Monday, 16 August 2010

More tai chi robots than you can shake a stick at

First there was Hubo, then QRIO and now here's another one: http://www.idemployee.id.tue.nl/g.w.m.rauterberg/Movies/tai-chic-robot%28wmv%29.wmv and another http://www.dailymotion.com/video/xa5l71_tai-chi-nao-robocup-08_tech. And also some more examples (most of which I've already highlighted before) from this thread over at kungfu magazine.

So It seems the benchmarking principle from my robotics framework is actually being used by  researchers out there.Which gives me a warm glowing satisfaction that my thought experiment framework isn't a million miles off :-)

Another question from my framework was the mininal requirements needed for a robot to use internal arts princples. To some extentet this is given lip service by Henrey Zhang (below), however it is unclear to me what internal arts principles the robot is actually using. Seems to me it's all about the human doing tai chi (or rather simply balancing and using weight transfer) and the robot detecting it - which is not the same thing at all.
Henry Zhang, an assistant professor of mechanical engineering technology, along with four students, spent several months building a hands-free, two-wheeled, self-balancing scooter based on the principles of Tai Chi, a form of Chinese martial arts that centers on focusing the mind and staying aware of your center of balance.

The scooter contains a platform to stand on, and the on-off, and turning functions are initiated by the rider via a remote control. Zhang said that what makes their machine unique is that there is no handlebar for driving or steering. Instead, these actions are controlled by the rider shifting his or her Tai Chi poses. The scooter is driven forward and backward through the rider’s self-balancing, and its turning is controlled by optical encoder signal feedback that monitors the angular displacement of the remotely triggered DC motor with gear reduction. http://www.lafayette-online.com/science-technology/2009/03/tai-chi-driven-scooter/

On the other-hand this article seems to point to some genuine reasearch into using tai chi as a means to train robotics, as to quote:
"We want to create robots that can work with humans, so we wanted robots to learn how humans move and interact with the world, which led us to go and try to better understand how humans move," Khatib told LiveScience.

The researchers investigated how people move by using sensors to follow human volunteers such as Khatib's students and a tai chi master visiting from China as they performed movements such as bending, walking and jumping.

"The tai chi master helps provide data on motions that are closer to optimal in performance for comparison," Khatib said.

Using this data, the scientists devised computer models that showed how people minimize effort. These allow robots to perform tasks without computing where exactly it will go in advance as conventional methods do, saving on the number of computations the robots have to carry out"

No comments:

Post a comment