Smart Home Assistance for humans with episodic memory decline problems using a fuzzy rule-based mechanism: The case of stove usage

Abstract

Nowadays, the fast pace of life is causing problems to many people concerning lapses of attention or episodic memory decline. These problems are to blame for many annoying incidents, like forgetting where the car is parked, to extremely serious situations, like an accident due to a fire provoked from a possible failure to remember to turn off the oven. In view of these problems, the aim of this paper is to assist humans with episodic memory decline problems or lapses of attention in carrying out everyday tasks and protect them from the dangers that their memory impairment may cause at home. Particularly, this paper presents a fuzzy inference system for smart homes that monitors the user’s everyday activity and calculates the degree of the emergency that is caused by an inconsistency in the monitored person’s actions and generates the appropriate alert. The input of the fuzzy inference system consists of variables that are associated with the monitored person’s activity like time, moves and apparatus and performs goal recognition and problem diagnosis in the sequence of actions. The presented system uses fuzzy sets and fuzzy rules to generate alert messages that notify the monitored person about the next action that s/he has to do in order to complete a particular activity. This paper presents of stove usage as a case study of research.