Position Estimation Using Vanishing Point Rotation Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
GPS navigation systems face significant position errors of 10 meters to 100 meters due to limitations in signal accuracy, necessitating additional correction methods for precise localization, especially in dynamic environments like vehicle navigation.
Innovation Solution
A method that determines a first rotation reference point from localization information and map data, and a second rotation reference point from image data, using a vanishing-point detection model, to calculate a rotation parameter for correcting localization information, thereby improving position estimation accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If GPS navigation system is used for position estimation, then position information can be obtained, but position error of 10 meters to 100 meters occurs
Solution Approach 1:
The patent introduces map data as an intermediary reference system. By determining rotation reference points from both localization information (GPS) and map data, and comparing them with rotation reference points detected from image data (vanishing points), the system creates a mediation layer that corrects GPS position errors through geometric relationships between multiple reference points.
Solution Approach 2:
The patent replaces reliance on purely mechanical/electronic GPS positioning with an optical-geometric system. By using image data from cameras to detect vanishing points and comparing these with map-based reference points, the system substitutes optical measurement and geometric calculation to correct the inherent errors in GPS positioning.
2Measurement precision
If additional correction methods are introduced to reduce position error, then position accuracy improves, but system complexity increases
Solution Approach 1:
The patent makes the rotation reference points serve multiple functions: they are used both for determining the vehicle's position and for correcting orientation/rotation information. By determining a first rotation reference point from map data and a second from image data, then using both to calculate rotation parameters, the system achieves multi-functional correction with a unified approach.
Solution Approach 2:
The system uses the vehicle's own camera and existing map data to perform self-correction of GPS position errors. The localization information and image data from the vehicle's own sensors are processed to generate correction parameters, making the system self-sufficient without requiring external correction infrastructure.
Data Source
AI summary
A method of estimating a position includes: determining a first rotation reference point based on localization information estimated for a target and map data; estimating a second rotation reference point from image data associated with a front view of the target; and correcting the localization information using a rotation parameter calculated based on the first rotation reference point and the second rotation reference point.


