A Voice-Controlled Robotic System Using ChatGPT for Intuitive Human-Robot Interaction

Industry 5.0 emphasizes the human-centered integration of advanced technologies within production systems, promoting adaptive and collaborative frameworks where humans remain central to system intelligence and decision-making. This work presents a natural language, voice-controlled framework that leverages ChatGPT to interpret spoken commands and facilitate intuitive control of a mobile robot simulated in MATLAB. The architecture integrates speech-to-text conversion, context-aware reasoning through generative AI, and real-time bidirectional communication between Python (ChatGPT) and MATLAB via a feedback loop. To evaluate the system’s effectiveness, three user-centered experiments were conducted with twelve participants issuing natural language commands under increasing task complexity. The experiments demonstrated the system’s ability to accurately interpret user intent and execute corresponding robot behaviors. User performance and experience were assessed using the NASA Task Load Index (NASA-TLX), revealing high scores in confidence and perceived performance, along with low ratings in effort, mental, and physical demand. Nonetheless, moderate frustration and temporal demand scores were observed, primarily due to response delays in voice processing. The third experiment, which involved the most complex navigation scenario, yielded the highest performance ratings, underscoring the system’s potential in challenging environments. Overall, the results highlight the framework’s usability and promise for real-world human-robot interaction and collaborative applications. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2026.