BSPBSM 2022
This project is about a P300 BCI paradigm of navigating a Zumo robot in order to follow the lines. Project has different submodules, which can be summarized as paradigm, offline preprocessing and classification, online preprocessing and classification, and controlling the robot.
The offline folder contains all the steps in the offline pipeline. In the p300_paradigm.py script, the visual platform to collect the training data is provided. There are four arrows placed on the screen, describing the possible directions that the Zumo robot can move towards to: right, left, forward, and backward. At every trial, one of the arrows are selected as target and emphasized with the red color. This signals the user to fixate on the given arrow. Then the arrows are highlighted with yellow color randomly. While the arrow that the user fixates is highlighted, a P300 signal is produced in the user's brain.
In the offline_preprocessing.py script, the raw data from numerous sessions are collected and concatenated. Later, the data is band-pass filtered, baseline corrected, common average referencing applied, and processed to smaller epochs. These epochs are later saved.