IMU-based sensor-to-segment multiple calibration for upper limb joint angle measurement-a proof of concept

Area of Science:

• Biomechanics
• Human Motion Analysis
• Wearable Sensor Technology

Background:

• Wearable inertial sensors offer a promising alternative for outdoor human motion tracking.
• Soft tissue artifact (STA) negatively impacts the accuracy of joint angle calculations from sensor data due to skin motion relative to bone.
• Sensor misalignment during movement further compromises the precision of measured joint angles.

Purpose of the Study:

• To propose and validate a novel sensor-to-segment calibration method for inertial measurement units (IMUs).
• To mitigate the effects of soft tissue artifact and sensor misalignment on joint angle accuracy.
• To enhance the reliability of human motion tracking using wearable sensors.

Main Methods:

• A new multiple calibration technique was developed, inspired by marker-based cluster calibration.
• The method involves performing multiple static postures of the upper limb across all anatomical planes.
• Alignment differences between sensor and segment frames are identified, calculated for each posture, and linearly interpolated.

Main Results:

• Experimental validation on a mechanical model and a human subject demonstrated significant improvements in joint angle measurements.
• The multiple calibration procedure notably enhanced accuracy compared to traditional technical calibration.
• Maximal error in shoulder internal-external rotation decreased by over 50% after applying the correction.

Conclusions:

• The proposed multiple calibration procedure effectively corrects joint angle values obtained from IMUs.
• This method significantly improves the accuracy of human motion tracking, particularly for movements affected by soft tissue artifact.
• The technique offers a substantial advancement for precise joint angle measurement in wearable sensor applications.