Tetraplegic paticnts (those who can't move their upper or lower body) are prisoners of their own bodies. Now a robot arm is to help them interact with their world. This research was completed by researchers at the Swiss Federal Institute of Technology Lausanne (EPFL). Professor Aude Billard and Jose del R. Millan worked together to create a computer program that can control a robot using electrical signals from a patient's brain.
First, the user wears an EEG cap to have their electrical signals inside their brain scanned(扫 描)effectively, which are then interpreted by the machine-learning algorithm (算法).The computer then sends signals to the robot arm to determine how it moves. As the robot arm performs a motion,the algorithm is looking to get feedback from the user when it makes a mistake: perhaps it moved too fast, or too violently. The end goal is that the robot can learn the right movements for a task in a given context. For example, you might want the arm to use a bit of force to throw a paper ball, but you might want it to be gentler when putting glass bottles.
In the team's research, they trained the robot arm to pick up a glass. The arm would move towards the glass and the user's brain would decide if they felt it was too close or too far away. The process is repeated until the robot understands the optimal route for the individual's preference - not too close to be a risk but not so far away to waste movement.
"Training an algorithm to read brain waves in a consistent fashion was the most challenging part, because the brain is not only focused on the hand but also processing many other things," said Millan. "This means our algorithm will never be 100 % accurate."
The researchers hope to eventually use their algorithm to control wheelchairs, which would allow people in wheelchairs to have greater control over their movements, speeds and general safety. However, this does require consistency over time to the algorithm.