Projects

Minimal Academic

This site, written using Hugo and inspired by Academic and Minimal Mistakes.

Abstract: Robot Self-Recognition Using Conditional Probability–Based Contingency

This abstract was accepted to the 2006 AAAI Conference that was held in Boston. It was presented as a poster presentation. If you would like to cite this article, see below.

Icarus

Naturalistic flight in virtual environments

Icarus

Since the beginning of time humans have watched birds soar through the air and have dreamed of doing the same. However, the laws of physics have conspired to make that dream difficult, if not impossible, for a single human under his or her own power. Fortunately, we are not constrained by physical law in virtual worlds so we can come close to fulfilling this dream for many. Our primary objective for this project is to allow a person to navigate a virtual world as if they are flying by flapping their arms like wings.

Narcissus

Robot Self-Recognition Using Conditional Probability–Based Contingency

External Project

An example of linking directly to an external project website using `external_link`.

Results and Data Description of Data Format Each of the raw data files below contains a number of different record types. One record per line. The fields of each record are separated by tabs. Ignore lines that don’t begin with one of the record type identifiers: ENTITY, ACTION, PERCEPT, TIME. The ENTITY record has the following fields: Global timestamp (double) Entity ID number (integer) Number of correlated actions (integer) Total number of actions (integer) Number of correlated perceptions (integer) Total number of perceptions (integer) Threshold of indicies to consider ENTITY self (double) Sufficiency index (double) Necessity index (double) The ACTION record has the following fields:

Videos Click on the thumbnails below to download the videos. Each of the videos is between 30 and 40 MB in size. The upper left image shows the raw video input with the sufficiency index (SI) and necessity index (NI) of each object overlaid on top of the image. The upper right image shows the color segmentation. The lower left image shows the motion segmentation. And finally, the lower right image is the camera view after we’ve applied the blur filter (to remove many of the noise artifacts).