John and Paul and I are here in Pittsburgh, PA, visiting Jessica
during CMU’s Family Weekend. Among other activities, we spent
several hours in Baker Hall watching a
live feed of the
2007 DARPA Urban Challenge. CMU’s robot car
Tartan Racing Boss
(#19) came in just behind Stanford U’s bot “Junior” (#3)
over the finish line in Victorville, CA. However, the final score
is not just based on speed completing the 6-parts of each of the 3
missions. We will know at 11 am tomorrow which team won. Two years
ago in the DARPA Grand Challenge, Stanford was first and CMU
second.
The Baker Hall auditorium was full of students and parents and faculty,
many wearing CMU Robotics shirts. Every time “The Boss” was mentioned
or shown on the big screen, there were cheers. Of the
11 finalists, 6 were expected to finish but only 3 had actually
crossed the line when we went off to find dinner.
The commentators were very clear on the differences they saw between the
6 final race contestants. One said that it was easy to forget that the
bots were empty – their behavior was so human, they didn’t seem
like machines. Each of the cars developed nicknames in addition
to their numbers and official names. Some of the remarks we heard
from the commentators:
- #3 Stanford University’s “Junior” – clean behavior, just
right - #19 CMU’s “The Boss” Tartan Racing – crisp and flashy, brash,
clean and effective - #32 Virginia Tech’s “Odin” Victor Tango – a sleeper,
gentler - #79 MIT’s “Talos” – a little spastic, agressive, a bully, lurching
- #26 Cornell’s “Skynet” – crisp and flashy
- #74 University of Pennsylvania and Lehigh University’s
“Little Ben” Ben Franklin – slow, sluggish, careful
Each of the car development teams had different objectives in
creating their bot. None of the cars created primarily for military
use made it into the final six. CMU’s team description includes: “General Motors, the world’s largest automaker, is aboard for the future of driver safety. CAT, the world’s largest equipment producer, is committed to the future of worksite automation. CMU is driven to push the technical limits of driverless machines.”
The race commentators tried to be objective; however, it felt like they
were most impressed by the Stanford and CMU bots, most affectionate
toward Little Ben, and just irritated by the pushy MIT bot with its
40 processors. The MIT car always went as fast as it could but had
at least two accidents (one with the Cornell car), which lead to
the comment that it was so smart
it was confusing itself. We had a fun time watching the race.
KatysBlog 