View Full Version : Re: Asimo robot gait

12-07-2007, 05:58 PM
Hello all,

I'm Post-Doc Researcher @ National Rehabilitation Center in Japan. I'm
currently working with the members of Robotics Laboratory @ Waseda University.

About the question regarding bended-knee, there are two reasons.

1) Problem in Inverse Kinematics.
What Stuart recently suggested. Joint angular velocity becomes infinity
at a singular point (with stretching leg).

2) To make COM height constant.
Please pay strong attention to the height of COM. Unlikely to human gait,
the COM height is constant for entire gait cycle, to avoid unexpected
balance loss. Bended-knee gait is neccessary for this style of
locomotion, because Ashimo doesn't have the joint to rotate the pelvis
in frontal plane (http://www.honda.co.jp/factbook/robot/asimo/200011/03.html
"text in Japanese but you'll see the figure").

Here I'll attach the URL of Biped Humanoid Robot Group of Takanisi Lab @
Waseda University. You'll see several movies tried to solve these

[Index page]
Ogura et al., Human-like Walking with Knee Stretched, Heel-contact and
Toe-off Motion by a Humanoid Robot, Proceedings of the 2006 IEEE/RSJ
International Conference on Intelligent Robots and Systems, 3976-3981.

FYI, their biped robot succeeded the stance phase initiating with "heel"
contact and terminating with "toe" off. (This is very difficult with the
locomotion of biped robot!!)


Yoshiyuki Kobayashi (Ph.D)

Research Fellow @
Department of Rehabilitation for Movement Functions
National Rehabilitation Center for Persons with Disabilities

4-1 Namiki Tokorozawa, Saitama
JAPAN 359-8555

+81-4-2995-3100 ext. 6641