Author(s) :

Eric Gu, Jolie Friedman.

Volume/Issue :

Abstract :

Telebot aims to create a mentally operated mobile robotic manipulator that bolsters human productivity by changing the way humans interact with the world. Human brainwaves recorded noninvasively through electroencephalography can be used to mentally operate a mobile robotic manipulator through machine learning. This study determines the most compatible and effective type of brain activity to use as mental commands by comparing users’ visualization of images versus actions, as well as generic visualizations versus personalized ones. There were 6 participants in this study to evaluate the efficacy. Each trial was conducted while each participant wore an EMOTIV EPOC X–14 Channel Wireless EEG Headset running the EMOTIV Brain-Computer Interface (BCI). The participants were asked to think of generic images, individualized images, generic actions, and individualized actions. Each participant established a neutral baseline and references for each mental command through five, eight-second calibrations and then performed four live tests, one for each mental command. The data recorded the thoughts that were sustained above the detectable threshold, the number of false starts, and other metrics. The result uses the main metric to analyze the general efficacy, which measures the uptime of the mental command being above the detectable threshold. The data showed that, on average, generic images are effective 58.8% of the time, individualized images 74.3%, generic actions 59.7%, and individualized actions 77.7%. These findings contribute to the development of mobile robotic manipulators, potentially transforming the way humans interact with the world.