Improving Smart Lighting with Activity Recognition Using Hierarchical Hidden Markov Model

Abstract

This paper has the aim of implementing the smart lighting systems that is able to analyze daily movement activities, analyze the performance of hierarchical hidden markov models as predictions and analyze the performance of smart lighting with activity analysis using hierarchical hidden markov models. The purpose is to answer the problems that occur, namely the smart lights only turn on if users are right under the lights so users need a smart light which is able to read the movement of people when approaching the lamp or not. Secondly, there are also smart lights, but when users are under the lights, it only lights up for a few seconds which should light up if there is a person below or a radius around the lamp so that a smart light is needed when someone is underneath and the lights will die it is outside the radius around the lamp. The model used is the hierarchical hidden markov model which is an extension of the hidden markov model which can solve the problem of evaluation, conclusion and learning with the algorithm used is the viterbi algorithm. The result obtained using HHMM are accuracy of 93%, 92% recall and 86% precision.