Lane Boundary Prediction Using Map and Sensor Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current digital maps lack information on road lane widths and precise lane courses, which is crucial for modern driver assistance systems, especially for oversized vehicles that require specific lane widths for safe navigation.
Innovation Solution
A method to determine the distance of lane boundaries from a vehicle using camera and radar data, combined with database information to predict the course and width of lanes, enabling precise lane detection and prediction for safe driving maneuvers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If digital maps provide only basic road course information, then map simplicity and storage efficiency are maintained, but lane width and precise lane course information are lost
Solution Approach 1:
The patent embeds lane boundary course information within the existing digital map road course structure. Each road segment in the database contains nested lane boundary definitions that describe the precise course and width of individual lanes, allowing detailed lane information to be stored within the conventional map data framework without requiring a complete restructuring of the map system.
Solution Approach 2:
The patent divides the road into multiple segments, with each segment containing specific lane boundary course information. This segmentation allows the map system to store detailed lane geometry for only those segments where precise lane information is available and necessary, rather than requiring complete detailed information for all roads, thus balancing information completeness with data management efficiency.
2Measurement precision
If lane width information is added to digital maps, then precision for driver assistance systems is improved, but data storage requirements and processing complexity increase
Solution Approach 1:
The patent pre-calculates and stores lane boundary course information in parametric form during map creation or updates. Rather than storing complete high-resolution lane geometry data, the system stores simplified mathematical representations (such as polynomial coefficients or control point coordinates) that can be efficiently evaluated to retrieve precise lane width and course information when needed, reducing storage requirements while maintaining measurement precision.
3Reliability
If precise lane boundary course information is stored in databases, then safety of driving maneuvers is improved, but system complexity and computational requirements increase
Solution Approach 1:
The patent replaces complex geometric calculations and manual lane analysis with pre-computed lane boundary course information stored in the database. Instead of requiring the vehicle's processing system to perform real-time lane detection and width calculation from raw sensor data or detailed map images, the system substitutes these computational tasks with direct retrieval and evaluation of pre-processed lane boundary parameters, significantly reducing onboard computational requirements while maintaining high reliability for safety-critical decisions.
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
The invention provides a method for determining a course of lanes (L1, L2) of a road for a vehicle (V). A distance (D1-D3) of at least one lane boundary (B1-B3, N1-N3) of at least one lane (L1, L2) from a predetermined point of the vehicle (V) is determined (S1). Further a course of the at least one lane boundary (B1-B3, N1-N3) is predicted (S2) based on information about a course of the road obtained from a database (DB) and based on the determined distance (D1-D3).