Neurobiological and neurorobotic approaches to control architectures for a humanoid motor system

Abstract

In the mammalian nervous system, the sensorimotor interface between cortex and spinal cord plays a key role in novel skill formation and motor learning. We seek to understand the principles of motor learning at this interface using a multidisciplinary approach. We believe this approach will prove relevant to the development of biomimetic control architectures for humanoid robots. Learning at this interface requires an understanding of the spinal output structures. Ultimately, these must form the basis of the algorithms needed for adaptive motor learning. These spinal structures interact with descending cortical control to produce accurate limb trajectories, and novel motor behavior. © 2001 Elsevier Science B.V. All rights reserved.