EIROB : An Emotionally Intelligent ROBot for Empathetic Support

Abstract

Social assistive robots enhance social interaction and well-being by providing companionship and support. Emotional Intelligence in these robots refers to their ability to identify, understand, and respond to human emotions. This study investigates how well a social robot named ‘EIROB’ can engage elderly persons by using Artificial Emotional Intelligence. EIROB employs a Multimodal Emotion Recognition Algorithm and a Multimodal Emotion Expression System to detect the user’s emotional state and generate appropriate responses through an effective dialogue manager. Interaction with EIROB is facilitated by an Android app, which uses the mobile phone’s inbuilt camera and microphone for facial expression and speech sentiment detection. The captured data is processed in the cloud that enables the model to be deployed remotely. This study aims to create an empathetic interaction model for senior citizens particularly during pandemics like COVID-19 to prevent feelings of isolation. The performance of the emotion recognition models is evaluated via accuracy, precision, recall, and F1-score to ensure the effectiveness of the system in recognizing and responding to user’s emotional states.