Page 43 - Robots Androids and Animatrons : 12 Incredible Projects You Can Build

machine would certainly know that it is intelligent. Another crite-
rion, more simple and direct, and the one that is used in this book,
is the ability to learn from experience.
Of course, we could abandon logical approximations entirely and
state that intelligence is achieved in systems that develop a sense
of humor. As far as I know humans are the only animals that laugh.
Perhaps humor and emotion will end up being the truest test of all.

Using neural networks in robots
So how do neural networks help our robotics work today? Well,
we’re a way off from creating competent AI, let alone putting it
into one of our robots. But neural technology can control robotic
function, and, in many cases, can perform superiorly to standard
central processing unit (CPU) control and programming. By using
neural networks in our robots, we can have our robots perform
small operational miracles without the use of a standard computer,
CPU, or programming. In Chap. 6 we will design a two-neuron
fuzzy logic system that can track a light source. Place this system
on a mobile robot, and the robot will follow a light source any-
where. Also in Chap. 6 we discuss BEAM robotics and Mark Tilden,
22 who designs transistor networks (nervous networks) that allow
legged robots to walk and perform other functions. Another neural
process that is making great strides is called subsumption archi-
tecture, which uses layered stimulus response.

Tiny nets

Small neural network programs can also be written in microcon-
trollers. For more information on these microcontrollers see Chap. 6.

Neural-behavior-based architecture

Behavior-based architecture, developed by Walter Grey, illustrates
that relatively simple stimulus-response neural systems when
placed in robotics can develop high-level, complex behaviors. Sub-
sumption architecture, an offshoot of behavior-based architecture
developed by Dr. Rodney Brooks at MIT, is also covered more fully
in Chaps. 6 and 8.

Team LRN
Chapter two