menu
robots A-Zrobot timelinegamestv & moviesfeaturescontactfacebooktwitteryoutubetumblrrss feed
bg logo logobump



Is MABEL Really The Fastest Bipedal Robot Ever?

Earlier this week you no doubt saw a biped robot called MABEL running at 6.8 miles per hour (10.9 kph). Researchers at the University of Michigan’s Department of Electrical Engineering and Computer Science claim that this is most likely the fastest biped robot developed so far.  Prudently, they’re not claiming it is definitely the fastest, which is wise for a couple of reasons.  To begin with, MABEL is attached to a boom which helps it maintain its balance along the Sagittal plane. Some might call that cheating.  And secondly, there may be another bipedal robot that can run just as fast (or faster) without the need for a boom, which we’ll get to after the break.

Video:

YouTube Preview Image

[source: University of Michigan’s MABEL website] via [Engadget, Physorg]

During Honda’s celebration of ASIMO’s 9th birthday (via this Impress Robot Watch article) Masato Hirose (Chief Engineer of ASIMO) claimed that ASIMO was able to achieve running speeds greater than 10 kph. However, they decided to limit ASIMO’s running speed to 6 kph because this was deemed fast enough for most purposes, and would be much more stable. Honda hasn’t released video footage of these running experiments, so there’s no concrete proof, but I don’t doubt they have pushed the limits of ASIMO’s running speed.

I certainly don’t want to take anything away from MABEL’s running; it does have the longest flight phase of any running robot (it is off the ground 40% of the time per stride), which is indeed very impressive.  The ground clearance is also important.  However, it seems to me that MABEL’s researchers will have to wait until their robot is running free from the boom before they may firmly claim the title of “fastest bipedal robot”.

If the balance made possible by the boom is trivial, then eliminating it should be trivial.  Otherwise MABEL isn’t really walking or running entirely by itself, which is what people expect robots to do (such as Boston Dynamics’ PETMAN, KAIST’s HUBO 2, or Toyota’s Partner Robot).