Lane Marking Localization Fusion for GPS Drift Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current GPS technology is inaccurate for determining a vehicle's lateral position within a roadway, experiencing significant drift due to signal strength and hardware precision issues, which is unacceptable for autonomous driving systems.
Innovation Solution
A system that compares environmental information from imaging devices against regional data stored in a database to improve accuracy, using multiple cameras to generate response maps and fuse location data, allowing for precise determination of a vehicle's position within a lane.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If GPS technology is used to determine vehicle lateral position, then the system can provide location information, but the positioning accuracy is poor with drift of 10 meters or more
Solution Approach 1:
The patent combines GPS positioning with visual recognition of lane markers and road features. The system merges data from GPS receivers, imaging devices (cameras), and processors to create a fused location estimate that compensates for GPS drift by using visual landmarks as reference points for lateral positioning within lanes.
Solution Approach 2:
The patent introduces visual lane markers and road features as intermediary reference points between the GPS system and the vehicle positioning. These visual elements serve as mediators that provide sub-meter accuracy lateral position information, correcting the coarse GPS lateral position data.
2Measurement precision
If inertial measurement units are added to improve positioning accuracy, then short-term precision improves, but drift accumulates over time making it insufficient for self-driving
Solution Approach 1:
The patent implements a feedback mechanism where the visual recognition system continuously monitors lane marker positions and road features, comparing them against expected positions to detect and correct drift in the inertial measurement unit data. This closed-loop feedback prevents long-term drift accumulation.
Solution Approach 2:
The system performs preliminary visual recognition and lane marker detection to establish reference positions before relying on inertial measurement for continuous tracking. This preliminary visual calibration prevents drift accumulation by periodically resetting the inertial navigation reference frame.
3Measurement precision
If multiple imaging devices are used to enhance positioning accuracy, then lateral position determination improves, but system complexity increases
Solution Approach 1:
The patent makes the imaging devices serve multiple functions: they detect lane markers for lateral positioning, identify road features for location verification, and provide visual feedback for driver awareness. This multi-functionality reduces the need for separate specialized sensors, managing system complexity while maintaining high positioning accuracy.
Data Source
AI summary
Various embodiments of the present disclosure provide a system and method for iterative lane marking localization that may be utilized by autonomous or semi-autonomous vehicles traveling within the lane. In an embodiment, the system comprises a locating device adapted to determine the vehicle's geographic location; a database; a region map; a response map; a plurality of cameras; and a computer connected to the locating device, database, and cameras, wherein the computer is adapted to receive the region map, wherein the region map corresponds to a specified geographic location; generate the response map by receiving information from the camera, the information relating to the environment in which the vehicle is located; identifying lane markers observed by the camera; and plotting identified lane markers on the response map; compare the response map to the region map; and iteratively generate a predicted vehicle location based on the comparison of the response map and the region map.