Anchor Lane Selection for Vehicle Tracking at Merge Points
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional lane tracking systems struggle with maintaining accurate vehicle positioning in scenarios where lane boundaries are not parallel, such as at entrance or exit lanes, and face difficulties with missing or obscured lane markers, leading to unwanted lateral maneuvers.
Innovation Solution
A method that combines digital map data, leading vehicle trajectory data, and vision system data for lane boundary detection, using a fusion calculation to select an anchor lane, which includes evaluating lane marker conditions and considering entrance or exit lanes, to improve lane tracking performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a typical lane tracking system positions the vehicle in the center of a lane with parallel boundaries, then the vehicle can be easily tracked in normal driving conditions, but the system exhibits unwanted lateral maneuvers and tracking failures at lane merge and split locations where boundaries are not parallel
Solution Approach 1:
The system changes the reference parameter for lane tracking from the center of the lane to a selected anchor lane boundary. This parameter change allows the vehicle to maintain proper positioning at lane merges and splits by tracking the appropriate boundary rather than attempting to center in the lane, resolving the contradiction between normal tracking ease and complex scenario reliability
Solution Approach 2:
The system segments the lane boundary detection into multiple independent sources (vision system detecting left and right boundaries separately, digital map data providing expected lane information, leading vehicle trajectory data providing additional lane position cues). By processing these segmented sources independently and fusing them, the system can reliably select an anchor lane boundary even when some sources are obscured or unavailable
2Adaptability or versatility
If the lane tracking system uses vision system data alone for lane boundary detection, then the system can adapt to real-time road conditions, but the system fails when lane markers are missing or obscured
Solution Approach 1:
The system merges multiple independent data sources (vision system detections of left and right lane boundaries, digital map data containing expected lane positions and geometries, and leading vehicle trajectory data) into a unified lane representation. This fusion allows the system to maintain reliable lane detection by compensating for missing or obscured markers using complementary information from other sources, while still adapting to real-time conditions through the vision system
Solution Approach 2:
The digital map data and leading vehicle trajectory data serve as intermediary references that mediate between the vision system detections and the final lane tracking output. When vision data is ambiguous or missing, these intermediaries provide expected lane positions to guide the tracking system, ensuring reliability while allowing the vision system to drive adaptations to actual road conditions
3Stability of the object's composition
If the lane tracking system keeps the vehicle centered between diverging lane boundaries at an exit lane split, then the vehicle maintains a centered position temporarily, but the system must eventually select a lane which causes a sudden, unwanted lateral maneuver
Solution Approach 1:
The system performs preliminary lane selection by identifying and selecting an anchor lane boundary before the vehicle reaches the merge or split location. Using digital map data and leading vehicle trajectories, the system determines the appropriate lane in advance and begins tracking the corresponding boundary early, allowing the vehicle to smoothly transition to the correct lane without sudden lateral maneuvers while maintaining stable positioning
Data Source
AI summary
A method for selecting an anchor lane for tracking in a vehicle lane tracking system. Digital map data and leading vehicle trajectory data are used to predict lane information ahead of a vehicle. Left and right lane boundary markers are also detected, where available, using a vision system. The lane marker data from the vision system is combined with the lane information from the digital map data and the leading vehicle trajectory data in a lane curvature fusion calculation. The left and right lane marker data from the vision system are also evaluated for conditions such as parallelism and sudden jumps in offsets, while considering the presence of entrance or exit lanes as indicated by the map data. An anchor lane for tracking is selected based on the evaluation of the vision system data, using either the fused curvature calculation or the digital map and leading vehicle trajectory data.