Volume : VIII, Issue : V, May - 2018

AN IoT BASED HOME AUTOMATION AND HUMAN PATH PLANNING USING EXPECTATION MAXIMIZATION ALGORITHM

Drisya Payyeri, Sreekanth M, Akash P

Abstract :

This project focuses on a method for learning models of people’s motion behavior in an Internet of things (IoT) enabled indoor/home environment which maximizes energy. As people move through their environments, they often engage in typical motion patterns, related to specific locations and time. An IoT (Internet of Things) based sensor, controller and actuator architecture is developed, and through which identification of person and learning their motion pattern is achieved. Learning is based on the expectation maximization algorithm in machine learning.This algorithm, at the same time groups the trajectories which belongs to one’s motion behavior and learn the characteristics and motions of persons in a very time crucial manner. Every node is equipped with a microwave radar human presence sensor, a Wi–Fi enabled microcontroller and relays to control multiple devices. Every node is connected to a central server using Wi–Fi based MQTT protocol. The MQTT protocol takes only low computing/ power requirement than conventional IP based communication and works with a generalized publish–subscribe scheme. This method is efficient for IoT environments due to the lower data rates. Every presence data is send to server and the algorithm learns with the new arrived data and predicts the next moves. The advantage of this method is efficiency in computing and lower requirement in storage of data.

Keywords :

Article: Download PDF   DOI : 10.36106/ijar  

Cite This Article:

Drisya Payyeri, Sreekanth M, Akash P, AN IoT BASED HOME AUTOMATION AND HUMAN PATH PLANNING USING EXPECTATION MAXIMIZATION ALGORITHM, INDIAN JOURNAL OF APPLIED RESEARCH : Volume-8 | Issue-5 | May-2018

Number of Downloads : 217

References :