-
Mateus de Oliveira Soares
Instituto Federal de Sergipe, SE
-
João Pedro Juchum Chagas
Universidade Federal de Campina Grande, CG
-
Raimundo Silvério Freire
Universidade Federal de Campina Grande, CG
-
Phillipe Cardorso Santos
Instituto Federal de Sergipe, SE
-
Stephanie Kamarry
Instituto Federal de Sergipe, SE
-
Roberto da Silva Macena
Instituto Federal de Sergipe, SE
Keywords:
Autonomous Wheelchair, Natural landmarks, tactile floor, image processing, assistive technology
Abstract
This work proposes the development and application of an autonomous navigation system for an electric wheelchair, based on tactile floor tracking. This approach is based on the installation pattern of tactile floors, a mandatory item in public buildings, which allows a simple and quick mapping of the environment, also being useful as natural landmarks for the development of an efficient localization system for the wheelchair . This locatization system is performed through computer vision based on the color segmentation of tactile floors. The proposed system is also composed of a human-machine interface by voice commands, which allows the individual to choose the location of a known environment to which he wants to move. The proposed approach was implemented in an electric wheelchair and tested at the Federal Institute of Sergipe, Campus Estância, where autonomous movements could be performed between different locations correctly. For safety reasons, the tests were conducted with an empty chair, in other words, with nobody on board.