Development of Brain-Controlled Assistive Feeding System

Abstract

Feeding difficulties and malnutrition are common phenomena in amyotrophic lateral sclerosis patients, locked in patients and people with upper limb disability. Feeding is often time consuming, unpleasant, and may result in choking or asphyxiation. Nowadays, robotic aids are applied to assist these people for eating. However, assistive robots that require movements from the user are not suitable for people with critical disabilities, including sensory losses, and/or difficulty in basic physical mobility. In this regard, a robotic system that can be controlled merely by brain signals is quite a remarkable aid. Therefore, based on the requirements for real-time assistive robot a prototype of an EEG-based feeding robot is proposed. The proposed feeding system enables the target group to eat independently. Experimental results show that the developed system is able to perform the required tasks, in real-time, with tolerable errors of around 17% in average. This amount of error can be further supervised to be reduced or in some cases even eliminated.