Design, Fabrication and Simulation of a 5-dof Robotic Arm using Machine Vision
Konstantinos Pagonis, Paraskevi Zacharia, Antreas Kantaros, Theodore Ganetsos, Konstantinos Brachos
Abstract
This paper studies the design, fabrication and simulation of a robotic arm that possesses the ability to use information that receives from its environment through machine vision. The main objective of this work is to design, construct and simulate a robotic system able to recognize and identify various target objects located in the robot's workspace using a camera. The image processing is implemented using a clever convolutional neural network (CNN). After studying the kinematics of the 5-dof robotic manipulator, a robotic arm is designed through a 3D design program. The mechanical parts of the robot are constructed through a 3D printer and are assembled with the servo motors. Next, the desired signals are being sent to the servo motors through a Matlab Simulink program and the Arduino Microcontroller with the intention to rotate the robot joints.