Kelly Cashion, Carly A. Gross



Download Project (719 KB)


Brain machine interface (BMI) also known as brain computer interface (BCI) is a field of research that has been explored in varying degrees throughout the last few decades. Initial research used invasive technology in order to read the signals from the human brain. These systems required surgery in order to connect the subjects to the sensors. Recent trends have moved toward non-invasive systems that make use of non-invasive physiological sensors such as electroencephalographs (EEG). EEG systems use a number of electrodes to read electrical signals on the scalp caused by brain activity. The patterns generated by certain thoughts can be classified and recognized by a BMI system using machine learning algorithms. These classified patterns can then be encoded as commands to prompt a certain response from a computer or machine. The completed system allows for control of the connected device using thought as the only input. The possible uses for a BMI system are as varied as the designs of computer programs and computer controlled devices. One of the most noteworthy applications of BMIs is in the field of medicine. BMIs offer the tools for the disabled to interact with the world, even if they are suffering from severe nerve damage between their brain and original limbs. In the case of a lost or paralyzed limb, BMIs offer the potential for patients to use a robotic limb, controlled with their natural thought patterns, to interact with the world. BMIs also offer potential modes of communication for patients who have no other way to convey their thoughts. With these applications in mind, this research focuses on control of a robotic arm using a 14-electrode EEG headset. Both pure EEG signals and electromyography (EMG) signals are encoded as controls for six possible actions performed by the robotic arm.

Publication Date


Project Designation

Graduate Research

Primary Advisor

Vijayan K. Asari

Primary Advisor's Department

Electrical and Computer Engineering


Stander Symposium project

Brain Machine Interface Using Electroencephalograph Data as Control Signals for a Robotic Arm