Crowdsourced Lane Line Maps Using Multi-Vehicle Point Clouds
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current advanced driver assistance systems (ADAS) and automated driving systems (ADS) do not adequately account for factors that affect occupant experience, particularly in the accurate interpretation of road markings like lane lines, leading to misalignment and inaccuracies in vehicle navigation.
Innovation Solution
A method and system for crowd-sourcing lane line map data using point cloud alignment and optimization techniques, including a point cloud registration algorithm and factor graph optimization to align observations from multiple vehicles, generating an optimized aligned point cloud and updating a map database for improved accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If current ADAS and ADS systems use sensors to detect road markings, then they can identify lane lines for navigation, but the accuracy is reduced due to misalignment caused by GPS errors and other factors
Solution Approach 1:
The patent combines observations from multiple vehicles (first observation, second observation, third observation) into a unified lane line map. By merging these observations and using a server to process and align them, the system achieves higher accuracy than individual vehicle systems could achieve alone, resolving the contradiction between measurement precision and reliability.
Solution Approach 2:
The system uses a server to receive observations from multiple vehicles, process them, and generate corrected lane line maps that are fed back to the vehicles. This feedback loop allows continuous improvement of navigation accuracy by using aggregated data from multiple sources to correct individual vehicle measurement errors.
2Measurement precision
If the system processes multiple vehicle observations to improve accuracy, then lane line mapping precision increases, but the system complexity increases
Solution Approach 1:
The patent introduces a server as an intermediary component that handles the complexity of processing multiple vehicle observations. Instead of each vehicle needing to process all observations itself, the server acts as a mediator that receives data from multiple vehicles, performs the complex alignment and merging operations, and returns corrected lane line maps to the vehicles, thereby reducing individual vehicle system complexity while maintaining high precision.
3Measurement precision
If the system uses point cloud alignment algorithms to correct misalignment, then measurement accuracy improves, but processing time and computational requirements increase
Solution Approach 1:
The system performs preliminary processing by receiving and storing observations from multiple vehicles before final lane line map generation. The server pre-processes these observations to create aligned point clouds, which then can be quickly used to generate the final lane line map. This preliminary action reduces the computational burden during final map generation and decreases overall processing time.
Data Source
AI summary
A method for crowd-sourcing lane line map data for a vehicle may include receiving a plurality of observations. The plurality of observations includes at least a first observation and a second observation. The method further may include generating a point cloud alignment vector based at least in part on the plurality of observations. The method further may include generating an optimized aligned point cloud based at least in part on the point cloud alignment vector. The method further may include determining a lane line map based at least in part on the optimized aligned point cloud. The method further may include updating a map database based at least in part on the lane line map.