1 of 18

SSVEP RC CAR

TRITON

NeuroTech

2 of 18

Background Information

Building an RC Car that is controlled by specific occipital brain signals (SSVEPs) through a BCI using EEG signals

BCI		Brain-Computer Interface → Systems that convert brain signals into device commands.
EEG		Electroencephalography → Non-invasive brain activity measurement tool.

Background	Outline	Software	Hardware	Next Steps

3 of 18

What is a SSVEP

Steady-State Visual Evoked Potentials (SSVEPs) are brain signals generated by the visual cortex in response to flickering stimuli at specific frequencies.

Reliably easy to detect with EEG, these signals match the frequency of the stimuli and are used in BCIs to identify where a person is focusing.

Background	Outline	Software	Hardware	Next Steps

4 of 18

SSVEP BCI Applications

Making BCIs more accessible
Wheelchair accessibility
Smart Device Accessibility
Vehicle Control Features
Immersive Game Applications

Background	Outline	Software	Hardware	Next Steps

5 of 18

Project Outline

TRITON

NeuroTech

6 of 18

SSVEP RC Car Outline

Nandini (HIGH LEVEL)

To give you all a high-level overview of the workflow of our project, we’ve made this schematic, which shows you exactly how we plan on remote-controlling a car with live EEG signals.

The system begins with the EEG headset, which detects neural responses triggered by the SSVEP stimuli on the given screen with a camera feed from the car. The raw EEG signals undergo some cleaning and bandpass filtering to isolate frequencies of interest in the range of 8 to 15 Hz. Then, the system computes the signal-to-noise ratio of that data and applies a matched filter detector to identify which frequency corresponds to which flickering stimulus.

Once the frequency is identified, it is mapped to specific motion command for the RC car. These commands are transmitted wirelessly to the RC car, which executes the desired actions, such as moving forward, backward, or turning, and so forth.

7 of 18

Software

TRITON

NeuroTech

8 of 18

GUI Demonstration

9 of 18

GUI

Components:

Camera feed from robot
1 12 hz flickering square.
1 15 hz flickering square.

We receive a feed from the robot and want to make it easy for users to control it. As a proof of concept, we’ve started with two commands: forward and backward. When a participant looks at the forward square, we expect the EEG data we are collecting to show a matching response. Thus we can then tell the robot to move.

Background	Outline	Software	Hardware	Next Steps

10 of 18

Data Collection

Unicorn Hybrid Black Portable EEG Headset

Live Bluetooth transmission of EEG signals

Unicorn Suite Software

Live bandpass filtering, notch filtering, and noise removal
Uses LabStreamingLayer to transmit data to our processing pipeline

Background	Outline	Software	Hardware	Next Steps

11 of 18

Extracting SSVEP Signal

Using a fast fourier transform, we will compute the power of different frequencies in the signal to detect which frequencies are the most prominent.

Peaks at stimulus frequencies (e.g. 12 and 15 Hz)

Raw EEG signal

Car motion command

Background	Outline	Software	Hardware	Next Steps

12 of 18

Hardware

TRITON

NeuroTech

13 of 18

ELEGOO Smart Robot Car V4.0

Camera Module

FPV camera feed in GUI

Programmable Arduino

Custom commands

Wifi & Bluetooth Connectivity

Remote control

Background	Outline	Software	Hardware	Next Steps

14 of 18

Controlling RC Car

Custom UI for remote car control

SSVEP signals will interface directly with this UI

Bluetooth module on Car receives input
Car executes command

SSVEP signal sends command to UI

UI sends signal to Car

Car executes command

Background	Outline	Software	Hardware	Next Steps

15 of 18

Next Steps

TRITON

NeuroTech

16 of 18

Plans for Winter Quarter

Optimizing signal detection:

Computing Fast Fourier Transform of EEG signal
Performing this processing on live EEG signals

Connecting frequencies of interest to car commands

Matching specific frequencies to desired car motions
Processing live EEG signals into car commands
Sending commands to remote-controlled car

Background	Outline	Software	Hardware	Next Steps

17 of 18

THANK YOU!

TRITON

NeuroTech

18 of 18

TRITON

NeuroTech