Navigation Filter Condensing Vision Data for Carrier Motion Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing navigation systems face challenges in accurately estimating the movement of carriers, such as vehicles or robots, relative to their environment, especially in scenarios like skidding, sliding, or when GPS signals are lost due to obstacles or multi-path phenomena, leading to errors that accumulate over time.
Innovation Solution
A method that calculates condensed measurements from characteristic elements in images acquired by vision sensors, combined with navigation measurements, to estimate the movement of the carrier, using an a priori estimate from the navigation filter to reduce processing complexity and ambiguity in 2D vision scenarios.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If vision sensors are used to acquire images for movement estimation, then measurement precision is improved, but device complexity increases due to the need for augmented navigation filters and image processing systems
Solution Approach 1:
The patent extracts and processes only the most relevant information from images - specifically condensing image data into a small number of measurements (e.g., 3-6 measurements) that capture essential movement characteristics. This selective extraction reduces the information processing burden on the navigation filter while maintaining estimation precision, directly resolving the contradiction between measurement precision and device complexity
Solution Approach 2:
The navigation filter is segmented into distinct functional modules: an image processing module that extracts condensed measurements from images, and a navigation estimation module that fuses these condensed measurements with other sensor data. This segmentation allows each module to be optimized independently, reducing overall system complexity while maintaining high measurement precision
2Measurement precision
If all characteristic elements from images are processed by the navigation filter, then measurement precision is improved, but processing complexity increases significantly
Solution Approach 1:
The patent applies the extraction principle by selecting only the most informative condensed measurements from the full set of image characteristic elements. Instead of processing all detected features, the system extracts a minimal set of measurements (e.g., displacement of characteristic points in specific directions) that provide sufficient navigation information, thereby reducing processing complexity while maintaining precision
Solution Approach 2:
The patent implements partial action by processing only a subset of available image information - specifically, condensed measurements derived from key characteristic points rather than all image pixels or features. This partial processing approach achieves adequate measurement precision without the computational burden of exhaustive image analysis, effectively resolving the complexity-precision trade-off
3Measurement precision
If navigation filter processes augmented information from vision system, then measurement precision is improved, but processing time increases
Solution Approach 1:
The patent extracts condensed measurements that can be computed rapidly from images and fed into the navigation filter. By focusing on essential movement parameters rather than comprehensive image analysis, the system maintains position estimation accuracy while reducing processing time to meet real-time navigation requirements
Solution Approach 2:
The patent performs preliminary processing of image data to generate condensed measurements before they are input to the navigation filter. This preliminary extraction of key features and measurements prepares the data in advance, reducing the computational burden on the navigation filter and enabling faster overall processing while maintaining precision
Data Source
Figure 1~3
AI summary
The present invention relates to a method (10) for estimating the motion of a carrier with respect to an environment (20) in relation to which said carrier is moving, the carrier bearing a navigation sensor and a vision sensor producing images of the environment, said method comprising steps of: - (12) identification, in images acquired by the vision sensor, of characteristic elements of the images representing characteristic elements of the environment, - (14) calculation of at least one condensed measurement as a function of characteristic elements of at least two images acquired at different instants, - (16) estimation of the motion of the carrier by a navigation filter, as a function of navigation measurements carried out and as a function of the condensed measurement, in which the condensed measurement is calculated as a function furthermore of an a priori estimation of the motion, as provided by the navigation filter.