Navegación basada en seguimiento de líneas mediante visión artificial para el robot Robogait Sport

Resumen

In this work, the design process, part selection, programming and integration of a line detection and following system is thoroughly described, adapting to the structure, hardware and control systems already present in a prototype called ROBOGait SPORT. It is a mobile gait analysis robot for humans and, more specifically, it analyzes the way athletes run. At the start of this work, the GPS system proposed to guide the robot was already developed, but it did not function correctly. Because of this reason, in this an alternative guiding mode from the one implemented in the original prototype it’s taken on. Taking advantage of the fact that tests with the subjects are done in athletic tracks, using the differentiated coloration of the lines that separate the runner’s tracks in the circuit, to delimit and guide the route of the robot using a computer vision system was proposed as a more effective method. To do this, the Raspberry Pi 5 microcontroller, Raspberry’s Camera Module 3 and the open source computer vision library OpenCV and the photographic sensor control library libcamera have been used.