Vision Aided Navigation Feature Tracking via Localized Image Adjustment

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Vision-based navigation systems face challenges in adapting to dynamic environmental conditions, such as changes in lighting or feature obscuration, which can lead to loss of tracked features and reduced navigation accuracy.

Innovation Solution

The system employs a dynamic localized parameter adjuster that applies targeted image enhancement techniques to specific regions of the image where tracked features are expected to appear, based on predicted locations and vehicle kinematics.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If vision-based sensors are calibrated for predefined environmental conditions, then navigation accuracy is maintained under those conditions, but navigation performance degrades when environmental conditions change dynamically

Engineering Contradiction:
Improvenavigation accuracyVSAvoidadaptability to dynamic environmental conditions
Core Design Contradiction:
ReliabilityVSAdaptability or versatility

Solution Approach 1:

The system dynamically adjusts image processing parameters based on real-time environmental conditions and predicted feature locations. The parameter adjuster modifies extraction parameters adaptively as the vehicle operates through changing environments, transitioning from static calibration to dynamic adaptation.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system changes image processing parameters (such as feature extraction thresholds, contrast enhancement levels, and detection sensitivity) based on predicted environmental conditions and actual feature detection results, allowing the navigation system to maintain accuracy across varying environmental conditions.

Inventive Principle:
Principle #35Parameter changes

2Reliability

If the system processes entire image frames to maintain feature tracking, then feature tracking reliability is improved, but processing time and computational resources increase

Engineering Contradiction:
Improvefeature tracking reliabilityVSAvoidprocessing time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The system divides the image processing task by focusing computational resources only on bounded regions around predicted feature locations rather than processing entire image frames. This segmentation maintains tracking reliability while significantly reducing processing time and computational load.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system applies enhanced processing quality (detailed feature extraction and analysis) only to local regions where features are expected to appear, rather than uniformly processing the entire image. This localized approach maintains tracking accuracy for critical features while reducing overall processing requirements.

Inventive Principle:
Principle #3Local quality

3Loss of information

If features are lost due to environmental changes, then navigation data quality decreases, but system complexity increases if comprehensive fallback mechanisms are implemented

Engineering Contradiction:
Improvenavigation data qualityVSAvoidsystem complexity
Core Design Contradiction:
Loss of informationVSDevice complexity

Solution Approach 1:

The system performs preliminary adjustments to image parameters based on predicted environmental conditions and vehicle kinematics before features are completely lost. This proactive approach maintains feature detectability and navigation data quality without requiring complex reactive fallback mechanisms.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system uses feedback from feature detection results and environmental condition predictions to continuously adjust image processing parameters. This closed-loop approach maintains navigation data quality by adapting to changing conditions in real-time without requiring pre-programmed fallback procedures for every possible scenario.

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP3745353B1Systems and methods for image aided navigation
Publication Date: 2025.06.04 HONEYWELL INTERNATIONAL INC
  • EP3745353B1 patent drawingFigure 1
  • EP3745353B1 patent drawingFigure 1A
  • EP3745353B1 patent drawingFigure 2A

AI summary

In one embodiment, a vision aided navigation system comprises: at least one image sensor configured to produce image frames of a surrounding environment; a feature extractor configured to extract at least one image feature from a first image frame; a navigation filter configured to output a navigation solution based navigation data from a navigation device and changes in position of the image feature in the images; a feature tracker to receive the image frames and predict a location of the image feature in a subsequent image frame; a dynamic localized parameter adjuster to adjust at least one image parameter of the subsequent image frame; and wherein the feature tracker is configured so that when the image feature cannot be identified in the subsequent image frame within a bounded region around the predicted location, the dynamic localized parameter adjuster adjusts the at least one image parameter within the bounded region.