[sci.space] Mars rovers

leonard@bucket.UUCP (Leonard Erickson) (12/19/89)

[long argument for getting around the time-lag and AI problems of
a Mars rover by slowing it down deleted]

Fine, the rover is creeping along at the feet feet per hour that you
suggest is safe. Now something unexpected happens. The surface shifts.
(loose rocks hidden by dust, a dust pocket, whatever]

The rover *must* make a real time response. If it isn't autonomous, it
can't. You can handle some things be designing in recovery gear (like
those levers for righting it that you mentioned) but the odds are
against you. What do you do it it manages to "high center" itself?

or if it is traversing a slope and starts to slip?

I'll agree that your rover will work on Mars the day after it has
operated for a year with only a time delayed radio link for support
in Antarica....  (if it wasn't for rain and floods, I'd have suggested
a desert)
-- 
Leonard Erickson		...!tektronix!reed!percival!bucket!leonard
CIS: [70465,203]
"I'm all in favor of keeping dangerous weapons out of the hands of fools.
Let's start with typewriters." -- Solomon Short

leonard@bucket.UUCP (Leonard Erickson) (12/19/89)

stolfi@jumbo.dec.com (Jorge Stolfi) writes:

>I find it amazing that people can argue so vehemently that teleoperated
>rovers Can't Possibly Work, as if the Russian Lunokhod rover had never
>existed.  Sure, it was only the Moon, not Mars; but are we to conclude
>that 10 years from now the US will not be able to even *try* improving
>on what the Russians did 20 years ago?  

But this is *not* the same thing. Engineers seem fairly confident that
teleoperated equipment can be done over the Earth-Moon distance with
it's 2-second lag. And much better than poor old Lunokhod.

But the difference between 2 seconds and 1200seconds is a quite different
matter. Even if you assume that the difficulty goes up linearly with the
lag you are talking about something that is *at a minimum* 600 times harder!
Personally, I suspect it goes up with the square.

Try this. All you have to do is tell the manipulator arm to pick up a rock
sample a foot or so more or less in front of the rover. So you tell it to
extend the arm out 12 inches. 20 minutes later you get to find out what your
next positioning command is... 

Sure, you can add sensor to let the arm track in on an object. But if you
do THEN IT ISN'T TELEOPERATED!! It's at least semi-autonomous. And you still
have to figure out what to do if it is homing in on the *wrong* rock...
-- 
Leonard Erickson		...!tektronix!reed!percival!bucket!leonard
CIS: [70465,203]
"I'm all in favor of keeping dangerous weapons out of the hands of fools.
Let's start with typewriters." -- Solomon Short