Recognition of human emotions in reasoning algorithms of wheelchair type robots

Abstract

This paper analyses the possibilities of integrating different technological and knowledge representation techniques for the development of a framework for the remote control of multiple agents such as wheelchair-type robots. Large-scale multi-dimensional recognitions of emotional diagnoses of disabled persons often generate a large amount of multi-dimensional data with complex recognition mechanisms, based on the integration of different knowledge representation techniques and complex inference models. The problem is to reveal the main components of a diagnosis as well as to construct flexible decision making models. Sensors can help record primary data for monitoring objects. However the recognition of abnormal situations, clustering of emotional stages and resolutions for certain types of diagnoses is an oncoming issue for bio-robot constructors. The prediction criteria of the diagnosis of the emotional situations of disabled persons are described using knowledge based model of Petri nets. The research results present the development of multilayered framework architecture with the integration of artificial agents for diagnosis recognition and control of further actions. The method of extension of Petri nets is introduced in the reasoning modules of robots that work in real time. The framework provides movement support for disabled individuals. The fuzzy reasoning is described by using fuzzy logical Petri nets in order to define the physiological state of disabled individuals through recognizing their emotions during their different activities.