Human emotion to control the quadcopter

• Main Author: Tan, Teck Leong
• Other Authors: Suresh Sundaram
• Format: Final Year Project
• Language: English
• Published: 2016
• Subjects: DRNTU::Engineering
• Online Access: http://hdl.handle.net/10356/66447
• Institution: Nanyang Technological University

Many research have been conducted on Electroencephalography (EEG) to enable user to interact with computer and extract useful data. The recent works on Brain Computer Interface (BCI) had witnessed the breakthrough of enabling human to have control over objects by using thoughts and emotions. Some studies have also revealed that the quality of the EEG signals can be affected by the number of acquisition channels and the non-stationarity properties in EEG. In this project, a non-invasive BCI is used to extract the motor imaginary tasks from the user. Then, a framework is deployed to translate the motor imaginary tasks into commands to drive the quadcopter in the intended direction. The framework comprises of the Common Spatial Pattern (CSP) and Meta-Cognitive Interval Type-2 Fuzzy Inference System Classifier (McIT2FIS) algorithms. The CSP is responsible for discriminating the features in the motor imaginary tasks while McIT2FIS manages the non-stationary in the motor imaginary tasks and performs the classification of move-left or move-right commands. The result revealed that CSP-McIT2FIS is effective in classifying the left and right motor imaginary tasks. The user was able to drive the quadcopter to the desired direction smoothly. In near future, more research and development will be invested to ensure the robustness in classify the motor imaginary tasks of any user. It is envision to enable people who are suffering from physically disabilities to train and employ their motor imaginary tasks to drive their wheel chair safely.