Goals and Introduction
The Erdos project provides software, curricular, and hardware-based information about using iRobot's Roomba vacuum cleaning robots in education, especially undergraduate education.
This material is based upon work supported by the National Science Foundation under Grants No. 0411176 and 0536173.
Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.
In summer 2009, we are working toward wireless, vision-based robots.
Here is a picture, as of 06/03/09 -- yes, it runs and sends data back and forth...!
But why? We draw inspiration from the wonderful Fluke bluetooth controller and camera for the Scribbler robot, all as part of the IPRE effort led by Doug Blank. The advantage of the system pictured above is its higher bandwidth. The disadvantage is cost: whereas the complete IPRE robot is about $150, the one above is about $350. We suspect that these systems will converge to the higher bandwidth and lower price in only a few years.
In 2008, we experimented with the IPRE Fluke/Scribbler as a research platform or a basis for a challenging undergraduate robotics course. In particular, we implemented FastSLAM, which includes Monte Carlo Localization as a subroutine, on the scribbler:
Complete details on this work by Elaine Shaver HMC '09 appear at this link.
In addition, we have long been interested in leveraging iRobot's platforms for the same purpose,
with laptops on-board. One success in summer 2008 was an OLPC-based FastSLAM-running Create:
All of the code for all of these platforms, including a pure-Python implementation of FastSLAM along with a visualizer, is available from our summer 2008 wiki pages.
Pre-2008 ERDos work is archived at this link.
Last updated June, 2009 by Z Dodds