Intelligent Instruments
Published on 6 Oct 2007 at 6:24 pm.
No Comments.
Filed under Computation, Exploration, Intelligent Systems, Inventions, Lego, NXT, Programming, Research, Robotics, Technology.
One of my main research thrusts recently has been my work on desiging intelligent instruments. The idea is to develop truly autonomous scientific platforms. Such systems would be able to explore the surface of the more distant worlds such as Europa or Titan, or dive deep into the oceans on Earth to search for undersea vents or to look for undiscovered life forms.
The basic idea is that the instrument has a set of models capable of describing the environment it is exploring. The instrument uses Bayesian inference to learn from the collected data. Once the inference process is completed for the most recent data set, the instrument considers all the possible meaurements that it could take and computes the expected gain of information based on what it already knows. Once the experiment or measurement has been chosen, the instrument collects the new data. The process then repeats until the instrument learned what it was programmed to learn with an acceptable level of precision.
I have been working to develop these ideas with the help of the LEGO Mindstorms NXT system. My first prototype is almost complete, and is a robotic arm equipped with a light sensor. The objective is for the robot to characterize a white circle on a black field by estimating its position and radius with a high degree of accuracy. The robot is not allowed to scan the scene, but is instead only allowed to take point measurements. Our preliminary tests show that this is performed with extreme efficiency, and promises to dramatically save on energy, time, data storage, and data transmission.
I recently gave a talk on this project at the MaxEnt 2007 workshop in Saratoga Springs NY. Below are links to a video of the talk, the slides, and the accompanying research paper. Enjoy!
Video: Designing Intelligent Instruments, K.H. Knuth
Slides: Designing Intelligent Instruments, K.H. Knuth
Research Paper:
Knuth K.H., Erner P.M., Frasso S. 2007. Designing intelligent instruments. K.H. Knuth, A. Caticha, J.L. Center, A. Giffin, C.C. Rodriguez (eds.), Bayesian Inference and Maximum Entropy Methods in Science and Engineering, Saratoga Springs, NY, USA, 2007, AIP Conference Proceedings 954, American Institute of Physics, Melville NY, In press.
Kevin Knuth
Albany NY
