Session: 02-10-01: Session #1: Robotics and Automation in Advanced Manufacturing

Paper Number: 94986

94986 - Assembly Automation Using an Industrial Robot 

Industrial automation is an increasingly prominent process that has been around for over 60 years. Industrial robots have been at the forefront, and their advancement has caused them to spread from industries such as automotive manufacturing, to healthcare and even silicon and circuit board manufacturing. The primary user of industrial robots is the automotive industry, which relies heavily on robots for material handling and additive manufacturing processes such as welding, and painting, has been at the forefront of industrial automation since the 1970’s. With the advancements to robot technology, the capabilities of industrial robots have been able to improve and expand into doing more complex operations such as material removal, and assembly tasks.

One of the most well-known brands of industrial robot is FANUC, which comprises over 60% of the Industrial Robotics market in North America. They offer a wide variety of industrial robots to their consumers, which include, Articulated, SCARA, Delta, and Cartesian coordinate robots. FANUC robots are uniquely designed to offered a completive edge over other manufacturers. These robots can incorporate built in iRVision camera software which allows quick, easy, and intuitive setup of vision guided robot tasks, as well as ROBOGUIDE simulation software. ROBOGUIDE is a powerful software that allows you to create a virtual work cell where you can easily simulate a realistic robot task. This task can be analyzed for performance and can allow a company to make a decision on whether or not automating a certain task is cost effective, without having to spend a singlemuch time and resources.  cent.

The focus of this paper is the robotic process of vision guided automated assembly. Standard robotic manufacturing processes rely on tooling to make sure components are in the correct position and orientation. Developing tooling costs money, and relies on the underlying assumption that during each cycle, the robot performs the same movements. Applying camera vision to the industrial robots allows the creation of a fully automated process that can be used on its own, or added into an existing process to improve precision by ensuring that any variation or drift is accounted for. The robot used for this study is the FANUC LR Mate 200iD/4S, which is a low payload, articulating, six-axis robot. The goal of the study is to incorporate the iRVision software present available on FANUC robots, with their simulation program ROBOGUIDE for the automated assembly task. Once the process is developed and simulated virtually, it will be implemented on a physical robot. The assembly process focused on in this paper is a circuit board assembly, which consists of a mock bare circuit board and mock components that include capacitors, microchips, and processors. These parts will be 3D printed and individual components will be placed in part trays to simulate real world scenarios. The robot will be responsible for locating the parts regardless of their position and/or orientation, as long as they are within its the camera’s field of view. The robot will also be responsible for locating a fiducial on the assembly circuit board which will allow it to calculate the position and orientation of the assembly fixture. The performance of the robot will then be evaluated by randomly placing and orienting the assembly fixture, and the part trays within the robot camera’s field of view. The robot will receive a passoperation will be considered successful  if it can successfully pick up parts in the correct order and place all of the parts in their respective locations on the assembly fixture. The performance of the virtual robot will be compared to the physical robot in order to compare a real-world and an ideal scenario.in terms of an accuracy of identification of the parts and cycle times.

Presenting Author: Vedang Chauhan Western New England University

Presenting Author Biography: Dr. Vedang Chauhan is an Assistant Professor in Mechanical Engineering Department at Western New England University (WNE) in Massachusetts, USA. He has more than 16 years of research and teaching experience in the field of Mechatronics Engineering, Applied Machine Learning, Machine Vision and Robotics. During his academic tenure, he has worked on the private and government funded industry research projects and published scholarly articles and acted as a reviewer for machine vision, robotics, sensors and machine learning journals.

Authors:

Timofey Dragun Western New England University
Seth Mascaro Western New England University
Jonathan Blanchard Western New England University
Vedang Chauhan Western New England University