Date of Award

5-2024

Document Type

Thesis

Degree Name

Master of Science (MS)

College/School

College of Science and Mathematics

Department/Program

Computer Science

Thesis Sponsor/Dissertation Chair/Project Chair

Weitian Wang

Committee Member

Hao Liu

Committee Member

Rui Li

Abstract

This paper describes a proposed model for human-robot collaboration (HRC), its implementation on a real-world representative application, and a final evaluation on the overall collaborative effort. Modern manufacturing places a strong emphasis on cost-reduction and increased throughput. A common means of scaling production and increasing productivity is achieved through the deployment of industrial robots, some of which work collaboratively with human workers. This contemporary pairing poses a number of challenges, including worker safety, process efficiency, reliability, among others. This paper details the approach of a human collaborating with a robot that utilizes a vision system, employs a dynamically generated finite state machine (FSM) and a Long-Short Term Memory (LSTM) neural network to complete basic assembly tasks, where the robot is “taught” the preferred assembly sequence of the human operator. Following this “learning” experience, the robot will be able to predict its human partner’s movements and cooperatively complete assembly tasks that align with the same sequence. This teaching-learning-prediction-collaboration (TLPC) forms the basis of the approach described herein. A statistical analysis of this approach’s efficacy is evaluated via a multifaceted survey that was conducted across a diverse population who were paired with the robot in completing an assembly. The proposed solution suggests potential for scalability through HRC given new assembly sequences, improved throughput, and increased operator comfort.

File Format

PDF

Available for download on Wednesday, November 04, 2026

Share

COinS