This paper presents a computer vision method that, taking information from an RGB-D camera, performs real time 3D object recognition to be used in cognitive robotics experiments, where the real time constraints are key. To this end, we have implemented and tested an algorithm that combines a deep neural network (YOLOv3 tiny) that processes RGB images and provides object recognition and 2D localization, with a point cloud analysis method to compute the third dimension. The proposed approach has been tested in real-time manipulation experiments with the UR5e robotic arm through ROS, and using a GPU to execute the method, showing that this combination allows for an efficient real-time learning using cognitive models.
An Approach to 3D Object Detection in Real-Time for Cognitive Robotics Experiments
Lect. Notes in Networks, Syst.
Iberian Robotics conference ; 2022 ; Zaragoza, Spain November 23, 2022 - November 25, 2022
2022-11-19
12 pages
Aufsatz/Kapitel (Buch)
Elektronische Ressource
Englisch
Real-time object determination for space robotics
Tema Archiv | 1988
|Platform for Robotics Experiments
British Library Conference Proceedings | 2005
|Experiments in Autonomous Robotics
NTIS | 1987
|Ada and real-time robotics: lessons learned
Tema Archiv | 1994
|