Autonomous Valet Parking Map Using Sign-Based Vehicle Positioning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing autonomous valet parking systems face challenges in precisely determining the position of a vehicle within a parking lot, especially in environments where GPS signals are unreliable or unavailable, such as multi-level parking structures.
Innovation Solution
A method and system that utilizes existing signs within a parking lot as markers, employing image recognition to extract characteristics and incorporate them into a digital map, enabling precise vehicle positioning and route guidance for autonomous parking.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If GPS signal is used for vehicle positioning in the parking lot, then positioning can be performed using existing infrastructure, but GPS signal cannot reach in the parking lot and positioning precision deteriorates
Solution Approach 1:
The patent introduces position markers as an intermediary element between the vehicle positioning system and the parking lot environment. These markers serve as mediators that reflect light patterns to enable precise positioning without relying on GPS signals that cannot penetrate the parking lot structure. The markers act as a bridge between the external satellite navigation system and the internal parking lot space.
Solution Approach 2:
The patent replaces the GPS satellite-based electromagnetic positioning system with an optical pattern recognition system using position markers. Instead of relying on satellite signals that cannot reach the parking lot, the system uses optical patterns reflected from markers on the ground, substituting a mechanical/optical measurement approach for the electromagnetic wave-based GPS system.
2Measurement precision
If position markers are installed in the parking lot to improve positioning precision, then vehicle position can be accurately identified, but device complexity and infrastructure requirements increase
Solution Approach 1:
The position markers serve multiple functions: they act as positioning references for autonomous vehicles, provide visual guidance for drivers, and can be integrated with existing parking lot signage and ground markings. This multi-functionality reduces the need for separate dedicated positioning infrastructure, thereby limiting the increase in device complexity.
Solution Approach 2:
The patent uses optical pattern parameters (color, shape, arrangement) of position markers to encode spatial information. By changing these optical parameters rather than installing complex electronic devices, the system achieves precise positioning with simpler infrastructure. The markers' visual characteristics serve as the primary data carriers for the positioning system.
3Measurement precision
If multiple position markers are deployed throughout the parking lot to ensure comprehensive coverage, then positioning accuracy is maintained across all areas, but manufacturing and installation costs increase
Solution Approach 1:
The patent divides the parking lot into multiple zones, each with its own set of position markers. This segmentation allows the system to achieve comprehensive coverage by strategically placing markers in key locations within each zone rather than uniformly distributing them throughout the entire area. The segmentation approach optimizes the number of markers needed while maintaining positioning accuracy across all parking lot areas.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables accurate and error-free autonomous valet parking by precisely identifying the vehicle's position within the parking lot, using existing signs to enhance positioning accuracy without the need for additional infrastructure.
Implementation Method 1
detecting, via automatic object recognition of the image, a sign in the image
Data Source
AI summary
Disclosed is a method for generating a parking lot map, including: obtaining image information and positioning information indicating a position at which the image is captured; detecting a sign from the image information; extracting characteristics for distinguishing the sign from another sign; and incorporating information correlating a position of the sign obtained based on the positioning information in a digital map of a parking lot with characteristics of the sign into the digital map of the parking lot.