Extension of the Rigid-Constraint Method for the Heuristic Suboptimal Parameter Tuning to Ten Sensor Fusion...

Area of Science:

• Robotics and Navigation
• Sensor Fusion
• Inertial Navigation Systems

Background:

• Magneto-inertial measurement units (MIMUs) rely on sensor fusion algorithms (SFAs) for orientation estimation.
• Accurate SFA parameter tuning is critical for reliable orientation estimation but typically requires laboratory settings and gold-standard equipment.
• Existing methods for parameter tuning are limited by their reliance on external orientation references.

Purpose of the Study:

• To evaluate the applicability of the Rigid-Constraint Method (RCM) for tuning multi-parameter SFAs without requiring orientation references.
• To assess the performance of RCM across diverse experimental scenarios and popular SFAs.
• To demonstrate a practical approach for optimizing SFA performance in real-world applications.

Main Methods:

• The Rigid-Constraint Method (RCM) was applied to estimate suboptimal SFA parameter values.
• The RCM's effectiveness was tested on 10 popular SFAs with multiple parameters under various experimental conditions.
• Performance was evaluated by comparing RCM-tuned parameters against optimal parameters derived using reference data.

Main Results:

• The RCM demonstrated effectiveness in tuning multi-parameter SFAs, achieving an average residual error of 0.6 degrees compared to optimal parameters.
• The maximum error increase observed was 3.7 degrees, indicating robust performance across different scenarios.
• Previous validation showed an average error increase of 1.5 degrees for single-parameter SFAs.

Conclusions:

• The Rigid-Constraint Method (RCM) offers a viable solution for tuning generic SFAs without external orientation references.
• RCM enables effective SFA parameter optimization across different experimental scenarios and SFA types.
• This approach facilitates improved orientation accuracy for MIMUs in practical, non-laboratory settings.