Tonight, ordering pizza around 10:30 pm. We’re pushing to get some stuff done before the break.
UPDATE:: Massive Success! We got timestamping for communication, improved our distance measurements by half, and made some killer behaviors: finding, turning, aligning, and tracking the ball, and then we added some approach vectors. Here’s two videos for your robot pleasure:
Great meeting today guys. Lot to do before Atlanta, but it’ll certainly be fun. For all of those who can make it, all-night pizza computer-science-4-life bender tomorrow night in the labs (wednesday). I’m so jazzed.
Couple of recent updates: the ‘ins’ script now fails to unmount the memstick when compilation fails (very good, thanks Jeremy). Also, we’re now exclusively using the UPENN head engine. Rock on. AiboConnect you can still choose between the two by calling ‘heads’ (no engine) and ‘headeng’ (head engine).
I am officially the wiz at I-O for the Aperios Operating System. I’ve just completed most of the internal work to make various OPEN-R objects (our modules–Vision, Motion, Communication) talk to each other in exactly the way I want them to–which is good! The motion system now tells the Vision module (and our Brain) the current state of the motion system and the current joint values every frame. This is a basic building block for better and better behaviors.
Predictably, we have it pose with one of our aibos. Here are a few more pictures.
Tonight I got the values of the joint angles for the normal walk forward. Now I have to figure out the best values for a trap (should be similiar to the trap I already have for the spin), and then readjust the values of the leg joints so the dog actually walks with the ball. This should primarily consist of readjusting the front leg values, but the back legs probably have to go down a bit as well.
I implemented the green-orange transition rules this evening and the AIBO was picking up orange.
The rules will need to be tweaked and tested and re-tweaked, but it’s a step in the right direction. Also, going to need to speed it up a bit as it worked a little slowly.
Long weekend ahead…
we’re nearly there on getting a near-real time OpenGL app running via AiboConnect that gives updates on the robot’s localization information to a client computer. When this thing starts to work well, it will be so rad. In the meantime, we have to really smooth out our distance readings.
Today Matt, Prof. Chown and I impressed half a dozen or so toddlers with neat AIBO tricks, including a goalie demo, a remote control demo, and locate ball demo. Chown showed off an AIBO in puppy mode. The kids also enjoyed playing with the assorted toy robots. The kids had a great time.
We’re going to Germany!!!
UPDATE:: Here’s a list of teams that qualified. Over 39 35 teams applied, twelve of which were guaranteed qualification from last year’s performance!
UPDATE x2:: Lots of Chatter going on in the four-legged league mailing list including a message sent by the head honcho of the league concerning the qualification process: Legged Qualification FAQ.
UPDATE x3:: 35 Teams applied. 12 Prequalified. 11 were let in by committee. Northern Bites is only one of three teams selected as first-year applicants to the World Championship!
