Real-time Accelerometer Calibration via Error Correction Model
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Accelerometers in electronic devices face calibration challenges due to sources of error like bias offset, sensitivity errors, and cross-axis errors, which are not effectively addressed by initial factory calibration, leading to inaccurate measurements over time due to mechanical stresses, age, and temperature changes.
Innovation Solution
An electronic device performs real-time calibration of on-board accelerometers by collecting acceleration measurements to form a data set, using an error correction model to determine calibration values for bias, sensitivity, and cross-axis errors, and updates these values as needed to maintain accurate measurements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If factory calibration is performed with high precision manipulation to known truth reference orientations, then measurement precision is improved, but manufacturing time and complexity increase
Solution Approach 1:
The accelerometer performs self-calibration by utilizing the known gravitational field as a reference. The system collects acceleration measurements while the device is stationary, automatically computes calibration parameters using an error correction model, and updates its own calibration values without requiring external manipulation or reference orientations, thereby eliminating the time-consuming factory calibration process
Solution Approach 2:
The patent changes the calibration approach from physical manipulation to mathematical computation. By collecting acceleration measurements in the gravitational field and applying an error correction model with calibration parameters, the system computes calibration values algorithmically rather than through physical positioning, significantly reducing manufacturing time while maintaining accuracy
2Measurement precision
If factory calibration is performed, then initial measurement accuracy is improved, but long-term reliability deteriorates due to mechanical stresses, age, and temperature changes
Solution Approach 1:
The system performs continuous calibration by repeatedly collecting acceleration measurements and updating calibration parameters over time. This ongoing process ensures that calibration values are continuously refined to account for drift caused by mechanical stresses, aging, and temperature changes, maintaining long-term reliability rather than relying on a single initial calibration event
Solution Approach 2:
The error correction model uses feedback from collected acceleration measurements to continuously adjust and refine calibration parameters. By monitoring the accelerometer's output and comparing it against expected gravitational field values, the system detects drift and automatically corrects it, ensuring long-term calibration stability despite environmental changes
3Measurement precision
If comprehensive calibration for bias error, sensitivity error, and cross-axis error is performed, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The error correction model serves multiple calibration functions simultaneously. A single computational framework handles bias error correction, sensitivity error correction, and cross-axis error correction by processing acceleration measurements through unified mathematical operations, eliminating the need for separate calibration systems for each error type and reducing overall device complexity
Solution Approach 2:
The patent replaces complex mechanical calibration systems with computational methods. Instead of requiring separate mechanical positioning systems for each error type, the invention uses algorithmic processing of acceleration measurements to correct all error types, significantly simplifying the device while achieving comprehensive calibration
Data Source
AI summary
An electronic device configured for real-time calibration of an on-board accelerometer. A plurality of acceleration measurements are collected from the accelerometer to form a data set. An accelerometer error correction model is maintained that includes bias error calibration parameters, sensitivity calibration parameters, and cross-axis calibration parameters that each specify respective weights for each of bias error, sensitivity error, and cross-axis error. Calibration values are determined for one or more of the bias error calibration parameters, the sensitivity calibration parameters, and the cross-axis error calibration parameters for the data set of acceleration measurements using the accelerometer error correction model. A true acceleration vector may be determined that corresponds to a subsequently received acceleration measurement using the determined calibration values.