Sparse Road Maps and Lane Measurements for AV Localization

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Autonomous vehicles face challenges in navigating roadways due to the sheer volume of data required for traditional mapping, which can limit navigation accuracy and efficiency, and there is a need for optimized map construction and storage solutions.

Innovation Solution

The use of cameras to construct and navigate with a crowdsourced sparse map, incorporating GPS and sensor data, and adaptive navigation systems that utilize lane measurements and lane markings for precise vehicle positioning.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If traditional mapping technology is used for autonomous vehicle navigation, then navigation coverage and completeness are improved, but data storage requirements and system complexity increase significantly

Engineering Contradiction:
Improvenavigation coverageVSAvoiddata storage requirements
Core Design Contradiction:
ReliabilityVSQuantity of substance

Solution Approach 1:

The patent extracts only the essential navigation-critical information from traditional dense maps, specifically road geometry, lane markings, and key landmarks, while discarding redundant detailed map data. This creates a sparse map that retains navigation functionality with dramatically reduced data storage requirements.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent segments the map data into discrete, manageable components such as road segments, lane elements, and landmark markers rather than storing a continuous dense map. This segmentation allows the system to store only navigation-relevant information at appropriate levels of detail, reducing overall data volume while maintaining navigation accuracy.

Inventive Principle:
Principle #1Segmentation

2Reliability

If traditional mapping technology is used for autonomous vehicle navigation, then map completeness is improved, but computation time and processing efficiency deteriorate

Engineering Contradiction:
Improvemap completenessVSAvoidcomputation time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent extracts only the essential navigation-critical information from traditional dense maps, specifically road geometry, lane markings, and key landmarks, while discarding redundant detailed map data. This creates a sparse map that retains navigation functionality with dramatically reduced data storage requirements.

Inventive Principle:
Principle #2Taking out (Extraction)

3Quantity of substance

If sparse map data is used for autonomous vehicle localization, then data storage requirements are reduced, but localization precision and measurement accuracy worsen

Engineering Contradiction:
Improvedata storage requirementsVSAvoidlocalization precision
Core Design Contradiction:
Quantity of substanceVSMeasurement precision

Solution Approach 1:

The patent applies local quality by storing high-precision data only where navigation-critical features are present (such as lane markings and road geometry at intersections) while using lower precision or no data in areas without navigation-relevant features. This selective precision maintains localization accuracy where needed while reducing overall data requirements.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The patent introduces intermediary reference landmarks and road geometry models that serve as mediators between the sparse map data and the vehicle's position estimation system. These intermediaries provide sufficient precision for localization without requiring dense map data, bridging the gap between sparsity and accuracy requirements.

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentEP3843002B1Crowdsourcing and distributing a sparse map, and lane measurements for autonomous vehicle navigation
Publication Date: 2025.12.17 MOBILEYE VISION TECH LTD
  • EP3843002B1 patent drawingFigure 1
  • EP3843002B1 patent drawingFigure 2A
  • EP3843002B1 patent drawingFigure 2B

AI summary

Systems and methods are provided for generating, distributing, and using a sparse map and lane measurements for autonomous vehicle navigation. For example, one implementation relates to a non-transitory computer-readable medium including a sparse map for autonomous vehicle navigation along a road segment. Another implementation relates to a method of generating a road navigation model for use in autonomous vehicle navigation. A third implementation relates to a system for autonomously navigating a vehicle along a road segment. A fourth implementation relates to a method of determining a line representation of a road surface feature extending along a road segment. A fifth implementation relates to a method of determining a line representation of a road surface feature extending along a road segment. A sixth implementation relates to a system for correcting a position of a vehicle navigating a road segment.