Pose-Driven Body Navigation Using Wearable SLAM in GPS-Denied Spaces
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional full-body bio-mechanical measurement systems struggle with accuracy and compatibility in GPS-denied environments, particularly during complex movements, and existing IMU-based solutions lack validation and fail to capture useful degrees of freedom for navigation.
Innovation Solution
A method and garment using distributed sensors and SLAM technology to determine body location and navigate without GPS or RF signals, incorporating strain sensors, cameras, and compasses to characterize body pose and movement, enabling simultaneous localization and mapping.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If GPS or RF signals are used for location determination, then position accuracy is improved, but the system fails in GPS-denied environments such as indoor or unfamiliar locations
Solution Approach 1:
The system segments the location determination task into multiple independent measurement components: pose estimation from wearable sensors, visual feature detection from cameras, and odometry calculations. Each segment operates independently and can function without GPS, with results integrated to provide robust location determination in diverse environments including GPS-denied areas.
Solution Approach 2:
The wearable measurement system serves multiple functions: it provides pose estimation for navigation, visual information for SLAM, and movement data for odometry. This multi-functional approach allows the same hardware to operate effectively across different environments (indoor, outdoor, GPS-available, GPS-denied) without requiring environment-specific configurations.
2Device complexity
If IMU-based solutions with Kalman filters are used, then location estimation is simplified, but useful degrees of freedom for navigation are filtered out
Solution Approach 1:
Instead of using aggressive filtering that removes all uncertain data, the system applies partial filtering that retains excessive information including minor degrees of freedom. This approach keeps potentially useful navigation data (such as small body movements or subtle orientation changes) that traditional Kalman filters would discard, allowing these partial signals to contribute to overall location estimation accuracy.
3Measurement precision
If treadmills and gait labs are used for validation, then measurement accuracy is improved, but real-world movement scenarios are not representative
Solution Approach 1:
The system transitions from static validation environments (treadmills, gait labs with fixed protocols) to dynamic real-world scenarios. The wearable sensors and cameras capture data during natural, unconstrained movements in varied environments, allowing the system to be validated under conditions that truly represent how it will be used in practice, while maintaining measurement accuracy through robust sensor fusion algorithms.
Data Source
AI summary
Method and body garment for pose-driven determination of a location of a body include: selecting a predetermined location of the body; determining a mode of locomotion of the body as the body moves; inferring the movement of the body and direction of the movement over a time period by combining physiological information of the body with calibration data; calculating a path travelled by the body from the predetermined location; capturing spatial environment of the body and its features, and a current location of the body within the spatial environment, by performing simultaneous localization and mapping (SLAM) on the path travelled by the body; and calculating navigation information of the body without using a location determining device.