NetNews Usenet Archive 1992 #18/NN_1992_18.iso/spool/sci/space/12112

Internet Message Format | 1992-08-21 | 5.4 KB

Path: sparky!uunet!cis.ohio-state.edu!magnus.acs.ohio-state.edu!slc3.ins.cwru.edu!agate!soda.berkeley.edu!gwh
From: [e-mail] (George William Herbert)
Newsgroups: sci.space
Subject: Re: With telepresence, who needs people in Earth orbit?
Date: 22 Aug 1992 06:46:29 GMT
Organization: Dis-
Lines: 94
Sender: [e-mail] (George William Herbert)
Message-ID: <[e-mail]>
References: <[e-mail]> <[e-mail]> <[e-mail]>
NNTP-Posting-Host: soda.berkeley.edu

[e-mail] (Gerald Cecil) writes:
>[e-mail] (George William Herbert) writes:

>>Have you talked to someone who builds or works with space robotics?
>No, I was hoping that there might be someone out there who would counteract
>hype from virtual-reality types.

Ok.

>>When anyone suggests replacing humans in the short term, they laugh.
>>Space robotics aren't as strong, for the most part, have less degrees
>>of freedom _and_ less limbs, less end effector dexterity, and more
>>likely failure points than a man in space.  They're safer, but often
>>can't do the job.  (note that an astronaut in EVA can't do everything
>>either.  both together are much more capable.)
>OK, fine.  But the only current `job' for SSF is apparently biological
>research on the long-term effects on microgravity, which means loading
>rats into a centrifuge.  I submit that that can be done with a conveyor
>belt.  Sure, you need people to (dextrously) put the station together
>(or at least to install the rats in their cages), but after that its all
>BF Skinner (push the bar, get the food pellet, wait for the scalpel).

Microgravity biology most certainly is not the only thing that's
going to be done on Freedom 8-)  There's going to be space
and earth science (both with maintenance needs), materials science,
(with maintenance and operational needs) and biology with maintenance
and operational tending needs.

Let's look at Bio first.  Fine control teleoperation isn't very advanced;
I'm not aware of anyone who's done even simple surgery or inserting an
IV line or needs via telepresense.  I'm not saying it's impossible, but
we need a much higher degree of sensors, feedback, mobility, and
reaction time than what we have now.  One hand gloved three-fingered
hand and fuzzy monocular vision aren't all that useful.  Especially
with a potential six second delay in the loop, much smaller than the average
rat's squirm time.  (Why six seconds?  Well, once to TRDSS, once down to
TDRSS ground station, once back to geostationary to a comsat and down
again to NASA's center (Johnson?) who's doing station ops, and then a
second while a NASA computer integrity checks the data.  And then back
again with your response 8-) (I agree that the delay is too long.
Talk to the NASA Freedom telerobotics people about it.  The people
at Rice and UTexas and Johnson who are playing with a prototype
system agree but haven't been able to shorten that delay yet).

And then there are the maintenance problems.

Ok.  Ignoring other detail problems, here's a neat experiment to try and
see how difficult this can be.  Imagine that you've got a bundle of
wires that's worked loose and needs to be tied to a structural member.
Simulate this with two sticks about 2 feet (0.6 meter) long.  Get some
friends to hold these.  Now, you have some string with which you're
going to attach the cable bundle to the structure.  Take one hand
(this is a one armed robot) and put three gloves on it (or one ski glove).
Lie down on your back, have your friends hold the two sticks nearly
at your full arms reach.  Tie the string around the two objects
with one hand and two fingers.  Oh, and you can't cheat and use gravity.
Your friends have to enforce not letting you use gravity.
If you can't do this (hint: about one person in three could last
time I saw the demo) try with five fingers.  Or with two hands with
two fingers each.

A more accurate version allows two arms with two fingers, but blindfolds
the person who's holding (someone else has to teleoperate him).

The scary part is that EVA ops aren't much easier.  But in a pressurized
environment, people have a lot of advantages.

>>Or are you just trying to start a flamewar?
>Well, I'm certainly getting sick of some of the topics that have been
>battered to death here.  We seem to be going round and round on how to
>supply SSF or get the crew away when things fall apart.  I'd still like
>to see a discussion of *why*, in the present scheme of things, people are
>necessary for Earth orbit operations.  Seems to me you could (in the spirit
>of many discussions in this group) free up a lot of $ (possibly some small
>fraction of which could be used to improve the dexterity of robots.)  This
>is, after all, sci.space, not sci.humansinspace.waiting4Soyuz

People are necessary because space is a hard plase to work.  Things
don't work the same there, in ways that we don't totally understand yet.
Robots, especially ones at a distance, are hard to use for aforementione
reasons.  They also break.

We can't get away (now) with just using robots in space.  We also can't
do as much with just people as you can with both people and robots.
If you think that space operations are going to increase in general,
it might be wise to fund both EVA and robotics more than they're funded
now.  Both have a long ways to go, and which one is best in the end
I cannot predict 8-)


   -george william herbert
   [e-mail]  [e-mail]