The human brain, through a process we loosely call thinking, integrates data collection, pattern recognition, object classification, and memory to obtain a higher understanding of what action needs to be taken and promptly takes action to respond to a threat or an opportunity. To mine the night sky effectively in the new era of time domain astronomy, we must construct “thinking” robotic telescopes. These next-generation robotic systems will not only have to be autonomous, but they must also have a continuously evolving knowledge of normal behavior and be capable of recognizing subtle anomalies in vast amounts of data. They must respond in real time by formulating queries and priorities, by commanding follow-up observations, and by learning to optimize the response. Just as the Internet and search engines have revolutionized the way we search for and collect information, discovery in time domain astronomy will rely on advanced data mining, visualization tools, and smart algorithms to automate the extraction of knowledge from the observations in real time.

To construct thinking telescopes for astronomy, we must develop and integrate technology in three key areas:

• Distributed networks of robotic monitoring and response telescopes
• Machine learning techniques for automated knowledge extraction in
  real time.
• Virtual observatories employing advanced database technology that provide context information in real time

	Hardware The initial hardware is designed around the concepts learned in developing RAPTOR. The system will start off using a 16" Optical Tube Assembly mounted on a fast-slewing mount. Eventually, the system will be developed into a series of four co-aligned 16" OTAs (each dedicated to a separate color filter) on a single mount and a 16-camera mosaic array on a single mount. More >>
	Database The Database will grow out of the SkyDOT project. An evolving database will sit on spinning disk and will be available for access automatically from the system and to outside user queries. More >>
	Machine Learning Advanced techniques of pattern recognition will examine the database at intervals, classifying behaviors in the time domain and searching for anomalous astronomical activity. Algorithms will be able to remove artifacts from the data. More >>
	Communications The Thinking Telescope will rely on the TALONS system used by the RAPTOR telescopes to provide communication between separate telescopes in the project and to outside resources. Additionally, outside users will be able to tap into alerts or targets of opportunity generated by the system. More >>