Lane Determination Using Visual Map Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current vehicle positioning methods, especially in self-driving technologies, face challenges in accurately determining the lane position with high precision, leading to reduced navigation accuracy and increased costs due to the reliance on satellite-based systems and high-investment hardware solutions.
Innovation Solution
A method and apparatus that utilize image processing and feature matching between in-vehicle camera-collected lane information and high-precision map data to determine the target lane, reducing the reliance on GPS/BDS for transverse positioning, thereby improving accuracy and reducing costs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If satellite-based positioning systems are used for vehicle positioning, then positioning coverage is improved, but positioning precision deteriorates
Solution Approach 1:
The patent combines satellite positioning results with visual recognition results from image processing to determine the final vehicle lane position. This fusion of multiple positioning methods resolves the contradiction by maintaining broad coverage from satellite systems while achieving high precision through image-based lane detection and matching with map data.
Solution Approach 2:
The patent introduces high-precision map data as an intermediary between satellite positioning and final lane determination. The map data provides pre-stored lane information that serves as a reference for matching detected lane markings, thereby improving positioning precision while maintaining the broad coverage capability of satellite-based systems.
2Measurement precision
If high-precision hardware solutions are invested in for lane positioning, then positioning accuracy is improved, but investment costs increase
Solution Approach 1:
The patent uses visual copying of lane markings from the real world through image capture and processing, then matches these visual copies with digital representations in high-precision map data. This approach achieves accurate lane positioning using existing camera hardware rather than requiring expensive specialized positioning hardware.
Solution Approach 2:
The patent replaces mechanical/hardware-based high-precision positioning systems with an information-processing approach using image recognition and data matching. This substitution achieves comparable or superior accuracy using software and standard sensors rather than expensive specialized hardware.
Data Source
Figure 1a~2a
Figure 2b~3
Figure 4a~4b
AI summary
Embodiments of this application disclose a lane determining method and a computing device. The method includes: obtaining image information of a road surface of a vehicle collected by the vehicle; identifying first lane information of at least one first lane of the road surface from the image information, the first lane information including a location relationship between a travelling lane in which the vehicle is located in the road surface and the at least one first lane, and a feature of the first lane; and performing feature matching between the feature of the at least one first lane and a feature of at least one second lane in a map in which the road surface is located to determine a target lane of the vehicle in the map, the target lane being a lane corresponding to the travelling lane in the map, and the feature of the at least one second lane being obtained from the map.