Multi-agent-based human computer interaction of e-health care system for people with movement disabilities

Abstract

This paper presents further development of intelligent e-health care system for people with movement disabilities. The research results present the development of multi-agent-based human computer interaction model and its integration into the system of fuzzy neural control of speed of two wheelchair type robots to providing movement support for disabled individuals. An approach of filtering of skin conductance (SC) and electrocardiogram (ECG) signals using Nadaraya-Watson kernel regression smoothing in R tool for emotion recognition of disabled individuals is described and implemented in the system. The unsupervised clustering by self organizing maps (SOM) of data sample of physiological parameters extracted from SC and ECG signals noninvasively measured from disabled was proposed. It was implemented to reduce teacher noise as well as to increase of speed and accuracy of learning process of multi-layer preceptron (MLP), which was applied in the multi-agent-based human computer interaction model for online prediction of e-health care state of disabled individuals.