Runway Side Stripe Detection With 3D Map Aircraft Positioning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional systems struggle to improve the reliability and accuracy of state-estimation during critical flight stages of aircraft using inertial and GNSS data fusion.
Innovation Solution
An Identification Engine that utilizes camera data and a 3D map to identify aircraft runway side stripes, employing a CenterNet deep neural network and image analysis algorithms for precise aircraft positioning and orientation determination.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional inertial and GNSS data fusion is used for state estimation, then the system structure is simple, but the reliability and accuracy of state estimation deteriorates
Solution Approach 1:
The patent combines multiple data sources (camera imagery, 3D map data, inertial data, and GNSS data) into a unified state estimation system. The visual odometry module processes camera data to estimate aircraft pose, which is then fused with inertial and GNSS data through a Kalman filter to produce a comprehensive and reliable state estimate, resolving the contradiction by merging simple components into a robust integrated system.
Solution Approach 2:
The patent introduces a visual odometry module as an intermediary that processes camera imagery and 3D map data to generate pose estimates. This intermediary component bridges the gap between simple sensor inputs and the complex state estimation requirement, enabling improved reliability without directly increasing overall system complexity by providing a modular processing layer.
2Measurement precision
If conventional inertial and GNSS data fusion is used for state estimation, then the device complexity is low, but the measurement precision deteriorates
Solution Approach 1:
The patent merges camera-based visual odometry with inertial and GNSS data fusion to achieve high measurement precision. The visual odometry module extracts feature points from imagery and matches them with 3D map data to compute accurate pose estimates, which are then combined with other sensor data through Kalman filtering to produce a precise and reliable state estimation.
Solution Approach 2:
The patent transitions from traditional 1D/3D sensor data (inertial and GNSS) to 2D image space analysis by processing camera imagery. The visual odometry module operates in the 2D image plane to detect and track feature points, then maps these 2D observations to 3D pose estimates, adding a new dimensional perspective that enhances measurement precision.
3Loss of information
If camera data and 3D map are used for visual odometry, then the information quality improves, but the processing complexity increases
Solution Approach 1:
The patent extracts only the essential and relevant information from camera imagery by detecting and tracking specific feature points rather than processing the entire image. The visual odometry module identifies salient features in the image, matches them with corresponding 3D map points, and uses only these extracted features for pose estimation, thereby maintaining high information quality while reducing processing complexity.
Solution Approach 2:
The patent segments the complex task of visual odometry into distinct modules: feature detection, feature matching with 3D map, and pose estimation. This segmentation allows each module to process only the necessary data for its specific function, improving information quality through specialized processing while managing overall system complexity through modular architecture.
Data Source
AI summary
Various embodiments of an apparatus, methods, systems and computer program products described herein are directed to an Identification Engine. The Identification Engine identifies a portrayal in image data of at least one side stripe of an aircraft runway at a geographical location. The Identification Engine applies a three-dimensional (3D) map of the geographic location to the portrayal of the at least one side strip in the image data. Based on applying the 3D map, the Identification Engine determines a current position of an aircraft in the 3D map with respect to the aircraft runway at the geographical location.