NASA Logo, National Aeronautics and Space Administration
Intelligent Systems Division Banner

Divers Use Mobile Agents to Develop Operational Procedures for Lunar Missions

The NASA Ames Mobile Agents team collaborated with Smithsonian Institution scientists in a series of underwater extravehicular activities (EVAs) at the Smithsonian Institution’s Carrie Bow Cay Field Station in Belize November 5-19, 2008. Using an ultrasonic communication system, divers communicated with each other, with support personnel in a boat topside, and with a computer system. Building on related experiments in Hawaii and New Mexico this past summer, the expedition showed that Mobile Agents is a useful “e-science” tool for explorers to document observations, while guiding them in their work.

Two experiments were performed: 1) A standardized undersea reef survey called a “transect” and 2) a replication of Apollo 12’s EVA #2. Both experiments involved formalizing the underwater operations in protocols that specified roles, tasks, locations, and schedule. The EVA protocol was represented as a Mobile Agents activity plan, by which the computer system could guide the undersea work and automatically record the association of data, tasks, and locations. The plan was rehearsed on land and then again underwater to establish smooth operations.

In the configuration found to be most practical, given the distortion of undersea speech, divers’ voice commands were repeated by a topside support operator to the Mobile Agents system (in a single-user configuration called iMAS). The software agents provided information and feedback directly to the divers through the communications system. Each diver wore a full-face diving mask with speakers and microphone connected to an Aquacom transducer. A surface station broadcast their communications, which were recorded. A third transducer topside connected to the iMAS computer so the software agents could speak to the divers on a voice loop. Undersea operations were recorded in high-definition video, thus capturing the full interactions among divers and computer agents. To enable marking locations of samples, photographs, and voice notes, a buoy with a GPS transmitter was towed by a diver, and this data was registered by the iMAS system.

The reef transect experiment included placing a grid and taking photographs over five meters and was completed in a one-hour dive. For the Apollo 12 EVA #2 experiment, divers wore extra weight to simulate lunar gravity (1/6 g), and they walked on a sandy bottom. The experiment replicated the layout of workstations and number of observations (photographs, samples) in the historical EVA, and by virtue of the limited air supply and decompression constraints provided a partial analog to lunar EVA logistics. This protocol enables comparing work efficiency and operations concepts across diverse lunar, volcanic, and undersea environments. Eight dives were required to replicate the original four-hour Apollo 12 EVA #2, involving walking 1.45 km and taking 259 photographs and sampling at 17 workstations. Preliminary analysis suggests that future work should focus on tools for creating plans and specifying voice command options in the field, plus consolidating, sharing, and analyzing data after EVAs.

BACKGROUND: NASA’s Moon and Mars Analogue Mission Activities (MMAMA) program aims to develop operation concepts for surface science scenarios relevant to the preparation of planned human missions to the Moon and Mars. The current project “Computer-Assisted Surface Science Scenarios to Develop Operational Procedures for Manned Lunar and Martian Missions Using the Individual Mobile Agents System” (PI: Brent Garry, Smithsonian) collects science data related to geologic research on lunar and Martian lava analog flows and establishes training scenarios using technology to assist Astronauts on an EVA.

Using the method of empirical requirements analysis, this investigation has three overarching questions:

  • How can geology be documented efficiently and effectively using computer technology?
  • What science-based field exercises can be developed to train Constellation astronauts?
  • How do simulated EVAs at three analog field sites compare to actual Lunar EVAs?

The iMAS system tracks the scientist’s progress through an EVA, makes predictions about ability to complete an EVA on schedule, provides spoken alerts when parameters exceed nominal thresholds, answers spoken queries about status, and records data for off-line analysis. iMAS is designed to increase astronaut self-sufficiency during an EVA, while improving safety and efficiency through context-sensitive monitoring, advising, and logging.

The Mobile Agents team has been developing agent-based systems for surface exploration since 2000 and has been testing these systems in authentic field conditions involving a variety of instruments and robots. iMAS is currently used in POGO experiments at JSC. The full networked version of Mobile Agents is used in spaceflight operations at JSC for automating routine aspects of file transfers between the International Space Station and ground support.

COLLABORATORS: Brent Garry and Laurie Penland (Smithsonian Institution); Randy Berthold (ARC Diving Safety Officer; UAS Project Office); John Dowding (ARC/TI; UC Santa Cruz), and Ron van Hoof (ARC/TI; Perot Systems)

NASA PROGRAM FUNDING: This project was funded in part by NASA’s MMAMA Program and an Intercenter Task Agreement, provided by The Biomedical Research & Countermeasures Projects Branch of the Human Adaptation and Countermeasures Division (HACD) at JSC.

Contact: Bill Clancey, Maarten Sierhuis

02/12/2009

+NASA Home

+Ames Home

Divers Use Mobile Agents to Develop Operational Procedures for Lunar Missions
+ Home + Organization + News + Research Areas + Publications + Software + Internal Systems Help
First Gov logo
NASA Logo - nasa.gov