Chicago Sun-Times: A step forward for the wheelchair-bound

by Sandra Guy

Despite the many advances of technology, no one has invented a wheelchair that knows where it’s going. Until now. Researchers Linda Fehr and Steve Skaar have come up with an idea for a motorized wheelchair that can “remember” paths laid out for it.p. Fehr is an electrical engineer in rehabilitation research at the Edward Hines Jr. Veterans Administration Hospital in west suburban Maywood, and Skaar, 49, is a professor of aerospace and mechanical engineering at the University of Notre Dame .

The two developed the prototype wheelchair using off-the-shelf parts—an achievement featured in the March edition of Technology Review magazine.

One of the first people to test the wheelchair was a VA patient who, because of a brain aneurysm, had been unable to function by himself for 30 years. He communicates by laboriously spelling out words on a keyboard with a pointer. Now he can use the pointer to respond to the wheelchair’s speech-scanning interface, and can whir around a trailer where the researchers test the wheelchair.

Skaar compares the wheelchair’s potential with that of the early PC: Much is yet unknown, but its users will chart the course.

The researchers are halfway through a two-year, $200,000 grant from the Department of Veterans Affairs’ Rehabilitation Research and Development Service in Washington, D.C.

They began their project by outfitting a motorized wheelchair with two video cameras, a pair of wheel rotation sensors and a computer. The computer was coded using an algorithm originally developed to allow guided missiles to follow a trajectory and home in on a target. Guillermo Delcastillo, a Notre Dame graduate research assistant , did much of the coding work.

Delcastillo and Tim Sheehan, an undergraduate research assistant , pushed the wheelchair along several trajectories to teach the chair to follow assigned paths. The wheelchair’s cameras tracked the paths based on elliptical markers the researchers put on walls. The wheelchair’s software then remembered the routes. The researchers took it a step further by piecing together the routes, and programming the wheelchair so it is “smart” enough to switch from one track to another in mid-stream.

The programming provides another advantage: Because the wheelchair knows its objective is to reach a particular destination, it moves directly to that point—the edge of a desk, for example. This avoids the typical tug-of-war that happens when a regular motorized wheelchair’s sensors believe the desk is an obstacle and try to keep its rider from getting too close to it.

Since the VA hospital has a large unit for patients paralyzed from spinal-cord injuries, Fehr worked to enable the wheelchair to “hear” spoken commands. The extra flexibility means that users can tell the wheelchair where to go in any number of ways, whether by blowing into a straw, using a bite switch or speaking aloud a desired destination.

The wheelchair thus could be used by people whose illnesses or disabilities cause them to lose sight or motor functions. The wheelchair could also be outfitted so that nurses or other workers at institutions could operate a remote control, automatically steering the wheelchair to an appointed destination.

The VA has applied for a patent on the wheelchair. The researchers also are talking with manufacturers that have expressed an interest, especially since few alternatives now exist. At least two dozen projects to develop a smart wheelchair are under way throughout the world, but few are being done in the United States and few have succeeded. So-called power wheelchairs now on the market start at $4,000.

Admittedly, the wheelchair has its limitations. It cannot be used outside, and can only proceed along the routes it knows.

But for millions of people who have suffered disabling strokes or diseases like amyotrophic lateral sclerosis, it could mean a newfound freedom.

March 26,2003

TopicID: 345