Orientation Estimation Using Multiple Adaptive Filters
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional adaptive filter systems for motion sensor fusion face challenges in accurately estimating orientation, particularly in complex motions, due to limitations in convergence accuracy and dynamic behavior, where a single filter may not optimize for different types of motions effectively.
Innovation Solution
The use of multiple adaptive filters with distinct tunings and weighting, each optimized for specific types of motion, receiving inputs from accelerometers, gyroscopes, and optionally magnetic sensors, to generate separate orientation estimates that can be combined for improved accuracy and convergence.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a single adaptive filter is used for motion sensor fusion, then the device complexity is reduced, but the measurement precision of orientation estimation deteriorates in complex motions
Solution Approach 1:
The patent divides the single filter system into multiple parallel adaptive filters (e.g., first adaptive filter for high dynamic motion, second adaptive filter for low dynamic motion). Each filter is segmented to handle specific motion characteristics, improving overall orientation estimation accuracy without requiring a completely complex unified system.
Solution Approach 2:
The system dynamically selects or weights the output of different adaptive filters based on the current motion characteristics detected by sensors. This dynamic adaptation allows the system to maintain high precision across varying motion conditions while keeping each individual filter relatively simple.
2Measurement precision
If multiple adaptive filters with different tunings are used, then the measurement precision of orientation estimation is improved, but the device complexity increases
Solution Approach 1:
Multiple adaptive filters are designed with different tunings to serve universal purposes across various motion scenarios. Each filter is optimized for specific conditions (e.g., high dynamic, low dynamic, vibration scenarios), allowing the system to maintain high convergence accuracy across diverse applications without requiring completely different filter designs.
Solution Approach 2:
A motion detection mechanism acts as an intermediary to analyze current motion characteristics and determine which adaptive filter output should be weighted more heavily. This intermediary component coordinates the multiple filters, managing their combined complexity while maximizing their individual precision contributions.
3Ease of operation
If a single adaptive filter tuning is used, then the ease of operation is improved, but the adaptability to different dynamic motions deteriorates
Solution Approach 1:
The filter system is segmented into multiple parallel adaptive filters, each with pre-configured tunings for specific motion types. This segmentation maintains operational simplicity by providing ready-to-use specialized filters while improving adaptability through the collective coverage of different motion scenarios.
Solution Approach 2:
The system dynamically adjusts the weighting or selection of different adaptive filter outputs based on real-time motion analysis. This dynamic behavior enables the system to adapt to various dynamic motions without requiring manual reconfiguration, maintaining ease of operation while achieving high versatility.
4Measurement precision
If multiple adaptive filters are used with separate computation paths, then the measurement precision is improved, but the use of energy increases
Solution Approach 1:
Instead of always processing through all multiple adaptive filters at full capacity, the system applies partial processing by selectively weighting or selecting from filter outputs based on current motion conditions. This approach maintains high precision when needed while reducing computational energy consumption during periods when full processing is not necessary.
Data Source
AI summary
Apparatuses, methods and systems apparatus for sensing an orientation are disclosed. One apparatus includes an accelerometer, wherein the accelerometer generates a sensed acceleration of the accelerometer and a gyroscope, wherein the gyroscope generates a sensed orientation of the gyroscope. The apparatus further includes a first adaptive filter, the first adaptive filter operative to receive at least the sensed acceleration and the sensed orientation of the gyroscope, and generate a first orientation (Q) of the apparatus, a second adaptive filter, the second adaptive filter operative to receive at least the sensed acceleration and the sensed orientation of the gyroscope, and generate a second orientation (Q′) of the apparatus, wherein a tuning for the first adaptive filter is different than a tuning for the second adaptive filter.