Jennifer, the DARwin-OP humanoid robot that's being taught to play ice hockey at the University of Manitoba is getting great press coverage, including the video segment above that ran on CBC News.
Adapting the robot to play a challenge like ice hockey is a significant challenge. Still, the way that the press portrays the robot as being "Canadian" is a bit misleading.
With humanoid robots already well on their way to challenge humans on the soccer field it was only a matter of time before they strapped on skates and took to the ice.
Naturally enough, the first group to make a serious attempt at humanoid robot ice hockey was the Autonomous Agents Laboratory at the University of Manitoba, which in addition to their robotics expertise also has been blessed by nature with an overabundance of winter weather perfect for the sport.
Robotis, one of the key R&D collaborators for the DARwIn-OP humanoid robot project, and the supplier of most of the standard hardware used for the robot, has started publishing an excellent series of basic tutorials via the ROBOTISCHANNEL on YouTube.
The first five videos in the basic tutorial series cover everything, step by step, from how to power up and power down the robot, through color calibration.
Michael Overstreet has been a good and respected friend since the first time we hooked up several years ago at RoboGames in California. So, I hope he doesn't mind if I make some frank, and well deserved, comments.
When we first met Michael seemed like a typical robot geek, very talented with lots of expertise, but a bit shy and withdrawn. You really had to push him to get him to tell you what he thought. I'm sure he had lots of valuable and useful things to share, but they didn't flow easily.
Over the years, with experience, learning, and success, Michael has really blossomed and come out of his shell. He's become a key member of the Cowtown Computer Congress - Kansas City's leading hackerspace, a frequent exhibitor and participant in Maker Faire events all over the US, and a strong proponent of the DARwin-OP humanoid robot platform.
The Hasegawa Group located at the Tokyo Institute of technology has been able to integrate artificial intelligence into a semi-humanoid robot enabling it to think, learn, and act by itself. The robot uses a self-replicating neural network to learn in much the same way as its human counterparts. It is able to extrapolate based on its past experience and knowledge combined with observation of its surrounding environment, then make educated guesses. In that sense, it learns from its mistakes, just like a child, and consistently corrects and adjusts until it can successfully perform required tasks.
’Robot Learns, Thinks, and Acts By Itself (Video)’ continues
I've asked robot builders from all over the world about their ultimate purpose in building a robot. German builders often respond saying they want to improve performance, to design a robot that is better, faster, more reliable, or more precise than others. Japanese builders, especially those involved with humanoids, typically say "I want to build a real Gundam." And, Americans jokingly say they want to design a robot that will "Bring me a beer."
As Andrew Alter at Trossen Robotics found out, there's a problem with that. What if your robot has a serious drinking problem? You better make sure that he can't figure out how to open the can...