Indoor Localization via Visual Feature Vectors
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing GPS technologies are ineffective for indoor localization and wayfinding due to poor performance indoors, and RF-based solutions face challenges such as infrastructure costs and precision issues, limiting their widespread adoption in environments like supermarkets and airports.
Innovation Solution
A system utilizing pre-processed floorplans stored on portable devices, which use visibility relationships and intervisibility models to localize users by encoding and decoding feature vectors, allowing users to identify their location and navigate through indoor environments without requiring absolute precision or additional infrastructure.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If GPS technology is used for localization, then outdoor positioning accuracy is improved, but indoor localization performance deteriorates
Solution Approach 1:
The patent introduces visual features (landmarks, signs, objects) as intermediary elements that mediate between the user and the indoor environment for localization. Instead of relying on GPS satellites that cannot penetrate buildings, the system uses visual cues within the indoor space that can be captured by the device camera and processed to determine position, effectively replacing the GPS satellite intermediary with local visual intermediaries.
Solution Approach 2:
The patent replaces the radio-frequency-based GPS mechanical system with a visual-based system using the device camera and image processing. This substitution allows the system to function indoors by leveraging the camera's ability to capture visual information through building structures, where RF signals fail, transforming the localization mechanism from electromagnetic wave triangulation to visual feature recognition.
2Reliability
If RF-based triangulation is deployed for indoor localization, then indoor positioning capability is improved, but infrastructure cost and complexity increase
Solution Approach 1:
The patent enables the mobile device to perform localization independently using its own camera and processing capabilities. The device captures images, extracts visual features, and determines position without requiring external RF beacons or infrastructure. This self-service approach transfers the localization function from the environment (RF beacons) to the device itself, eliminating complex infrastructure requirements.
Solution Approach 2:
The patent leverages the universal camera component already present in most mobile devices for its primary photography function, repurposing it for localization. This multi-functionality approach allows the same hardware (camera) to serve both user documentation purposes and positioning requirements, eliminating the need for dedicated localization infrastructure.
3Device complexity
If RF fingerprinting is used for indoor localization, then infrastructure requirements are reduced, but manual effort for map creation increases
Solution Approach 1:
The patent performs preliminary action by pre-processing the floorplan image to extract visual features, create feature vectors, and build the localization database before actual use. This preprocessing step, which includes identifying landmarks, signs, and objects and storing their spatial relationships, is done once and then reused for multiple localization queries, significantly reducing the time required for actual positioning operations.
Solution Approach 2:
The patent creates a visual copy of the indoor environment by capturing a floorplan image and extracting feature representations from it. This copied visual model serves as the localization database, replacing the need for manual RF signature mapping. The system copies visual information from the floorplan and uses it for positioning, eliminating time-consuming manual field surveys.
4Measurement precision
If high precision localization is achieved indoors, then navigation accuracy is improved, but system complexity and computational requirements increase
Solution Approach 1:
The patent segments the indoor environment into discrete visual features (landmarks, signs, objects) and represents the space as a graph of interconnected features. This segmentation allows the system to process localization by comparing captured images against individual feature templates rather than analyzing the entire environment at once, reducing computational complexity while maintaining precision.
Solution Approach 2:
The patent uses partial action by selecting and processing only the most distinctive and informative visual features from the environment for localization. Rather than analyzing all visual information equally, the system identifies key landmarks and signs that provide sufficient positioning accuracy with minimal processing, avoiding excessive computational requirements while achieving high precision.
Data Source
AI summary
A user proceeds from one location to another location inside of a building by traveling in a sequence of several hops in response to different visual cues. A portable handheld device may provide the visual cues to the user. The user reaches the destination through the sequence of hops using the portable handheld device.