Applications developmet for a person robot system
Technologies such as robotics have benefitted and made an impact on several industrial. Combining artificial intelligence (AI) enables the robot to perform many complex tasks such as couriers service to bring business to the front of customers. The usage of robotics also applies to businesses operat...
Saved in:
Main Author: | |
---|---|
Other Authors: | |
Format: | Final Year Project |
Language: | English |
Published: |
Nanyang Technological University
2021
|
Subjects: | |
Online Access: | https://hdl.handle.net/10356/153209 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Nanyang Technological University |
Language: | English |
Summary: | Technologies such as robotics have benefitted and made an impact on several industrial. Combining artificial intelligence (AI) enables the robot to perform many complex tasks such as couriers service to bring business to the front of customers. The usage of robotics also applies to businesses operating in retail stores to perform tasks such as answering simple questions from customers and assisting with inventory monitoring. These tasks performed by robots can benefit workers in the retail store to focus more on the critical task. Due to the Covid-19 pandemic, workers in the retail store can reduce the risk of infection by facilitating interaction with customers by substituting robots to serve the customers in front of the stores.
This report will present the process of building the application that incorporates the existing platform robot, the Misty robot, and its integrated hardware to build an application to bring services to customers. The robot will combine Rasa open-source framework, which uses conversation AI technologies and Natural Language Processing to understand the natural language from user utterance through speech. The Rasa open-source framework uses the DIET model (Dual Intent Entities Transformer) to perform both classification and entities recognition. In this project, the Rasa DIET model is enhanced by modifying the pipeline to perform sentiment analysis on user utterance to accurately predict customer sentiment and allow the misty robot to react and respond accordingly.
The final part of the report in Appendix C provides an example of a use case demo to simulate the input utterance and response result from Rasa when users provided speech input to misty robot. The scenario will assume that misty is deployed as a sales assistant to answer customer queries in retail stores. |
---|