Real-Time Inertial Sensor Fusion for High-Speed Human Motion Tracking in Unconstrained Environments

View at publisher Summarise with VIRAT
Authors: Aderajew Tilahun, Jeronimo Cox, Tomonari Furukawa, Gamini Dissanayake
Cited by: 0
References: 23

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

This paper presents an inertial sensor fusion approach for estimating high-speed human motion using a wearable Inertial Measurement Units (IMUs) in unconstrained environments and over long durations. The method targets dynamic activities in unlimited fields such as athletic activities, where traditional camera-based motion tracking systems are impractical due to limited field-of-view or occlusions. Accelerometers, gyroscopes and magnetometers are strategically mounted and modeled to maximize measurement accuracy through the past work of inertial sensor fusion using Extended Kalman Filter (EKF). A torque-free dynamic model is employed as the motion model, while IMU threshold-based motion phase detection adaptively adjusts process noise, enabling robust tracking across varying dynamic conditions. The system was experimentally validated through human lower-limb and benchmarked against marker-based ground truth from an optical motion capture system. Experimental results show that the proposed approach significantly improves joint-angle estimation accuracy during high-speed motions. Compared to the traditional methods, the proposed approach reduces the mean RMSE across the three axes by up to 60.5%, 19%, and 62.1% for the hip, knee, and ankle joints in jumping, and by 62.5%, 57.2%, and 59.4% during leg swing, demonstrating strong robustness and potential suitability for full-body tracking in high-speed, unconstrained scenarios.

Fields of Study & Subjects

Inertial Sensor and Navigation Target Tracking and Data Fusion in Sensor Networks Context-Aware Activity Recognition Systems

Publication details

Journal / source
IEEE International Conference on Industrial and Information Systems (ICIIS)
Pages
49–54
Publication date
16 January 2026
Language
EN