MEMS Gyroscope Bias Compensation via No-Motion Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
MEMS gyroscope sensors suffer from bias errors due to unwanted offset signals, leading to integration errors and reduced accuracy in orientation tracking, with existing solutions relying on additional sensors and complex computational methods.
Innovation Solution
A system that detects and compensates bias errors using raw gyroscope data by analyzing RMS values and spectral power density to determine no-motion conditions, allowing for the estimation and subtraction of bias signals without relying on external sensors or extensive processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If existing bias compensation approaches using auxiliary sensors and sensor fusion algorithms are employed, then bias error compensation is achieved, but device complexity and computational requirements increase
Solution Approach 1:
The gyroscope sensor compensates for its own bias errors using only its own output signal, without requiring auxiliary sensors or external systems. The sensor performs self-diagnosis by detecting no-motion conditions and self-correction by estimating and removing bias offsets, achieving autonomous bias compensation that eliminates the need for additional sensors and complex sensor fusion algorithms
Solution Approach 2:
The patent extracts and removes the harmful bias offset signal from the gyroscope output by identifying no-motion periods, estimating the bias during these periods, and subtracting it from the signal. This extraction approach eliminates the need for auxiliary sensors while directly addressing the bias error problem
2Measurement precision
If existing bias compensation approaches using sensor fusion algorithms are employed, then bias error compensation is achieved, but processing power requirements increase
Solution Approach 1:
The system uses the gyroscope's own signal to compensate for bias, eliminating the need for computationally intensive sensor fusion algorithms that combine multiple sensor inputs. The self-service approach processes only the gyroscope signal, dramatically reducing computational requirements while maintaining bias compensation effectiveness
Solution Approach 2:
The patent applies bias compensation only during detected no-motion conditions rather than continuously processing all sensor data. This partial action approach reduces computational load by estimating bias only when the sensor is stationary, avoiding the need for continuous complex algorithm execution while still achieving effective bias removal
3Measurement precision
If bias compensation is performed continuously, then bias errors are reduced, but false corrections during actual motion occur reducing accuracy
Solution Approach 1:
The system performs preliminary detection of no-motion conditions before applying bias compensation. By first identifying periods when the sensor is truly stationary through RMS value analysis and spectral power density evaluation, the system ensures that bias estimation and correction are applied only when appropriate, preventing false corrections during actual motion
Solution Approach 2:
The patent uses feedback mechanisms to monitor the gyroscope output signal characteristics (RMS values, spectral power density) and adjust bias compensation accordingly. The system continuously monitors motion indicators and only applies compensation when feedback confirms a no-motion state, ensuring reliable and accurate bias removal without interfering with actual motion detection
Data Source
AI summary
Various embodiments of the invention provide for automatic, real-time bias detection and error compensation in inertial MEMS sensors often used in handheld devices. Real-time bias correction provides for computational advantages that lead to optimized gyroscope performance without negatively affecting user experience. In various embodiments, bias non-idealities are compensated by utilizing raw output data from the gyroscope itself without relying on additional external sensors.