Indoor autonomous robot: Perception of the environment through computer vision

  • Felipe Fabricio UTFPR
  • André Schneider de Oliveira PUCPR
  • A. S. Marco Teixeira PUCPR


Mobile robots are a trend. We went through several waves of automation techniques in the world, we had industries, the internet, and now we see the advancement of Artificial Intelligence (AI). Another wave that is emerging is mobile robotics. These devices can overcome the barrier of a computer and perform tasks in the real environment, replacing people in repetitive or dangerous tasks. This project aims to develop an autonomous robot to operate in an environment of collaborative companies, commonly known as coworking. For such an action, the equipment needs to understand the environment in which it is located, so that it can interact with it. This work focuses on the perception of the environment, with the proposal of an object identification approach with 22 classes. The desired operating environment for the equipment is in Curitiba, Brazil, at FINEPSENAI. The startup 33Robotics develop this project.
Palavras-chave: Office Robots, Deep Learning, Object Detection, Transfer Learning, 33Robotics
FABRICIO, Felipe; OLIVEIRA, André Schneider de; TEIXEIRA, A. S. Marco. Indoor autonomous robot: Perception of the environment through computer vision. In: SIMPÓSIO BRASILEIRO DE ROBÓTICA E SIMPÓSIO LATINO AMERICANO DE ROBÓTICA (SBR/LARS), 15. , 2023, Salvador/BA. Anais [...]. Porto Alegre: Sociedade Brasileira de Computação, 2023 . p. 543-548.