An expert system for grasp selection by a high dexterous robotic hand

Abstract

The paper describes a new approach to grasp synthesis by a highly dexterous robotic hand. The adopted design methodology is based on an AI technique called assumption-based truth maintenance system (ATMS). The validity of this approach has been ascertained with reference to a high-dexterity 3-finger, 10-degrees of freedom robotic hand (the University of Bologna Hand), but the underlying methodological issues can be specialized to many other robotic dexterous hands. The developed prototype system, while referring to some working hypotheses, has shown a satisfactory operative level.