Embedded voice-controlled AI assistant for robotic arm operation in industrial automation

Abd Manap, Nurulfajar and Teow, Chean Yang and Putra, Azma (2025) Embedded voice-controlled AI assistant for robotic arm operation in industrial automation. Journal of Telecommunication, Electronic and Computer Engineering (JTEC), 17 (4). pp. 7-14. ISSN 2180-1843

Text 00306311220251121212844.pdf Download (506kB)

Abstract

The integration of Artificial Intelligence (AI) and Natural Language Processing (NLP) into Human–Machine Interfaces (HMI) has become increasingly significant in advancing Industry 4.0. This paper presents the design and implementation of an embedded voice-controlled AI assistant for robotic arm operation in industrial automation. The system employs a Raspberry Pi as the embedded platform, combined with Google’s Gemini Large Language Model (LLM) to interpret voice commands and execute precise movements on a six degrees-of-freedom (6-DoF) robotic arm through Pulse Width Modulation (PWM) control. The assistant architecture integrates speech-to-text conversion, context-aware NLP processing and servo-based actuation, providing a natural and hands-free interaction between humans and machines. Performance evaluation demonstrates a command recognition accuracy of 90% and an average execution time ranging from 3–10 seconds under laboratory conditions. The results highlight the feasibility of deploying LLM-powered voice assistants on embedded hardware for enhanced efficiency and usability in industrial automation. Future work will address robustness against noisy environments, enabling multilingual support and extending applicability to real-world industrial settings.

Item Type:	Article
Uncontrolled Keywords:	Natural language processing, Voice control, Industrial automation, Large language model, Robotic arm
Divisions:	Faculty Of Electronics And Computer Technology And Engineering
Depositing User:	Sabariah Ismail
Date Deposited:	03 Feb 2026 03:38
Last Modified:	03 Feb 2026 03:38
URI:	http://eprints.utem.edu.my/id/eprint/29404
Statistic Details:	View Download Statistic

Actions (login required)

View Item