Robot Control Using Motor Imagery in Real Time

Abstract

Brain-computer interface (BCIs) provides the communication between the user and computer or external devices. The objective of this project is to develop an EEG(Electroencephalogram) Data acquisition system (DAQ) that will be able to read brainwaves from an individual so that we can further analyse this data in MATLAB, and use their classification result to give command to arduino microcontroller to perform an action. In this project, the EEG Data is acquired from the developed EEG Data acquisition system and using the acquired EEG signal to differentiate an individual’s motor imagery task i.eto differentiate between user thinking right or left hand motor imagery movement and then based on the classification result the command had been send to the arduino which enables the motor Of the robot to perform an action, .e to moves the robot to left or Right. The project has been implemented in three parts. The first part of the project involves acquiring the EEG Data from the implemented EEG DAQ by amplifying the signal using instrumentation amplifier and operational amplifiers as well as a notch filter, high pass filter and a low-pass filter to make the data noise free and in properly readable form . The second part consists of analysing the data that has been received from the EEG DAQ through arduino to the system Using Discrete wavelet transform (DWT) as feature and Linear discriminant analysis (LDA)and support vector machine (SVM) as the classifier in MATLAB. The last part of the project involves the use of MATLAB togive the command to the robot through serial communication to arduino. which, is acting as a microcontroller for robot control.