Autonomous Driving Map Construction With Virtual Intersection Centerlines
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing map construction methods for autonomous driving, such as obstacle grid maps, indoor maps, and topology maps, fail to provide accurate road boundaries, lane boundaries, road traffic direction information, and lane traffic direction information, especially in complex scenarios like automated valet parking with extreme perception conditions.
Innovation Solution
A method that involves obtaining road information, intersection information, and lane information from manual driving track data and obstacle grid maps, and generating a virtual topology center line based on intersection entry and exit point information to create an autonomous driving map that includes road boundaries, lane boundaries, and traffic direction information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional obstacle grid map construction methods are used, then the map can be generated from vehicle motion posture and image data, but the map accuracy is low and cannot provide road boundaries, lane boundaries, or traffic direction information
Solution Approach 1:
The map construction process is segmented into multiple specialized modules: obstacle grid map construction, road map construction, and topology map construction. Each module focuses on extracting specific types of information (obstacles, road boundaries, lane markings, traffic directions) separately, allowing for higher precision in each aspect rather than attempting to capture all information in a single undifferentiated process
Solution Approach 2:
The invention transitions from traditional 2D grid-based representation to a multi-layered map structure that includes semantic information dimensions. By adding layers for road boundaries, lane markings, and traffic flow directions to the base obstacle grid, the system captures information in multiple dimensional categories simultaneously, enabling comprehensive navigation guidance
2Ease of manufacture
If indoor map construction methods are applied to autonomous driving, then a grid map can be generated by recording movement tracks, but the map cannot provide road boundaries, lane boundaries, or traffic direction information needed for vehicle guidance
Solution Approach 1:
The system creates a universal map framework that can handle multiple functions simultaneously: obstacle detection, road boundary identification, lane marking recognition, and traffic flow analysis. This multi-functional map structure serves both navigation and guidance purposes, eliminating the need for separate specialized maps for different functions
Solution Approach 2:
The system performs preliminary extraction of road boundaries, lane markings, and traffic directions during the map construction phase itself, rather than requiring separate post-processing steps. By pre-extracting and storing this guidance information in the map structure, the system enables real-time navigation decisions without additional computational overhead during vehicle operation
3Device complexity
If topology maps are constructed from random vehicle driving tracks, then a simplified map can be generated, but the map accuracy is low and cannot meet complex scenarios like automated valet parking
Solution Approach 1:
The system applies different levels of detail and processing to different regions of the map based on local requirements. High-precision road boundary and lane marking data are extracted and stored for critical areas such as intersections and parking zones, while less detailed representations are used for open road sections. This localized quality adjustment optimizes both accuracy where needed and overall system efficiency
Data Source
AI summary
A map construction method and a related apparatus are provided. The method includes: obtaining, based on manual driving track data and/or an obstacle grid map, road information, intersection information, and lane information of a region through which a vehicle has traveled; obtaining road traffic direction information based on the manual driving track data and the road information, and obtaining lane traffic direction information based on the lane information and the road traffic direction information; obtaining intersection entry and exit point information based on the intersection information and the lane traffic direction information; and performing, based on the intersection entry and exit point information, an operation of generating a virtual topology center line to obtain an autonomous driving map of the region through which the vehicle has traveled, where the virtual topology center line is a traveling boundary line of the vehicle in an intersection region.