Technology involved
from: http://neverendinglist.com/a-blank-page/
The intelligent controlling mechanism of a typical mobile
robot is usually a computer or microprocessor system and
hence much of the initial work considering the future ethics
and rights of robots has apparently focused only on this
specific sub-class of intelligent robots (e.g. Arkin 2009).
Research is however, now ongoing in which biological
neuronal networks are being cultured and trained to act as
the brain of a physical, real world robot—either completely
replacing or operating in tandem with a computer system.
From a medical viewpoint, studying such neuronal
systems can help us to understand biological neural structures
in general and it is to be hoped that it may lead to
basic insights into problems such as Alzheimer’s and
Parkinson’s Disease. Other linked research meanwhile is
aimed at assessing the learning capacity of such neuronal
networks (Xydas et al. 2008). To do this a hybrid system
has been created incorporating control of a mobile wheeled
robot solely by a culture of neurons—a biological brain.
A brain, the human version in particular, can be said to
be a complex computational platform (Lloyd 1991). It
rapidly processes a plethora of information, is adaptable to
noise and is tolerant to faults. Recently however, significant
progress has been made in the practical integration of
biological neurons and electronic components by culturing
tens of thousands of brain cells in vitro (Bakkum et al.
2003). These technologies blur the distinction between
synthesized brains and those which can be regarded as
being achieved through a normal biological route.
The brains are created by dissociating/separating the
neurons found in cortical tissue using enzymes and then
culturing them in an incubator, providing suitable environmental
conditions and nutrients. In order to connect a
brain with its robot body, the base of the incubator is
composed of an array of multiple electrodes (a multi
electrode array—MEA) providing an electrical interface to
the neuronal culture (Thomas et al. 1972).
HOW CAN YOU READ THIS - THERE'S NO PICTURES
Once spread out on the array and fed, the neurons in
such cultures spontaneously begin to grow and shoot
branches. Even without any external stimulation, they
begin to re-connect with nearby neurons and commence
both chemical and electrical communication. This propensity
to spontaneously connect and communicate demonstrates
an innate tendency to network. The neuronal
cultures themselves form a layer over the electrode array
on the base of the chamber making them accessible to both
physical and chemical manipulation (Potter et al. 2001).
The Multi Electrode Array enables output voltages from
the brain to be monitored from each of the electrodes,
allowing the detection of the action potential firing of
neurons near to each electrode as voltage spikes representative
of neural charge transfer. It is then possible to
separate the firing of multiple individual neurons, or small
groups, from a single electrode (Lewicki 1998).
With multiple electrodes an external picture of the
neuronal activity of the brain can be pieced together.
in Implications and consequences of robots with biological brains
also
rat-brained robots take their first steps
navigating with the robot brain
computing with instinct
The intelligent controlling mechanism of a typical mobile
robot is usually a computer or microprocessor system and
hence much of the initial work considering the future ethics
and rights of robots has apparently focused only on this
specific sub-class of intelligent robots (e.g. Arkin 2009).
Research is however, now ongoing in which biological
neuronal networks are being cultured and trained to act as
the brain of a physical, real world robot—either completely
replacing or operating in tandem with a computer system.
From a medical viewpoint, studying such neuronal
systems can help us to understand biological neural structures
in general and it is to be hoped that it may lead to
basic insights into problems such as Alzheimer’s and
Parkinson’s Disease. Other linked research meanwhile is
aimed at assessing the learning capacity of such neuronal
networks (Xydas et al. 2008). To do this a hybrid system
has been created incorporating control of a mobile wheeled
robot solely by a culture of neurons—a biological brain.
A brain, the human version in particular, can be said to
be a complex computational platform (Lloyd 1991). It
rapidly processes a plethora of information, is adaptable to
noise and is tolerant to faults. Recently however, significant
progress has been made in the practical integration of
biological neurons and electronic components by culturing
tens of thousands of brain cells in vitro (Bakkum et al.
2003). These technologies blur the distinction between
synthesized brains and those which can be regarded as
being achieved through a normal biological route.
The brains are created by dissociating/separating the
neurons found in cortical tissue using enzymes and then
culturing them in an incubator, providing suitable environmental
conditions and nutrients. In order to connect a
brain with its robot body, the base of the incubator is
composed of an array of multiple electrodes (a multi
electrode array—MEA) providing an electrical interface to
the neuronal culture (Thomas et al. 1972).
HOW CAN YOU READ THIS - THERE'S NO PICTURES
Once spread out on the array and fed, the neurons in
such cultures spontaneously begin to grow and shoot
branches. Even without any external stimulation, they
begin to re-connect with nearby neurons and commence
both chemical and electrical communication. This propensity
to spontaneously connect and communicate demonstrates
an innate tendency to network. The neuronal
cultures themselves form a layer over the electrode array
on the base of the chamber making them accessible to both
physical and chemical manipulation (Potter et al. 2001).
The Multi Electrode Array enables output voltages from
the brain to be monitored from each of the electrodes,
allowing the detection of the action potential firing of
neurons near to each electrode as voltage spikes representative
of neural charge transfer. It is then possible to
separate the firing of multiple individual neurons, or small
groups, from a single electrode (Lewicki 1998).
With multiple electrodes an external picture of the
neuronal activity of the brain can be pieced together.
in Implications and consequences of robots with biological brains
also
rat-brained robots take their first steps
navigating with the robot brain
computing with instinct
Comments
Post a Comment