Vision Based Motion Determination Drift Correction

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing navigation systems face challenges in providing drift-less localization, especially in environments where Global Navigation Satellite Systems (GNSS) and magnetometer performance deteriorate, such as indoors and urban canyons, leading to a 'black out' in absolute localization and aiding.

Innovation Solution

A method and system using stereo camera systems or LIDAR to determine motion by comparing images over time, extracting features, and applying algorithms like the eight-point algorithm and absolute orientation to calculate rotation and translation, improving accuracy and robustness through feature matching and outlier rejection.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If incremental navigation sensors (gyroscopes, accelerometers, odometers) are used to determine position and attitude, then the system can operate without infrastructure, but drift in localization estimate accumulates over time

Engineering Contradiction:

Improveability to operate without infrastructureVSAvoidlocalization estimate accuracy

Core Design Contradiction:

Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The system continuously compares observed feature positions across multiple images with predicted positions based on incremental motion estimates, generating feedback signals that correct drift accumulation in real-time

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

Visual features from the environment serve as an intermediary reference frame that mediates between incremental navigation sensors and absolute position determination, enabling drift correction without direct infrastructure

Inventive Principle:

Principle #24Intermediary (Mediator)

2Measurement precision

If absolute navigation sensors (GNSS, magnetometers) are used to provide drift-less localization, then positioning accuracy is maintained, but the system fails in environments like indoors, urban canyons, and under foliage

Engineering Contradiction:

Improvepositioning accuracyVSAvoidavailability in challenging environments

Core Design Contradiction:

Measurement precisionVSReliability

Solution Approach 1:

The system uses naturally occurring visual features in the environment as self-provided reference points, eliminating dependence on external infrastructure like GNSS satellites or magnetometer signals that may be unavailable

Inventive Principle:

Principle #25Self-service

3Adaptability or versatility

If stereo or monocular vision systems are used as incremental navigation systems, then the system can operate without GNSS infrastructure, but drift accumulation occurs without absolute localization aid

Engineering Contradiction:

Improveoperation without GNSSVSAvoidlocalization drift

Core Design Contradiction:

Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The system pre-identifies and tracks multiple visual features across the image sequence, establishing a reference framework before drift becomes significant, enabling proactive correction rather than reactive compensation

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentEP2116975B1Method and apparatus for vision based motion determination

Publication Date: 2014.04.02 HONEYWELL INTERNATIONAL INC

Figure 1
Figure 2
Figure 3

AI summary

A method for determining motion is provided. The method determines a rotation of an object from a first time to a second time by analyzing a first 2D image obtained at the first time and a second 2D image obtained at the second time. Then, the method determines a translation of the object from the first time to the second time based on the determined rotation, 3D information relating to the first image, and 3D information relating to the second image.