Author

Clint Saidy

Date of Award

Spring 2021

Document Type

Open Access Dissertation

Department

Mechanical Engineering

First Advisor

Ramy Harik

Second Advisor

Abdel-Moez Bayoumi

Abstract

The manufacturing sector is continuously reinventing itself by embracing opportunities offered by the industrial internet of things and big data, among other advances. Modern manufacturing platforms are defined by the quest for ever increasing automation along all aspects of the production cycle. Furthermore, in the next decades, research and industry are expected to develop a large variety of autonomous robots for a large variety of tasks and environments enabling future factories. This continuing pressure towards automation dictates that emergent technologies are leveraged in a manner that suits this purpose. These challenges can be addressed through the advanced methods such as [1] large-scale simulation, [2] system health monitoring sensors and [3] advanced computational technologies to establish a life-like digital manufacturing platform and capture, represent, predict, and control the dynamics of a live manufacturing cell in a future factory.

Autonomy is a desirable quality for robots in manufacturing, particularly when the robot needs to act in real-world environments together with other agents, and when the environment changes in unpredictable or uncertain way. This dissertation research will focus on experimentally collecting sensor signals from force sensors, motor voltages, robot monitors and thermal cameras to connect to such digital twin systems so that more accurate real-time plant descriptions can be collected and shared between stakeholders. Creating a future factory based on an Industrial Internet-of-Things (IIoT) platform, data-driven science and engineering solutions will help accelerating Smart Manufacturing Innovation. Besides, this study will examine the ways of sharing knowledge between robots, and between different subsystems of a single robot, and implement concepts for communicating knowledge that are machine logical and reliable. My work will focus on applying the proposed methodology on more diverse manufacturing tasks and materials flows, including collaboratively assembly jobs, visual inspection, and continuous movement tasks. These tasks will require higher-dimensional information such as, analog plant signals, and machine vision feedback to be fed into and train the digital twin.

Share

COinS