Patients who suffer from disease or injury that leave them unable to move have little hope of independent mobility. But that may be about to change. Researchers are developing a thought-controlled robotic wheelchair.
Spanish scientists have begun work on a new brain-computer interface, or BCI, capable of converting thought into commands that a wheelchair can execute.
Other researchers have already had some success with hard-wired brain computer interfaces, but they're powered by large computers and are physically plugged into the brain.
The Spanish researchers hope to develop a small, mobile interface that works with electroencephalogram electrodes, or EEG, placed on the scalp.
"We are planning to use non-invasive devices to record the rhythms from the surface of the skull," says Javier Minguez, a researcher at the University of Zaragoza in Spain. "We also plan to use this system with a school for disabled children that we collaborate with and (we) prefer to use non-invasive techniques with these children."
The Spanish Ministry of Education and Science has invested 180,000 euros in the "Biomedical Evaluation Of Robots to Assist Human Mobility" project. The goal is to bring mobility and a degree of independence to people with limited motor capabilities as the result of injury, disability or old age.
While EEGs have a reputation for providing very crude signals, advances in decoding algorithms yield patterns that are precise enough to control the movements of a wheelchair.
"You're not going to be using EEGs to control a robotic arm to play the piano or anything," says Dawn Taylor, an assistant professor at Case Western Reserve University's Department of Biomedical Engineering, who isn't involved in the project. "But you can certainly turn right and left and stop and go using that sort of signal
Two 800-MHz Intel computers mounted on the wheelchair will process these readings and send instructions to the wheels. After about a week's training the software will adapt to patients' thought patterns for simple commands such as "left" and "right."
The team hopes to use a combination of thought and mapping software to enable more complicated "macro" commands such as "Go to the kitchen."
"The important issue is to have a good selection of the mental tasks for each user, so that they produce discriminable EEG patterns," Minguez says.
A front-mounted laser will work as a sensor, detecting obstacles ahead and changes in the environment, like furniture that has been moved. This is particularly important for people with limited head and neck mobility, as they often cannot clearly see the way ahead, especially at ground level.
"One of the real advantages of using robotic wheelchairs is that their navigation technologies can override mechanical and human errors," says Minguez.
The biggest problem the team faces is miniaturizing the BCI and making it mobile.
"Right now, BCI technology is in labs with large computers attached to people," says Dawn. "So, it's not at the stage of being miniaturized and portable yet."
The first working prototype is expected in 2008 or early 2009, Minguez says. The scientists have already developed a wheelchair with a navigation system that can follow a map.
In 2010, the team hopes to test the system in collaboration with El Colegio Público de Educación Especial Alborada, a local school for the disabled.
