Indoor Navigation for People with disability(Wheelchair)

Abstract

Purpose: There are no better ways to navigate indoor areas. It is certainly a necessity for people with disability, specially wheelchair users. The purpose of this research is to develop an indoor navigation software which uses smartphone and machine learning to help users navigate indoor efficiently. Relevant Research Context: There are some research work done for indoor navigation which uses magnetometer. It is more accurate than Wi-Fi and GPS and saves energy. My research mentor uses accelerometer for tracking wheelchair user’s mobility for improving healthy life style of wheelchair users. Description: First step, building a mobile application (currently using Android) that can collect accelerometer, gyroscope and magnetometer data. Second step, based on those data, we will create segments about stationary, moving and turning. We will use those segments as training data for machine learning algorithms. Third step, by using machine learning techniques, we will accurately determine a wheelchair’s moving status. Fourth step, we will use path finding for better user navigation. Last step, we will combine wheelchair’s moving status and path finding to determine the real time updating location of the user and instruct him accurately. Conclusion: people with disabilities, especially wheelchair users, have hard time navigating in indoor areas. This application will help those people easily navigate indoors.