SciTech

New omnidirectional robot utilizes only two moving parts

Credit: India Price/Online Editor Credit: India Price/Online Editor

Researchers always strive for efficiency in their robotic designs. Achieving this goal proves to be quite the challenge, however, as the demand for more moving parts and power in modern robots increases.

Nonetheless, the Hollis research group at Carnegie Mellon University demonstrated that this trend need not be the case by engineering the Spherical Induction Motor robot (SIMbot), a robot that balances on a single spherical ball bearing.

The SIMbot consists of only two moving parts — its vertical body and a rotor for movement. The body of the SIMbot rests atop the rotor like a pen atop a ballpoint, and stands about six feet tall.

The hollow metal rotor, coated with poly-urethane, spins under the body in all directions, albeit with some friction from the bearings it is connected to; Hollis suggested in an article for CMU News that future iterations of the robot could essentially eliminate friction by replacing these ball bearings with a thin cushion of air.

The SIMbot uses fascinating technology to function. Motion in the rotor is induced by actuators and rollers that can be adjusted to change spinning direction and yaw (i.e. the spinning motion of the rotor about a vertical axis); the robot achieves brisk walking speeds with this.

The robot also expends energy to keep itself balanced and to return to the upright position. This makes it difficult to tip over the robot, even after being kicked and shoved. It uses an Inertial Measuring Unit to detect its body orientation and correct it; the measuring unit may be replaced by a gyroscope in other versions of the ballbot. In the unlikely case of a system failure, the SIMbot has three legs that it can rest on for support.

Recent graduate Greg Seyfarth, who obtained a master’s degree in Carnegie Mellon’s Robotics Institute, wrote his thesis on the development and inner workings of the SIMbot, which gets much more technical.

Robotics Institute research professor Ralph Hollis designed the ballbot, on which the SIMbot is based, in 2005. He then patented the design some years later.

To add to this, Professor Masaaki Kumagai of Tohoku Gakuin University in Japan and EHP Zurich also made essential leaps in the SIM field with the BalllP robot and Rezero robots, respectively.

The SIMbot differs from the conventional robot in many ways. Other robots often have much lower centers of gravity and slower acceleration and deceleration capabilities to avoid tipping. If they do have wheels, they usually number no less than three, to ensure stability. Furthermore, whereas other robots need to produce yaw to turn, the SIMbot simply changes the direction of rotation of the rotor, allowing it to turn in any direction.

The SIMbot opens offers unique advantages in several fields. Such technology could be harnessed in medical fields, package delivery, and locomotion. Degrees of Freedom (DOF) arms can be added to the vertical body to equip the robot with more capabilities, as was done on the allbot in 2011, making it the only ballbot in the world with arms. Thanks to its fewer moving parts, the SIMbot would be easier to maintain and produce, making it a viable option for such varying fields.

In a university press release, Hollis commented on the financial benefit of the SIMbot, stating “This motor relies on a lot of electronics and software. Electronics and software are getting cheaper. Mechanical systems are not getting cheaper, or at least not as fast as electronics and software are.”

The SIMbot shows us what innovation in engineering at Carnegie Mellon can achieve. It overcomes many of the limitations of conventional robots, while also offering newfound technology and opportunity.

“Even without optimizing the motor’s performance, SIMbot has demonstrated impressive performance,” Hollis said in the same university press release. “We expect SIMbot technology will make ballbots more accessible and more practical for wide adoption.”

This research was supported by the National Science Foundation and Grants-in-Aid for Scientific Research (KAKENHI) in Japan.

It would be unsurprising to see the spread of Hollis and his research group’s technology, possibly to consumer markets, in the coming years — something I suggest we all look out for.