Onboard and External Magnetic Bias Estimation for UAS through CDGNSS/Visual Cooperative Navigation

Area of Science:

• Robotics and Control Systems
• Geophysics and Navigation

Background:

• Accurate heading estimation is crucial for Unmanned Aerial System (UAS) flight autonomy, georeferencing, and mapping.
• Onboard and external magnetic disturbances significantly degrade heading accuracy in UAS navigation.
• Existing navigation filters often struggle with substantial magnetic interference.

Purpose of the Study:

• To develop and validate a novel calibration technique for combined estimation of onboard and external magnetic disturbances in small UAS.
• To improve heading estimation accuracy by estimating onboard horizontal bias and external magnetic declination.
• To enable robust UAS operation in magnetically disturbed environments.

Main Methods:

• Cooperative approach utilizing multiple UAS (deputy UAVs).
• Integration of drone-to-drone carrier phase differential GNSS and visual measurements.
• Non-linear least squares problem solved using the Levenberg-Marquardt method for bias estimation.
• Acquisition of visual and GNSS data at varying heading angles.

Main Results:

• Demonstrated flight validation using customized quadrotors.
• Achieved calibrated heading estimates with angular errors below 1°.
• Significant improvement in accuracy compared to non-calibrated magnetic heading and standard onboard navigation filters.

Conclusions:

• The proposed calibration technique effectively mitigates magnetic disturbances for improved UAS heading accuracy.
• The method enhances flight autonomy and supports precise georeferencing and mapping applications.
• Cooperative sensing and advanced estimation techniques offer a robust solution for UAS navigation in challenging magnetic environments.