SMART-PAD

The motivation for SMART-PAD stemmed from the challenges volleyball players face in monitoring movement quality and mitigating ACL injury risks. Unlike camera-based tracking solutions, which introduce latency and complexity, this system embeds sensing directly into the player’s equipment — specifically a kneepad — to provide real-time, low-latency insights. This design choice makes the system lightweight, portable, and personalized while remaining accessible to both athletes and coaches.

To achieve this, I designed a multi-layer IoT architecture consisting of a sensor layer, transmission layer, data processing layer, and visualization layer. Five ESP32 DW3000 UWB boards served as the backbone for localization, with three anchors and one tag establishing spatial references, while an additional IMU provided heading and movement context. Position data was streamed via serial connections to a central hub, where Python scripts handled trilateration, jerk calculations, and error correction. Thread-safe queues and asynchronous worker threads enabled trilateration to run independently from visualization updates, reducing latency and ensuring smooth motion tracking.

The data pipeline was carefully tuned for speed and reliability. Polling rates on the UWB boards were iteratively optimized, with the RNG delay value reduced to 3–4ms to balance responsiveness with data integrity. Further refinements, such as decreasing serial read timeouts, improved throughput and responsiveness, bringing system responsiveness to under 500ms. This iterative approach allowed me to incrementally build from noisy IMU-only data to a fully functional UWB-based system capable of reliable real-time performance.

On the frontend side, I implemented a live dashboard using Nuxt.js and Vue.js, with D3.js powering the interactive charts. The interface visualizes player trajectories, heading, and motion metrics in real-time, alongside analytics such as jerk and acceleration spikes that can flag risky or injury-prone movements. This enables coaches to instantly assess not only where a player is positioned but also how their body mechanics may contribute to strain. The dashboard design focused on clarity and responsiveness, ensuring the visualizations matched the rapid pace of in-game movement.

Overall, SMART-PAD demonstrates how embedded sensing, real-time computation, and intuitive visualization can converge to create an impactful sports technology tool. Beyond immediate player tracking, the system lays the groundwork for future extensions such as long-term injury trend analysis, integration with cloud storage, and machine learning models for predictive insights. It represents a scalable, data-driven approach to athlete safety and performance optimization, born from a personal passion for volleyball and a technical curiosity in IoT systems.