Intersection Traffic Information via Trajectory Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for providing traffic information at intersections, especially for automated vehicle control, are inadequate as they rely on human observation of right-of-way rules or light-signaling systems, which are not effective for automated driving scenarios.
Innovation Solution
A method and apparatus for automatically detecting vehicle trajectories using environment sensors like cameras, radar, and lidar to identify paths and provide traffic information, which can be used for monitoring and controlling traffic flow, particularly by dividing detection into non-overlapping time intervals corresponding to specific traffic light phases or times of day.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If human drivers observe right-of-way rules or light-signaling systems, then traffic control is maintained, but automated vehicle control cannot effectively use this information
Solution Approach 1:
The patent replaces human observation and manual traffic control with an automated sensor-based system. Environment sensors (cameras, radar, lidar) automatically detect vehicle trajectories and generate traffic information without human intervention, enabling automated vehicles to access real-time traffic data while eliminating the information loss barrier between human operators and automated systems.
Solution Approach 2:
The system enables self-service traffic information provision where the sensor system automatically monitors, detects, and provides traffic information without requiring human drivers to manually observe or report traffic conditions. The system serves itself by autonomously generating and transmitting traffic data to automated vehicles.
2Productivity
If traffic information is provided continuously, then real-time control is enabled, but processing load and data volume increase
Solution Approach 1:
The patent segments the continuous traffic monitoring process into discrete time intervals, where trajectories are detected and paths are identified in separate processing stages. This segmentation allows the system to handle continuous data flow by dividing it into manageable chunks, reducing instantaneous processing load while maintaining real-time responsiveness.
Solution Approach 2:
The system performs preliminary trajectory detection during time intervals before the actual path identification is needed. By pre-processing and storing trajectory data, the system prepares information in advance, reducing the computational burden during critical decision-making moments and allowing more complex processing when needed.
3Measurement precision
If multiple time intervals are used for trajectory detection, then detection accuracy improves, but system complexity increases
Solution Approach 1:
The patent divides the detection process into multiple non-overlapping time intervals, with each interval handling specific trajectory detection tasks. This segmentation improves accuracy by allowing focused analysis of vehicle movements during specific periods while managing complexity through structured, modular processing of each time interval independently.
Solution Approach 2:
The system uses periodic time intervals for consistent trajectory detection cycles. By repeating the detection-process-identification cycle in regular intervals, the system maintains high accuracy through multiple sampling points while simplifying management through predictable, rhythmic operation that can be easily synchronized with traffic light phases.
Data Source
AI summary
A method for providing traffic information at an intersection, wherein trajectories of a plurality of vehicles are detected over time intervals, paths are identified on the basis of the trajectories and these paths and/or information derived from them are provided as traffic information.