Lane-Level Slowdown Detection Using Vehicle Trajectory Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Navigation and mapping service providers face challenges in detecting lane-level dangerous slowdown events with confidence and low latency, as traditional methods often rely on outdated or inaccurate data, and manual monitoring is resource-intensive and prone to errors.
Innovation Solution
A system that collects and processes probe and sensor data from vehicles to detect lane-level slowdown events by splitting trajectories, classifying event types based on speed reductions and locations, and delivering messages with confidence values and severity factors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional static incident reporting methods are used, then service providers can report real-time incidents on road segments, but the data quickly becomes outdated and inaccurate for lane-level events
Solution Approach 1:
The patent segments road data from segment-level to lane-level granularity. Instead of treating entire road segments as single units, the system divides them into individual lanes and detects slowdown events at the lane level using probe data from multiple vehicles. This segmentation enables more precise and timely detection of localized traffic events.
Solution Approach 2:
The system implements continuous feedback loops where probe data from vehicles is constantly collected, processed, and used to update traffic conditions in real-time. The slowdown detection system monitors speed variations, provides feedback about detected events, and continuously refines its detection algorithms based on incoming data streams, ensuring data remains current and accurate.
2Measurement precision
If lane-level data collection resources are increased, then detection accuracy improves, but the technical challenges and resource requirements significantly increase
Solution Approach 1:
The patent leverages multi-functional probe vehicles that already perform various navigation and mapping functions. These vehicles serve multiple purposes: they collect segment-level probe data for general traffic monitoring, gather lane-level data for detailed slowdown detection, and provide positioning information for map updates. This universal approach avoids dedicated complex lane-level collection systems while achieving precise detection.
Solution Approach 2:
The system uses the vehicles' own onboard sensors and navigation systems to collect the data needed for slowdown detection. Rather than requiring external dedicated infrastructure, the probe vehicles self-generate the necessary lane-level data through their existing GPS, speed sensors, and navigation equipment, reducing the need for additional complex collection devices.
3Measurement precision
If location sensor accuracy is improved, then lane-level positioning precision increases, but the cost and complexity of sensor systems increase
Solution Approach 1:
The patent combines multiple positioning data sources including GPS coordinates, map-matching algorithms, and trajectory analysis to achieve lane-level positioning accuracy. Instead of relying on a single high-precision sensor, the system merges data from standard GPS receivers with processed trajectory information and digital map data, achieving enhanced positioning precision through data fusion rather than expensive specialized sensors.
Data Source
Figure 1
Figure 2
Figure 3A
AI summary
An approach is provided for detecting lane-level dangerous slowdown events based on probe data and/or sensor data. The approach, for example, involves splitting probe data, sensor data, or a combination thereof into at least one vehicle trajectory, wherein the data is collected from one or more vehicles traveling on a road segment. For each vehicle trajectory of the at least one vehicle trajectory, the approach also involves processing said each vehicle trajectory to detect a slowdown event based on a speed reduction greater than a threshold reduction. The approach further involves classifying a slowdown event type of the slowdown event based on a final driving location, a final driving speed, or a combination thereof of the at least one vehicle trajectory. The approach further involves providing the slowdown event, the slowdown event type, or a combination thereof as an output for the road segment.