Biologicaly inspired on-line generation of complex movements using primitives

The objective of this work is to demonstrate the possibility of using primitives for on-line generation of complex movements that ensure motion of bipedal humanoid robots in unstructured environment where on-line generation of motion is required. Primitives represent simple movements that are either reflex or learned. Each primitive has its parameters and constraints that are determined on the basis of the movements capable of performing by a human. The set of all primitives represents the base from which primitives are selected and combined for the purpose of performing the corresponding complex movement. It is convenient to describe primitives on the kinematic level (shape of motion is easily visible) but motion have to be learned and executed as set of synchronized driving torques (to system are forwarded corresponding motor control variables). Shape of executed motion serve as a visual supervision of motion appropriateness.