Spherical Robot Can Climb Over Obstacles

Writing by Evan Ackerman on Thursday, 15 of October , 2009 at 12:01 am

Spherical robots are nifty designs, and they have several advantages over their wheeled counterparts: they adapt well to all kids of terrain (even water) and are completely sealed, making them ideal for planetary exploration. The big sticking point (as it were) of spherical robots is that they tend to get hung up on obstacles. They move by changing their center of gravity, but they can’t shift their internal mass outside of their own diameter, which means that they have a finite amount of leverage to play with and therefore can’t get themselves over obstacles of a given height, no matter how much power their motors might output.

Greg Schroll, a graduate student at Colorado State, has developed a spherical robot that uses gyroscopes instead of a movable mass for both drive power and steering. The gyros just have to sit there, and depending on which way they’re being spun, they exert torque on the robot to move it forward or turn it. The big advantage to this is that the torque exerted by the gyros increases as they spin faster, meaning that more powerful motor does increase the power of the robot, giving it the ability to literally jump out of holes:

Greg has been named one of Popular Science’s 10 Most Brilliant Innovators of 2009 for his work on this robot (among other things, like those badass air cannons in the video).

[ Popular Science ]

Comments (4)

Category: Research

Add this post to

Del.icio.us -

Digg

4 Comments

Comment by Anonymous

Made Thursday, 15 of October , 2009 at 12:19 pm

What’s the innovation? Ben Brown built his Gyrover back in 1997.

Comment by Foreigner1

Made Friday, 16 of October , 2009 at 10:49 am

…What’s the “robotic” aspect to that thing…?

I’ve remarked this before- Most items I see even on botjunky are not so much robots, but rather remote-controlled vehicles of some sort or another.
Even the bipedal versions that from time to time are presented, mostly work on the bases of remote control. Same with this ball here – Robot??? First thing he does is take the remote control to let it do things! Where’s the autonomy in that ball…?

Sorry to nag, but I think the term “robot” has undergone some hefty devaluation and now stands more for anyting remote-controlled with a camera bolted on it.
And I think that is wrong.

Comment by Anonymous2

Made Monday, 19 of October , 2009 at 10:18 am

The Gyrover is balanced with gyroscopes but not propelled with them.

Comment by J. Brad Harris

Made Thursday, 22 of October , 2009 at 9:19 pm

The gyro’s aren’t doing any propulsion or stabilisation in this device. You can clearly see the two gyros at the 3:01 mark (front and back), and then one of the drive motors at the 3:30 mark. The gyros simply provide additional rotational inertia which allows the drive motors to create more leverage than they could without the gyros. Since the weight of the pendulum alone can only provide so much leverage being trapped within the body of the sphere, Greg decided to find another way of increasing that inertia. Since inertia is a function of angular velocity, he added angular velocity via the gyros to the front/back of the unit and voila. Intuitive once you think about it, but as far as I know he’s the first.