Swarm Point-Cloud Matching for Traffic Light Lane Assignment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods struggle to accurately assign traffic lights to specific lanes at intersections due to complex and diverse intersection situations, leading to high error rates, especially when high-precision vehicle localization is unavailable.
Innovation Solution
A method utilizing swarm data in a backend server to determine the ego vehicle's lane by projecting offset vectors onto orientation vectors, employing the Iterative Closest Point (ICP) algorithm for point cloud matching and DBSCAN clustering to correct traffic light positions and assign them to lanes, using a global coordinate system.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional single-vehicle detection methods are used to assign traffic lights to lanes, then the system can operate with minimal data processing, but the accuracy and reliability of lane assignment deteriorates due to complex intersection situations
Solution Approach 1:
The patent combines detection data from multiple vehicles (swarm data) to improve lane assignment accuracy. By merging observations from several vehicles passing through the same intersection, the system achieves more reliable traffic light-to-lane assignments than single-vehicle detection, resolving the contradiction between reliability and data processing complexity.
Solution Approach 2:
The system creates a virtual representation (copy) of the intersection scene by aggregating detection data from multiple vehicles. This synthesized model allows the backend server to analyze traffic light positions and lane assignments without requiring each individual vehicle to have complex processing capabilities, thus improving reliability while managing complexity.
2Measurement precision
If high-precision vehicle localization is used to determine lane assignment, then the accuracy of traffic light assignment improves, but the system becomes more complex and less robust to localization failures
Solution Approach 1:
The system uses feedback from multiple vehicles' detection data to collectively determine lane assignments. Instead of relying on single-vehicle localization precision, the system aggregates observations from multiple sources, making it robust to individual localization failures while maintaining overall accuracy through consensual detection results.
Solution Approach 2:
The patent segments the lane assignment task into separate detection components performed by individual vehicles, then aggregates these segmented results at the backend server. This segmentation allows each vehicle to use simple localization while the collective data provides high precision, improving robustness to individual localization errors.
3Reliability
If the system processes swarm data from multiple vehicles to improve lane assignment, then the accuracy of traffic light detection improves, but the data processing time and computational resources increase
Solution Approach 1:
The system performs preliminary detection and temporary storage of traffic light data at each vehicle before aggregation. By pre-processing detection results locally and only transmitting necessary aggregated data to the backend server, the system reduces overall processing time while maintaining the reliability benefits of swarm data analysis.
Data Source
Figure 1~2
AI summary
The invention relates to a method for assigning traffic lights to the corresponding lanes of a roadway from swarm data in a back-end server, having the steps of: - receiving consolidated data for each traversal of a traffic light by a vehicle of a swarm for a specified intersection, wherein the consolidated data for each intersection traversal contains at least one traffic light image with the positions of the individual light signals, - matching the traffic light images, consisting of point clouds, of different traffic light traversals for a specific combination of intersection and approach direction in that individual pairs of traffic light object points of the point clouds are mapped together, and - correcting the traffic light positions on the basis of the results of the matching.