To move scientific inquiry into a regime where optimal use can be made of the remarkable instruments and data analysis techniques that are currently being developed. To accomplish this, the instrument and the data analysis algorithm must act as a single system: an Intelligent Symbiotic Instrument-Analysis System (ISIAS).

To unite the instrument with its analysis system so that they can interact symbiotically will require the analysis system to perform both inference and inquiry. The inference engine will not only make inferences from data, but will also model the instrument during data acquisition. This will be performed using Bayesian machine learning methods. In addition, the inquiry engine, built relying on a new area of mathematics called the inquiry calculus, will allow the analysis system to interact symbiotically with the instrument by requesting measurements that will significantly improve the analysis results, thus empowering the ISIAS with the ability to perform active exploration and experimentation during sampling to ensure adequate coverage. In this way the ISIAS will be able to focus on relevant, interesting and informative measurements and will lay the groundworkw for developing opportunistic and contingent science operations on remote platforms. This approach to intelligent systems will be demonstrated on a Department of Defense (USNO) instrument called the dispersed Fourier Transform Spectrometer (dFTS) developed by Arsen Hajian, and on the 3D Laser Scanner developed by Joseph Lavelle's Group at NASA Ames to scan Space Shuttle Tiles for damage.

Results
At this point we have developed a basic inference engine for the dFTS, in addition to developing the mathematics underlying the inquiry engine so that calculations can be computed with questions.

Testing limitations
Entropy calculations, which are required by this project, are difficult to perform accurately from data.

Availability of documentation
Five papers on the novel mathematics underlying the inquiry engine have been published (three key papers are listed below)

• Knuth K.H. 2004. Lattice duality: The origin of probability and entropy. In press: Neurocomputing.
• Knuth K.H. 2003. Intelligent machines in the 21st century: Automating the processes of inference and inquiry. Phil. Trans. Roy. Soc. Lond. A, Millennium Issue. 361(1813):2859-73.
• Knuth K.H. 2002. What is a question? In: C. Williams (ed.), Bayesian Inference and Maximum Entropy Methods in Science and Engineering, Moscow ID 2002, AIP Conference Proceedings 659, American Institute of Physics, Melville NY, pp. 227-242.

This methodology will allow us to construct new computers that manipulate questions as easily as logical statements. Thus automated systems will be able to decide which questions to ask. This will result in a new generation of small, highly capable, active instruments. ISIAS will perform active exploration and experimentation during sampling ensuring adequate and efficient coverage.