Visual Map Matching for Vehicle Positioning Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current positioning systems, such as GPS, are not accurate enough to reliably locate a vehicle within an intersection, which is a common site for fatal accidents, and existing methods for generating maps are not efficient for improving intersection safety.
Innovation Solution
A method using a video camera to capture images of visible features at an intersection, detecting and comparing points in successive images to build a map, with a predictive filter like a Kalman filter to improve accuracy and confidence in feature detection and vehicle positioning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If GPS positioning system is used to locate the vehicle, then the system is simple to implement, but the positioning accuracy is insufficient (error of tens of metres)
Solution Approach 1:
The patent combines GPS positioning with visual map matching technology. The system integrates GPS coarse positioning with image-based feature detection and map comparison to achieve precise positioning. The processor combines GPS coordinates with detected landmark features from images to determine the vehicle's exact position, resolving the contradiction between simplicity and accuracy.
Solution Approach 2:
The patent introduces a visual map as an intermediary between the GPS system and the final positioning result. The map contains pre-stored landmark features that serve as reference points. The system captures images, detects features, compares them with the map, and uses this intermediate visual information to correct and refine GPS positioning accuracy.
2Measurement precision
If video camera and image processing are used to detect feature points, then the positioning accuracy is improved, but the processing time and computational complexity increase
Solution Approach 1:
The patent performs preliminary actions by pre-generating a visual map containing landmark feature points before the vehicle reaches the location. The map is created offline using images captured during vehicle movement, with features detected and stored in advance. When positioning is needed, the system only needs to compare current images with the pre-existing map, significantly reducing real-time processing time.
Solution Approach 2:
The patent extracts only the essential landmark features from images rather than processing all image data. The system identifies and extracts key feature points (landmarks) that are sufficient for positioning, discarding redundant information. This selective extraction reduces computational complexity while maintaining positioning accuracy.
3Reliability
If more landmark features are detected and stored on the map, then the positioning reliability is improved, but the data processing load and storage requirements increase
Solution Approach 1:
The patent applies local quality by storing detailed landmark feature data only at critical locations (intersections and key points) rather than uniformly across all areas. The system identifies locations where positioning reliability is most needed and concentrates map data resources there, reducing overall data volume while maintaining reliability where it matters most.
Solution Approach 2:
The patent segments the map into multiple sections or zones, each containing relevant landmark features for that specific area. The system divides the geographical space into manageable segments and stores features locally for each segment, reducing the amount of data that needs to be processed and stored globally while maintaining comprehensive coverage.
Data Source
Figure 1
Figure 2~3
Figure 4
AI summary
A method of generating a map comprising the location of points relating to features, comprising moving a vehicle (100) past a scene, capturing (202) an image of a scene from the vehicle (100), detecting (204) points relating to features in the captured image, and generating the map (230) by recording the position of the points. The map may be of a road intersection, a car park or a single car parking space (800). In addition, a parking assistance apparatus for a vehicle (100) is disclosed, comprising a video camera (102) arranged to, in use, capture images of parking space (800) as the vehicle (100) is driven past the space, the parking space (800) having visible lines (801, 802, 803) demarcating the space, and a processor (103) arranged to, in use, map the lines demarcating the space using the method of any preceding claim and captured images from the video camera and guidance means to guide the vehicle into the space.