Cross-Device Magnetometer Calibration for Indoor Magnetic Positioning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Magnetometer calibration errors in indoor positioning systems (IPS) due to soft and hard iron effects lead to inaccuracies in magnetic sensor measurements, affecting the quality of indoor mapping and localization, especially in GPS-denied environments.
Innovation Solution
A method and system for equalizing calibration errors in magnetic sensor measurements by partitioning data into sets based on mobile device orientation, location, and timestamp, identifying pairs of similar measurements, and estimating calibration errors using cost functions to modify and align sensor data to a common reference frame.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If magnetometer calibration is performed to compensate for soft iron and hard iron effects, then measurement accuracy is improved, but calibration process complexity and time requirements increase
Solution Approach 1:
The system performs preliminary calibration of magnetometers during the mapping phase before positioning operations. Calibration data is collected and stored in advance, then applied to compensate for soft iron and hard iron effects during subsequent positioning operations, eliminating the need for complex manual calibration procedures at positioning time.
Solution Approach 2:
The system automatically performs calibration using sensor data collected during normal device movement and mapping operations. The calibration process is conducted autonomously without requiring manual intervention or specialized calibration procedures, with the system using its own sensor measurements to estimate and correct calibration parameters.
2Adaptability or versatility
If multiple mobile devices undergo different magnetometer calibration processes, then each device can be calibrated independently, but different magnetic sensor measurements may be associated with the same location
Solution Approach 1:
The system transforms magnetic sensor measurements from individual device coordinate systems into a common reference frame by applying device-specific calibration parameters. This parameter transformation aligns measurements from multiple devices to a unified coordinate system, ensuring consistent location representations across all devices despite their independent calibration histories.
Solution Approach 2:
The system introduces a common reference frame as an intermediary that mediates between different device coordinate systems. Each device's calibrated measurements are transformed through this intermediate reference frame to achieve mutual alignment, resolving the inconsistency problem without requiring direct coordination between devices.
3Productivity
If calibration errors are not equalized, then the mapping process can proceed with existing sensor data, but residual calibration errors affect indoor positioning accuracy
Solution Approach 1:
The system implements a feedback mechanism where calibrated sensor data is continuously compared with expected values at known locations. Residual calibration errors are detected and corrected by adjusting the calibration parameters, creating an iterative process that maintains positioning accuracy while preserving mapping efficiency.
Solution Approach 2:
The system replaces manual calibration procedures with automated computational methods that process sensor data mathematically. Instead of physical calibration operations, the system uses algorithms to estimate calibration errors and apply corrections, maintaining productivity while improving precision.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Improves the accuracy of indoor positioning by reducing residual calibration errors, enhancing the quality of magnetic fingerprint maps, and improving localization in indoor environments.
Implementation Method 1
Magnetometers generally perform sensor measurements by sensing the magnetic field of the Earth
Implementation Method 2
The soft iron is caused by ferromagnetic objects (any object attracted by a magnet), and changes the direction of an existing magnetic field
Implementation Method 3
The hard iron effect is cause by any magnet (either natural or electric) which generates its own magnetic field that is added to the Earth magnetic field
Data Source
AI summary
Methods and systems estimate calibration errors of magnetic sensor measurements collected at mobile devices. Each measurement is associated with a location of one of the mobile devices and has a calibration error. Data from the sensor measurements is partitioned into sets. Each set is associated with a respective calibration error associated with the measurements that generated the data in the set. Pairs of data items are identified, where each pair includes a data item from a first of the sets corresponding to a measurement, associated with a first location, that generated the data in the first set, and a data item from a second of the sets corresponding to a measurement, associated with a second location that is approximately the same as the first location, that generated the data in the first set. The calibration error associated with each of the sets is estimated based in part on the pairs.