misasia logo
The system analyzes thought patterns and translates them into wheelchair movement in a fraction of a second By Martyn Williams
30 Jun 2009

TOKYO, 29 JUNE 2009 - Researchers in Japan have developed a brain-machine interface (BMI) system that allows for control of a wheelchair using thought.

The system processes brain thought patterns and can turn them into left, right and forward movements of the wheelchair with a delay as short as one-thousandth of a second. That's a vast improvement over other systems that can take as long as several seconds to analyze and react to the user's thoughts.

It was developed by scientists at the BSI-Toyota Collaboration Center, a research and development center established in 2007 by Japanese government-related research unit RIKEN, Toyota Motor, Toyota Central R&D Labs and Genesis Research Institute.

The system measures the electrical activity in a person's brain using electroencephalography (EEG) data gathered from five sensors above the areas of the brain that handle motor movement. It seeks to interpret the measurements to achieve control of the wheelchair.

It can also adapt to a particular user's thought patterns to improve accuracy to as high as 95 per cent, the researchers said. Training on the system for 3 hours a day for a week is enough to have it tuned in to a user's motor-control thought patterns.

In a video released of the experiments a researcher is shown navigating a wheelchair left and right between six chairs in a room using the technology. A laptop computer mounted on the wheelchair is all that's needed to interpret the researcher's thought patterns.

To perform an emergency stop, the researcher just had to puff out his cheek: a sensor mounted there detected the movement and brought the wheelchair to a halt.

The group says plans to use the technology in a range of applications are already under way. First uses will likely center on the fields of medicine and nursing care management with the BMI interface decoding brain waves related to imaginary hand and foot control.

But researchers are confident that they can extend its use to detection of brain waves generated by various mental states and emotions with further research.

Earlier this year researchers at Honda Motor reported success in development of a BMI system that allows a person to control a robot through thought alone. The system allows a researcher to think one of several predefined movements, such as the robot lifting its right arm, and hopefully have the robot follow through with the same action. Honda said its system achieves a 90 per cent success rate.

Japanese car makers have built-up expertise in robotics from their development of highly automated production lines, and have been seeking to channel some of this knowledge into humanoid robots. Japan faces a rapidly aging society and home-help robots are seen as a potential answer to an anticipated shortage of health care workers in the years ahead.

Comments

Be the first to comment.


Post your comment

  • Please use English to post and reply to comments
  • Please do not use offensive language in the form of racial or ethnic slurs, abuse or personal insults
  • We welcome opinion and debate geared towards finding solutions
  • Please keep comments relevant to the topic
  • All comments are moderated
** Mandatory Field

Name
    **

Email
    **

Country


Comments
Maximum characters allowed: 2000
Disclaimer: All the content posted in this category comes independently from readers of Fairfax Business Media (FBM) Asia publications, unless specified otherwise. Fairfax Business Media (FBM) is not responsible for the opinions of its readers and the content posted by them does not represent the views and opinions of FBM.

Feature

Ovum logo

Government

The race for the Indian ID database begins

The US$4 billion Unique ID (UID) project, headed by former Infosys co-chairman Nandan Nilekani, is the Indian government’s most ambitious e-governance initiative ever.
By Hansa Krishnamurthy Iyengar | 15 Mar 2010

RSS Feeds

Add this section to your favourite feed reader.