Robot Control based on Neural Network Systems


CONNY - 6715

Keywords computer-integrated agriculture, inspection systems, neural networks, process control, robotics, space applications


Start Date: 30 April 1992 / Duration: 36 months / Status: ongoing

[ contact / participants ]


evaluation of the suitability and advantages of neural network techniques for complex robot control tasks and development of a set of corresponding neural network modules

Objectives and Approach

Modules are developed by utilising existing neural network techniques as well as through their enhancement and adaptation to the particular applications. Three demonstrators have been chosen:

These involve focusing on three task areas:

Results taken from the development environment part of the GALATEA project (5293) are exploited in this project.

The research and concept definition of the required neural network technology and specifications of the three demonstrators and the Development Environment have been completed. The concept definition involved developing new neural network based solutions.


CONTACT POINT

Albrecht Kellner
MBB/ERNO Raumfahrttechnik
Hünefeldstr. 1-5
D-28199 Bremen
tel: +49 421 539-4987
fax: +49 421 539-4603
telex: 245548 ERNO D

Participants

Coordinator
MBB/ERNO Raumfahrttechnik [D]
Partners
Cram [I]
Mimetics SA [F]
Thomson-CSF -Laboratoire Électronique de Rennes [F]
Universidad Politecnica de Cataluña [E]
University College London [UK]
Associate Partner
Framentec SA [F]


HPCN synopses home page HPCN acronym index HPCN number index
All synopses home page all acronyms index all numbers index

CONNY - 6715, December 1993


please address enquiries to the ESPRIT Information Desk

html version of synopsis by Nick Cook