Lane Departure Detection via Probe Data Integration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current lane keeping systems in vehicles rely solely on onboard cameras and do not effectively share data to detect lane departure events across multiple vehicles, limiting their ability to provide proactive warnings and improve traffic safety in autonomous driving scenarios.
Innovation Solution
A method and system that utilizes map data and probe data from various sources, including vehicle sensors and user devices, to detect lane departure events, categorize them as intentional or unintentional, and generate proactive warning messages for vehicles, leveraging interconnected transport infrastructure for enhanced safety.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If lane keeping systems use only onboard vehicle cameras to detect lane departure, then the system is simple and independent, but it cannot share data across multiple vehicles and provide proactive warnings
Solution Approach 1:
The patent combines individual vehicle probe data with infrastructure-based sensor data to create a collaborative lane departure detection system. Multiple data sources (vehicle cameras, road-side sensors, probe data from multiple vehicles) are merged to improve detection accuracy and enable proactive warnings to multiple vehicles simultaneously
Solution Approach 2:
The system creates universal lane departure detection capabilities that serve multiple vehicles and infrastructure components. The probe data collection and processing system is designed to be multi-functional, supporting both individual vehicle safety and fleet-wide traffic management
2Loss of time
If the system collects and processes probe data from multiple vehicles and infrastructure sensors, then proactive lane departure warnings can be provided, but the data processing complexity and computational requirements increase
Solution Approach 1:
The system performs preliminary processing of probe data by categorizing lane departure events as intentional or unintentional in advance. This pre-categorization allows for faster real-time decision-making and warning generation when actual lane departure events occur
Solution Approach 2:
The patent introduces an intermediary processing layer that aggregates and analyzes probe data from multiple vehicles and infrastructure sensors before generating warnings. This intermediary system manages the complexity of multi-source data integration while enabling rapid response times
3Measurement precision
If the system categorizes lane departure events as intentional or unintentional, then warning accuracy improves, but the detection and measurement complexity increases
Solution Approach 1:
The system uses parameter changes in probe data (such as speed variations, steering angle changes, temporal patterns) to distinguish between intentional and unintentional lane departures. By monitoring multiple parameters simultaneously, the system achieves accurate classification without requiring overly complex detection mechanisms
Data Source
AI summary
An approach is provided for detecting lane departure events based on map data and probe data. The approach, for example, involves map-matching probe data to a lane of a road segment. The probe data is collected from one or more sensors of at least one vehicle and/or at least one user device that traversed the road segment. The approach also involves processing the probe data to detect at least one lane departure event. The approach further involves categorizing the at least one lane departure event as an intentional lane departure event or an unintentional lane departure event. The approach further involves creating a lane departure warning message for the road segment based on the at least one categorized lane departure event, and/or road segments associated with multiple lane departure warning messages within a certain time period. The approach further involves providing the lane departure warning message as an output.