LARN: Indoor navigation for elderly and physically challenged

Abstract

Automatic navigation systems have a wide range of applications and are usually implemented in many robotic systems. The area of interest here is an automated powered wheelchair for the elderly and physically challenged individuals, which empower them to navigate inside their homes without having to continuously steer the wheelchair. Such a system dramatically reduces the amount of continuous physical effort the user has to apply. This is achieved by mapping the floor plan of a house into number of square grids, with each grid having a unique address. The Location Aware and Remembering Navigation (LARN) algorithm, which is based on the Dijkstra's shortest path algorithm, calculates the shortest path to a set of predefined locations inside the house. With sufficient hardware to measure the orientation of the wheelchair and distance traversed, LARN can dynamically calculate the shortest path to any of the predefined locations. The greatest advantage of implementing LARN is the fact that the system would require no wireless devices to function. This paper discusses the implementation of the LAN algorithm on MATLAB. Simulation and testing of the algorithm is done in MATLAB 2008 and 2010 in both Intel i3 and i5 processors.