-
Matheus Loureiro
Programa de Pós-Graduação de Engenharia Elétrica, Universidade Federal do Espírito Santo, ES
-
Fabiana S. V. Machado
Programa de Pós-Graduação de Engenharia Elétrica, Universidade Federal do Espírito Santo, ES
-
Ricardo Mello
Programa de Pós-Graduação de Engenharia Elétrica, Universidade Federal do Espírito Santo, ES
-
Anselmo Frizera
Programa de Pós-Graduação de Engenharia Elétrica, Universidade Federal do Espírito Santo, ES
Keywords:
Autonomos Navigation, Computer Vision, Mobile Robot, Object Count, Pose Estimate, RGB-D Camera
Abstract
Robotic devices capable of interacting with the environment are increasingly common, with autonomous robots as an example of this. However, it is necessary for these devices to be able to extract as much information as possible from the environments that they are envolved. To this, computer vision techniques are used to process images of the environment. So, the objective of this work is to propose a strategy that uses computer vision techniques to identify objects to count and estimate their position, from an RGB-D camera coupled to an autonomous mobile robot. This strategy was validated through testing in 3 different scenarios, which explore the robot’s autonomous navigation capability and camera image capture. The results of all scenarios presented a relevant object count and a average error less than 0.19 meters in the pose estimate.
