Datapages, Inc.Print this page

Robotic Scouting for Human Exploration

Deans, Matthew C.1; Fong, Terry 1; Lee, Pascal 2; Hodges, Kip V.3; Helper, Mark 4; Landis, Rob 5; Riley, Steve 5; Bualat, Maria 1; Pacis, Estrellina 6; Kobayashi, Linda 1
1 Intelligent Robotics Group, NASA Ames, Moffett Field, CA.
2 Mars Institute, Moffett Field, CA.
3 School of Earth and Space Exploration, Arizona State University, Tempe, AZ.
4 Geological Sciences, University of Texas, Austin, TX. (5) Mission Operations Directorate, NASA Johnson Space Center, Houston, TX. (6) SPAWAR SSC, San Diego, CA.

From November 3-6, 2008, we conducted a full-scale test of a "robotic recon" system at NASA Ames. During this test, the K10 "Red" robot was used to establish a preliminary geologic map of an outdoor test area and to help plan a short 30-min follow-up EVA in shirtsleeves in that area. The goal of the test was to improve NASA's understanding of how robotic scouting can help plan EVAs, and also how robots might best be used to complement humans.

When humans return to the Moon around 2020, crews will initially be on the lunar surface less than 10% of the time. During the 90% of time between crew visits, however, robots will be available for surface operations under ground control. A central challenge, therefore, is to understand how robotic systems can be used to improve overall science return and mission productivity. One possible method is to use robots to perform scouting in advance of human activity.

Robotic rover scouting involves using a planetary rover to collect ground-level data. Scouting is well understood to be an essential phase of field work, particularly for geology, and can be: (1) traverse-based (examining stations along a route); (2) site-based (examining stations within a bounded area); or (3) survey-based (systematically collecting data along defined transects). Robot-mounted instruments can be used to examine the surface and subsurface at resolutions (e.g., (m to cm scale) and at viewpoints not achievable from orbit. The data can then be used to plan subsequent human or robot activity.

During our test, K10 Red carried three recon instruments: a scanning lidar (acquires 3D measurements of terrain at mm resolution), a color panorama capture system (consumer digital camera on a pan/tilt), and a high-resolution terrain imager (downward-facing consumer digital camera). We remotely operated K10 Red with a ground control team, which was located in the NASA Ames "FutureFlight Central" facility. The ground control team included a science team (strategic level planning), flight control team (tactical operations), and robot support team (diagnosis and repair).

In this paper, we summarize the objectives for the test, describe the test setup and protocol, and present results and lessons learned. In addition, we identify key issues and open questions that remain to be addressed, as well as suggest directions for future work.


AAPG Search and Discovery Article #90090©2009 AAPG Annual Convention and Exhibition, Denver, Colorado, June 7-10, 2009