Implementación de una interfaz cerebro computadora para el control de una silla de ruedas dentro de de un ambiente virtual en UNITY utilizado SSVEP

A brain-computer interface (BCI) is a system that establishes direct communication between the brain and a computer by recording the electrical activity of the brain. It makes use of Steady State Visual-Evoked Visual Evoked Potentials (SSVEP). The data obtained from the signals with respect to visual stimulation are filtered and used to train an artificial intelligence model, which identifies and classifies the user's activity intention, responsible for issuing control commands. The detection of brain neural activity is performed through OpenBCI, making use of the SSVEP protocol, the signals are stored and subsequently processed in Python interpreted programming language. For the acquisition of EEG signals, the user is instrumented by placing electrodes and the OpenBCIsystem, the impedance of the electrodes is determined and it is verified that they are in the required magnitude. They are recorded in 3 channels located in the occipital region: PO3, PO4, and OZ with a sampling frequency of 250 Hz and with a Notch filter to normalize response peaks that change with impedance. The visual stimuli are generated by an application using the Psychtoolbox-3 toolbox, running in the Matlab environment, where the EEG signal processing and command identification is performed by the method of filter bank. At the end of this project, it is expected to validate what has been done in a virtual environment in Unity 3D that simulates an apartment with the furniture where a wheelchair is located so that it can be guided by the user through the classification of EEG signals using the SSVEP mental strategy and the use of the OpenBCI system, As observed later on, where the user sees the stimulus on a screen to which the wheelchair is directed, it will move in that direction. It should be noted that the system due to some interference in the signal acquisition can identify movement towards another direction, such as towards the left and can randomly take it upwards, downwards or to the right. All this is due to errors that may occur when taking the subject’s signal, but in general it can be said that despite the fact that the interface may throw certain errors, it has a percentage of successes above the range considered adequate, thus correctly implementing the interface.

• Autores: Rivera Larrañaga, Johnny Esteban
• Tipo de recurso: Trabajo de grado de pregrado
• Fecha de publicación: 2023
• Institución: Universidad Antonio Nariño
• Repositorio: Repositorio UAN
• Idioma: spa
• OAI Identifier: oai:repositorio.uan.edu.co:123456789/8406
• Acceso en línea: http://repositorio.uan.edu.co/handle/123456789/8406
• Palabra clave: Interfaz cerebro computador; Silla de ruedas virtual; Inteligencia Artificial; 56.23 R621i; Brain Computer Interface; Virtual Wheelchair; Artificial Intelligence
• Rights: openAccess
• License: Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0)