Crowdsourced Lane Line Mapping with Point Cloud Alignment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current advanced driver assistance systems (ADAS) and automated driving systems (ADS) struggle to accurately interpret road markings, such as lane lines, due to factors like GPS/GNSS errors and weather conditions, leading to misalignment and suboptimal vehicle navigation.
Innovation Solution
A method and system for crowd-sourcing lane line map data using vehicle observations, involving point cloud alignment and optimization algorithms to align vehicle trajectories and correct point clouds, utilizing computer vision and iterative optimization techniques to generate accurate lane line maps.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional ADAS/ADS systems use standard sensors (cameras, radar, LiDAR) to detect road markings, then they can identify basic lane line positions, but the accuracy deteriorates due to GPS/GNSS errors and weather conditions causing misalignment
Solution Approach 1:
The patent merges data from multiple vehicles (first vehicle and second vehicle) to create a more reliable lane line map. By combining observations from multiple sources and using a server to process and align the data, the system achieves higher accuracy and reliability than single-vehicle systems can accomplish alone.
Solution Approach 2:
The server acts as an intermediary between multiple vehicles and the lane line mapping process. It receives observations from multiple vehicles, performs point cloud alignment using correction vectors, and generates unified lane line maps. This intermediary processing enables accurate alignment despite GPS errors and weather conditions affecting individual vehicle sensors.
2Measurement precision
If the system processes observations from multiple vehicles to improve accuracy, then lane line mapping precision improves, but the computational complexity and processing time increase
Solution Approach 1:
The patent segments the complex processing task into distinct stages: receiving observations from multiple vehicles, generating point cloud alignment vectors, determining correction vectors, aligning point clouds, and generating lane line maps. This segmentation allows the server to process data systematically and efficiently, reducing overall computational complexity.
Solution Approach 2:
The system creates a simplified representation (point cloud alignment vector) of the complex relationship between multiple vehicle observations. By using correction vectors to align point clouds, the system copies and transforms the essential geometric relationships without processing every raw data point in full detail, thus reducing computational burden while maintaining precision.
Data Source
AI summary
Systems and methods are provided for crowd-sourcing lane line map data for a vehicle. The systems include a server communication system, a map database, a server controller that is programmed to: receive observations using the server communication system, generate a point cloud alignment vector based at least in part on the observations, determine confidence scores for the observations based at least in part on the point cloud alignment vector, designate a subset of the observations having confidence scores in excess of a confidence score threshold as anchor points, generate an optimized aligned point cloud based at least in part on the point cloud alignment vector and the anchor points, determine a lane line map based at least in part on the optimized aligned point cloud, and update the map database based at least in part on the lane line map.