IRJET- Cognitive Approach Towards Detecting Human Emotions along with Robot Control using Brain Comp

Page 1

International Research Journal of Engineering and Technology (IRJET)

e-ISSN: 2395-0056

Volume: 07 Issue: 08 | Aug 2020

p-ISSN: 2395-0072

www.irjet.net

COGNITIVE APPROACH TOWARDS DETECTING HUMAN EMOTIONS ALONG WITH ROBOT CONTROL USING BRAIN COMPUTER INTERFACE Naren K1, Ritesh Kumar2, MD Amir3, Manjunatha Siddappa3, Ravikumar K M4 1,2,3Student,

Department of Electronics and Communication, SJCIT, Chikkaballapur, Karnataka, India Professor, Department of Electronics and Communication, SJCIT, Chikkaballapur, Karnataka, India 4Principal, SJCIT, Chikkaballapur, Karnataka, India ---------------------------------------------------------------------***--------------------------------------------------------------------4Assisstant

Abstract – Brain computer Interfaces (BCIs) are systems

to waves characterized by different frequencies and amplitudes called EEG waves. EEG waves are mainly of 5 types which are as follows: Delta waves (0.5 to 3 Hz), Theta waves (4 to 7 Hz), Alpha waves (8 to 12 Hz), Beta waves (12 to 30 Hz) and Gamma waves (above 30 Hz). These are

that can bypass conventional channels of communication to provide direct communication and control between the human brain and physical devices by translating different patterns of brain activity into signal commands in real time. However, it is essential to establish methods to recognize the brain state accurately in order to implement BCI, and a number of challenges still remain. Emotions are intrinsic to the way humans are interacting with each other. A human being can understand the emotions of another human being to a certain extent and behave in the best manner to improve the communication in a certain situation however a machine cannot. Using EEG-based emotion detection, the computer can literally observe the user’s mental state. The applied domains for these studies are varied, and include such fields as communication, education, entertainment, and medicine.

Brain Computer interfaces (BCI) use the different variations in these EEG waves to communicate with external assistive devices. The BCI system should also be able to classify different EEG signals of brain activity as accurately as possible and the BCI user should learn to produce distinct brain signals to perform the different task. The robot mentioned in this paper can be used as a stepping stone to create wheelchair and other assistive devices. Several measures have been devised for classifying the human emotion. Such classification is generally divided into two perspectives namely Dimensional and Discrete perspectives. Discrete perspectives analyses emotion in a way that every specific emotion (e.g. fear, sad, happy, etc.) maps to its own unique parameter of environment, physiology and behavior. In Dimensional perspective, human emotions are organized in few fundamental dimensions. The most commonly assumed dimension is arousal and valence proposed by Russell in his bipolar model of emotion classification. In this dimension, valence represent from negative to positive whereas arousal represent from not excited to aroused or dull to intense.

As a result of various forms of illnesses or accidents such as spinal cord injury (SCI) or a form of motor neurons disease or ALS, many people suffer from a severe motor function loss and are forced to accept a reduction in the value of life, depending on the care of others. BCI can provide logistic support to those suffering from said disease. The intention of project paper is to develop a novel BCI system that accurately detects and isolates emotions into valence and arousal states and to develop a novel BCI based robot which can be used as a wheelchair to assist the disabled people in their daily life to do some work independent of others

2. OBJECTIVE AND PROBLEM STATEMENT The main goal of BCI is to enhance the communications between the people and computers. Since the greater part of computers are not able to understand the person's emotion, most of the time they can't react to the person's needs naturally and accurately. So, here we worked on detecting emotions (Valence and Arousal) efficiently by working on the pre-processed data obtained from DEAP database to detect the state of mind of a person.

Key Words: EEG, BCI, SVM, KNN, DEAP Dataset

1. INTRODUCTION The human brain is formed by the billions of neurons which are interconnected among themselves; the interaction between these neurons are represented as thoughts and emotional states. For every interaction between neurons a minuscule electrical discharge is created; alone these charges are impossible to measure from outside the skull. However, when the activity is created by hundreds of thousands simultaneous discharges, it is aggregated into waves which can be measured.

Independent mobility is core to being able to perform activities of daily living by oneself. Millions of people around the world suffer from mobility impairments and hundreds of thousands of them rely upon powered wheelchairs to get on with their activities of daily living However, many patients are not prescribed powered wheelchairs at all, either because they are physically unable to control the chair using a conventional interface, or because they are deemed

As a result of different patterns of neural interaction, it can be classified into different brain states. These patterns lead

© 2020, IRJET

|

Impact Factor value: 7.529

|

ISO 9001:2008 Certified Journal

|

Page 373


Turn static files into dynamic content formats.

Create a flipbook
Issuu converts static files into: digital portfolios, online yearbooks, online catalogs, digital photo albums and more. Sign up and create your flipbook.