Accelerometer Calibration Using Magnetometer Steady-State Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Accelerometers in mobile devices require periodic calibration to maintain accuracy due to drifting gains and offsets over time, especially influenced by temperature, and accurate calibration is essential for reliable operation with magnetometers in applications like digital compasses.
Innovation Solution
A method using magnetometer readings to detect steady-state orientations, filter out accelerometer samples from movements, and apply least squares fitting algorithms to determine gain and offset values for each axis, ensuring accurate calibration and drift tracking.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If accelerometer calibration is performed using traditional methods without steady-state detection, then calibration can be performed quickly, but the accuracy of calibration parameters is degraded due to inclusion of motion-induced erroneous readings
Solution Approach 1:
The patent introduces magnetometer readings as an intermediary tool to detect steady-state conditions. The magnetometer independently determines device orientation, which serves as a mediator to identify when the accelerometer has settled into a stable reading, filtering out motion-induced errors without requiring manual intervention or complex motion analysis algorithms.
Solution Approach 2:
The system implements feedback by continuously monitoring magnetometer readings to detect when steady-state conditions are achieved. This feedback mechanism automatically triggers the calibration process only when the device is stationary, ensuring high accuracy while avoiding unnecessary calibration attempts during motion, thus optimizing both precision and time efficiency.
2Productivity
If accelerometer readings are collected during device motion for calibration, then calibration can be performed faster with more data samples, but the reliability of calibration parameters deteriorates due to motion-induced errors
Solution Approach 1:
The patent applies preliminary action by using magnetometer readings to pre-detect and confirm steady-state conditions before initiating accelerometer data collection for calibration. This preliminary verification ensures that only reliable, motion-free data is captured, maintaining high calibration parameter reliability while enabling rapid calibration when conditions are appropriate.
3Device complexity
If calibration parameters are determined without filtering motion-induced readings, then the calibration process is simpler and requires fewer computational resources, but the accuracy of orientation detection applications is reduced
Solution Approach 1:
The magnetometer serves as an intermediary filtering mechanism that automatically identifies and separates valid calibration readings from motion-induced erroneous readings. This approach maintains relatively simple calibration process complexity while dramatically improving orientation detection accuracy by ensuring only steady-state data is used for parameter determination.
Data Source
Figure 1~3
Figure 4~5
Figure 6
AI summary
A method and system are provided for obtaining data for calibrating an accelerometer. The method and system operate by using at least one magnetometer reading to detect that a first orientation is being maintained; obtaining a plurality of accelerometer readings at the first orientation; using at least one magnetometer reading to detect that a plurality of additional orientations are being maintained and, for each orientation, obtaining a plurality of accelerometer readings at that orientation; determining calibration parameters comprising, for each axis of the accelerometer, at least one of a gain value and an offset value, using the plurality of accelerometer readings at the first and plurality of additional orientations; and applying the calibration parameters to subsequent accelerometer readings.