Adaptive Sensor Fusion for Orientation Data Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing mobile computing devices face challenges in efficiently determining orientation data using sensors, as current sensor fusion techniques often require static filters that struggle with adapting to changing noise environments, leading to accuracy and power efficiency issues.
Innovation Solution
Implementing independent adaptive filters for accelerometer and magnetometer data, allowing for dynamic adjustment of filter parameters such as time constant and gain, and using a Kalman filter to improve the accuracy and speed of orientation determination while reducing power consumption.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If static first order filters are used for sensor data filtering, then the system is simple to implement, but the filter cannot quickly adapt to changing noise in the operating environment
Solution Approach 1:
The patent applies dynamics by transitioning from static filter coefficients to dynamic coefficients that adapt in real-time. The filter coefficients are continuously updated based on the observed noise characteristics in the operating environment, allowing the filter to automatically adjust its behavior without manual intervention. This resolves the contradiction by making the filter adaptive while maintaining a relatively simple structure through the use of recursive least squares estimation.
Solution Approach 2:
The patent implements feedback mechanisms where the filter continuously monitors the sensor data and noise characteristics, then uses this information to adjust its coefficients. The recursive least squares algorithm provides feedback by comparing predicted values with actual measurements and updating coefficients accordingly. This feedback loop enables the filter to adapt to changing noise conditions while maintaining computational efficiency.
2Measurement precision
If sensor fusion techniques are implemented to improve orientation accuracy, then measurement precision improves, but power consumption increases
Solution Approach 1:
The patent applies parameter changes by dynamically adjusting filter coefficients based on operating conditions and noise characteristics. Instead of using fixed, conservative parameters that require continuous high-power processing, the system adapts parameters like filter order and coefficients to match current conditions. This allows accurate orientation determination while reducing computational load and power consumption during stable conditions.
Solution Approach 2:
The patent uses dynamic adaptation of filter parameters based on real-time noise assessment. When noise levels are low and conditions are stable, the filter uses simpler models requiring less power. When noise increases or conditions change rapidly, the filter dynamically increases its complexity to maintain accuracy. This dynamic parameter adjustment resolves the contradiction between precision and power consumption.
3Reliability
If multiple sensors are fused to determine orientation, then completeness and reliability of orientation data improves, but system integration complexity increases due to different sensor formats
Solution Approach 1:
The patent applies universality by creating a unified filter framework that can process data from multiple sensor types (accelerometers, gyroscopes, magnetometers) through a common mathematical model. The recursive least squares estimation approach provides a universal interface that handles different sensor formats and characteristics, allowing diverse sensors to be integrated without complex individual processing pipelines for each sensor type.
Solution Approach 2:
The patent uses the adaptive filter as an intermediary layer between raw sensor data and orientation output. This intermediary process standardizes the integration of multiple sensors by translating their different formats and characteristics into a unified orientation estimate. The filter acts as a mediator that reconciles differences in sensor data formats, timing, and characteristics, simplifying the overall system integration.
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
This approach enhances the accuracy and speed of orientation data determination, adapts better to changing noise conditions, and conserves power by dynamically adjusting filter parameters based on device motion and noise levels.
Implementation Method 1
An accelerometer measures acceleration forces including gravity
Implementation Method 2
A gyroscope measures relative orientation of a device, for example based on the principles of conservation of angular momentum
Implementation Method 3
A magnetometer is an instrument used for measuring the direction and strength of a magnetic field
Data Source
Figure 1~2A
Figure 2B
Figure 2C
AI summary
Aspects of the disclosure relate to computing device technologies, such as systems, methods, apparatuses, and computer-readable media for improving orientation data. In some embodiments, a magnetic vector filter receives magnetometer data from a magnetometer and gyroscope data from a gyroscope and determines the magnetic vector. In another embodiment, a gravity vector filter receives accelerometer data and gyroscope data and determines the gravity vector. Further techniques are described for adjusting filter parameters, based at least in part on the detected change in the variability of the first signal parameter. Furthermore, orientation may be synchronized to a common sensor input, such as a gyroscope.