Self-scaling robots edge nearer
They can be a bane to a temperamental youth but they could also make robots
bigger, stronger and more independent. What are they? Hormones, and a team at
the University of Southern California's Information Science Institute have
included hormone-like messaging in a robotic control system.
The discovery could lead to significant advances in self-scaling robots -
robots that can rebuild themselves. Robots with this ability could explore
deep space independently: shipped to distant planets as a mass of tiny dumb
units, they would build themselves into whatever kind of machinery proved
appropriate on arrival.
Behnam Salemi and his colleagues at the ISI have found a way to make such
resourceful robots. These include walking creations that will adapt their gait
if they lose a leg, and others that inch along like a caterpillar regardless
of what length they have become. Both use software inspired by biological
In nature, hormones trigger events in different parts of the body
simultaneously. 'For example, when a human experiences sudden fear, a hormone
is released by the brain, causing different actions,' said Salemi. Fear
releases hormones that might make your mouth open and your leg muscles twitch
into action. Although the brain triggers the hormone, the muscle control is
located in the mouth and legs, Salemi said.
'Hormones are propagated signals that can be modified or delayed-or which can
even disappear along the way,' said project supervisor Peter Will. This
adaptability lets hormones accomplish tasks that a single-destination message
In a segmented or multi-module robot, a message based on digital hormones will
reach all connected sections. The hormone message tells all modules about the
condition of the others without carrying specific commands. Each module can
then interpret this knowledge in its own way.
Each module is an autonomous unit, with its own on-board power supply and
motors. The modules connect via two-pronged plugs through which the hormones
propagate. 'You don't even give it information about what gait it should
adopt,' said Salemi. The human 'operator' just tells the robot which direction
to move and the robot will figure out the best gait, given its current
This article appears courtesy of New Scientist.
consciousness, phlogiston, philosophy, vitalism, mind, free will, qualia,
analog computing, cultural relativism
Everything that can happen has already happened, not just once,
but an infinite number of times, and will continue to do so forever.
(Everything that can happen = more than anyone can imagine.)
This archive was generated by hypermail 2b30 : Mon May 28 2001 - 09:59:45 MDT