Intersection Turn Path Splines From Probe Data for Vehicle Guidance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous and semi-autonomous vehicles face challenges in navigating intersections due to the complexity of available paths, which existing digital maps struggle to accurately represent, leading to inefficiencies and potential navigation errors.
Innovation Solution
The method involves generating two- or three-dimensional intersection turn path spline curves using probe data, specifically employing cubic Bézier curves and B-splines to filter and smooth the data, creating a representative turn path that can guide vehicles through intersections effectively.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional digital maps are used to represent intersection paths, then the system is simple and easy to implement, but the navigation accuracy and path representation precision deteriorate
Solution Approach 1:
The system pre-generates spline curves (Bézier and B-splines) representing turn paths through intersections before actual navigation occurs. These pre-computed paths are stored and retrieved during autonomous vehicle operation, eliminating the need for complex real-time path calculation while providing accurate path representation.
Solution Approach 2:
The patent introduces spline curves as an intermediary mathematical representation between the discrete probe data points and the continuous turn path. These splines act as a mediator that smooths and connects raw probe data into realistic vehicle trajectories, improving path accuracy without requiring direct complex data processing during navigation.
2Measurement precision
If detailed probe data from multiple probes is collected to improve path accuracy, then the measurement precision improves, but the data processing complexity and computational load increase
Solution Approach 1:
The system extracts only the essential geometric features from probe data by fitting spline curves that capture the dominant turn path characteristics. Rather than processing all raw probe data points, the extraction process isolates the key trajectory information needed for accurate navigation, reducing computational complexity while maintaining precision.
Solution Approach 2:
The patent transforms raw probe data into a different parameter space using spline curve representations. By changing from discrete coordinate points to continuous mathematical functions with controlled parameters (control points, knots), the system achieves smooth path representation with reduced data complexity and easier processing.
3Ease of operation
If smooth spline curves are generated to represent realistic vehicle trajectories, then the navigation guidance quality improves, but the computational effort and processing time increase
Solution Approach 1:
Spline curves representing turn paths are pre-generated and stored in the digital map database before autonomous vehicle operation. When the vehicle approaches an intersection, the system simply retrieves the pre-computed spline path corresponding to the desired turn, eliminating time-consuming real-time spline fitting while maintaining smooth trajectory guidance.
Solution Approach 2:
The system creates simplified copies of turn paths using standardized spline representations that can be rapidly retrieved and applied. Instead of generating unique complex trajectories for each situation, the system uses copied spline models that approximate realistic vehicle behavior, reducing computational effort while preserving navigation guidance quality.
Data Source
AI summary
A method is provided for gathering probe data and using the gathered data to create map intersection turn paths. Methods may include: receiving probe data from a plurality of probes passing through an intersection; identifying a turn maneuver of the intersection including an incoming road link and an outgoing road link; generating a curve of the turn maneuver based on a geometry of the intersection and a scaling factor; filtering the probe data based on the generated curve to obtain remaining probe data; fitting a spline to the remaining probe data to generate a turn path representative of the turn maneuver; and providing for guidance of a vehicle based on the spline of the turn path. Embodiments including apparatuses and computer program products for creating map intersection turn paths are also provided.