Abstract Details

Title: 3 Things Your Robot Should Know

Authors: Rob Seaman

Abstract Submitted by: Rob Seaman

Institute/Affiliation: National Optical Astronomy Observatory

Abstract Type: talk

Abstract Information:

Observational astronomy is an ancient pursuit, following an exponential trend since Galileo of ever greater capabilities. Or rather, a series of exponentials of growing apertures, the opening of new windows on the electromagnetic universe, photographic to photoelectric to digital instrumentation, balloons and spacecraft rising above atmospheric murk, to more avant-garde empirical endeavors such as multi-messenger facilities no longer recognizable as telescopes. At the same time the operational logistics of astronomy have been evolving. New observing modes and new scheduling paradigms have multiplied. Multi-year surveys supply last minute targets-of-opportunity. Astronomical software has mutated and ramified a millionfold from FORTH and FORTRAN to the Virtual Observatory (http://www.usvao.org). The beat goes on, but with accelerating syncopation.

The majority of astronomical data nonetheless continues to be collected more-or-less directly by humans. Perhaps not the bottleneck, we are often the cork. The next revolution (evidenced by this conference) is toward autonomous technologies such as telepresence, robotic control, and complex telescope networks - power tools for observers. These technologies are intrinsically complex and interdependent, emergent in nature and benefit from network externality - that is, the scientific value will increase with the number of interconnected nodes.

Such a wholesale cybernetic re-imagining of astronomy will only succeed if layered on an upgraded foundation. Three key elements of this new infrastructure will be discussed, particularly in the context of the astronomical time domain. These are 1) standardized transient celestial event messaging (http://voevent.org), 2) the efficient representation of data via compression technologies (http://heasarc.gsfc.nasa.gov/fitsio/fpack), and 3) traceability in timekeeping signals - NTP, GPS, and the uncertain future of UTC (http://ucolick.org/~sla/leapsecs).