A team of French scientists working with collaborators at the University of Michigan (U-M) and Ohio State University have created a robot that walks and balances just as a human does.

They say it is the first of its kind, and can catch its balance without having to rely on big, clunky feet to do so.

Honda has been developing its humanoid robot Asimo, which stands for Advanced Step in Innovative Mobility, during the past five years.

The Japanese company claims its robot is also the only one of its kind able to climb stairs, as well as to walk an incline and decline of up to 30 degrees.

But the U-M, Ohio State and French scientists say their robot, called “Rabbit,” is the first robot that can walk and balance like a human. Rabbit is based in Grenoble, France.

The U-M and Ohio State researchers have worked for more than five years with their French colleagues to come up with a way to dynamically balance the machine while walking.

At the end of last year they got it to run six steps.

U-M professor of electrical engineering and computer science Jessy Grizzle, one of the scientists behind Rabbit, said the robot did not have any feet.

Instead, its legs end like stilts so that it pivots on a point when it moves forward, meaning that if you nudge the Rabbit, it steps forward and catches its balance, much like a human would.

“If you build a robot that pivots on a point you must understand how the different parts interact dynamically, or else it will fall over,” Grizzle said.

Other “bipedal robots,” or two-legged walking machines, walk flat-footed, which stops them from falling.

Grizzle said the theory behind Rabbit, recently published in the “International Journal of Robotics Research,” involved a real understanding of the mechanics of walking and balance, which enabled scientists to predict in advance how the robot will move.

“The concept of stability is reduced to two formulas,” he said.

“It’s a matter of understanding enough about the dynamics of walking and balance so that you can express with mathematical formulas how you want the robot to move, and then automatically produce the control algorithm that will induce the desired walking motion on the very first try.”

It could prove useful in designing cost-effective human prosthetics and rehabilitative walking aids for spinal injury patients, Grizzle said.

“If you can take properties of a patient, their height, weight, how the valid leg functions, etc., maybe you could more quickly have the prosthetic adapt its characteristics to the person, instead of the person adapting his gait to the prosthetic–which is essentially what happens now.”

The theory behind the Rabbit could also be used in machines in homes that can climb stairs, or for designing robots that work in rough terrain in exploratory missions.

More here.