Inspiration
My main motivation was to replicate the Turkish Get-Up (TGU), one of the most difficult functional exercises, in order to assess human movement ability. In particular, in places without access to expert training, rehabilitation, or diagnostics, I aimed to empower people with wearable AI that tracks strength, mobility, and coordination with real-time biomechanical data.
What it does
It is a wearable piece of hardware and a Flutter mobile application. Using IMU and load sensors, the device measures acceleration, gyro movement, leg load, and joint angles (such as knee). The mobile app receives this data via BLE (ESP32-S3). The app shares the advantages and distinctiveness of the movement, shows these parameters in real-time, and offers a detailed video tutorial on how to perform TGU correctly.
How it was Build
For cross-platform smartphone development, we employed Flutter. The flutter_blue_plus plugin was used to stream data from the ESP32S3 chip and manage BLE connectivity. The hardware consists of knee angle sensors, load cells (via HX711), and an IMU (MPU6050). Four pages make up the app: (1) a video and form guide; (2) live BLE data; (3) advantages and distinctiveness; and (4) safety and cautionary instructions. Material theming with video_player for media support was used in the design of the user interface.
Challenges we ran into
One of the main challenges was combining smooth UI updates with steady real-time BLE data streaming. For ESP32 compatibility, Flutter's small BLE libraries needed to be modified. Optimization was needed to handle frequent sensor updates while creating a clear, easy-to-use multi-tab interface. It was also challenging to precisely map joint movement and calibrate knee angle sensors.
Accomplishments that we're proud of
We developed a basic functional prototype that integrates real-time mobile feedback with wearable biomechanics. With accurate form detection and real-time sensor values, the app successfully leads users through the TGU. We take pride in the strong BLE connection and the smooth tab switching.
What we learned
We gained knowledge of how to create modular sensor-driven Flutter user interfaces, manage BLE communication in Flutter, and incorporate external hardware data into mobile interfaces. We also improved our knowledge of sensor fusion and pose dynamics.
What's next for TGU Insight: Motion Analytics for Functional Strength
We aim to integrate advanced AI/ML to evaluate form quality and give corrective feedback. We’ll add more exercises like squats, lunges, and planks using similar sensor mapping. Long-term, we’ll connect the app to cloud analytics for tracking performance trends and injury prevention.
Log in or sign up for Devpost to join the conversation.