June 29, 2009

Well well, we are officially in Seattle. Our flight was fairly uneventful. Dang I am such a nerd. I spent half the time listening to the air traffic controller radio chatter on the on board headset. Everything was fairly smooth. The shuttle got us to our hotel just fine, and we checked in with no problem. Then at about 2 am on Sunday morning Ugonna opens his bag only to realize he grabbed the wrong bag.

Now, I'm not angry at Ugonna, that was an honest mistake. It looked just like his bag. No, I've got a beef with the United Airlines baggage claim personnel. So, after Ugonna dropped the bombshell, we ended up calling a cab to take us back to the airport (of course Ugonna had been calling the airport and airline before hand with no answer). Apparently at 2:30 am the baggage claim area of the Seatac airport is completely devoid of life. So we stuck around in the area until about 4:30 am when someone from United finally showed up and helped Ugonna get his bag back. During this time I tried to sleep, but they REALLY design those airport chairs so you can't sleep in them. We ended up getting back to the hotel at about 5 am (another cab, ended up costing about $100 total for both rides). Taylor and I got about 3 hours of sleep and Ugonna opted to just not sleep at all. But at least it was a learning experience, like for example I learned that I get really cranky when I don't sleep.

The conference itself has been going pretty well. We spent Sunday (what kind of a conference starts on a Sunday?) and this morning in workshops. I really enjoyed Professor Stoytchev and Jivko's talks. The rest of the presentations were interesting as well, but the best stuff were the discussions we had after each session. It was really amazing listening to what all the people in the field think about robotics. Although one topic really dismayed me. Most of the people in our workshop (it was on robotic manipulation and there were about 50) thought that robot's of the future would be preprogrammed with 3D models of objects and that's how they would know how to use them. WHAT?! How does that make ANY sense? Okay, so first of all it's not even possible to program every object in existence so that a robot could interact with objects successfully. Second, objects have a tendency to break and deform in so many various ways that there is no way to program for all of those. As well, people (younger children namely) are constantly taking existing objects and making new, completely different ones. The point of all this is that a robot needs to be able to learn and adapt to objects. Sure you could use a database of objects to bootstrap its learning, but in the end it still needs to learn. As the conversation went on, I was ready to chime in and say something when Professor Stoytchev spoke up and set them all straight.

This afternoon we listened to talks about various robotics projects. The introduction talk was about the navigation techniques of fruit flies and how they might apply to flying robots. The other talks were very, very dense. There was one even where I had absolutely no idea what even the general idea of what the speaker was talking about. It turns out that out of all the talks today, only one was even remotely related to our research area. I guess that's the difference between general robotics and developmental robotics. Developmental focuses on learning, whereas general does just about everything. I really, really prefer the developmental, the kinematics and inertial matrices a lot of the people talked about are crazy hard to use. Tomorrow looks like all talks again. Hopefully its not too boring (today kinda was). But tomorrow evening is going to be amazing. We are getting ferried out to an island in the harbor to have a very fancy dinner and on the way back will be able to see the Seattle skyline at night. I'm psyched. I'll report on it tomorrow.

ZayiraJordan - 09 Jul 2009 - 10:22

No photos?