3D Road Geometry Estimation Using Fresnel Segment Modeling

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current road geometry estimation methods for autonomous vehicles rely on mathematical approximations that incur significant errors, which are not suitable for the high precision and safety-critical operations required in autonomous driving.

Innovation Solution

A system and method that utilize Fresnel integrals to accurately model road geometry, allowing for precise estimation of road curvature and heading, and extend this to 3D road geometry modeling by dividing roads into segments with uniform length, ensuring G2-continuity and minimal approximation errors.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If mathematical approximations are used for road geometry estimation, then computation speed is improved, but measurement precision deteriorates due to significant approximation errors

Engineering Contradiction:
Improvecomputation speedVSAvoidroad geometry estimation accuracy
Core Design Contradiction:
ProductivityVSMeasurement precision

Solution Approach 1:

The patent changes the mathematical parameters from standard polynomial approximations to Fresnel integrals with specific parameter transformations. This allows the system to maintain computational efficiency while achieving high precision in road geometry estimation by transforming the problem into a different mathematical domain where both speed and accuracy can be preserved.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent segments the road geometry into multiple road segments, each modeled independently using Fresnel integrals. This segmentation allows for localized high-precision modeling while maintaining overall computational efficiency through divide-and-conquer strategy, resolving the contradiction between detailed accuracy and computation speed.

Inventive Principle:
Principle #1Segmentation

2Measurement precision

If high precision road geometry modeling is implemented, then measurement precision is improved, but device complexity increases due to computational requirements

Engineering Contradiction:
Improveroad geometry estimation accuracyVSAvoidcomputational system complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent uses Fresnel integrals as a mathematical copy or representation of the road geometry that can be computed efficiently. Instead of directly modeling complex real-world road variations, the system creates a simplified mathematical copy using Fresnel functions that preserves essential geometric features while reducing computational complexity.

Inventive Principle:
Principle #26Copying

3Ease of manufacture

If standard polynomial approximations are used, then ease of manufacture is improved, but manufacturing precision deteriorates due to approximation errors affecting autonomous vehicle safety

Engineering Contradiction:
Improveimplementation simplicityVSAvoidroad geometry modeling accuracy
Core Design Contradiction:
Ease of manufactureVSManufacturing precision

Solution Approach 1:

The patent substitutes traditional mechanical/mathematical approximation methods (polynomial fitting) with a different mathematical system based on Fresnel integrals. This substitution maintains the ease of implementation through standardized mathematical functions while dramatically improving the precision of road geometry modeling to meet safety-critical requirements.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentEP4350642A1Three-dimensional road geometry estimation
Publication Date: 2024.04.10 EMBARK TRUCKS INC
  • EP4350642A1 patent drawingFigure 1
  • EP4350642A1 patent drawingFigure 2A
  • EP4350642A1 patent drawingFigure 2B

AI summary

A system and method including identifying lane line data associated with a road within the sensor data; modelling a geometry of the road as a sequence of road segments, each road segment being defined by parameters including a curvature rate and a road grade rate; generating, based on a mathematical representation of the modelled road geometry, an approximation of each road segment; and generating, based on the generated approximation of each road segment, a three-dimensional representation of the road including the sequence of segments.