Noninvasive brain-actuated control of a mobile robot by human EEG
read more
Citations
A review of classification algorithms for EEG-based brain–computer interfaces
Brain Computer Interfaces, a Review
A Review of Classification Algorithms for EEG-based Brain-Computer Interfaces: A 10-year Update
Control strategies for active lower extremity prosthetics and orthotics: a review
A survey of signal processing algorithms in brain-computer interfaces based on electrical brain signals.
References
Brain-computer interfaces for communication and control.
An Behavior-based Robotics
Behavior-Based Robotics
Spherical splines for scalp potential and current density mapping
Motor imagery and direct brain-computer communication
Related Papers (5)
Talking off the top of your head: toward a mental prosthesis utilizing event-related brain potentials
Frequently Asked Questions (13)
Q2. What are the future works mentioned in the paper "Noninvasive brain-actuated control of a mobile robot by human eeg" ?
Their results open the possibility for physically disabled people to use a portable EEG-based brain-machine interface for controlling wheelchairs and prosthetic limbs. However, the authors will need to scale up the number of recognizable mental states to provide a more flexible and natural control of these robotics devices. This could be done by estimating local field potentials of small cortical areas from the scalp potentials recorded with a sufficiently high number of electrodes ( 32, 64, or more ) [ 21 ]. The Gaussian classifier embedded in the BMI would work upon the local field potentials of selected cortical areas instead of using EEG features.
Q3. How do the authors initialize the center of the prototypes?
To initialize the center of the prototypes and the diagonal covariance matrix of the class the authors run a clustering algorithm— typically, self-organizing maps [17]—to compute the positionof the four prototypes per class.
Q4. What is the class-conditional probability density function of class for sample?
The authors assume that the class-conditional probability density function of class for sample is a superposition of several Gaussians(1)where denotes the number of prototypes (Gaussians) of the class and are the activation level and the amplitude of the ith prototype of the class , respectively.
Q5. What was the effect of the brain-actuated robot?
one of their subjects reported that when he tried to stay in an idle state, he had a tendency to anticipate the next behavior the robot should execute and, instinctively, concentrated on the corresponding mental state—thus delivering a wrong mental command.
Q6. What is the reason for the relatively clear responses of the robot?
A second reason for the relatively clear responses is that averaging class-conditioned probabilities before using Bayes’ rule [see (3) and (4)] helps to stabilize the responses of the classifier.
Q7. What is the criterion for a rejection of a false positive?
This rejection criterion keeps the number of errors (false positives) low, which is desired since recovering from erroneous actions (e.g., robot turning in the wrong direction) has a high cost.
Q8. What is the reason why the robot did not perform the correct behavior?
Note that even if the classifier responses were not always the correct one (there are several “unknown” responses and even an error at step 7), the robot still performed the correct behavior because of the control strategy implemented by their finite state automaton.
Q9. How long did the subject learn to control the mobile robot?
After this initial training, subjects learned to control mentally the mobile robot for 2 days with the interface operating in mode II.
Q10. How did the subjects conduct the second set of experiments?
In order to evaluate quantitatively the performance of the brain-actuated robot, subjects “A” and “B” also carried out a second set of experiments.
Q11. How many control commands do subjects emit in order to reach the target?
in order to reach the target as fast as possible, subjects do emit more control commands than the minimum (almost twice).
Q12. What is the effect of the classifier on the performance of the subjects?
As additional evidence that subjects are not using EMG activity, which is broad-band, if the authors apply machine-learning techniques for the selection of those relevant features that best differentiate the mental tasks, the authors find that the classifier performance improves with only a small proportion of features, which are not grouped in a cluster [20].
Q13. Why is the number of control commands significantly larger in the case of mental control?
in the case of mental control the number of control commands is significantly larger due to the less accurate control of the robot.