This thesis contains the implementation of the Matlab interface used to communicate to a 6-axis manipulator. The TM5-700 is a cooperative robot on whose arm is mounted an Eye-in-Hand camera. This interface is then integrated into a framework where also an HTTP server is needed to retrieve the pictures taken from the robot camera. This vision system is then used to detect a target object to be subsequently picked up by the robot. To do so the binocular vision system model is employed in order to retrieve the the depth information of the 3D position of the target object. The ORB algorithm is utilized to locate the target object from the two pictures. Once the 3D target position is retrieved, through the Matlab interface is possible to produce a script external to the robot that gives the robot instructions for the pick up routine. At the end the experimental results of the deployment of the framework are presented, together with an estimation of the depth perception error.

This thesis contains the implementation of the Matlab interface used to communicate to a 6-axis manipulator. The TM5-700 is a cooperative robot on whose arm is mounted an Eye-in-Hand camera. This interface is then integrated into a framework where also an HTTP server is needed to retrieve the pictures taken from the robot camera. This vision system is then used to detect a target object to be subsequently picked up by the robot. To do so the binocular vision system model is employed in order to retrieve the the depth information of the 3D position of the target object. The ORB algorithm is utilized to locate the target object from the two pictures. Once the 3D target position is retrieved, through the Matlab interface is possible to produce a script external to the robot that gives the robot instructions for the pick up routine. At the end the experimental results of the deployment of the framework are presented, together with an estimation of the depth perception error.

Sviluppo Interfaccia Matlab per Braccio Robotico Controllato Tramite Sistema di Visione

FANTINI, VALERIA BIANCA
2022/2023

Abstract

This thesis contains the implementation of the Matlab interface used to communicate to a 6-axis manipulator. The TM5-700 is a cooperative robot on whose arm is mounted an Eye-in-Hand camera. This interface is then integrated into a framework where also an HTTP server is needed to retrieve the pictures taken from the robot camera. This vision system is then used to detect a target object to be subsequently picked up by the robot. To do so the binocular vision system model is employed in order to retrieve the the depth information of the 3D position of the target object. The ORB algorithm is utilized to locate the target object from the two pictures. Once the 3D target position is retrieved, through the Matlab interface is possible to produce a script external to the robot that gives the robot instructions for the pick up routine. At the end the experimental results of the deployment of the framework are presented, together with an estimation of the depth perception error.
2022
Development of a Matlab Interface for a Vision-Guided Robotic Arm
This thesis contains the implementation of the Matlab interface used to communicate to a 6-axis manipulator. The TM5-700 is a cooperative robot on whose arm is mounted an Eye-in-Hand camera. This interface is then integrated into a framework where also an HTTP server is needed to retrieve the pictures taken from the robot camera. This vision system is then used to detect a target object to be subsequently picked up by the robot. To do so the binocular vision system model is employed in order to retrieve the the depth information of the 3D position of the target object. The ORB algorithm is utilized to locate the target object from the two pictures. Once the 3D target position is retrieved, through the Matlab interface is possible to produce a script external to the robot that gives the robot instructions for the pick up routine. At the end the experimental results of the deployment of the framework are presented, together with an estimation of the depth perception error.
Vision
Robotic Arm
Matlab Interface
Object Detection
File in questo prodotto:
File Dimensione Formato  
Fantini_ValeriaBianca.pdf

accesso aperto

Dimensione 3.96 MB
Formato Adobe PDF
3.96 MB Adobe PDF Visualizza/Apri

The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/20.500.12608/53807