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 (2-Aug-94)
Message-ID: <1994Aug3.181948.23659@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: Wed, 3 Aug 1994 18:19:48 GMT
Lines: 159

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

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

Status Report (2-Aug-94)

We came, we saw, we walked.

At the conclusion of today's activities, Dante had met and exceeded the
objectives specified for the project.  The goals of the project,
reviewed and accepted at the three NASA project reviews conducted during
the development of Dante II, have been to have the robot descend into
the crater of the volcano and to provide observations of the crater
floor back to the scientists at the Alaska Volcano Observatory.  For the
purposes of the project, successfully completing the descent was defined
as traversing more than 600 feet into the crater, and providing
observations of the crater floor was defined as continuously streaming
data from the onboard gas sensors while in proximity to one or more
active fumeroles and providing video images of areas of the crater floor
which could not be seen from the rim.

At the end of the day Dante had descended more than 660 feet into the
crater of Mt. Spurr, and had walked off the sloped wall of the crater
and onto relatively level ground in the fumerole field on the crater
floor.  This exceeded the distance goal set for the project by 10%.
Dante spent all last night (approximately 11 hours) sitting in front of
an active fumerole opening, with the on board gas sensors exposed to the
emerging gas plume for the entire time.  During today's operations the
robot explored four additional fumerole sites, including the locations
of extinct as well as active openings.  At each of these locations video
images of the fumeroles and the fumerolic encrustations surrounding the
vent opening were transmitted from Dante back to the control center for
review by AVO scientists.  In addition, Dante was able to view the area
beneath an overhanging rock bulge on the crater walls to examine terrain
which is not visible from above, as well as provide a detailed video
survey of the crater floor which was transmitted to observing geologists
at the NASA Ames Research Center.  These activities again exceeded the
performance criteria specified for the project.

In the process of preparing for the deployment of the robot into the
volcano crater, Dante was carried on a flatbed truck, left in a gravel
pit, had components moved by airplane, automobile and pickup truck,
lifted by multiple cranes, barged across the rough salt-water of Cook
Inlet, poked and prodded by photographers, stored in a power plant,
airlifted 30 miles by helicopter, left on a dirt airstrip, sat upon by
an aircraft, and had components chewed by bears.  Once inside the
crater, Dante has been required to cross terrain that was harsher than
originally expected, to walk longer distances than anticipated, been
subjected to impacts by falling rocks, been exposed to fumerolic gases
for over 36 hours, been required to walk through deep mud and snow (both
of which were encounter by the robot for the very first time during this
descent), had the vision capabilities obscured by crater plume and
clouds for long periods, and been subjected to rain which caused the
temporary loss of the laser scanner, all while remaining untouched by
human hands for the last five days of operations.  At the end of all of
this, the robot continues to perform well, and has experienced no major
system or component failures.

The last three and a half days of descent to the crater floor have given
Dante's developers an opportunity to demonstrate and evaluate a large
suite of advanced robotic technologies in one of the harshest and
unforgiving applications imaginable.  This has included the laser
scanner perception system with telepresence interface, the TCX
communications architecture, self contained computer architecture to
eliminate reliance on off-board computation links, limited
communications bandwidth control, multi-level control contexts which can
be switched in real time, field-worthy deployable systems, distributed
remote telepresence control via satellite, autonomous behavior-based
control, single-chip motor controller, robust multi-day operations, and
hardware and software for autonomous rapelling mechanisms.  Information
regarding the performance of these, and several other, component
technologies has been gathered during Dante's deployment to allow the
team to analyze the performance and effectiveness of the technologies
and their contribution toward achieving the project goals. While some of
these technologies are refinements of existing capabilities which have
been tailored to this application, most of them are unique new
technologies which have been specifically developed for this project and
which will advance the state of the art in robotics.

The robot also gathered volcanic science information and observations
from the interior of the crater.  The hydrogen sulfide and sulfur
dioxide gas sensors were continuously providing data from the moment the
robot took it's first steps on the crater rim to the present (and will
continue to do so all the way back to the rim).  During the exploration
of this portion of the crater, no sulfur compounds were detected in
either the ambient air or adjacent to the fumeroles.  The lack of
gaseous sulfur compounds is a clue that the heat source under the crater
may have moved or be completely capped (although it is possible that
these gases are being emitted in one of the regions of the crater floor
which was not explored).  It is believed that the white plumes being
emitted by the fumeroles are almost entirely water vapor.  Although the
sulfur gases were not detected at the visited sites, the visual
information provided by the robot revealed fumerolic sublimates
surrounding the vent openings which are indicative of recent activity
(ie. within the past year).  In addition, indications of continuing
hydrostatic activity beneath the crater floor was indicated by the
evidence of ongoing formation and collapse of the fumerole vents as they
migrate around the crater floor.

One of the more significant aspects of the project was the opportunity
to gather information on the effects of the human interface on the
operators over an extended period of time.  After four and a half days
of continuous operations (with at least two more to go), the operators
have gained several insights into methods for improving both the
operator interface and the capabilities of the robot which would
decrease the fatigue experienced by the operators after many hours of
activity.  This has included changes in the interface displays, as well
as the suggested addition of a walking context in between the
capabilities of the frame context and blind walking modes (which have
been the primary methods of locomotion during this descent) which would
eliminate the need for manual maintenance of body posture and attitude
during frame-based control movement.  The descent also highlighted the
need for the next version of the blind walking context (version 2.0?)
and scan-based walking which would further alleviate some of the
operator overhead associated with extended activities, and it identified
the limit where, without increasing the level of system autonomy, the
wear on human operators is too high for extended missions.

The expectation from this point is that the robot will require
approximately two days to ascend back up and out of the crater, and back
to the shipping pallet.  It is anticipated that the ascent will take
less time than the descent due to two factors: extended route planning
needs to be performed (Dante will just follow it's own footprints back
to the rim), and the autonomous walking controller will be enabled
during a greater percentage of the return walk (during the descent, the
robot walked under autonomous control for about half of the distance
covered).  This walking mode is faster than the more manually- intensive
frame context, and due to the geometry of the ascent path it can be
enabled for a greater period of time.

------------------------------------------------------------------------------
-- 
_______________________________________________________________________________
 "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
