Newsgroups: comp.robotics
Path: brunix!cat.cis.Brown.EDU!agate!ames!kronos.arc.nasa.gov!doctor
From: doctor@kronos.arc.nasa.gov (Terry Fong)
Subject: Dante II - Mt. Spurr Expedition Status (28-Jul-94)
Message-ID: <1994Jul31.081805.28418@ptolemy-ethernet.arc.nasa.gov>
Summary: status information on CMU/NASA volcano robot explorer
Keywords: Dante, teleoperations, volcano, Alaska
Sender: usenet@ptolemy-ethernet.arc.nasa.gov (usenet@ptolemy.arc.nasa.gov)
Nntp-Posting-Host: tardis.arc.nasa.gov
Organization: NASA Ames Intelligent Mechanisms Group
Date: Sun, 31 Jul 1994 08:18:05 GMT
Lines: 176

The following is a status update on the Dante II project. Dante II is
an eight-legged robot (developed by the CMU Robotics Institute with
sponsorship from NASA) which is currently descending into the active
crater of Mt. Spurr, an Alaskan volcano 80 miles west of Anchorage,
Alaska. The primary objective of the Dante II project is to develop and
evaluate new techniques and technologies which can be applied to space
and planetary exploration. 

Throughout the Mt. Spurr expedition, Dante II is being remotely
operated from a base station in Anchorage using control station
software developed at CMU and NASA Ames. Live video coverage is
periodically available via NASA Select. Full mission details including
current status, real-time images/video, and technical data is available
via a WorldWideWeb server provided by the NASA Ames Intelligent
Mechanisms Group. This site can be accessed on the Internet using
information browsers such as NCSA Mosaic and the URL:

	http://maas-neotek.arc.nasa.gov/dante

Questions regarding the Dante II project may be addressed to me via email 
(terry@ptolemy.arc.nasa.gov). Reponse time may be slow.

Terry Fong
NASA Ames Research Center

------------------------------------------------------------------------------

Daily Report (28-Jul-1994)

Another 90% day.  After arriving at the rim, Shawn Fitzpatrick worked
for about 45 minutes to install the feed horn on the satellite dish
assembly and power up the transmitter.  Shortly before 11:00ADT the
satellite connection between the crater rim and the control center in
Anchorage was established and the network between the computers on
board Dante and at the control station in Anchorage was brought up.
Once this network was established, the operators began the initial
checkout of the robot, which involved evaluating the condition of the
onboard computer systems and verifying that they had survived the
transport from Anchorage to the rim.

Within thirty minutes of bringing up the link, the first video images
were received from the remote pan and tilt camera mounted on the rim
of the volcano, follow three minutes later by the first images from
the robot-mounted cameras.  Within another fifteen minutes the first
remote actuation was achieved, when the pan and tilt camera suspended
from the sensor arch was repositioned to look at the deployment team.
The initial computing system checkout was completed shortly
thereafter.  Just after noon ADT, the brakes on the legs were released
by the operators and the verification process for the sensing and
actuation systems of the robot was started.

During the course of this checkout, three issues were identified.  The
weather seal on the vertical actuator housing for leg #2 (this is the
small motor which rises above the "hip" of each leg on a stalk, and
drives the vertical motion of the leg; leg #2 is the outside right leg
on the front - downhill - side) had not functioned properly and water
had entered the actuator housing.  When the robot was powered up on
the rim, this caused a cascading set of secondary problems as the
electronics within the housing shorted out.  These issues were
eventually solved by opening and draining the actuator enclosure.
This was followed by a thorough drying with a heat gun and verifying
the electronics before closing and resealing the enclosure.  It is
suspected that the water entered the enclosure when Dante spent at the
gravel pit test site, when it was exposed to a substantial downpour
during the night.

The second issue involves the communications link from the robot to
the rim station.  At the volcano crater the network is basically an
Ethernet system within the body of the robot, which is then extended
over unshielded twisted pair embedded within the interior of the
tether out to the rim station computers.  Tutankhamen communications
boards at either end of Dante's tether convert the physical layer of
the network from coaxial cable to the twisted pair.  The Tutankhamen
on board Dante was sporadically trashing the signal and providing
erroneous data up to the rim station.  An option to replace the board
before proceeding is being discussed.

And third, the load cell on leg #2 began to give data values
completely out of bounds.  This sensor normally measures the amount of
force applied to the leg (ie. how much weight the leg is supporting).
Dante's eight legs are arranged in two set of four, known as the
"inner frame" and the "outer frame", and the robot walks by supporting
all of it's weight on one of the frames while the other frame picks
up, moves forward, sets down, and takes up the support of the body
weight.  The load cell provides data back to the control software on
board the robot and at the operators console which allows them to
determine when the leg is in contact with the ground and how much of
the body weight each leg is supporting (under nominal conditions, each
leg supports 440 pounds).  While it is possible to derive this
information algorithmically (ie. take the load values from the other
seven legs and subtract from the known weight of the vehicle, adjusted
by the weight supported by the tether), this method is somewhat slower
and would eliminate one layer of fault tolerance in the system (ie. it
would require the seven remaining load cells be perfectly functional
during the entire mission; any additional load cell failure would
remove any possibility of determining the load the first leg).

The load cell in the #2 leg was providing useless data, and could not
be reliably used by the operators or the autonomous controller.  It
was determined that it could not be repaired, and needed to be
replaced with one of the spare load cells brought from CMU.  As
mentioned, the robot is capable of walking with one of the load cells
damaged, but with a lower level of redundancy and higher computational
and operator overhead.  While this level of operations could be
tolerated once the entry into the crater was initiated, is was
determined to be unnecessarily risky to start the walking traverse
with a bad load cell which could be easily replaced.

Given these three issues, the project decided to postpone the start of
Dante's walk into the crater until tomorrow morning.  At that time the
dried actuator can be rechecked, the Tutankhamen board can be replaced
if necessary and the faulty sensor can be replaced, thus maintaining
the overall level of system redundancy as the traverse is initiated.
All other systems examined during the vehicle checkout were verified
and are believed to be fully operational.  It is expected that the
three items will be cleared fairly quickly, perhaps as quickly as
within two hours after the arrival of the team on the crater rim, and
at that point Dante will start walking into the crater.

Two other items which occurred today are of note.  The first is a
viewing of the vehicle checkout operations on the crater rim by an
audience at the National Air & Space Museum and at Ames Research
Center.  The museum displayed live video of the checkout which was
transmitted via ACTS satellite from the control center in Anchorage
direct to NASM.  From there, the video was echoed to NASA Select at
Goddard Space Flight Center and sent up to the NASA Select Satellite.
Simultaneously, the robot status information and VEVI information was
transmitted from Anchorage to Fairbanks to the ARC to NASM over the
NASA Science Internet.  The audiences at the remote operations sites
were able to view the checkout operations, and remotely operate the
pan and tilt camera at the top of Dante's sensor arch while the
checkout was proceeding.  We understand that the events were very
successful, with considerable interest and participation, and they
helped to demonstrate the multitude of technology developments which
are supporting this project (including advanced satellite
communications, advanced land-line networks, remote operations
technology, virtual reality control developments, video and voice
digitization and compression, etc) in addition to the fundamental
robotics technology and operations.

At the conclusion of the remote demonstration, the ACTS link was
disassembled and removed from the project communications configuration
as the ground station was required by another project.  For the
remainder of the Dante field deployment, all digitized video and
telemetry transmission from Anchorage to ARC and NASA Select will be
provided by the NASA Science Internet.

Finally, while all of this was going on, a minor event out at the
mountain again reminded the project team that they were working on an
active volcano in an extremely hazardous and dangerous environment.
During the day the constant rock fall from the crater walls that has
been mentioned in earlier reports continued.  As has also been
mentioned, both the robot preparation site and the path that Dante
will take to descend into the volcano are within the rock fall impact
zones (although in areas of lower rock fall density than the north and
west walls).

During the checkout operations, a small rock slide occurred on the
east wall two hundred feet above the robot preparation site.  One
small rock, about two inches in diameter, from the slide fell all the
way to the robot and struck the right outer leg tower with a very loud
"CLANG!"  Although there was no damage, it served as a reminder that
if it had been going a little faster, or been a little larger, or
fallen a few feet to either side, serious damage could have occurred
to either one of the cameras or sensors on the robot or to any of the
team members in the area at the time.  This just reinforced the need
to get the robot walking and get the team members out of the
environment as quickly and safely as is possible.

------------------------------------------------------------------------------

-- 
_______________________________________________________________________________
 "Every once in a while declare        Terry Fong <terry@ptolemy.arc.nasa.gov>
  peace. It confuses the hell out      NASA Ames, M/S 269-3, Moffett Field, CA
  of your enemies" -- Rule of Acq. #76      (415) 604-6063, (415) 604-6081 lab
