Mobile Pose Estimation Using Horizontal Vector Association
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional vector positioning methods for intelligent mobile devices suffer from false associations between extracted semantic information and map vector elements, leading to significant positioning errors.
Innovation Solution
A method that involves projecting map vector elements into an image coordinate system, ensuring one-to-one correspondence, and associating perception vector elements with projection vector elements based on horizontal coordinate consistency and minimal positional differences, followed by pose estimation using weighted particle poses to improve accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If vector elements are associated based on minimum distance between perception vector elements and map vector elements, then the association process is simple, but false associations occur frequently leading to large positioning errors
Solution Approach 1:
The patent introduces projection vector elements as an intermediary between perception vector elements and map vector elements. Map vector elements are first projected into the image coordinate system to generate projection vector elements, which then serve as the basis for association with perception vector elements. This intermediary step enables the use of horizontal coordinate consistency as an association criterion, significantly reducing false associations while maintaining operational feasibility.
2Measurement precision
If horizontal coordinate consistency is used as the association criterion, then positioning accuracy is improved, but the computational complexity increases due to projection operations
Solution Approach 1:
The patent performs preliminary projection of map vector elements into the image coordinate system before the association process. By pre-computing the projection vector elements and their horizontal coordinates, the system establishes a reference framework that simplifies the subsequent association operation. This preliminary action transforms a complex real-time computation problem into a more manageable comparison task, reducing the computational burden during actual positioning.
Data Source
AI summary
A mobile device positioning method and a mobile device positioning apparatus are disclosed. Perception vector elements are associated with projection vector elements based on horizontal positions of the perception vector elements extracted from an image and horizontal positions of the projection vector elements obtained through projection of vector elements in a map. In this way, when the horizontal positions of the associated perception vector elements and the horizontal positions of the associated projection vector elements are separately arranged in a same order, horizontal position distribution of all the arranged perception vector elements is consistent with that of corresponding projection vector elements in the arranged projection vector elements, and pose estimation is performed on a mobile device based on an association result.