Map-Based Lane-Keeping Control Without Visible Lane Markings
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle steering assistance systems fail to maintain lane control, especially in curves, during emergency situations when lane markings are not detected or unreliable, leading to potential accidents.
Innovation Solution
A method and device that utilize a digital map for lane-keeping control, combining rough and fine localization using GPS and cross-correlation to determine the vehicle's position and course, enabling precise lane-following even without visible lane markings, and activating emergency stop maneuvers if the driver is unfit to drive.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If lane markings are used for lane-keeping control, then the control is simple and direct, but it fails when lane markings are not detected or unreliable (e.g., in curves or emergency situations)
Solution Approach 1:
The patent introduces digital map data as an intermediary between the vehicle and the lane markings. When lane markings are not detectable, the system uses pre-stored digital map information about lane geometry and course as a substitute reference, allowing the lane-keeping control to continue functioning reliably without directly observing physical lane markings.
Solution Approach 2:
The system performs preliminary localization by rough matching vehicle position with digital map data before detailed lane-keeping control is needed. This preliminary action of identifying the relevant lane section in advance from digital maps enables the system to prepare the correct reference course before it is actually needed for control, ensuring continuity when markings disappear.
2Reliability
If map-based lane course determination is used when lane markings are not detected, then lane-keeping control can continue, but precise vehicle localization in the digital map is required
Solution Approach 1:
The localization process is segmented into two distinct stages: rough localization using GPS coordinates to identify the general area and relevant lane section in digital maps, followed by fine localization using cross-correlation of lane course with recorded vehicle path. This segmentation allows each stage to use appropriately matched precision methods.
Solution Approach 2:
The system uses feedback from cross-correlation between the digital map lane course and the actually recorded vehicle path to refine and verify the vehicle's position in the digital map. This feedback mechanism continuously adjusts the localized position based on the mismatch or match between expected and actual vehicle trajectories, improving positioning precision.
3Measurement precision
If cross-correlation of lane course with recorded path is performed for fine localization, then precise vehicle position is achieved, but computational complexity increases
Solution Approach 1:
The system applies cross-correlation only partially - specifically on the identified portion of the lane in the digital map that is relevant to the current vehicle position and direction, rather than correlating the entire digital map data set. This partial action reduces computational load while maintaining the precision benefits of cross-correlation where it is most needed.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables safe and precise lane-keeping control during emergency situations, including curves, without relying on detected lane markings, reducing the risk of accidents and ensuring the vehicle remains within the lane during braking.
Implementation Method 1
a rough position of the vehicle (1) in the digital map (2) is determined for a rough localization of the vehicle (1) in the digital map (2) by means of a global navigation satellite system
Implementation Method 2
a course of the lane (FS), in particular of the identified portion (A) of the lane (FS), is compared with a recorded course of a portion (WA) of a path travelled by the vehicle (1), wherein the comparison of the course of the lane (FS), in particular of the identified portion (A) of the lane (FS), with the recorded course of the portion (WA) of the path travelled by the vehicle (1) is carried out by cross-correlating the course of the lane (FS), in particular the identified portion (A) of the lane (FS), with the recorded course of the portion (WA) of the path travelled by the vehicle (1)
Data Source
AI summary
A method for operating a vehicle includes carrying out a lane-keeping control of the vehicle along a course of a lane travelled in by the vehicle. When lane markings are detected, the course of the lane is determined on a basis of detected lane markings. When lane markings are not detected, the course of the lane is determined in a mapped-based manner on a basis of data from a digital map where a rough localization of the vehicle in the digital map and a fine localization of the vehicle in the digital map is performed.