NeuraGrip

Inspiration

We wanted to explore whether brain signals could be used to detect when a person wants to move a limb and turn that intention into a physical action.

That led us to NeuraGrip, a proof-of-concept system that combines brain signals, muscle activity, and physical feedback.

The main idea is simple: when a user wants to move their limb, the EMOTIV Insight detects the associated EEG signal and sends it to the ESP32. The system can then use that input to control a physical response.

What it does

NeuraGrip uses an EMOTIV Insight EEG headset to detect when the user intends to move a limb. The EEG signal is sent to the ESP32, which serves as the core controller for the system.

A MyoWare muscle sensor is used to detect muscle activation. A servo motor can provide mechanical movement, while a TENS unit demonstrates the possibility of providing electrical stimulation.

The overall concept is:

User intends to move → EMOTIV detects intention → EEG signal → ESP32 → Physical response

The EMOTIV's role is specifically to detect the user's intention to move. The MyoWare provides a separate measurement of muscle activity, allowing us to explore the relationship between intended movement and actual muscle activation.

NeuraGrip is a hackathon proof of concept and is not intended to diagnose or treat any medical condition.

How we built it

We built NeuraGrip using several off-the-shelf components:

• EMOTIV Insight — detects EEG signals associated with the user's intention to move a limb.
• ESP32 — serves as the intended embedded controller that receives the EEG signal.
• MyoWare Muscle Sensor — detects electrical activity produced by muscle activation.
• Servo motor — provides mechanical movement as a physical response.
• TENS unit — demonstrates the potential for electrical stimulation as another form of physical feedback.

The core idea is to use the user's brain signal as an input for a hardware system rather than simply collecting EEG data.

Challenges we ran into

One of our biggest challenges was connecting the ESP32 to our computer. My computer does not have a USB Type-A port, so I ultimately was not able to connect and program the ESP32.

We also had to work with physiological signals, which can be noisy and affected by factors such as sensor placement, movement, and electrical interference.

Another challenge was figuring out how the different components could work together while keeping the system simple enough to build during a hackathon.

Accomplishments that we're proud of

We're proud that we designed a system that combines EEG, muscle sensing, embedded computing, and physical feedback into one concept.

The most important part of our project is the connection between the user's intention and the physical system:

Intention to move → EEG detection → ESP32 → Physical response

Instead of requiring a traditional physical button or switch to control the system, NeuraGrip explores using the user's brain activity as the input.

We're also proud that we incorporated a muscle sensor to provide another layer of information about the user's actual muscle activity.

What we learned

We learned that working with biological signals is much more challenging than working with standard electronic inputs. EEG and muscle signals can be noisy and require careful sensor placement and signal processing.

We also learned how important it is to clearly define how hardware components communicate with one another when building a system that connects biological signals to physical devices.

Most importantly, we learned how brain activity can potentially be used as an input for controlling hardware and how different physiological signals can be combined to create a more interactive system.

What's next for NeuraGrip

Our next step is to successfully connect the EMOTIV Insight to the ESP32 and reliably send the detected movement-intention signal to the controller.

We also want to improve the accuracy of the EEG signal detection and integrate the servo and TENS components into the system so that a detected movement intention can trigger an appropriate physical response.

In the future, we could expand the system to recognize different limb-movement intentions and create different responses for each one.

Our ultimate vision is to create a system that can turn the intention to move into physical action.

Built With

• python