Vandapel_1999abstr

 

THE EVA ROBOTIC ASSISTANT EXPERIMENT

ON THE HAUGHTON-MARS 99 EXPEDITION

 

Nicolas Vandapel, Pascal Lee, Dimitrios Apostolopoulos,

Geoffrey Briggs, Bill Clancey, Brian Glass,

Steve Hoffman, Kimberly Shillcutt, Maarten Sierhuis,

Mike Sims, H. Thomas and W. “Red” Whittaker

vandapel@cs.cmu.edu

 

Humans and robots will both be needed to explore and settle Mars, as the two present specific capabilities and performances that may in many instances be regarded as complementary rather than redundant. But what role should and can humans and robots play given the current status of robotic technology and optimistic timeframes for human exploration (human missions to Mars beginning within the next 2 decades)?

 

In response to the growing interest in understanding how humans and robots might work together on Mars, we plan to carry out baseline observations and field simulations in human-robot integration on the 1999 Haughton-Mars Project expedition to Devon Island, Arctic Canada. A variety of tasks will be performed, including an inventory of all activities that support field work to assess the range of possible robotic needs in field exploration, a systematic recording of the metrics of extravehicular (i.e., outdoor) activity (EVA) in those areas likely to require field robotic support, and a definition of the information technology architectures likely to be needed to achieve a well-integrated human-robot exploration system. Among the planned simulations, we will have a roboticist on an all-terrain vehicle (ATV) accompany field scientists on traverses across the Mars-like terrains of Devon Island acting as if he/she were a robotic assistant. Tasks to be performed by this EVA Robotic Assistant may range from navigation and supplies caching to sample curation and geologic interpretation. The exercise should help identify specific robotic needs of astronaut explorers on Mars and help guide future robotic research in this area.

 

            An integrated, synergistic human-robot exploration system is envisioned as a possible outcome of this research. Preliminary results from the 1999 field season are presented.