Centrifuge coating for low-waste solution processing of transparent nanostructured electrodes

Centrifuge coating was implemented to fabricate nanostructured conductive layers through solution processing at room temperature. This coating procedure allows fast evaporation, thereby fixing the nanomaterials in their dispersed state onto a substrate by the centrifuge action. Material wastes were minimized by mitigating the effects of particle reaggregation. Using this method, we fabricate single-wall nanotube coatings on different substrates such as polyethylene terephthalate, polydimethylsiloxane, and an acrylic elastomer with no prior surface modification of the substrate. The effects of the choice of solvents on the morphology and subsequent performance of the coating network are studied. © 2002-2012 IEEE.

Fast Hands-free Writing by Gaze Direction

We describe a method for text entry based on inverse arithmetic coding that relies on gaze direction and which is faster and more accurate than using an on-screen keyboard. These benefits are derived from two innovations: the writing task is matched to the capabilities of the eye, and a language model is used to make predictable words and phrases easier to write.

Suspended multiwalled carbon nanotubes as self-aligned evaporation masks

We describe the nanofabrication study of self-aligned electrodes on suspended multiwalled carbon nanotube structures. When metal is deposited on a suspended multiwalled carbon nanotube structure, the nanotube acts as an evaporation mask, resulting in the formation of discontinuous electrodes. The metal deposits on the nanotubes are removed with lift-off. Using Al sacrificial layers, it was possible to fabricate self-aligned contact electrodes and control electrodes nanometers from the suspended carbon nanotubes with a single lithography step. It was also shown that the fabrication technique may also be used to form nano-gapped contact electrodes. The technique should prove useful for the fabrication of nano-electromechanical systems.

Probabilistic inference for fast learning in control

A novel framework is provided for very fast model-based reinforcement learning in continuous state and action spaces. It requires probabilistic models that explicitly characterize their levels of condence. Within the framework, exible, non-parametric models are used to describe the world based on previously collected experience. It demonstrates learning on the cart-pole problem in a setting where very limited prior knowledge about the task has been provided. Learning progressed rapidly, and a good policy found after only a small number of iterations.