A robotic wheelchair with cerebral control combines artificial intelligence to make it easier to maneuver for people using only their thoughts. The approach, known as "shared control" could help people with paralysis to gain new mobility transformed brain signals into commands simple more complicated.
The wheelchair developed by a team of researchers at the Federal Institute of Technology based in Lausanne, has software that can start with a simple command like "go left" and assess the immediate area to find how to fulfill the order without hitting anything. This software can also understand when the driver wants to move toward a particular object, like a table.
The wheelchair developed by a team of researchers at the Federal Institute of Technology based in Lausanne, has software that can start with a simple command like "go left" and assess the immediate area to find how to fulfill the order without hitting anything. This software can also understand when the driver wants to move toward a particular object, like a table.
Currently there are several technologies that allow patients to control computers, appliances and other devices captured signals through the nerves, muscles or brain. Electroencephalography (EEG) has emerged as a promising way for paralyzed patients to control computers or wheelchairs. The user needs to use a special cap and follow a few hours a day training for about five days. Patients can control the chair with only imagine that they are moving a body part. Thinking about moving his left hand, for example, command the chair, turn left. Orders can also be caused by certain mental tasks, like arithmetic.
However, the EEG has limited accuracy and can detect a few different orders. The maintenance of these mental exercises when it comes to maneuver a wheelchair around a cluttered environment can be very tiring, says José del Millán, director of brain-machine interfaces noninvasive Federal Institute of Technology, who led the project. "People can not sustain that level of mental control over long periods of time," he points out. The necessary concentration also creates more noise in the signals, making them more difficult to interpret by a computer.
shared control solves this problem because patients do not need to be continually instructing the wheelchair to advance it and only have to think about the command once and the software does the rest. "The wheelchair can take care of small details, making it more natural," Miller said.
The wheelchair is equipped with two web cameras to help detect obstacles and can avoid them. If drivers want to get closer to an object rather than around it, can give an order of annulment. The chair will stop in front of the object.
Millán In the prototype, 16 electrodes monitor brain activity of the user. For the time has not been tested in any patient paralyzed.
Damien Coyle, researcher at the Brain-Computer Interface and Assistive Technology, University of Ulster, says that the EEG signals can be slow and may be difficult to work with them. As a result, he adds, many researchers are finding ways to use shared control, and draft Miller is a good example of this being implemented. "The more control you have shared, the better the brain-computer interface, and the faster you can move a person from one place to another," says Coyle.
Millán team is developing the capacity to recognize objects to make the chair intelligent enough to automatically stand before a table or desk sure is close enough and it stops at the proper angle.
Source: Research Andalucia
0 comments:
Post a Comment