NetNews Usenet Archive 1993 #1/NN_1993_1.iso/spool/sci/space/18653

Internet Message Format | 1993-01-04 | 11.1 KB

Xref: sparky sci.space:18653 comp.robotics:2759
Newsgroups: sci.space,comp.robotics
Path: sparky!uunet!elroy.jpl.nasa.gov!kelvin.jpl.nasa.gov!baalke
From: [e-mail] (Ron Baalke)
Subject: Dante Advisory #5
Message-ID: <[e-mail]>
News-Software: VAX/VMS VNEWS 1.41
Sender: [e-mail] (Usenet)
Nntp-Posting-Host: kelvin.jpl.nasa.gov
Organization: Jet Propulsion Laboratory
Date: Mon, 4 Jan 1993 16:38:00 GMT
Lines: 255

Charles Redmond
Goddard Space Flight Center
Greenbelt, Md.                        Jan 2, 199
(Phone:  [phone])

DANTE ADVISORY #5
(all other things being, equal, the FINAL Dante Advisory)

At 5:10 pm Eastern Standard Time Saturday, January 2, the
Erebus project team located on the ice at the foot of the
Mt. Erebus volcano in Antarctica called off any further
exploration by the 8-legged rappelling robot Dante because
of a physical break in the fiber optical cable which
connects the robot with the computers providing its machine
intelligence.

The team reported the results of a day's troubleshooting
during a one-hour-long video conference which included their
colleagues located at the remote robot control site at the
Goddard Space Flight Center in Greenbelt, Md.  The
conference was called following what had been a series of
seemingly minor problems which had cropped up during the
past two days in attempts to deploy the robot down the 750-
foot deep, nearly vertical incline, from the rim of the
continuously active volcano to the lava lake below.

At the time the team called off any further mission
operation, the robot Dante remained suspended approximately
21 feet below the rim of the crater having been successfully
launched yesterday.  Dante had moved the 21 feet on its own
during initial testing of the robot control mechanisms and
had been halted in its further progress by what, at the
time, seemed to be computer network problems.

The team reported that further trouble-shooting of the
computer network problem disclosed kinks in the fiber
optical cable which connects the sensors and motor
mechanisms of the robot with the computer systems which
provide Dante with a depth perception capability and a self-
navigation capability.  The team, in examining the fiber
optical cable kinks, caused a break which stopped all
communications between the robot and the controlling
computers.  In this passage, transmitted from Antarctica
early this morning Eastern time, Dave Lavery, the project
supervisor and the NASA telerobotic program manager, reports
how this situation arose:

"By late in the afternoon, the problem had been isolated to
the fiber optic cable which stretches between the robot and
the control station two kilometers away.  It was found that
the passive deployment system which releases the fiber from
the robot as it walks had formed multiple kinks in the fiber
which had reduced data communications to the robot.  At 1955
hours, while removing the fiber optic cable from the
deployment mechanism, the fiber was severed entirely,
cutting off all communications with the robot.  Without
integrity in this cable, the robot cannot operate."

During the conference today, the Erebus project team located
in Antarctica indicated they have a rescue plan for the
robot Dante which the team simulated in practice sessions
held in Pittsburgh before their arrival in the Antarctic on
Dec. 15.  This rescue plan calls for using the Dante robot
carrier Geryon to hoist the robot up from its current
position to the top of the volcano rim.  From there, the
team will place Dante on Geryon and then move back down the
approximately 1.5 miles from the rim to the base camp.

Once at the base camp, Dante and Geryon will be disassembled
and placed into the shipping crates which were used to get
them from Pittsburgh to the Antarctic.  The rest of the base
camp will be similarly broken down and prepared for shipment
back through McMurdo Sound station.  The team estimated that
the rescue of the robot to the rim position could take from
one to two days beginning tomorrow, Sunday, Jan. 3 Eastern
Time.  The trip down the mountain to the base camp could be
done in one or two days, depending on local weather
conditions.

The team must also break down the communications gear and
antenna which allowed for transmission of video and computer
data through the Tracking and Data Relay Satellite and the
computers which are presently located in the base camp hut.
This process could take as much as five days.

The team reported that one of the considerations in their
decision to call the demonstration project off at this point
was the time period remaining until they must leave the side
of the volcano.  By January 15 the team has to be back at
McMurdo station because of impending ice-up of the Ross Sea
and a worsening of general weather conditions.  This time
constraint has existed all along but became of higher
importance when the break in the fiber optical cable
occurred.

The team indicated they had contacted McMurdo Station and
determined that a replacement fiber cable was not available,
nor were connectors for this type of cable which might have
allowed for a repair operation of this cable.  The
supporting team at the Goddard payload operations control
center was able to find a supplier for identical cable but
could not guarantee delivery of that cable to the Antarctic
in time to make the repair, perform the rest of the descent
down into the volcano and still get the robot back up to the
top and pack everything up in time to depart on January 15.

Dave Lavery said during the conference that the team
considered the project an "unqualified success" in one of
the three primary objectives, that of testing the remote
control of a robot.  The Carnegie-Mellon University team
located at Goddard had successfully commanded the Dante
robot the previous day while the robot was poised at the rim
awaiting its deployment into the volcano crater.

The robot demonstration project had three objectives: to
test telerobotic capabilities; to test the use of such
sophisticated hardware in a very harsh and demanding
environment; and to test the use of advanced computer
programs which would enable machines such as the Dante robot
to act under a form of machine intelligence.  According to
Lavery, the first two objectives of this experiment were
met.  The robot never got to a point where it was under
operation of its own autonomous control systems.

NASA and the National Science Foundation undertook this
demonstration project to develop technology  and
telecommunications capabilities which NASA could use in
future explorations of the Moon or Mars and which the NSF
might apply to its ongoing research activities in the
Antarctic.

Part of the test involved transferring control of the robot
from the Mt. Erebus team to team members  located at the
Goddard payload control center.  This  portion of the
project tested the "telepresence"  capabilities of such
robots for future NASA exploration  missions and is the
portion of the demonstration which was tested and called an
unqualified success.

Carnegie-Mellon University and the New  Mexico Institute of
Mining and Technology are partners with  NASA and the NSF as
robotics and volcano experimenters for  this project.  Team
members from Carnegie-Mellon were located both in the
Antarctic and at Goddard.  New Mexico Tech members were
located at the Mt. Erebus portion.  The robot included six
different sensors which were to have provided significant
and new date about the physical and chemical composition of
gasses and aerosols being released into the atmosphere by
the Mt. Erebus volcano.

The following sets of quotations come from the video
conference which began at 5:10 pm EST Saturday, Jan. 2, and
represent the comments from the team leader, Dave Lavery,
and the two co-principal investigators of this project --
Professor Phillip Kyle, representing the science side of the
project and William "Red" Whittaker, representing the
robotic side of the project.

David Lavery, NASA Telerobotics program manager:

"There is obvious disappointment over what is a component
failure, but we're proud of what we've accomplished.  We've
gone further than anyone said was possible and much farther
than anyone said we would get.

"We've made tremendous progress and compressed five years of
work into one year.

"The mission is an unqualified success in terms of the
telerobotic aspects.  The robot works.  The prototypes are
worthy contenders for inclusion in any further planetary
exploration.  The aspects of sending autonomous robots on
planetary exploration mission has been proven as well.

"We were given in January 1992, when we started this
project, odds of about 20 percent of probably successful
completion.  This was a very, very risky venture.  This has
been a true adventure.  Our spirits remain undaunted."


Phillip Kyle, New Mexico Inst. of Mining and Tech. and co-
principal investigator:

"This has whetted the appetite of volcanologists everywhere.
This was the ultimate challenge.  There is no more nastier
volcano.  It has been a remarkable achievement.

"I think of this as one small step and expect we will be
seeing robots exploring inside volcanoes around the world in
a few years.

"This demonstration project showed we could integrate
science into a robot."


William "Red" Whittaker, Carnegie-Mellon Univ. and co-
principal investigator:

"We are calling it off because of a fiber optical cable
break for which we have no workaround.

"The achievement has been tremendous.  We've written over
150,000 lines of new software code.  We've used the ground
station for the first time.

"This has been a dream program.  It was inspirational work
and I'm really proud of what we've done.  We've made a bold
leap and pushed this technology from the laboratory into the
real world.

"This is just one example of what must be thousands of
similar applications.  This was the real thing.  This robot
left the laboratory and had a real job with a real science
customer.

"We've learned a great deal during this experience in
Antarctica, especially about self-reliance and ingenuity.

"This has really been a dream year.  It called for the best
in a lot of people"



Participants in the conference from Antarctica
were the following individuals:

David Lavery, NASA Headquarters
Steve Thompson, NASA Goddard Space Flight Center
William Whittaker, Carnegie-Mellon University
Eric Hoffman, CMU
Dan Christian, CMU
David Wettergreen, CMU
Phillip Kyle, New Mexico Institute of Mining and Technology
Nelia Dunbar, NMIMT
Bill Macintosh, NMIMT
Sara Krall, Antarctic Support Associates-National Science
Foundation
Ken Sims, ASA-NSF

Participants located in the temporary robot control center
at Goddard during this conference:

James Osborn, Carnegie-Mellon University
Paul Keller, CMU
Jay West, CMU
Chris Fedor, CMU
Jim Christo, NASA Goddard Space Flight Center
Bob Rodriguez, NASA GSFC Bendix Field Engineering Company
     ___    _____     ___
    /_ /|  /____/ \  /_ /|     Ron Baalke         | [e-mail]
    | | | |  __ \ /| | | |     Jet Propulsion Lab |
 ___| | | | |__) |/  | | |__   M/S 525-3684 Telos | Choose a job you love, and
/___| | | |  ___/    | |/__ /| Pasadena, CA 91109 | you'll never have to work
|_____|/  |_|/       |_____|/                     | a day in your life.