Vision Positioning System Using Visual Landmarks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing visual positioning systems face challenges in environments with multiple nearby sources of multipath signals or lack of clear sky views, such as indoors, where GNSS receivers are ineffective, and require infrastructure like cellular towers that may not be available or functional.
Innovation Solution
A vision positioning system that determines the position, orientation, and movement of cameras relative to a coordinate system using visually identifiable unique objects in images, with a database that stores and updates position coordinates for these objects, allowing cameras to operate accurately even in environments with no infrastructure.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If GNSS receivers are used to determine global position, then positioning accuracy is improved in open environments, but the system becomes ineffective in environments with multipath signals or blocked sky views
Solution Approach 1:
The patent introduces visual features (landmarks, objects, or artificial markers) as intermediary elements between the camera and the environment. These features serve as mediators that reflect light back to the camera, enabling position determination through image analysis rather than direct satellite signal reception. This intermediary approach allows the system to function in environments where GNSS signals are blocked or degraded by multipath effects.
Solution Approach 2:
The patent replaces the electromagnetic signal-based GNSS system with an optical image-based vision system. Instead of relying on radio frequency signals from satellites, the system uses visible light reflected from visual features in the environment. This substitution enables positioning in indoor and urban canyon environments where electromagnetic signals are blocked but visual features remain accessible.
2Measurement precision
If cellular triangulation systems are deployed to provide positioning, then position determination is achieved, but infrastructure requirements increase system complexity
Solution Approach 1:
The system enables cameras to determine their own positions autonomously by analyzing images of visual features in the environment. Each camera independently identifies features, matches them against a database, and calculates its position without requiring external infrastructure or coordination with other system components. This self-service capability eliminates the need for cellular towers or other centralized positioning infrastructure.
Solution Approach 2:
The patent creates a digital copy or map of the visual environment by storing images and feature databases. Instead of requiring physical infrastructure like cellular towers, the system uses a digital representation of visual features that can be distributed and accessed by multiple cameras. This copying approach replaces complex physical infrastructure with information-based solutions.
3Adaptability or versatility
If visual positioning systems are implemented without infrastructure, then adaptability to various environments is improved, but measurement precision may deteriorate without reference points
Solution Approach 1:
The system performs preliminary action by pre-populating a database with images and visual features from the environment before actual positioning operations. This pre-processing step creates a reference framework that enables accurate position determination during operation. The database serves as a prepared reference that cameras can query to achieve precise positioning without requiring real-time infrastructure.
Solution Approach 2:
The system dynamically adapts to different environments by allowing the visual feature database to be updated and expanded as cameras encounter new features. The system can learn and adapt to changing environments, adding new visual landmarks to the database as needed. This dynamic capability maintains positioning accuracy across diverse and changing environments without requiring pre-deployed infrastructure for each location.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A system determines an otherwise unknown position and orientation of a camera in a working environment, relative to an associated coordinate system, based on visually identifiable unique objects in images taken by the camera. The system utilizes a database that includes or may be updated to include position coordinates for unique objects of interest. The system identifies a plurality of objects within one or more images taken by the camera at a given location, and enters the database either to determine position coordinates for the respective identified objects or to add position coordinates to the data base for the respective identified objects, or both. The system may also update the database to include newly identified unique objects and objects that are altered between images, to determine position and orientation of the camera in a changing environment. Sensors may be included to add additional information relative to the position and orientation of the camera.