 2007: Small Volume Non-back Drivable Actuation Units
 


Since they reduce battery consumption, non-back drivable actuation units are classical solution for prosthetic hands. In order to reduce the overall size of the SmartHand, novel mechanisms are being under evaluation and testing.

A first prototype of capstan based system has been designed and tested, showing good results. A debuged version is currently being designed.

The final actuation unit will embed on the motor axes a 5 milion cycles potentiometer and two limit switches for position control.

 

  



First prototype of small size non-back drivable actuation unit