Boundary Crossing Detection via Vector Projection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for detecting boundary crossings on digital road maps in GNSS toll systems are computationally intensive and unreliable, especially when vehicles turn around near boundaries, leading to inaccurate toll calculations.
Innovation Solution
A method that determines the intersection point of a boundary with a road segment and projects position vectors onto a reference direction to detect changes in direction or sign, validated by checking projection values and normal distances, suitable for real-time implementations on OBUs and central servers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If elaborate calculation methods are used for boundary detection, then detection reliability may improve, but computational complexity increases making real-time implementation difficult
Solution Approach 1:
The patent extracts only the essential geometric relationship between the boundary and road segment by determining the intersection point and projecting position vectors onto the reference direction. This extraction eliminates unnecessary complex calculations while retaining the core detection functionality, enabling real-time implementation on OBUs with limited computational power.
Solution Approach 2:
The boundary detection process is segmented into distinct simple steps: determining the intersection point of boundary with road segment, calculating position vectors from intersection point to vehicle position, projecting these vectors onto the reference direction, and detecting sign changes in the projection sequence. This segmentation allows each step to be computed independently and efficiently.
2Reliability
If continuous distance determination from boundary point is used, then boundary crossing can be detected, but position determination inaccuracies lead to unreliable detection when vehicles turn around near boundaries
Solution Approach 1:
Instead of measuring the distance from the boundary point to the vehicle (which is sensitive to position inaccuracies), the patent inverts the approach by determining the intersection point of the boundary with the road segment and calculating vectors from this fixed reference point to the vehicle position. This inversion makes the detection less sensitive to small position errors near the boundary.
Solution Approach 2:
The patent introduces the intersection point as an intermediary reference that lies on both the boundary and the road segment. This intermediary provides a stable, well-defined reference frame that is less affected by position determination inaccuracies compared to using the boundary point directly, enabling reliable detection even when vehicles turn around near boundaries.
Data Source
AI summary
Method for detecting a boundary crossing of an object, the movement of which is represented by a sequence of positions on a digital road map ith vectorized road segments and at least one boundary, the method comprising: determining an intersection of the boundary with a road segment and a reference direction in the orientation of this road segment, successively projecting vectors, each determined between one of the positions and the intersection point, onto the reference direction, in order to obtain a sequence of projections, and detecting a boundary crossing if a change of direction or sign occurs in the sequence of projections.