Object Detection Algorithms and Implementation in a Robot of Service

Abstract

The recognition of objects and the proper handling of them is a common problem in service robots. This article presents a set of computational vision algorithms to improve an object detection system using an RGBD sensor. Within this set of algorithms is the computational vision algorithm to identify a plane, a second algorithm to determine the position of an object and a third algorithm that performs the Principal Components Analysis to identify the orientation of the objects and to perform a correct manipulation. Several tests are presented that validate our approach. This group of algorithms is used by the service robot Justina at the upcoming RoboCup 2018 international tournament [1].