Am Montag, 2. März 2020 16:55:02 UTC schrieb linas:
>
> Ah!
> -- So, if you just need a motion planner that works (as opposed to messing 
> with the theory of it) then, as a practical solution, I would recommend 
> just grabbing something off the shelf and wiring it in, ad hoc.
>

Yes, I'm aware of ASP (and several other approaches) that do specific 
things very fast. But most of them can't handle uncertainty or can't handle 
variables, both of which we make use of in this project (e.g. robot 
perception and its interpretation of human behaviour introducing 
uncertainty)..and I'm not looking for a solution where I have to keep 
around *n* knowledge bases to make *n* different software packages happy 
just because they tend to be excellent at a specific niche. (I already have 
to integrate too many for comfort)

If path planning turns out to be too slow, I'd rather apply some 
abstraction (from waypoints to hubs and spokes).

What I'd also like to try is some 'smart' (in the smartphone sense) 
chaining, where we try to evaluate the easy parts first and use those to 
constrain the harder parts, and where we allow for more finely-grained 
scoping (e.g. "only dogs qualify for this variable, we only have two dogs, 
don't consider 500 waypoints here").


Best,
Alex

-- 
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 [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/opencog/9a695663-1243-4848-8103-6832c358536b%40googlegroups.com.

Reply via email to