A Unified Method for Robust Self-Calibration of 3-D Field Sensor Arrays

Self-calibrating an array of 3-D field sensors, such as three-axis magnetometers and accelerometers, requires estimation of two variable sets-each sensor's intrinsic model that maps its input field to the corresponding measurement and each sensor's coordinates relative to a common frame of reference within the array. In this work, we propose the first unified self-calibration method for arrays of same-type 3-D field sensors, which is robust to anomalous sensor measurements unlike previous algorithms. The method breaks down the array calibration task into three steps of more easily subproblems, first estimating the intrinsic variables of each sensor independently, second computing the sensor coordinates with respect to a common reference frame, and last refining both these intrinsics and orientations jointly to minimize physically meaningful sensor estimation errors. Each stage has been carefully designed to maintain robustness to anomalies without compromising estimation quality. The performance of our method is compared against other state-of-the-art algorithms on both simulation and real data from a magnetometer array and accelerometer array, demonstrating significant improvements in accuracy and precision of the estimated array variables in versatile real-world self-calibration environments.

Automated summary

This research introduces a unified self-calibration method for arrays of same-type 3-D field sensors, which breaks down the array calibration task into three steps, maintains robustness to anomalies, and achieves significant improvements in accuracy and precision of estimated array variables in versatile real-world self-calibration environments compared to state-of-the-art algorithms.