Getting your Trinity Audio player ready...

A child in a wheelchair approaches a building, perhaps the home of a friend. A single concrete step stands between her and the door. There is no curb cut, no ramp, and she lacks the strength and coordination to wheel the chair up the step the way some adults can.

Instead, she manipulates a joystick, causing a pair of robotic arms attached to the sides of the chair to rise, bend, then land on top of the step. The robotic arms anchor the chair like a pair of ski poles, and the motorized wheelchair climbs the step with its front wheels.

The arms then swing back in the opposite direction and push off from the ground, lifting the chair the rest of the way onto the step.

It may take a few years before you see such a scene, but it’s one that researchers in a robotics laboratory at the University of Pennsylvania are determined to make possible.

They have a patented design for an “all-terrain wheelchair” that could give new mobility to people with disabilities. Now all they need is somebody to polish the design and manufacture the thing.

In theory, the chair would not only scale a 12-inch step, but cross over potholes, tree limbs and other obstacles, plus get through terrain that’s hard for other wheelchairs to negotiate, such as beaches, fields, gravel paths or steep grades.

The device is still a prototype, and a rather austere-looking prototype at that. It resides in a corner of GRASP, Penn’s General Robotics and Active Sensory Perception laboratory, a room dotted with computers, robots and gizmos in assorted stages of development.

The wheelchair’s seat is an ordinary plastic classroom chair whose legs have been removed. It is mounted on a four-legged aluminum frame that has two rubber tires in back and two casters in front. Attached to the frame are six motors, two for each of the robot arms and two for the back wheels. Electric wires cascade from the back, and the whole thing is tethered to a personal computer that tells the motors what to do.

The Penn researchers are hoping someone will take their design and make it commercially viable, not to mention more aesthetically pleasing. The plastic chair would be replaced with something more useful to people with disabilities. The mechanical arms, ideally, would be detachable and capable of maneuvering objects and opening doors. Batteries would replace the array of wires, and a computer chip would replace the PC.

The chair was created under the supervision of Vijay Kumar, an associate professor in Penn’s School of Engineering and Applied Science. Kumar said he got the idea of a “walking” wheelchair in 1989 or 1990, after observing an elderly man struggling to manipulate his wheelchair over a step in a park. The old man got over the obstacle only with the help of a companion.

Kumar is a mechanical engineer and a scholar in “legged locomotion,” or, as he puts it, “things that walk.” He realized that what the man needed was not wheels, which are designed to work on flat surfaces, but some kind of robotic legs that could climb steps and cross over obstacles.

Kumar and a group of Penn undergraduates started tackling the problem in the GRASP lab in 1990. The first idea they considered was a chair that had four independently moving robotic legs instead of wheels. The chair would walk around like a horse.

After trying two types of four-legged designs, neither of which reached the prototype stage, Kumar decided the idea was unworkable, and instead tried using a combination of wheels and legs, hybrid locomotion. He obtained a three-year, $180,000 grant from the Whitaker Foundation in Rosslyn, Va., which supports biomedical engineering research.

The current wheelchair is the second prototype built in the GRASP laboratory. The first had motors only for the robotic parts, not for the wheels, and was built using a modified child’s wheelchair. Because the seat had a relatively high center of gravity, however, the first prototype could not tilt back as far as the researchers wanted. So a second prototype was made, using a seat with a lower center of gravity.

Mathematical formulas determined the optimum dimensions for all the components and how force should be distributed at all the points where the chair makes contact with the ground. Those painstaking processes took months to complete. One student even built a master’s thesis around solving the dimension problem.

As currently designed, the chair can only support the weight of a child, a maximum of 60 pounds, which is about how much the vehicle itself weighs. Venkat Krovi, a graduate research assistant who has worked on the project since 1992 and runs its Web site, said overcoming road obstacles can be particularly difficult for children.

“As adults, people figure out ways of climbing the curbs. I’ve seen some adult users power their way up. … But children, unfortunately, can’t do that,” Krovi said.

Krovi compared the all-terrain wheelchair’s capabilities to those of a four-wheel-drive vehicle. “You’re no longer just pushing the wheelchair, so to speak, with your wheels, but you’re also pulling yourself along,” he said.

The device cannot climb a continuous flight of stairs. The steps must be a certain distance apart, such as stadium stairs.

(GRASP is working on a robot based on the wheelchair design that will be only about 15 inches tall and will climb flights of stairs, Kumar said. That robot could be useful for firefighting, bomb squads and other dangerous situations.)

In 1995, Penn received a patent on the wheelchair, but the device has served primarily to demonstrate the capabilities of the robotics lab.

“We haven’t been doing as good a job of advertising as I would like to do,” Krovi said. “But then, I don’t have the time to (market the design) other than creating a good Web site. The Web site has actually caught a lot of attention. I have a whole bunch of e-mail from people who said, `Hey, when can I buy this?’ “

Kumar, too, would like to see the product go to market. He said it would likely be expensive–perhaps $10,000 or more.

The main selling point of the design, Krovi said, is its ability to operate in several environments, unlike a beach buggy or an indoor wheelchair lift.

“What people don’t realize is this technology could enable them to do more, like go out on a beach. The government isn’t going to lay a road on a beach,” he said.

No matter what happens, the experience gave students a way to bring classroom theory into practice and the exhilaration of having accomplished a goal.

Said Krovi: “It was a learning experience for me because you build something. The first time when this machine walked, we were like proud parents. Maybe that was the reason we called it Baby. We were like, `Come to Papa.’ “