Wearable Human Attitude Visualization Tracking System Based on MEMS Inertial Sensors

To meet the demands for high precision, stability, and visualization in applications, this study presents a wearable human attitude tracking and visualization system based on MEMS sensors. The system employs a nine-axis inertial sensor combined with a gradient descent calibration method and data fusion algorithm to achieve high-precision and high-stability human attitude estimation. A human model is created using OpenSim and Matlab to track movement, enabling the visualization of human attitude. Test results demonstrate excellent static and dynamic performance: attitude angle drift is under 0.1° during static tests lasting over 8 hours, and the convergence time when rotating 90° rapidly is less than 0.45 seconds. Additionally, the system can display human motion attitude in real-time through a visualization interface. This wearable human attitude tracking and visualization system, based on inertial sensors, demonstrates high accuracy, good stability, and an intuitive display, showing great potential for applications in smart elderly care, remote healthcare, and other fields.