Wearable Tracking System Cross-Device Calibration via Sensor Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing wearable tracking systems for augmented reality (AR), virtual reality (VR), and mixed reality (MR) face challenges in tracking user position when using devices from different manufacturers, often requiring pre-stored information or markers, which complicates setup and reduces user experience.
Innovation Solution
A wearable tracking system comprising a wearable device with a tracking sensor and an external camera, where a processor determines a user's body pose by combining movement data from both sources, allowing marker-free calibration and easy setup regardless of manufacturer compatibility.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If wearable tracking systems use devices from different manufacturers, then device compatibility and user flexibility improve, but tracking accuracy and setup complexity worsen due to lack of pre-stored information and markers
Solution Approach 1:
The patent introduces a calibration object as an intermediary element that enables tracking between devices from different manufacturers. The calibration object contains identifiable features that both the wearable device's tracking sensor and the external camera can detect, serving as a common reference medium that bridges compatibility gaps between different device ecosystems without requiring pre-stored manufacturer-specific information
Solution Approach 2:
The system performs preliminary calibration by capturing images of the calibration object at known positions and orientations before actual tracking begins. This preliminary action establishes reference data and transformation matrices that enable accurate tracking subsequently, eliminating the need for markers during operation and allowing devices from different manufacturers to work together accurately
2Measurement precision
If wearable tracking systems require pre-stored information or markers, then tracking accuracy improves, but setup complexity and ease of operation worsen
Solution Approach 1:
The calibration object is designed to be self-contained with identifiable features that automatically provide calibration information when captured. The system performs self-calibration by processing images of the calibration object through the processor to generate tracking references, eliminating the need for manual configuration or pre-stored information by operators
Solution Approach 2:
The system creates a digital copy or representation of the calibration object's features from captured images and stores this as reference data. This copied information serves as the tracking reference, replacing the need for physical markers during operation and simplifying setup while maintaining accuracy
3Device complexity
If wearable tracking systems use single-source tracking (only wearable sensor or only external camera), then system complexity reduces, but tracking reliability and accuracy worsen
Solution Approach 1:
The patent merges tracking data from two independent sources: the wearable device's tracking sensor and the external camera. The processor combines information from both sources to determine body pose, creating a more reliable tracking system that compensates for limitations of individual sources while maintaining manageable system complexity through integrated processing
Data Source
AI summary
A wearable tracking system is provided. The wearable tracking system includes a wearable device, an external camera, and a processor. The wearable device is adapted to be worn on a body part of a user. The wearable device is configured to detect a body movement of the body part based on a tracking sensor. The external camera is configured to obtain a body image of the body part. The processor is configured to obtain a first moving track of a body moving track of the body part based on the body movement, obtain a second moving track of the body moving track of the body part based on the body image, and determine a body pose of the body part based on the first moving track and the second moving track.