Road Feature Point Extraction for Precise Autonomous Vehicle Control

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current autonomous driving systems face challenges in accurately extracting and providing relevant road information, leading to imprecise vehicle control, as they often extract unnecessary data and fail to identify critical road features like acceleration lane start and end points, affecting operations such as acceleration, deceleration, and lane changing.

Innovation Solution

A road feature point extraction method that uses high-precision map information and navigation data to determine and extract specific road feature points relevant to vehicle control, such as acceleration lane, deceleration lane, and main road widening/narrowing points, reducing data analysis and enhancing accuracy by combining navigation and map information redundancy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of information

If general road information is extracted from map information, then the vehicle can obtain road data, but useless information is extracted and really necessary information is not extracted

Engineering Contradiction:
Improveloss of useful road informationVSAvoidquantity of extracted data
Core Design Contradiction:
Loss of informationVSQuantity of substance

Solution Approach 1:

The patent extracts only the necessary road feature points (acceleration lane start/end points, deceleration lane start/end points, main road widening/narrowing points) from the map information, rather than extracting all general road information. This selective extraction ensures that useful information is captured while minimizing unnecessary data processing.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent applies different extraction criteria to different road feature types based on their specific characteristics. For example, acceleration lanes require extraction of start and end points for longitudinal control, while main road widening requires identification of specific location points. This localized approach ensures each feature type is extracted with the appropriate level of detail.

Inventive Principle:
Principle #3Local quality

2Reliability

If general road information is extracted, then road data is available, but the information is not targeted and cannot accurately control vehicle operations

Engineering Contradiction:
Improveaccuracy of vehicle controlVSAvoidcomplexity of information processing
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent extracts only the specific road feature points necessary for accurate vehicle control operations. By identifying and extracting only acceleration lane start/end points, deceleration lane start/end points, and main road widening/narrowing points, the system provides targeted information that directly supports longitudinal and lateral control without unnecessary data complexity.

Inventive Principle:
Principle #2Taking out (Extraction)

3Measurement precision

If navigation information and map information are combined, then the accuracy of extracting road feature points is improved, but the data processing complexity increases

Engineering Contradiction:
Improveaccuracy of road feature point extractionVSAvoidcomplexity of information integration
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent combines navigation information (current position, heading) with map information (road geometry, feature points) to accurately determine the vehicle's location and extract relevant road feature points. This integration allows the system to identify acceleration lanes, deceleration lanes, and main road widening/narrowing points with high precision by correlating real-time navigation data with pre-stored map data.

Inventive Principle:
Principle #5Merging (Combining)

Data Source

PatentEP3859278B1Road feature point extraction method and system
Publication Date: 2024.06.05 GREAT WALL MOTOR CO LTD
  • EP3859278B1 patent drawingFigure 1~2
  • EP3859278B1 patent drawingFigure 3~5
  • EP3859278B1 patent drawingFigure 6~8

AI summary

Disclosed are a road feature point extraction method and system. The road feature point extraction method comprises: acquiring map information for the current position of a vehicle (S110), wherein the map information comprises the attributes of the current road and the attributes of the next road section; comparing the attributes of the current road with the attributes of the next road section to determine the road type of the next road section (S120); and extracting, in conjunction with the road type of the next road section and the map information corresponding to the next road section, a road feature point representing a road scene on the next road section, and outputting the name of the extracted road feature point and information of the relative distance between the road feature point and the vehicle (S130). By determining a road feature point capable of precisely representing a road scene and extracting the name and position information of an important road feature point relating to vehicle control, without the need for separately collecting overall road information in the vicinity of the current position of a vehicle, the vehicle can be precisely controlled.