AR Navigation via Pose-Anchored Traversability Graph
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current navigation systems, such as GPS, face significant accuracy issues when used indoors, making them inadequate for pedestrian-scale navigation.
Innovation Solution
A computing device equipped with a location sensor system, pose-tracking engine, graphing engine, and navigation engine that creates a pose-anchored traversability graph using measured environmental parameters to provide accurate pedestrian-scale navigation directions, even in environments where GPS is unreliable.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional GPS systems are used for navigation, then outdoor navigation is provided, but position tracking accuracy deteriorates significantly indoors
Solution Approach 1:
The patent introduces an intermediary system consisting of infrastructure sensors (cameras, barometers, magnetometers) and a map of the environment as a mediator between the GPS system and the indoor navigation need. This intermediary enables position tracking indoors by capturing environmental parameters and comparing them against a pre-stored map, thereby resolving the contradiction between maintaining navigation reliability and achieving measurement precision in indoor environments where GPS signals are unavailable
Solution Approach 2:
The patent replaces the mechanical/electromagnetic GPS signal-based system with an optical and environmental parameter-based system. Instead of relying on satellite signals, the system uses cameras to capture visual features, barometers to measure pressure changes for elevation, and magnetometers to detect magnetic field variations, substituting the GPS mechanism with environmental sensing and comparison against a digital map
2Measurement precision
If environment parameters are captured and processed in real-time, then navigation accuracy is improved, but computational complexity increases
Solution Approach 1:
The patent applies preliminary action by pre-capturing and storing environmental parameters (camera images, barometric pressure data, magnetic field measurements) and creating a detailed map of the environment before navigation is needed. This pre-processing reduces real-time computational complexity during actual navigation, as the system only needs to compare current sensor readings against the pre-existing map rather than processing raw environmental data from scratch
Solution Approach 2:
The patent segments the navigation problem into distinct sensor modalities (visual features from cameras, pressure data from barometers, magnetic field data from magnetometers) and processes each independently. This segmentation allows parallel processing of different environmental parameters, reducing overall computational complexity while maintaining high position tracking accuracy through multi-sensor fusion
Data Source
Figure 1~2
Figure 3~5
Figure 6~8
AI summary
A computing device includes a location sensor system including sensor(s) configured to measure one or more parameters of a surrounding environment, a pose-tracking engine configured to determine a current pose of the computing device based on the one or more measured parameters of the surrounding environment, a graphing engine configured to access a traversability graph including a plurality of vertices each having a local coordinate system, a navigation engine configured to identify a nearby vertex of the traversability graph to the current pose, and configured to determine a path of traversable edges between the nearby vertex and a destination vertex of the traversability graph, and a display configured to visually present as an overlay to the environment a navigation visualization corresponding to the path.