Robots Podcast, the best podcast related to robots online (and I’m not just saying that because I won some delicious chocolates from them), has done a great interview with Dr. Andrea Thomaz (Georgia Tech’s Socially Intelligent Machines Lab) about robot learning. One thing I found somewhat interesting is that Dr. Thomaz does not refer to Simon as “he” or “she” but as an “it”. I suppose like Honda’s stance on ASIMO, it is important not to assign human qualities to machines. But I can’t help it, Simon is just so darned cute!
Simon’s color-changing ears could be used to communicate quite easily despite no morphological model because humans use color signals in every day life, particularly when driving. One can imagine Simon’s ears changing from red to green to reflect when it is busy or ready to start a new learning session. Additionally, color theory suggests that cold colors are used to reflect sadness while warm colors are often associated with happiness. Blue might mean Simon is “sad”, while bright yellow might be his “happy” color. These signals would be made more clear with motions like drooping or perking up. In addition, shifting colors could accentuate facial expressions in communicating the robot’s “mood”. An angry scowl would be made more effective with hot red ears, while a shocked or surprised expression could blink a rainbow of color. Simon will eventually get eye brows and lips, which will also help communicate its internal state.
The second interview is with Dr. Sethu Vijayakumar. There’s some really cool stuff concerning the iLimb by Touch Bionics and the future of haptic control (like Deka’s Luke Arm prosthesis, they’re working on sensing through the prosthetic to determine grasping strength). He provides a great response to critics of Honda’s ASIMO and similar humanoids, who often question the validity of building a robot that runs in circles or kicks a soccer ball.
[source: Robots Podcast]