Brain Controlled UGV: Desenvolvimento de veículo terrestre não-tripulado controlado por interface cérebro-computador

Abstract

This study aims to develop a Brain-Computer Interface (BCI) responsible for con trolling the movements of an Unmanned Ground Vehicle (UGV), demonstrating that it is possible to develop a Brain-Robot Interface (BRI) using locally available or low-cost tools within the context of UFPB. Significant advances in Human-Computer Interaction (HCI) have been observed over the past decades, among which the development of BCIs stands out, offering the potential to overcome the learning curve barrier present in existing HCI technologies, regardless of how natural and intuitive they may be. Similarly, the use of mobile robots has become increasingly common in the context of Industry 4.0, reflecting the growing interest from industries, companies, and academia in incorporating mobile robots into their activities and research. In this context, the present study is situated at the intersection of these two highly promising fields, aiming to create a proof of concept for a UGV controlled by Electroencephalography (EEG). In accordance with the Design Science Research Methodology (DSRM), the BCI and the UGV were developed separa tely and subsequently integrated. The BCI was primarily developed using tools provided by Emotiv, specifically the Emotiv Insight EEG headset and the Emotiv SDK. Similarly, the constructed mobile robot features differential traction and is powered by two 6V DC motors, controlled via an H-bridge driver and an ESP-WROOM-32, and powered by a 9V alkaline battery. Communication between the BCI and the UGV is achieved through the Message Queuing Telemetry Transport (MQTT) protocol. The results indicate a positive response to the research question, confirming that it is feasible to develop a BRI using the tools already employed at UFPB, while still leaving considerable room for improvement in future work.