Visual Servoing with Two Cameras for Object Grabbing Using Yaskawa SIA10D

• Main Author: Julyando
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/36536
• Institution: Institut Teknologi Bandung

The use of robots for industry is increasing rapidly, especially for manufacturing processes where robots are programmed to do work according to the input given. Visual servo is a camera-based robot control where the camera will track the object to be carried out. The type of end-effector installed on the robot will determine the work that the robot will do to the object. But in principle, the end-effector must move towards the object until it has a certain offset so that it can do the job properly. In this thesis, the writer will design robot motion control system program that is capable of tracking and retrieving predetermined objects which will then be placed in a designed location For the tracking process, two cameras are used and installed in two different places. The first camera is mounted on the robot to determine the position of the x and y of the object, while the second camera is mounted in front of the robot at a specified distance to determine the position of z from the end-effector to the object. By knowing the values of x, y, and z, the robot can take action on the object. From the results of the experiment it was found that the robot can track accurately enough and every time the retrieval process takes place in about 20 seconds for each different object. The use of two cameras allows the robot to determine the position of the end-effector in motion in a good direction so that avoiding blind-grabbing when objects are not caught by the camera installed in the end-effector.