I have absolutely no foundation for calling this the "world's cleverest" jumping robot, except that we've covered a whole bunch of jumping robots, and this one is easily one of the most brilliant designs that we've seen. We first met this little guy at ICRA, where it showed off its ability to jump, land without smashing itself to pieces, stand up again, turn, and then make another jump. Cool, and now it's just gotten some serious upgrades.
The epic cleverness of this robot (which was developed by Jianguo Zhao at Michigan State University) comes from the fact that it uses just one single pager motor to jump, self-right after landing, and then orient itself to make its next jump in the right direction. Wondering how all of that is possible? Just watch:
Since we first saw this robot, it's gotten some substantial improvements. It has nearly doubled its jumping height to just under a meter, which is about 14 times the height of the robot itself. It can turn much, much faster, at 36 degrees per second, up from 2 degrees per second. And the self-righting system is significantly more robust. All of this stuff has happened without the robot increasing in size or weight, which is fairly remarkable, and it's so efficient that it can jump hundreds of times without needing to recharge.
A robot like this has all sorts of potential uses, although most of them fall into the (by now familiar) categories of search and rescue, environmental monitoring, and military surveillance. It'll be straightforward enough to mount a payload (like a wireless camera) onto this little guy, but in order to be really useful, it's probably going to have to learn how to right itself on non-flat surfaces. I'm not worried, though: I'm sure there's some sort of clever little tweak that'll make this robot able to jump from rough surfaces, clear tall buildings in a single bound, solve the world's energy problems, and play the piano, all on one pager motor.
UPDATED November 28, 2011, 10:10 a.m. Corrected affiliation of the robot's creators.