OpenCog dev thru Nao

[Ahmad M.]

Good day Ben and all the other devs and users,

I strongly believe openCog to be a good complete framework for an AGI, albeit it needs work on being implemented and more tweaks here and there.

At the moment it is only a skeleton and we need to design its body, and to show its performance, we need to embody it in a humanoid robot. I suggest we use an Aldebaran Nao, as Xiamen University folks have done. Before I begin development work, I would like to seek advice from the OpenCog devs here. Is it possible to use OpenCog as an embedded cognition system on the Nao robot, given the hardware limitations of the bot ?

[Ben Goertzel]

Using Nao is certainly possible, but I am more interested at the
moment in seeing what we can do with OpenCog on inexpensive robots
powered by Raspberry Pi 3 or similar...

This is something we (Hanson Robotics, iCog Labs, etc.) will be
exploring over the course of 2017...

> --
> You received this message because you are subscribed to the Google Groups
> "opencog" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to opencog+u...@googlegroups.com.
> To post to this group, send email to ope...@googlegroups.com.
> Visit this group at https://groups.google.com/group/opencog.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/opencog/2be7e1aa-e1e0-43f0-8bea-ccbafa04d417%40googlegroups.com.
> For more options, visit https://groups.google.com/d/optout.



--
Ben Goertzel, PhD
http://goertzel.org

“Our first mothers and fathers … were endowed with intelligence; they
saw and instantly they could see far … they succeeded in knowing all
that there is in the world. When they looked, instantly they saw all
around them, and they contemplated in turn the arch of heaven and the
round face of the earth. … Great was their wisdom …. They were able to
know all....

But the Creator and the Maker did not hear this with pleasure. … ‘Are
they not by nature simple creatures of our making? Must they also be
gods? … What if they do not reproduce and multiply?’

Then the Heart of Heaven blew mist into their eyes, which clouded
their sight as when a mirror is breathed upon. Their eyes were covered
and they could see only what was close, only that was clear to them.”

— Popol Vuh (holy book of the ancient Mayas)

[Ahmad M.]

I was having difficulty fathoming how an AtomSpace would be stored within Nao, not to mention how algorithms such as MOSES running on extensive networks, is implausible for the Nao Atom processor.

Therefore, I am surprised at the notion of running OpenCog on low-level hardware such as Raspberry Pi.

[Ben Goertzel]

Atomspace is just a container, it doesn't take especially much memory
or processing power... unless you put a lot of data in it obviously!

Running MOSES for learning on a Pi isn't going to be a good idea, OTOH
*evaluating* MOSES models that were learned elsewhere is not expensive

The URE rule engine (and so PLN) can run on Raspberry Pi, but
obviously complex inferences will take a long time...

So it's possible to make OpenCog do stuff on embedded processors but
for many applications it may be better to have e.g. an OpenCog on the
Pi that stores a relatively small amount of immediately important
knowledge and does only simple processing (say, perception, action
selection and very simple inference) , and that connects to an OpenCog
on the cloud that does more serious thinking...

-- Ben

> https://groups.google.com/d/msgid/opencog/f873eabe-944f-4100-a348-41a7e9113964%40googlegroups.com.