Inspiration
We envisioned a system that would let users think about moving their hand and actually see it move. A device that doesn’t just assist, but teaches the brain and muscles to communicate again. Something that could be built cheaply, worn comfortably, and used anywhere from clinics to someone’s living room.
What it does
MindGrip translates brain activity into real, physical movement. When the user thinks about closing their hand, an EEG headset detects neural signals and sends them to a Python-based interface that communicates with an Arduino. The Arduino then drives a DC motor connected to lightweight tendons threaded through a hybrid 3D-printed and fabric glove. The result is a soft exoskeleton that gently curls and opens the user’s fingers, helping individuals with stroke, lupus, or tremor-related weakness perform hand exercises and daily tasks with ease.
How we built it
We designed MindGrip using a combination of hardware, software, and rapid prototyping. Brain signals are captured from an Emotiv EEG headset and processed in Python, which translates the data into OPEN and CLOSE commands. These commands are sent via serial connection to an Arduino Uno, which drives a DC gearmotor. The motor winds a lightweight thread-based tendon system running through a hybrid 3D-printed and soft glove structure, causing the fingers to flex and extend.
Challenges we ran into
Our biggest challenge came when our main servo motor broke mid-project, leaving us without our primary actuator. We pivoted to using a DC gearmotor, which meant redoing the wiring, code, and mechanical design on the spot. The fishing line tendons we started with were too thick to pull smoothly, so we replaced them with strong sewing thread that could glide through our 3D-printed channels. Calibrating the EEG headset to filter false readings was also difficult. Despite the setbacks, we stayed adaptable, solved each issue creatively, and managed to complete a working prototype just in time for the demo.
What we learned
This project taught us how to merge biomedical signals, mechanical design, and embedded systems in a single workflow. None of us had experience connecting EEG data directly to physical motion before, so building a thought-controlled glove from scratch was both technical and eye-opening. We learned to improvise under pressure when our servo broke, we replaced it with a DC motor in under an hour. We also learned the value of iteration and teamwork. Every member contributed from a different field, and combining those perspectives made MindGrip possible.
What's next for MindGrip
Our next step is to evolve MindGrip from a prototype into a fully functional assistive device. We plan to integrate force and motion sensors to measure user progress and provide data-driven rehab feedback that can be sent to the patient's doctor. We also want to work with rehabilitation centers and occupational therapists to conduct pilot trials.
Log in or sign up for Devpost to join the conversation.