Autonomous Vehicle HD Map Realignment for Precise Positioning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles at levels 3 and 4 face challenges in accurately calculating their position due to inaccuracies in GPS and HD map alignment, leading to potential safety risks and control transfer issues.
Innovation Solution
The system uses GPS information and HD map data to determine a vehicle's position, set candidate spatial regions, and realign based on road property values to enhance precision positioning, using algorithms like Graham scan and point-in-polygon to improve accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If GPS information and HD map alignment is used to determine vehicle position, then positioning speed is improved, but positioning accuracy deteriorates due to GPS inaccuracies and complex road structures
Solution Approach 1:
The positioning process is divided into multiple stages: initial GPS-based positioning, candidate spatial region identification, position assessment operation, and final position determination. This segmentation allows the system to use fast GPS positioning for initial location while subsequently refining accuracy through multi-step verification processes.
Solution Approach 2:
Candidate spatial regions serve as intermediary elements between GPS coordinates and final vehicle position. The system identifies multiple candidate regions based on GPS data, then performs position assessment operations to determine which candidate region contains the actual vehicle position, thereby mediating between imprecise GPS data and accurate positioning.
2Measurement precision
If multiple candidate spatial regions are evaluated using position assessment operations, then positioning accuracy is improved, but computational complexity increases
Solution Approach 1:
The system performs preliminary actions by pre-identifying candidate spatial regions based on GPS information and road network data before executing the position assessment operation. This preliminary filtering reduces the search space and prepares data structures that facilitate faster and more efficient position determination.
Solution Approach 2:
The position assessment operation dynamically evaluates multiple candidate spatial regions by comparing GPS coordinates with road network geometry. The system adaptively determines vehicle position by assessing which candidate region best matches the GPS data, allowing flexible computational effort based on positioning uncertainty.
3Reliability
If real-time position realignment is performed based on road property values, then positioning reliability is improved, but processing time increases
Solution Approach 1:
The system implements feedback by continuously comparing determined vehicle position with road network data and performing realignment when discrepancies are detected. Road property values such as lane markings and road geometry provide feedback signals that trigger position corrections, ensuring reliable positioning while maintaining real-time performance through selective realignment.
Data Source
AI summary
An apparatus for controlling driving of a vehicle is introduced. The apparatus may comprise, a processor, a memory storing instructions, that when executed by the processor, are configured to cause the apparatus to, determine, based on GPS information and HD map information, a first position of the vehicle, determine, based on the determined first position, a first road property value, determine, based on the determined first road property value, a spatial region of the vehicle, determine, based on the determined spatial region, a second position of the vehicle, determine, based on the determined second position, a second road property value, adjust, based on the second road property value and the first road property value, a position of the vehicle in an HD map, output a signal associated with the adjusted position of the vehicle, and control, based on the signal, driving of the vehicle.