[rescue] Anyone Familiar with VMEbus?

[John Hudak]

One doesn't need 'modern hardware' to control that arm.  I was doing it wan
a DEC LSI11 and later with a 11/73.  There was a Intel 86x box in the lab
which could have been used as well.  Most of the code I developed was in
assembler, and some C.  Yea, standard approach is to take the small voltage
outputs of the d/a's and have that as inputs to the servo amps.  My gig was
applying/developing modern control theory approaches to construct observers
to refine the time to position and accuracy in 3-space, avoiding the
dreaded overshoot problem.   Had to prove to one of my instructors that you
can't change physics/law of nature with software..the beauty was in the
math model...but that is another story.....


On Fri, Sep 17, 2021 at 2:44 PM Mouse <mouse at rodents-montreal.org> wrote:

> >>> [...PUMA arm...robotics control...]
> >> I don't recall what we were using for interface hardware once it
> >> left the host system (I wasn't dealing with it much past that
> >> point), but I *think* there was no VME anything involved.  [...]
> > Building new control hardware for PUMA robots is a favorite college
> > project, it seems.  Lots and lots of college robotics courses do
> > exactly that, with that hardware.
>
> Nice!  That must be fun for the hardware hackers.
>
> This grad student's work was more about the software, though.  There
> was hardware, but the point was to run the control loop on the host.
> This being the late '80s, we didn't have now-modern hardware for it; we
> were using Qbus MicroVAXen (hence the finding a bug in the KA630), one
> of the (I gather) few sites actually using multiprocessor MicroVAX-IIs
> (one CPU for the host OS, the other for the real-time low-level control
> loop).  I don't recall much about the interface electronics we used,
> but IIRC the OP said something about wanting to talk to the arm with a
> VT420.  If so, that means not just the minimal interface electronics we
> had but some smarts - we wanted the interface from the host to the
> robot to be about as dumb as possible, since the whole point of the
> research was the control software.
>
> My memories are thirty years old and somewhat fuzzy, but I _think_ we
> used a hand-built Qbus board as the interface hardware.  I don't recall
> how much there was between that and the motors, but there must have
> been some driver transistors at a bare minimum, because I'm fairly sure
> it was still logic levels when it left the host enclosure.
>
> /~\ The ASCII                             Mouse
> \ / Ribbon Campaign
>  X  Against HTML                mouse at rodents-montreal.org
> / \ Email!           7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B
> _______________________________________________
> rescue list - http://www.sunhelp.org/mailman/listinfo/rescue