Navegação autônoma de cadeira de rodas motorizada em ambiente previamente conhecido

Abstract

Autonomous systems consist in a sequence of rules designed to make decisions and perform tasks without the need for full-time human supervision. Autonomous wheelchair is an assistive technology that facilitates the mobility of people with mobility impairments. In order to assist the mobility of these people, an intelligent embedded system was developed for the autonomous navigation of a motorized wheelchair in a known environment. The onboard electronics will consist of two microcontrollers, a Raspberry that will be responsible for all system control and monitoring of the sensors, and an ESP32 that will drive the motors for correct locomotion through fuzzy control logic. In addition to these devices, the system also consists of encoder sensors and a color sensor applied in the generation of data for the navigation technique called odometry and recognition of artificial landmarks through tactile floors, helping to determine the absolute positioning of the wheelchair. The virtual site map is generated using the discrete map technique and loaded into the embedded system database, so that a trajectory algorithm is implemented to devise a possible path between a starting point and a chosen destination (sic).