EEG controlled Mobile Robot using Neurosky Mindwave Mobile 2 headset
Kaustubh Prabhu
Kolhapur, Maharashtra
- 0 Collaborators
Analyze EEG signals using a single channel eeg device (headset), classify it to control a wheelchair(prototype Arduino). Using Fuzzy rules making an algorithm for controlling for controlling bot using other features of a headset. ...learn more
Project status: Under Development
Robotics, Artificial Intelligence
Intel Technologies
Intel Opt ML/DL Framework
Overview / Usage
People suffer from a Neuro motor disability, they are in a condition where they areunable to perform any action due to paralysis in the body. The only way they can communicate is if someone can read their thoughts. The EEG helps with the same,this technique reads brain activity and now through it, we can analyze and understand that person. Their fundamental need is a wheelchair thus this project is meant for controlling a wheelchair prototype robot
Methodology / Approach
Steps given below were followed in order to make ourproject:
•Installation of Jayrock and Jayrock.Json and were usedfor connecting with ThinkGear socket connector and to store the collecting data is CSV format. Serial Socket calls were set for controlling the bot.
•Collecting data for our project: data was collected from 30 different persons with help of Neurosky MindwaveMobile device and was stored in a file. Different wave-forms were recorded from the device like Alpha, Beta,Theta, Delta, Gamma along with the blink strength,meditation and attention. The Raw EEG was also stored
.•The raw data was processed through EMD and then HHT in matlab. The data was Stored into a CSV file for training. The overlap was so high that classifying it was difficult.•That is the reason why we shifted to creating a fixed algorithm. Here we plotted the Power EEG data and using fuzzy logic we decided upon the rules for the motion
Technologies Used
Intel optimized libraries: numpy, scipy, tensorflow
Other libraries: keras,
Other tools: matlab, visual studio