Desenvolvimento de um sistema de rastreamento inercial em tempo real sem fio para membros superiores

Abstract

Quantitative analysis of human movement is an essential tool for clinical diagnosis, physical rehabilitation, and robotic interface control. However, commercial motion capture systems present portability restrictions and operational complexity that limit their widespread use. This work presents the development and validation of an accessible, portable, and robust motion tracking system based on a wireless inertial measurement unit (IMU) network, focused on high-fidelity kinematic reconstruction of upper limbs. The hardware platform uses ESP32 microcontrollers integrated with MPU9250 sensors, operating with a hybrid TCP/UDP communication protocol communicating with Python software, optimized to ensure data integrity and real-time transmission speed. For precise spatial orientation estimation, the Madgwick sensor fusion algorithm was implemented using quaternions. The metrological validation of the system was performed by comparing it to an industrial robotic manipulator (Dobot CR10A), adopted as the gold standard for positioning. The trials included static and dynamic angular sweep tests, as well as functional tests. The results demonstrated a high correlation between the proposed system and the robotic reference, proving the reliability of the measurements and the system's capability to reconstruct complex movements with precision and stability, validating its viability as a robust tool for biomechanical analysis.