Detecting neural responses to use them in the robotic technology as a new controlling signal.
The ability to detect errors is crucial in human-robot interactions when a human tries to control a robot by his or her thoughts. Several studies have reported that erroneous actions elicit distinct neural responses, called Error related Potentials (ErrPs), which can be measured using electroencephalography (EEG). Interestingly, ErrPs can be observed both when the errors are committed by the user, as well as when the errors are introduced by the interfacing device such as a robot. In this project, we would like to detect the abovementioned neural responses and use them in the robotic technology as a new controlling signal. For this purpose, we will record brain signals while the user monitors actions of a robot arm to see if the desired actions are performed. Advanced signal processing and pattern recognition algorithms will be applied to evaluate whether or not the correct action has been achieved. Subsequently, if ErrPs detected, it will be used as a corrective signal to correct the erroneous action of the robot. Moreover, the detected ErrPs will be used to adapt the intelligent robotic controller such that the likelihood of committing the same error in future is reduced.