Johns Hopkins professor helps robots to feel better

Henry Baumgartner
ASME NEWS

Haptic technologies, which allow humans to interact with robots and virtual environments through the sense of touch, are attracting intense interest these days.

At The Johns Hopkins University in Baltimore, Alison Okamura, an assistant professor of mechanical engineering, is setting up a laboratory to further investigate this field.

Okamura, an ASME member, explained that her Haptic Exploration Laboratory is involved in two primary areas of research: the devel-opment of haptic interfaces for virtual environments, so that humans can actually feel the textures and forces that are virtually present, and robotics applications, which can either involve the robot's own processing of its haptic data or a human who can feel sensations based on what the robot encounters.

Okamura received her Ph.D. from Stanford just last summer. While there, she worked on haptic systems for robots — "how they can find small bumps and ridges" — as well as vibration feedback, which is a way to enhance the realism of virtual environments.

"A lot of virtual environments feel soft and squishy," Okamura said; "it's difficult to make them feel hard. Vibration feedback is a way to make this possible."

While she is still working on similar problems, "being at Johns Hopkins gives me a tremendous opportunity to work on medical applications," she noted. Medical simulations can give doctors an opportunity to try out their operating skills on virtual patients, for instance, before putting a scalpel to real flesh and blood.

Johns Hopkins professor Alison Okamura is investigating haptic technologies.

The robotics applications are also important. Using reality-based modeling, it is possible to take data from surgical tools in an actual operation and replicate the same forces and vibrations in a realistic simulation. A robot performing an operation can also compare current data with the database to help it understand things like how many layers of tissue it has already punctured in an operation to insert radioactive seeds in a cancerous prostate gland.

Such a scenario also brings up the question of what is the proper level of human-robot coordination — at what point does the human take over? Okamura cited the example of retinal surgery, where "the human tremor level is as big as the veins you're working on."

Another advantage the robot has is that humans are confined to what they can feel with their fingers, while robots can have sensors located anywhere, even, say, on the tip of a needle.

Other possible applications being explored include using haptics in the schools. To bring home the difference between mass and weight, it would be possible to virtually change gravity and give kids a chance to haptically bounce a ball on the moon.

The Office of Naval Research is interested in the technology for use in undersea salvage and exploration of artifacts that lie under the sea. NASA, too, is considering using haptic technology in the robots it plans to send to Mars, enabling earthbound researchers to feel the texture of alien rocks and soils.

And we haven't even mentioned how the field could transform video games.

Okamura is currently working on development of a robotic finger with a rotating sphere at the tip. "A sphere like this could move all over a surface. It would be excellent for observation," she said. The first step is to get the system to recognize features on a hard surface, then move on to more difficult soft surfaces.

Perfecting the mechanical devices and software needed to simulate the human sense of touch is a challenge that could take decades, but Okamura is eager to conduct some of the basic research.

"Human beings, obviously, have amazing tactile sensing ability," she said. "What we've done so far with robots doesn't even come close. There's a lot of work to be done."

back to news & features

front page | features | columns | meetings & courses | milestones | calendar | ME Magazine
about ASME NEWS | ASME.ORG | ME Magazine Online | news update | ASME NEWS archive
© 2001 by The American Society of Mechanical Engineers