Lane Map Estimation Using Vehicle Clustering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Vehicle sensors often fail to provide adequate data for identifying lane markings due to sensor faults or environmental conditions, which can impact driving safety, especially in semi-autonomous or autonomous vehicles.
Innovation Solution
A system that uses a computer in a host vehicle to receive data from various sensors and other vehicles, define vehicle clusters, identify lane boundaries, and generate a lane map using curve fitting techniques, enabling accurate lane detection even in adverse conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If vehicle sensors are used to detect lane markings, then driving safety can be improved, but sensor faults and environmental conditions cause inadequate data for identifying lane markings
Solution Approach 1:
The patent uses other vehicles as intermediary sources to provide lane marking data when the host vehicle's sensors fail. Data from surrounding vehicles acts as a mediator to compensate for sensor deficiencies, allowing the system to maintain reliable lane detection even when direct sensing is inadequate due to faults or environmental conditions.
Solution Approach 2:
The system merges data from multiple sources including the host vehicle's sensors and sensors from other vehicles to create a comprehensive lane marking detection system. By combining these data streams, the system overcomes the limitations of individual sensors and maintains reliable lane identification under various conditions.
2Measurement precision
If data from other vehicles is used to identify lane markings, then lane detection accuracy is improved in adverse conditions, but system complexity increases
Solution Approach 1:
The system implements a universal data processing framework that can handle multiple data sources (host vehicle sensors, other vehicle sensors) through a common architecture. This multi-functional approach allows the same system to process diverse input types without requiring separate specialized subsystems, thereby managing complexity while maintaining detection accuracy.
Solution Approach 2:
The system uses feedback mechanisms to validate and fuse data from multiple vehicles. By continuously comparing and reconciling data from different sources, the system can identify reliable lane marking information while filtering out inconsistent data, improving accuracy without proportionally increasing complexity through structured data validation.
Data Source
AI summary
A computer can receive, from a vehicle sensor, data about a plurality of second vehicles, define two or more vehicle clusters based on location data of second vehicles, each cluster including two or more of the second vehicles determined to be traveling in a same lane, identify two or more lane boundaries according to clusters, and use lane boundaries to generate a lane map.