Vehicle Lateral Lane Positioning Using Map-Aided Road Marking Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for determining a vehicle's lateral position relative to road lanes lack sub-lane accuracy, with existing approaches either being cost-intensive, limited to recognizing only one or two lanes, or suffering from distance errors in lane marking detection.
Innovation Solution
A method that combines GPS data with lane geometry map information and imaging sensors to detect road markings, using orthogonal lines and intersection points to determine the vehicle's approximate lateral position, considering passable lanes and driving maneuvers, and adjusting for situations where lane changes are not possible.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If high-precision GPS and digital lane geometry map are used, then lateral position determination accuracy is improved, but cost increases significantly
Solution Approach 1:
The patent combines imaging sensors (camera) with digital map information and GPS data to create a fused lane detection system. The camera detects lane markings while the digital map provides lane geometry and the system merges these data sources with confidence weighting to achieve sub-lane accuracy without requiring high-precision GPS alone, thus reducing system cost while maintaining accuracy.
Solution Approach 2:
The patent introduces an intermediary processing layer that uses digital map information as a reference framework. The camera-detected lane markings are matched against pre-stored lane geometry from digital maps, and the system uses this intermediary map-based reference to correct and refine GPS-based position estimates, achieving high accuracy without directly relying on expensive high-precision GPS hardware.
2Device complexity
If imaging sensors are used without digital map, then system cost is reduced, but lane detection accuracy and coverage are limited
Solution Approach 1:
The patent merges imaging sensor data with digital map information in a fused detection system. The camera captures lane markings while the digital map provides lane geometry, number of lanes, and edge marking information. These complementary data sources are combined with confidence weighting to achieve accurate lane detection and sub-lane position determination that neither source could achieve alone.
3Adaptability or versatility
If imaging sensors detect lane markings, then lane identification is enabled, but distance errors reduce positioning accuracy
Solution Approach 1:
The patent implements a feedback mechanism where detected lane markings are continuously compared against pre-stored lane geometry from digital maps. The system calculates confidence levels for each detection and uses this feedback to weight and fuse the data sources. When detection confidence is high, sensor data dominates; when confidence is low or distance errors are large, the digital map reference corrects the position, maintaining sub-lane accuracy.
Solution Approach 2:
The patent uses digital map lane geometry as an intermediary reference framework. The camera-detected lane markings are matched against this intermediary map data, and the system uses the map-based reference to correct distance errors in sensor measurements. This intermediary reference layer enables accurate position determination even when direct sensor-to-vehicle distance measurements have errors.
4Device complexity
If only detected lanes are used for positioning, then system simplicity is maintained, but position relative to all lanes cannot be determined
Solution Approach 1:
The patent makes the system universally applicable to all lanes by using digital map information that contains comprehensive lane geometry for the entire road section. The camera and processing system can identify and position the vehicle relative to any lane on the road, not just the currently detected lane, because the digital map provides a complete lane framework that the system can reference for multi-lane positioning and navigation.
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
The invention relates to a method, a device (20) and a computer-readable storage medium with instructions for determining the lateral position of a vehicle relative to the lanes of a road. The device (20) has an image processing unit (22) for detecting (10) road markings, and a position-determining unit (23) for determining (11) a starting position for the vehicle. An evaluation unit (24) is designed to ascertain (12) that the detected road markings are not sufficient to establish a lateral position of the vehicle by comparing the detected road markings and road marking information from a lane geometry map for the starting position determined for the vehicle. In this case, the evaluation unit (24) establishes (13) an approximately lateral position of the vehicle while taking into account information from the lane geometry map.