Camera-Based Lane Topology Detection for Navigation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing navigation systems for motor vehicles often provide inaccurate lane information, especially in heavy traffic or when lane markings are obscured, leading to incorrect navigation instructions.

Innovation Solution

A method using a camera to determine the number and topology of lanes by analyzing vehicle columns and direction indicators, allowing for real-time identification of right-turn, left-turn, and straight-ahead lanes, even when traditional lane markings are not visible, and providing precise vehicle positioning for improved navigation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If navigation systems use database data for lane information, then route guidance can be provided, but the lane information may be outdated and inaccurate, especially in heavy traffic or when lane markings are obscured

Engineering Contradiction:
Improvelane information accuracyVSAvoidoutdated database data
Core Design Contradiction:
ReliabilityVSLoss of information

Solution Approach 1:

The system continuously captures images of the road environment using a camera and processes these images to detect current lane markings, vehicle columns, and direction indicators. This real-time feedback loop updates the lane topology information dynamically, replacing outdated database data with current observed conditions, thereby ensuring accurate and reliable navigation guidance.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The navigation system performs its own lane detection and topology determination using onboard camera and image processing algorithms. Instead of relying solely on external database information, the system independently observes and determines current lane conditions, vehicle positions, and lane directions, making the navigation information self-updating and self-correcting.

Inventive Principle:
Principle #25Self-service

2Measurement precision

If the system uses image analysis to determine lanes, then real-time lane information can be obtained, but the complexity of image processing increases

Engineering Contradiction:
Improvelane detection accuracyVSAvoidimage processing complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The image processing is divided into distinct functional modules: vehicle column detection, lane marking detection, direction indicator detection, and lane topology determination. Each module handles a specific aspect of the analysis, making the overall complex task manageable and modular. This segmentation allows for specialized algorithms in each area while maintaining overall system organization.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The image processing system performs multiple functions using the same camera input: it detects lane markings, identifies vehicle columns, determines direction indicators, and infers lane topology. This multi-functionality reduces the need for separate sensors and processing systems, managing complexity while achieving comprehensive lane information extraction.

Inventive Principle:
Principle #6Universality (Multi-functionality)

3Measurement precision

If the system determines lane topology using direction indicators of vehicles, then accurate lane direction can be identified, but the system becomes more dependent on detecting multiple vehicle states

Engineering Contradiction:
Improvelane direction determinationVSAvoiddirection indicator detection
Core Design Contradiction:
Measurement precisionVSDifficulty of detecting and measuring

Solution Approach 1:

The system determines lane direction by analyzing the direction indicators of only a subset of vehicles in the detected vehicle columns, rather than requiring analysis of all vehicles. By examining the turn signals of representative vehicles and using threshold-based logic, the system achieves sufficient accuracy for lane topology determination without the excessive complexity of analyzing every vehicle's state in detail.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentEP3204889B1Method for a motor vehicle provided with a camera, device and system
Publication Date: 2021.03.31 CONTINENTAL AUTOMOTIVE GMBH
  • EP3204889B1 patent drawingFigure 1~2
  • EP3204889B1 patent drawingFigure 3~4
  • EP3204889B1 patent drawingFigure 5~6

AI summary

The invention relates to a method for a motor vehicle (100) provided with a camera (101), said method comprises the following steps: an image (102) from the camera (101) is provided, said image representing the surroundings (103) of the motor vehicle (100); at least one line of vehicles (104) is detected in the image; at least one driving lane (108) is detected in accordance with the detected at least one line of vehicles (104); a state of at least one driving direction display (112) in the image (102) is detected; a lane topology (114) is detected for the detected at least one vehicle lane (108) in accordance with the state of the at least one driving direction display (112).