GPS Correction Using Image Recognition for Lane Positioning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional GPS devices lack the precision needed for accurate lane recognition in autonomous navigation, especially in complex road scenarios, and high-precision DGPS devices are expensive, limiting their widespread adoption in vehicles.
Innovation Solution
A GPS correction system utilizing image recognition information, which includes a GPS module, an image recognition device, a road map storage unit, and an information processing device that matches line recognition information with road map data to correct positional data, enabling precise lane recognition and longitudinal position estimation without requiring high-precision GPS devices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a DGPS device is used to achieve high-precision position information (positional deviation of about 1m), then measurement precision is improved, but device cost increases significantly
Solution Approach 1:
The patent introduces image recognition information as an intermediary to bridge the gap between low-precision GPS and high-precision navigation requirements. The system uses camera-based lane recognition to provide additional spatial context that corrects GPS positional errors, achieving DGPS-level accuracy without the high cost of hardware corrections
Solution Approach 2:
The patent replaces the mechanical/GPS-signal-based positioning system with an optical/image-based recognition system for precision correction. By substituting camera-based lane detection for GPS signal processing, the system achieves higher precision at lower cost
2Ease of manufacture
If a conventional GPS device is used, then device cost is reduced, but measurement precision deteriorates (positional deviation of 20 to 30m)
Solution Approach 1:
The patent merges GPS positioning data with image recognition data to create a hybrid navigation system. By combining the broad coverage of GPS with the high precision of visual lane recognition, the system achieves accurate lane-level positioning using inexpensive components
Solution Approach 2:
The patent creates a composite information system that fuses GPS coordinates with image-based lane detection data. This composite approach combines the strengths of both data sources to achieve precision unattainable by either system alone
3Reliability
If GPS signal is blocked (e.g., in tunnels), then position information availability is reduced, but navigation reliability is required to maintain continuous tracking
Solution Approach 1:
The patent prepares for GPS signal loss by maintaining image recognition capabilities that can independently determine position. The system continuously processes visual data so that when GPS signals are blocked, the navigation can seamlessly transition to image-based positioning without interruption
Data Source
AI summary
Provided is a GPS correction system and method which corrects position information obtained from a low-precision GPS device using image recognition information. The GPS correction system using image recognition system includes: a GPS module; an image recognition device having a line recognition function; a road map storage unit configured to store road map information including line characteristic information or a road map receiving unit configured to receive the road map information; and an information processing device configured to compare the line recognition information acquired through the image recognition device to the line characteristic information, correct a current position measured by the GPS module, and calculate traveling lane information.