Exploring new approaches to machine hosted
neural-network simulation, and the science
A programmer who is obsessed with giving experimenters
a better environment for developing biologically-guided
neural network designs. Author of
an introductory book on the subject titled:
"Netlab Loligo: New Approaches to Neural Network
Simulation". BOOK REVIEWERS ARE NEEDED!
Can you help?
Demonstrates some basic capabilities of air muscles.
See the video below...
This entry is mainly just to show how people are beginning to accept unconventional actuator-capabilities.
I doubt this is cutting edge. It certainly doesn't help their credibility that they are claiming to have developed these air-muscles, which have been around now for decades. I make allowances because (1.) The person manning the trade-show booth is probably just a guy from the sales and marketing department and isn't one of the engineers, and (2.) There are likely to be some language issues, in that he may have simply meant that they had developed their technology in support of (i.e., "around") air muscles.
Recent use of air muscles is one example of how the design-priorities surrounding actuator-technology are changed in systems capable of implementing complex feedback. Though air-muscles have been around a long time, they have been obscured. Their lack of positional precision, and repeatability have made them useless for inclusion in conventional systems. While accurate, repeatable positioning are important in conventional "dead reckoning" systems, they become much less important when systems are able to get feedback on, and adapt to, positional variation.