This workshop will try to map out the potential for artificial neural networks (ANNs) in the fields of robotics and systems control. Robotics and systems control are among the most important possible applications of ANNs, for two reasons. First, they are of enormous economic significance; higher levels of factory automation are essential to higher levels of productivity in the American economy, as well as such areas as military and space applications. The largest potential applications of ANNs in these areas will not be possible for many years, because a lot of very fundamental research will be needed first; this workshop will try to map out what is needed. Second, the difficulties of ANNs in controlling robots are similar to the difficulties faced by organic brains in controlling movement to achieve goals over time; it is hoped that a better understanding of the former will lead to a better understanding of the latter. This workshop will attempt to bring together a number of scattered research groups doing work which is important to these areas, and produce a report on future directions which can guide NSF and stimulate good proposals in this field.
