Posture Estimation Bias Correction Using a Gravity Reference

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing posture estimation methods using inertial measurement units (IMUs) face accuracy issues due to bias errors, especially when the object's posture changes are small, leading to decreased estimation accuracy.

Innovation Solution

A posture estimation method that calculates the posture change amount based on angular velocity sensor output, predicts posture information, limits bias error components around a reference vector, and corrects the predicted posture using error information from both the angular velocity and acceleration sensors, particularly utilizing a gravitational acceleration vector as the reference.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If a bias error correction technique using Kalman filter is applied to estimate accurate posture, then estimation accuracy is improved, but when posture changes are small, the bias error monotonously increases and estimation accuracy decreases

Engineering Contradiction:

Improveposture estimation accuracyVSAvoidbias error stability

Core Design Contradiction:

Measurement precisionVSReliability

Solution Approach 1:

The patent applies dynamics by making the bias error correction adaptive rather than static. The Kalman filter dynamically adjusts the bias error correction based on the actual posture change detected by the angular velocity sensor. When posture changes are large, the filter trusts the angular velocity data more; when posture changes are small, it relies more on the acceleration sensor data, preventing monotonic bias error accumulation.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The patent implements feedback by using the output of the acceleration sensor to continuously monitor and correct the posture estimation. The acceleration sensor provides feedback on the actual gravitational acceleration direction, which is used to adjust the bias error in the angular velocity sensor output, preventing unbounded error growth during small posture changes.

Inventive Principle:

Principle #23Feedback

2Productivity

If angular velocity sensor output is used to calculate posture change amount, then posture prediction is achieved, but bias error component around reference vector increases monotonously

Engineering Contradiction:

Improveposture calculation speedVSAvoidbias error accumulation

Core Design Contradiction:

ProductivityVSMeasurement precision

Solution Approach 1:

The patent introduces the acceleration sensor as an intermediary to mediate between the angular velocity sensor and the final posture estimation. The acceleration sensor measures gravitational acceleration directly and provides an independent reference that mediates the bias error in the angular velocity integration, preventing monotonic error accumulation while maintaining the speed benefits of angular velocity-based prediction.

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent changes the parameters used for posture estimation by combining angular velocity data with acceleration sensor data. The system dynamically changes the weighting and trust placed in different sensor outputs based on the detected posture change magnitude, adjusting the estimation parameters to prevent bias error accumulation while maintaining calculation efficiency.

Inventive Principle:

Principle #35Parameter changes

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 of posture estimation by effectively managing bias errors and maintaining estimation precision even when the object's posture changes are minimal, thereby improving the reliability of posture calculations.

Implementation Method 1

calculating a posture change amount of an object based on an output of an angular velocity sensor

Methodology Applied

Scientific EffectAngular velocity sensing:

Implementation Method 2

the reference vector may be a gravitational acceleration vector

Methodology Applied

Scientific EffectGravitation: Gravitation

Data Source

PatentUS11971429B2Posture estimation method, posture estimation device, and vehicle

Publication Date: 2024.04.30 SEIKO EPSON CORP

AI summary

A posture estimation method includes calculating a posture change amount of an object based on an output of an angular velocity sensor, predicting posture information of the object by using the posture change amount, limiting a bias error in a manner of limiting a bias error component of an angular velocity around a reference vector in error information, and correcting the predicted posture information of the object based on the error information, the reference vector, and an output of a reference observation sensor.