PCAR 2006 is a ‘International Symposium on Practical Cognitive Agents and Robots’ that was held very recently in Perth, Australia.
One of the invited speakers was 4-Legged League Technical Committe Advisory Member Masahiro Fujita, who also happens to be the literal god of the Sony robots Aibo and Qrio.
Our good friend Michael Quinlan and his NUBot teammates presented a paper entitled ‘Impact of Tactical Variations in the RoboCup Four-Legged League’. Also, Kai Xu, a new technical committee member this year for the 4-Legged league and captain of China’s WrightEagle, went and presented ‘Legged Robot Gait Locus Generation Based on Genetic Algorithms’. Both papers sound interesting and I’d love to get a copy of them.
To top it off, they held an ‘exhibition’ of 4-Legged teams squaring off against each other. Sadly, the reigning champion NUBots did not participate but WrightEagle, UTS Unleashed, and a new team from Australia, the UWArriors from the University of of Western Australia. If someone who attended should stumble upon this page, please let us know how the exhibition went. It’s never too early to start scouting for RoboCup 2007.
Though a far stretch from being useful in an actual game, I am still to this day impressed with these slow, ineffective behaviors. I believe this was done in early March 2006.
Considering the vision system was running the worst camera settings (everything was really blurry), and was running at about four frames a second (instead of 30 it is now), I think these simple, slow grabs are actually pretty impressive. Moreover, all the behavior code was written in C++, meaning that we had to literally turn the robot on/off every time we wanted to make a change. Ah, how crappy the process was back then.
Just announced today: The GermanOpen 2007 will be held in Hannover over April 16-21. Apparently the event will be held in ‘the world’s largest industrial fair’. That’s pretty cool. The event will be hosting lots of other leagues beyond our own, and I found this little note particularly interesting:
The RoboCupJunior competition in Hannover is restricted to German teams. They have to qualify for participation in the RoboCup German Open in regional events held in Magdeburg (February, 23rd-25th) and Vöhringen (March, 2nd-4th). Details and a call for participation for these qualifying events will be issued separately.
They are serious about their RoboCupJunior in Germany. And I’m not a hater at all–RoboCupJunior was the sole reason I stayed sane throughout the entire RoboCup 2006. Especially the Dancing.
UPDATE::link above re-linked to English version, linked in the Links section to your right as well.
Well now that I have, on behalf of the Northern Bites, under the pseudonym ‘nbites’, within the very sacred walls of the Robot Lab, uploaded TWO videos in the last TWO weeks–that kind of modernity deserves a link on the side of the this blog. So here it is, the link to the Northern Bites’ very own YouTube channel.
Here’s a video I took a couple months back from the first CS320 lab: getting a robot to walk. What was really fascinating was the extremely varied walks that resulted.
varied reports suggest that robocup.bowdoin.edu is inaccessible from the bowdoin wireless network: in the lab, in seth glickman’s room, and possible elsewhere. Yes, this means that the blog, and svn, trac, etc.
Just plug an ethernet cord in as a temporary measure.
I put this note in the blog to get Jeremy to put his thanksgiving laziness behind him and fix this bug.
UPDATE:: though Jeremy and I spent an hour debugging this, we later found this email from IT:
From: it@bowdoin.edu
To: all_students@list.bowdoin.edu
Date: Tue, 28 Nov 2006 18:33:28 -0500
Subject: wireless connectivity issue
We are currently experiencing problems with the wireless network and logging into Bowdoin web services (Blackboard). Until this is resolved please connect to the network using an Ethernet cable.
Thank you,
Information Technology
Let me know if anyone gets another saying that it’s working now. That would be useful.
UPDATE::Problem seems to be fixed in the lab. Anyone else having issues?
So I feel like I’ve gotten line “detection” and line “formation” to good places. Sadly that pink dot that looks like recognition of a corner joint is actually just an outlier. Landmark detection is the next chore: now that we have bunches of lines, how to figure out if these lines make a corner? And which corner? Oh, plenty more work to do.
So the the real focus of our team right now is localization: teaching the dog to know where it is on the field. A critical part of localization is a decent Vision system: having the dog to correctly identify landmarks. We’re pretty good at identifying big things like the posts and goals (though the new goals are giving us issue), but we’re pretty lousy at detecting line intersections on the field.
Here are the various issues plaguing our line recognition:
* Thresholding. The way we threshold–identifying RoboCup colors from the millions of colors that show up on the Aibo’s camera–heavily relies on segmentation. I don’t have time to explain segmentation, but let’s just say that it’s great for every kind of color object except for really small white sections that make up the lines.
* Landmark Detection. We can identify line as individual segments with some success–we can’t identify when they intersect. That is to say we can recognize two lines on the screen, but can’t recognize that they form a corner of the field. This should be one of the easier tasks.
* The lines and center circle in the lab. The physical lines on our Lab’s field are pretty yellowish (made from masking tape) and the center circle isn’t actually a circle. You have to stand way back to mistake it for a circle. It’s instead just a circle-ish grouping of line segments.
* The camera settings. We are forced to use the most blurry settings — which make the field brighter — because the lighting in our lab is so dim. Our lighting upgrade may still be months away.
* We don’t have a PS3 or even a Wii.
Search
About
You are currently browsing the Northern Bites Blog weblog archives for the month November, 2006.
