Control Engineering

Paul Grayson

Team #02 Albertabot
June 29, 2007

Team #02 on my list is ALBERTABOT.   Two young men from Fredericton Canada told DARPA "Our robot entry seeks to create a low cost and reliable navigation system using only passive computer vision to complement GPS data.  We are attempting to eliminate the need for specialized active sensors such as laser scanners, LIDAR, RADAR and rely solely on computer vision to model and navigate in environment.  Our robot will sense the environment though real time vision processing of 12-52 synchronized video cameras mounted around the robot.  Our goal is to create a practical and low cost environment sensing system that can be easily added to the interior of a vehicle's windows or on roof and bumper racks.  Our team will focus on the sensor and navigation, and plans to either use professional vehicle modifications by 3rd party firms specializing in conversion for handicapped drivers or will purchase an already computer controlled vehicle."

While it sounds like they had a very good plan, something has happened to them.  Their website http://www.albertabot.org is not working and while my e-mail to them at info@albertabot.org has not bounced, it also has not been answered.  There has also been no reply from the editors of their home-town newspapers.  If you have any information on ALBERTABOT or how to get in touch with them let me know.

In the past when a team goes silent, like this one has, they have either made such remarkable progress that they have been hired away from the race and now are sequestered employees of a large commercial firm generating trade secrets for their employer or they ran into a road block they couldn't get past and gave up. 

Many teams are formed to make use of a single brilliant idea but lack the resources to be good in all areas necessary to complete the entire project.  This does not diminish the brilliance of a team's idea or the strategic value of an unusual approach.  If I can I would like to document these and share them with you here.  Often there is more to be learned from something that didn't work than something that did.  I am curious about what can be learned from ALBERTABOT and the other 35 teams that have been de-listed by DARPA.

Here at AMERICAN INDUSTRIAL MAGIC the Friday support staff work-bee made great progress organizing the things donated to the team.  Our way to establish order and make some sense out of the huge number of items donated to our team has been to arranged it all as a typical hardware store is organized.  We are arranging our inventory into departments, on shelves, labeling the shelves with hardware-store-like shelf tags.  BJ is in charge of making the map or floor plan of where the various departments are located such as Computers, Motors, Sheet Metal, Bolts, Sign Making, Plumbing, Paint, etc.  To exercise the new way of doing things, every time a volunteer asks me where something is I send them to BJ.  We are also "working out of the tool box" which means instead of laying a tool down you put it back where it belongs.  At the end of the day to clean up you have only one tool to put away, the one in your hand.  With large numbers of people using the same set of tools and the chaos that surrounds building a prototype it is the only way to keep from loosing track of the tools.

Work continues on our Automatic Guided Vehicle AGV WENDY DARLING.  We are considering the sheet metal work that it will take to lower the fenders.  This along with the Texas bumpers would give our M215 a low-rider appearance.  My team members and I are working on this project for what we can learn about making US Army supply trucks driverless and the lives that driverless systems can save.  If you want to make a difference, one way to do it is to help convert AIM's Army truck in your spare time, contact me at pgrayson@aimagic.org  to get started.

Posted by Paul Grayson on June 29, 2007 | Comments (0)