Topological reasoning about dextrous manipulation

An integrated symbolic-numeric scheme for dextrous manipulation that uses a topological approach is considered. A comprehensive summary and discussion of previous work and its extension to an intelligent robot prehension scheme are presented. A topological model of multifingered hands is reviewed and the notion of hand functionality, is introduced. It is shown that both hand posture and hand functionality can be computed from symbolic, high-level task requirements and object attributes, and can be transformed into numeric, low level, joint space variables. Numeric specifications for hand posture and a numeric description of hand functionality are derived. The realization of dextrous manipulation is considered as a combination of two processes, a postural transformation process and a functional transformation process, from a given high-level, symbolic description of task/object into joint variables. A complex task to be performed by an entire hand may be decomposed into simpler tasks to be concurrently performed by subconfigurations in either tip prehension, palm prehension, or a combination of both