Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!swrinde!zaphod.mps.ohio-state.edu!sol.ctr.columbia.edu!emory!emcard!wa4mei!ke4zv!gary
From: gary@ke4zv.UUCP (Gary Coffman)
Newsgroups: sci.electronics
Subject: Re: morse code
Summary: toy bicycles
Message-ID: <1719@ke4zv.UUCP>
Date: 26 Dec 90 06:59:06 GMT
References: <1990Dec12.231058.23895@engin.umich.edu> <1990Dec14.012315.7858@zoo.toronto.edu> <516@aupair.cs.athabascau.ca> <13605@will.ntt.JP>
Reply-To: gary@ke4zv.UUCP (Gary Coffman)
Organization: Gannett Technologies Group
Lines: 40

In article <13605@will.ntt.JP> kitagawa@will.ntt.jp (Masahiro Kitagawa) writes:
>CW/Morse-code itself may not be so interesting as a modulation/coding
>technique any more. But how experienced operators decode morse-code is
>very interesting.
>
>After operating CW intensively in those contests, every hiss sound is
>decoded as Morse-code for me. Even wind seems calling like 'DE ....'. 
>This symptom continues almost for a day, until I get enough sleep. My
>ear/nerve/brain system seems over adaptive to morse-code :-) Some kind
>of matched filter must be formed in my brain. Or it has been already
>built in my brain after years of over dose to CW. (addictive) And a
>hour of intensive operation may activate it.
>
>As a little bit old-fashioned ham operator, I like CW very much.
>
>As a scientist, I am very much interested in how human brain is
>adapted to CW/Morse-code. (Though I am amature in neuro science.)
>
>As an engineer, I want to emulate CW decoding mechanism of human brain
>by electronics such as DSP, neural network, software, ... 
>Interesting enough !
>
>As someone pointed at, CW is like bicycle. Both are easy for human,
>but very difficult for machine to manipulate. Is there already a robot
>which can ride bicycle very well ?

I haven't heard of an anthropomorphic robot that could mount a bicycle
built for humans and ride off. I have seen a child's toy motorcycle
that rides about by itself. I have even seen a toy unicycle that
can stay upright. And of course there are model helicopters that
can hover under gyro control, a three axis stablization problem.

It would be terribly inefficient to design an anthropomorphic robot
to ride a bicycle. There is no need to go to such Rube Goldberg lengths
when a robot can be directly connected to the wheels. It would be terribly 
inefficient to design a machine to read on-off keyed Morse Code. It's been 
done, but there are other non-human readable codes that are more efficient 
for a machine to use. PSK encoded ASCII with FEC for example.

Gary KE4ZV