Aerial Vehicle Pose Localization Using Visual Route Features

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Traditional navigation systems for aerial vehicles, such as drones, rely on GNSS and IMU, which are susceptible to signal interference and drift, leading to inaccurate pose estimation in complex environments.

Innovation Solution

Utilizing absolute visual localization (AVL) to determine pose by analyzing route features with onboard cameras, integrating with IMU and GNSS data for enhanced navigation, and employing neural networks for image processing to improve accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If GNSS and IMU are used for navigation, then positioning is provided, but accuracy deteriorates in complex environments due to signal obstructions and drift

Engineering Contradiction:
Improvenavigation reliabilityVSAvoidpose estimation accuracy
Core Design Contradiction:
ReliabilityVSMeasurement precision

Solution Approach 1:

The patent introduces visual landmarks as an intermediary reference system between the aerial vehicle and the environment. By detecting and matching visual features in the environment against a pre-built map, the system establishes a reliable pose estimation mechanism that does not depend on GNSS signals or IMU integration, thereby resolving the accuracy problem in complex environments.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent replaces the mechanical/sensor-based navigation system (GNSS+IMU) with a vision-based localization system. Instead of relying on inertial measurements and satellite signals that suffer from drift and signal loss, the system uses image processing and feature matching to determine pose, substituting the navigation approach to eliminate the fundamental accuracy limitations.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

2Adaptability or versatility

If multiple sensor modalities are integrated, then navigation coverage is improved, but system complexity increases

Engineering Contradiction:
Improvenavigation coverageVSAvoidsystem complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent extracts and focuses solely on vision-based localization for pose estimation, separating this function from the traditional GNSS/IMU navigation stack. By taking out the visual localization component as a distinct system that operates independently, the patent achieves navigation coverage in GNSS-denied areas without requiring complex integration of multiple sensor modalities, thereby reducing overall system complexity.

Inventive Principle:
Principle #2Taking out (Extraction)

Data Source

PatentEP4645282A1System and method for determining the pose of an aerial vehicle using absolute visual localization of route features
Publication Date: 2025.11.05 NURENDA TECHNOLOGIES
  • EP4645282A1 patent drawingFigure 1
  • EP4645282A1 patent drawingFigure 2
  • EP4645282A1 patent drawingFigure 3A

AI summary

The system and methods of the various embodiments may enable an aerial vehicle to determine its pose using absolute visual localization of route features and either a keypoint-based pipeline or a template based pipeline. This may result in the aerial vehicle being able to determine its pose when traditional methods of pose detection fail.