Free Space Boundary Curves With Adaptive Control Points

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing vehicle assistance systems face challenges in efficiently processing and storing grid-based environment models due to high computational resource requirements and large memory needs, especially when sharing data between vehicles, and parametric curve representations with fixed control points fail to accurately match complex boundary shapes.

Innovation Solution

A method that dynamically adjusts the number of control points in a parametric curve based on their modification relevance, determined by factors like distance and angle relative to the vehicle, to optimize the representation of the free space boundary, allowing for more precise and efficient modeling of the environment.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If grid-based representation is used to model the environment, then the representation accuracy is improved, but the computational resources and memory requirements increase significantly

Engineering Contradiction:
Improveenvironment representation accuracyVSAvoidcomputational resource requirements
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent transforms the environment model from a fixed grid-based representation to a parametric curve representation defined by control points. This parameter change allows the model to maintain high accuracy while reducing computational complexity, as the parametric form requires fewer parameters to describe the same environmental boundaries compared to grid cells.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent applies different levels of detail to different parts of the environment model by dynamically adjusting the number and distribution of control points along the parametric curve. Regions with higher curvature or importance receive more control points, while simpler regions use fewer points, optimizing both accuracy and computational efficiency locally.

Inventive Principle:
Principle #3Local quality

2Device complexity

If a fixed number of control points is used in the parametric curve, then the device complexity is reduced, but the approximation quality of complex boundary shapes deteriorates

Engineering Contradiction:
Improvenumber of control pointsVSAvoidboundary shape approximation quality
Core Design Contradiction:
Device complexityVSManufacturing precision

Solution Approach 1:

The patent makes the number of control points dynamic rather than fixed. The system automatically adjusts the number and distribution of control points based on the local complexity of the boundary being modeled. This dynamic adaptation allows the model to use more control points where needed (complex boundaries) and fewer where sufficient (simple boundaries), resolving the contradiction between model simplicity and accuracy.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent segments the parametric curve into multiple sections, each with its own set of control points. This segmentation allows independent optimization of each section, enabling the model to allocate computational resources (control points) efficiently across different parts of the boundary based on their individual complexity requirements.

Inventive Principle:
Principle #1Segmentation

3Device complexity

If equidistant control points are distributed along the parametric curve, then the device complexity is reduced, but the approximation precision of locally complex regions deteriorates

Engineering Contradiction:
Improvecontrol point distributionVSAvoidlocal boundary approximation precision
Core Design Contradiction:
Device complexityVSMeasurement precision

Solution Approach 1:

The patent implements non-uniform distribution of control points along the parametric curve, concentrating more control points in regions with high curvature or local complexity while using fewer points in simpler regions. This local quality adjustment ensures that approximation precision is optimized where it matters most without unnecessarily increasing the total number of control points.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The control point distribution transitions from a static equidistant arrangement to a dynamic, adaptive distribution that responds to the local geometric characteristics of the boundary. The system continuously adjusts the spacing and number of control points based on measured curvature or complexity metrics, enabling precision adaptation to local features.

Inventive Principle:
Principle #15Dynamics

Data Source

PatentEP4310798A1Method for determination of a free space boundary of a physical environment in a vehicle assistance system
Publication Date: 2024.01.24 APTIV TECHNOLOGIES AG
  • EP4310798A1 patent drawingFigure 1
  • EP4310798A1 patent drawingFigure 2
  • EP4310798A1 patent drawingFigure 3

AI summary

The present disclosure relates to a computer-implemented method, apparatus, computer program and a vehicle comprising the apparatus for determining in a vehicle assistance system a free space boundary of a physical environment surrounding the vehicle. The method comprises the steps of: a. obtaining a parametric curve approximating spatial information representing the free space boundary, wherein the parametric curve is defined by a plurality of control points; b. determining a modification indicator for a control point of the plurality of control points, the modification indicator indicating a modification relevance of a section of the parametric curve affected by the control point, wherein determining the modification indicator comprises determining a position modification indicator indicating a relative position of the control point towards a position of the vehicle; and c. modifying the section of the parametric curve affected by the control point based on the modification indicator.