Augmented Reality Point-of-View Synchronization via Controller Cradle
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing augmented reality systems face challenges in accurately synchronizing the point of view of a 3D virtual model with respect to a physical model, particularly in ultrasound visualization training, due to misalignment errors and computational overhead from methods like visual hand tracking and ARuCO markers.
Innovation Solution
A system that determines rotational and positional offsets between a headset and a physical model by combining sensed variable offsets between the headset and a controller with fixed offsets between the controller and the physical model, using a cradle to hold the controller at known positions, without requiring visual hand tracking or ARuCO markers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If visual hand tracking is used for alignment, then hardware requirements are reduced, but alignment precision deteriorates due to small angular errors (2-3°) resulting in large location errors
Solution Approach 1:
The patent introduces a controller as an intermediary device that serves as a stable reference point between the headset and the physical model. The controller is held at fixed and known rotational and positional offsets from the physical model by a cradle, providing a reliable mediation point that eliminates the precision errors inherent in direct visual hand tracking methods
Solution Approach 2:
The patent replaces the optical-based visual hand tracking system with a mechanical positioning system using a cradle to physically hold the controller at fixed offsets. This mechanical substitution provides stable, known geometric relationships that are far more precise than visual tracking, while the computational overhead is reduced by using simple rigid-body transformation mathematics rather than complex visual processing
2Extent of automation
If ARuCO marker techniques are used for alignment, then automated calculation of pose offsets is achieved, but computational overhead increases and additional cameras are required
Solution Approach 1:
The patent extracts and removes the complex ARuCO marker detection and computational overhead by replacing it with a simpler mechanical reference system. The controller held in a cradle provides the necessary automated reference without requiring marker detection algorithms, additional cameras, or significant computational resources for pose estimation
Solution Approach 2:
Instead of using visual markers that require complex detection, the patent creates a physical copy/reference system where the controller in the cradle physically embodies the reference pose. This physical reference copy provides the same automated calculation capability but through simple geometric relationships rather than visual pattern recognition
3Measurement precision
If Thompson's positioning dock is used for calibration, then sensor calibration is achieved, but it requires a known location of the positioning dock which is not available in this application
Solution Approach 1:
The patent creates a universal calibration system that works regardless of the physical model's location or orientation. By using the controller as an intermediary with fixed offsets from the physical model, the system becomes adaptable to any physical model position, making the calibration process universally applicable rather than location-dependent
Solution Approach 2:
Instead of trying to determine the physical model's position relative to a known reference (as in Thompson's approach), the patent inverts the problem by using the controller as a known reference that is physically connected to the physical model. The fixed offsets from controller to physical model are used to derive the model's position, reversing the calibration logic to work without knowing the model's initial location
Data Source
AI summary
An augmented reality point of view synchronisation system has a VR headset having an augmented display and a controller operably interfacing the headset. The system is configured to sense variable relative rotational and positional offsets of the controller with respect to the headset. The system also has a cradle configured to position the controller at fixed relative rotational and positional offsets with respect to a physical model. The system is configured to augment a view of the physical model with a 3D virtual model using the augmented display wherein the point of view of the 3D virtual model is determined according to the variable and fixed relative rotational and positional offsets.