Gå til hovedinnhold

Digital Twin for Sharebox Locker: 3D Visualization and AI-Powered Troubleshooting

Hopp til hovedinnhold

Sharebox, a leading provider of key exchange and access management solutions, collaborates with Twilligent to enhance customer support and streamline troubleshooting processes. Their physical lockers facilitate secure and convenient key sharing for various industries. This thesis proposes developing a digital twin for a specific Sharebox locker model.

Aim

The aim of this thesis is to create a digital twin of a Sharebox locker that:

  1. Provides a real-time 3D visualization of the locker's internal components and their states. This will include motor positions, door status, and any potential errors.
  2. Incorporates a large language model (LLM) to develop an intelligent support bot. This bot will leverage the 3D visuals and real-time data to assist customers in troubleshooting and resolving issues.

Objectives

  1. 3D virtual locker reflecting the locker state. Locker data can be hard coded with a few predefined demos showcasing different errors. Use the Unity Game Engine to build the 3D virtual environment.
  2. Implement a large language model (e.g. ChatGPT or Google Gemini) capable of understanding natural language queries and providing contextually relevant responses based on documentation from Sharebox.
  3. Support Bot Development: Integrate the LLM with the virtual locker and develop a user-friendly interface for interacting with the support bot and troubleshooting the locker.
  4. Evaluate the effectiveness of the digital twin and support bot in improving customer support and reducing downtime.

Supervisor

Twilligent will provide supervision and guidance throughout the project, leveraging their expertise in digital twin technology.

Expected Outcomes

1.      A functional digital twin of a Sharebox locker in Unity, capable of showcasing predefined error scenarios through 3D visualization and real-time component states.

  1. An AI-powered support bot capable of assisting customers with troubleshooting.
  2. A valuable tool for Sharebox to enhance customer support and reduce locker downtime.

Keywords

Digital twin, Sharebox, 3D visualization, large language model, troubleshooting, customer support, Twilligent

Oppdragsgiver

Twilligent

Twilligent empowers industrial companies with digital twins. They are based in Arendal and works delivers a software as a service platform for digital twins. Through their strong connection to UiA, the company aims to be at the forefront of research on digital twins and artificial intelligence. Twilligent was founded in 2021. Despite their relatively short history, they have several ongoing projects and have taken a significant position in the market as a provider of digital twins.

Oppgaveforslag

Type: Fra virksomhet
Publisert: 2024-09-26
Status: Ledig
Grad: Uspesifisert

Fagområder

Dette forslaget er åpent for flere studentoppgaver.

Kontaktperson