Vehicle Environment Mapping with Cylindrical and Planar Structures

Solution Overview

Problem

Current vehicle environment detection systems face inefficiencies in data handling and navigation due to large map data sizes and the inability to perform lane-level navigation with traditional mapping technologies, which are not suitable for autonomous vehicles.

Innovation Solution

A method for generating an efficient sparse map representation by identifying cylindrical and planar structures in a 3D environment, projecting them onto a 2D coordinate system, and comparing them with predefined reference data to determine vehicle position and environment type, reducing data size and computational complexity.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of information

If traditional mapping technology is used to store significant portions of observed spatial features, then comprehensive environment representation is achieved, but data size becomes large and computational complexity increases

Engineering Contradiction:

Improveenvironment representation completenessVSAvoidmap data size

Core Design Contradiction:

Loss of informationVSQuantity of substance

Solution Approach 1:

The patent extracts only the essential geometric features (cylindrical structures and planar structures) from the complete environment representation. Instead of storing all spatial features, it selectively identifies and stores only those features that are critical for vehicle localization and navigation, thereby reducing data size while maintaining sufficient representation completeness

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The patent segments the environment into distinct geometric primitives (cylindrical structures and planar structures) rather than storing the complete point cloud or mesh data. This segmentation allows the system to represent complex environments using simplified geometric models, reducing data storage requirements while preserving key spatial characteristics

Inventive Principle:

Principle #1Segmentation

2Measurement precision

If traditional mapping technology is used for navigation, then general location information is available, but lane-level navigation capability is not achieved

Engineering Contradiction:

Improvepositioning precisionVSAvoidmapping system complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent changes the representation parameters from traditional map data (roads, intersections, landmarks) to geometric structure parameters (cylindrical and planar structures with specific coordinates and orientations). This parameter transformation enables lane-level positioning precision by capturing the geometric characteristics of road infrastructure while maintaining manageable system complexity

Inventive Principle:

Principle #35Parameter changes

3Loss of information

If 3D environment data is processed using traditional methods, then complete spatial information is obtained, but computational complexity and processing time increase

Engineering Contradiction:

Improvespatial feature completenessVSAvoidprocessing speed

Core Design Contradiction:

Loss of informationVSProductivity

Solution Approach 1:

The patent extracts only the essential geometric features (cylindrical structures and planar structures) from the complete environment representation. Instead of storing all spatial features, it selectively identifies and stores only those features that are critical for vehicle localization and navigation, thereby reducing data size while maintaining sufficient representation completeness

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The patent segments the environment into distinct geometric primitives (cylindrical structures and planar structures) rather than storing the complete point cloud or mesh data. This segmentation allows the system to represent complex environments using simplified geometric models, reducing data storage requirements while preserving key spatial characteristics

Inventive Principle:

Principle #1Segmentation

Data Source

PatentEP3696507B1Representing an environment by cylindrical and planar structures

Publication Date: 2022.04.20 VEONEER SWEDEN AB

Figure 1
Figure 2
Figure 3~4

AI summary

The present disclosure relates to a vehicle control system (3) adapted for an ego vehicle (1) travelling on a road (2). The vehicle control system (3) comprises a main control unit (4) and an environment detection sensor arrangement (5) that is adapted to provide an initial detected representation of the environment. The main control unit (4) is arranged to provide a processed representation of the environment by identifying cylindrical structures and planar structures extending along a vertical extension in a coordinate system, and to confer a first identification that corresponds to a cylindrical structure, and a corresponding coordinate in a 2-dimensional projection x-y of the coordinate system 13, for each identified cylindrical structure. The main control unit (4) is further arranged to confer a second identification that corresponds to a planar structure, and a corresponding coordinate in the 2-dimensional projection for each identified planar structure.