Stereo Camera Lane Detection with Map Fusion

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current lane departure warning systems rely heavily on vision sensors for detecting lane markings, which can be inaccurate and unreliable, especially in adverse conditions such as fog or night, and do not effectively estimate lane positions beyond the camera's view.

Innovation Solution

A vehicle driving assistance device equipped with a stereo camera that captures stereo images, performs disparity detection, and combines this data with map information to detect and estimate lane positions, even in areas not directly visible, using a processor to generate lane images and control steering, braking, and suspension units for lane departure avoidance.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If vision sensors are used for detecting lane markings, then lane detection can be performed, but detection accuracy deteriorates in adverse conditions such as fog or night

Engineering Contradiction:
Improvelane detection reliabilityVSAvoidlane marking detection precision
Core Design Contradiction:
ReliabilityVSMeasurement precision

Solution Approach 1:

The patent combines multiple sensing approaches (vision sensors for visible lane markings, map information for geographic lane data, and sensor fusion algorithms) to create a complementary detection system. When vision sensors fail in adverse conditions, map information provides alternative lane position data, maintaining overall system reliability and precision.

Inventive Principle:
Principle #5Merging (Combining)

2Speed

If only camera data is used for lane detection, then real-time detection is possible, but lane estimation beyond camera view is not effective

Engineering Contradiction:
Improvelane detection speedVSAvoidlane position information beyond view
Core Design Contradiction:
SpeedVSLoss of information

Solution Approach 1:

The system pre-acquires and stores map information containing lane position data for areas beyond the current camera view. This preliminary preparation allows the system to immediately provide lane estimation for regions ahead of the vehicle without waiting for camera data, eliminating information loss in the forward direction.

Inventive Principle:
Principle #10Preliminary action

3Measurement precision

If map information is integrated with camera data, then lane detection accuracy is improved, but system complexity increases

Engineering Contradiction:
Improvelane position detection precisionVSAvoiddata integration system complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent introduces a sensor fusion module that acts as an intermediary between camera data and map information. This mediator processes and aligns the two data sources, handling coordinate transformations and data reconciliation, thereby improving lane detection precision while managing system complexity through modular architecture.

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentEP3088268B1Vehicle driving aid device and vehicle having same
Publication Date: 2020.02.12 LG ELECTRONICS INC
  • EP3088268B1 patent drawingFigure 1
  • EP3088268B1 patent drawingFigure 2
  • EP3088268B1 patent drawingFigure 3A

AI summary

The present invention relates to a vehicle driving assistance device and a vehicle having same. The vehicle driving assistance device according to an embodiment of the present invention comprises: a stereo camera; an interface unit exchanging data with at least one device inside a vehicle; and a processor performing lane detection for a first region in front of the vehicle based on a stereo image received from the stereo camera, and deducing lanes for a second region in front of the vehicle based on map information from an interface unit. Accordingly, the lanes in front of a vehicle can be perceived based on a captured image.