The role of hidden layers in learning motor control in autonomous systems: Jahir Pabon and David Gossard. Mechanical Engineering Department, Massachusetts Institute of Technology, 77 Massachusetts Avenue Room 3-449, Cambridge, MA 02139
- Book ID
- 103926260
- Publisher
- Elsevier Science
- Year
- 1988
- Tongue
- English
- Weight
- 95 KB
- Volume
- 1
- Category
- Article
- ISSN
- 0893-6080
No coin nor oath required. For personal study only.
✦ Synopsis
Autonomous systems is a research area of large practical importance. Such research has two potential benefits: increased insight into perception and control strategies used by humans, and the identification of subsets of those strategies that can be efficiently implemented in autonomous robots.A central problem inherent to autonomous systems is the need for an internal reference frame in which sensory inputs can be interpreted. In particular, an autonomous system must be able to interpret sensory information in a way which takes into account the relative positions of sensory and motor components with respect to the system's external environment. It is hypothesized that, in natural systems, sensory information is transformed into a consistent internal representation that serves as an internal invariant re{crence frame. It is desirable that artificial systems learn this representation in order to compensate for unforeseen changes in the environment or in the system itself following growth or damage.