Selective automation and skill transfer in medical robotics: a demonstration on surgical knot-tying.
Int J Med Robot
; 8(4): 384-97, 2012 Dec.
Article
en En
| MEDLINE
| ID: mdl-22605676
ABSTRACT
BACKGROUND:
Transferring non-trivial human manipulation skills to robot systems is a challenging task. There have been a number of attempts to design research systems for skill transfer, but the level of the complexity of the actual skills transferable to the robot was rather limited, and delicate operations requiring a high dexterity and long action sequences with many sub-operations were impossible to transfer.METHODS:
A novel approach to human-machine skill transfer for multi-arm robot systems is presented. The methodology capitalizes on the metaphor of 'scaffolded learning', which has gained widespread acceptance in psychology. The main idea is to formalize the superior knowledge of a teacher in a certain way to generate support for a trainee. In our case, the scaffolding is constituted by abstract patterns, which facilitate the structuring and segmentation of information during 'learning by demonstration'. The actual skill generalization is then based on simulating fluid dynamics.RESULTS:
The approach has been successfully evaluated in the medical domain for the delicate task of automated knot-tying for suturing with standard surgical instruments and a realistic minimally invasive robotic surgery system.
Texto completo:
1
Colección:
01-internacional
Base de datos:
MEDLINE
Asunto principal:
Robótica
/
Técnicas de Sutura
Límite:
Humans
Idioma:
En
Revista:
Int J Med Robot
Año:
2012
Tipo del documento:
Article
País de afiliación:
Alemania