Journal
AUTONOMOUS ROBOTS
Volume 12, Issue 3, Pages 301-312Publisher
KLUWER ACADEMIC PUBL
DOI: 10.1023/A:1015221832567
Keywords
walking machines; central pattern generators; neural networks; legged locomotion; non-linear oscillators; biologically inspired robots
Ask authors/readers for more resources
A newborn foal can learn to walk soon after birth through a process of rapid adaptation acting on its locomotor controller. It is proposed here that this kind of adaptation can be modeled as a distributed system of adaptive modules (AMs) acting on a distributed system of adaptive oscillators called Adaptive Ring Rules (ARRs), augmented with appropriate and simple reflexes. It is shown that such a system can self-program through interaction with the environment. The adaptation emerges spontaneously as several discrete stages: Body twisting, short quick steps, and finally longer, coordinated stepping. This approach is demonstrated on a quadrupedal robot. The result is that the system can learn to walk several minutes after inception.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available