Robot Modelling and Control towards Digital Twin Technology

Tan, Xiao Ze (2024) Robot Modelling and Control towards Digital Twin Technology. Final Year Project (Bachelor), Tunku Abdul Rahman University of Management and Technology.

Text TAN XIAO ZE_Full Text.pdf Restricted to Registered users only Download (28MB)

Abstract

This project addresses the problems in manufacturing machinery maintenance, inefficient robot evaluation, and integrating virtual twins with physical robots. By providing detailed operational insights, reducing resource-intensive physical trials, and improving synchronization and data accuracy, it enhances efficiency and accuracy in maintenance, evaluation, and simulation processes. The project aims to create accurate virtual representations of physical robots, generate motion sets between them, and evaluate the effectiveness of data exchange. This involves translating robot attributes into virtual twins, transferring motion between virtual and physical robots, and ensuring seamless communication. Since the emergence of Industry 4.0, the concept of digital twin (DT) has transformed engineering and technology. This project demonstrates the implementation of the DT concept for a robot. A DT is a virtual replica of a real-world object that integrates realtime data and simulation models for monitoring, analysis, and optimization. The study involves creating a virtual model of a 6 degree of freedom (6 DOF) robotic arm, incorporating key DT characteristics. Leveraging the features of the 3dExperience platform, this virtual model can control an actual 6 DOF robot. This means that motion commands from the virtual model can be transmitted to the real robot for execution. Conversely, the actions performed by the physical 6 DOF robotic arm can be relayed to the 3dExperience software, where the same motions can be visualized. The project involves several key phases. It starts with building a virtual twin robotic arm using 3dExperience platform, covering design, simulation, and control. CATIA in 3dExperience is used for initial design, followed by constructing mechanical components and simulating robot movements. A Bidirectional Communication Module enables two-way communication: it facilitates controlling the physical robot through 3dExperience and also enables the representation of the physical robot’s motion on the virtual model within 3dExperience. A user-friendly interface is developed, giving users the option to choose between two functions: controlling the robot using motion sets exported from 3dExperience or exporting the physical robot’s motion sets to 3dExperience’s virtual model for virtual motion illustration. As a result, the project successfully develops a virtual twin for the xArm6 Robotic Arm, accurately replicating its physical characteristics and behaviors while enabling seamless data exchange between virtual and physical environments. Analysis reveals an average error of approximately 7.67% in motion simulation times, highlighting the virtual twin’s capability for realistic simulations despite moderate discrepancies.