Low Complexity Robots

jbp jbppiss at cheerful.com
Wed Nov 10 11:24:30 EST 1999


IMHO, top down complexity running on serial computers won't yield any decent
results in the near future. Where can I find _my_ source code? Point made.
: )

Joseph Birr-Pixton
http://www.robotic.co.uk
Take the piss out of my address to email me

>Mark W. Tilden, mwtilden at math.uwaterloo.ca, beamed to us:
>> Date: 10 Nov 1999 01:02:03 GMT
>> Organization: University of Waterloo
>
>> In article [...], Steve <rstevew at armory.com> wrote:
>>> This is Mark Tilden's BEAM robotics. It's malarkey.
>>> Cute malarkey, but still malarkey. The thing keeps
>>> thrashing is all, it doesn't think any more than a
>>> pair of dice thinks. It just tries all the numbers
>>> till something happens. It makes many of the set of
>>> all possible mistakes. It is never used for anything
>>> difficult because it will fail. His stuff is at lanl.gov.
>>> He has a following and it strongly resembles a cult
>>> for those frightened of learning microcontrollers or
>>> bored of the hard work of mainstream AI.
>
>> Dear Steve.  This is inaccurate to say the least.
>> Chaotic control systems are an entrenched part of
>> conventional AI and robotics, and not just our work.
>> Such control aspects are used ubiquitously, and there
>> are numerous papers and books on the subject going back
>> many years.  As for the "frightened cult", they are neither.
>> Most pick up on BEAM educational robotics because it's fast
>> to learn, has a high success rate, and is very inexpensive
>> compared to processor-based robotics. Furthermore, the
>> efforts of many BEAM enthusiasts show that hard work
>> is the last thing they're frightened of, as a cruise
>> through some of their websites would quickly confirm.
>
>> In conclusion I am always willing to discuss with people who
>> have informed opinions.  Please let me know when you get one.
>
>Steve has the following informed opinion, because I just gave it to him:
>
>http://www.geocities.com/Athens/Agora/7256/mind-fpc.html Mind.Forth is
>the way to go if we are to make intelligent robots in our lifetimes.
>
>The colliding worlds of robotics and AI owe an enormous debt to
>Richard Steven Waltz, because in early 1998 he so strongly attacked
>http://www.scn.org/~mentifex/ Project Mentifex (Mindmaker), that
>Jeff Fox of Forth fame felt obliged to defend the PD AI project.
>At that point, I resumed the Forth coding of 1995-era Mind.Forth and
>brought it to its current status as a top-down high-complexity AI.
>
>Hurrah! Godspeed! And Happy New Year to Richard Steven Waltz --
>a prime mover and shaker in both fields:  AI and Robotics.
>
>> Mark W. Tilden [...]
>
>--- please promote open source AI -- put this link on the Web ----
><a href="http://www.geocities.com/Athens/Agora/7256/mind-fpc.html"
> TITLE="for documentation click on modules in flow-chart">
>Mind.Forth AI Source Code</a> -- for porting to other languages.<BR>






More information about the Neur-sci mailing list