
With governments and public agencies across the country scrambling to meet the requirements for wheelchair access mandated by the Americans with Disabilities Act, it's quite likely that some public official somewhere has wondered whether it would not be cheaper simply to build wheelchairs that could climb stairs and curbs.
This was not an option at the time the ADA was passed. But a group of researchers in Penn's School of Engineering and Applied Science have been working on it.
Their work is part of a larger field of research aimed at working around the most notable deficiency of wheeled vehicles: They cannot cross irregular terrain or surmount obstacles well. The engineers working in this field take their cues from nature, trying to design machines that can walk as nimbly as goats or spiders do.
Vijay Kumar, associate professor of mechanical engineering, explained, "Insects and goats represent the ends of the spectrum of models for walking vehicles." Insects are rigid creatures whose legs, like those on a tripod, support their weight and keep their bodies stable as they walk. When goats walk, there is a brief unstable period when all their feet are off the ground, a trait that humans share.
The challenge in designing a chair that can "walk" over curbs is to get it to support its own weight in a stable fashion, so Dr. Kumar and his team opted for the insect model. That in turn gives rise to a new set of challenges, for, as Dr. Kumar put it, "The spider is perhaps the strongest creature on earth," with legs that can support hundreds of times their own weight. By comparison, it is difficult and expensive to design and build a legged vehicle whose legs can support even four times their weight. Wheels, on the other hand, easily bear heavy loads, and the principles of wheeled locomotion are well understood.
So Dr. Kumar and his associates, working in Penn's General Robotics and Active Sensory Perception (GRASP) Lab, went to work on a hybrid: a four-wheeled, motorized wheelchair equipped with a pair of "legs." The first working prototype was completed two years ago, and a second prototype was completed last year in the GRASP Lab.
Actually, legs may not be the right term, for they function more like extensions of the arm. While the chair functions as an ordinary motorized wheelchair on dry flat surfaces, the legs help pull the chair along on slick or uneven surfaces, much like a pair of cross-country ski poles. When climbing a stair or curb, the legs act as crutches for the chair, first pulling the front end up, then rotating behind the chair to push its rear end onto the raised surface. When descending a stair or curb, the legs prop up the front end and gently lower it, then steady the rear as it descends.
In addition, the researchers envision the chair's legs eventually performing more of the functions we use our arms for, such as holding open a door for the chair to pass through or grasping and lifting objects, including the user: "The arms could also be used to help the user mount and dismount from the chair, giving a greater degree of independence," research assistant Venkat Krovi explained. Such functions require that the legs be able to rotate horizontally as well as vertically, which is a proposed modification.
The prototype has successfully scaled platforms as high as 12 inches in the lab, thus proving that the basic idea is sound. But there is still much work to do before the chair is ready for commercial adaptation. For instance, the vehicle itself will not tip over backwards, even with a rider, while climbing a foot-high step. As Mr. Krovi explained, "Motorized wheelchairs are designed to be heavy at the bottom to avoid tipping." But the rider's seat tilts back to an angle that causes discomfort. Mr. Krovi believes that this problem can be solved by mounting the seat on a semi-elliptical track so that it can slide forward or back as the base tilts, thus keeping the rider level.
Other problems that remain include incorporating the computerized controls for the arms on board the chair, and fitting the controls and batteries needed to power the chair under the seat.
Dr. Kumar and his research team have applied for and are about to receive a patent on their all-terrain wheelchair technology. Mr. Krovi reports that word of the technology has begun to spread among wheelchair users and rehabilitation-medicine professionals, thanks largely to information found on the World Wide Web (http://www.cis.upenn.edu/~venkat/wheel.html), which includes actual video footage of a curb climb (see photographs). But so far, there has been no interest expressed in developing it commercially. However, Mr. Krovi said, "We remain hopeful that some commercial enterprise will see the potential in this technology and seek to exploit it."






Images courtesy of Venkat Krovi/GRASP Lab