XR Pose Estimation Using Relative Elevation to Correct Drift
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
XR systems face challenges in accurately tracking the pose of head-mounted displays (HMDs) and hand-held controls due to pose estimation drift and environmental conditions affecting computer vision-based systems, leading to user experience degradation and motion sickness.
Innovation Solution
Incorporating altimeters into the HMD, hand-held control, and compute pack to determine relative elevations, which are used to correct pose data drift and maintain accurate tracking, even in adverse lighting conditions, using atmospheric pressure data to calibrate and correct IMU data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If computer vision-based systems are used to track HMD and hand-held control pose, then visual feedback and spatial awareness are improved, but pose estimation drift occurs under adverse lighting conditions
Solution Approach 1:
The patent combines computer vision-based ground truth sensors with altimeter sensors to create a hybrid pose estimation system. The vision system provides accurate spatial awareness when conditions permit, while the altimeter provides continuous vertical position data that corrects drift. These complementary systems are merged through sensor fusion to achieve both visual feedback and tracking stability.
Solution Approach 2:
The altimeter acts as an intermediary that mediates between the vision system and the IMU. When the vision system loses tracking under adverse lighting, the altimeter provides continuous vertical position data that serves as a bridge, preventing complete pose estimation failure and reducing drift during vision system interruptions.
2Speed
If IMU data is used for continuous pose tracking, then tracking frequency and responsiveness are improved, but pose estimation drift accumulates over time
Solution Approach 1:
The patent implements feedback by using the altimeter's continuous vertical position measurements to correct IMU-derived pose estimates. The altimeter data feeds back into the pose estimation algorithm, continuously adjusting for drift accumulation while preserving the high-frequency IMU tracking responsiveness. This feedback loop maintains both speed and precision.
Solution Approach 2:
The system performs preliminary correction by using ground truth vision data when available to establish accurate baseline pose information before drift accumulates. This preliminary accurate measurement serves as a reference that the altimeter later uses to correct gradual IMU drift, preventing error accumulation before it significantly degrades precision.
3Measurement precision
If ground truth sensor system is used to correct IMU pose data, then pose estimation accuracy is improved, but system complexity increases
Solution Approach 1:
The altimeter serves multiple functions: it provides continuous vertical position tracking, corrects IMU drift in the vertical dimension, and supplements vision system data when lighting conditions are adverse. This multi-functionality reduces the need for additional dedicated sensors, thereby limiting the increase in system complexity while maintaining improved pose estimation accuracy.
Solution Approach 2:
The patent changes the operational parameters of existing sensors by utilizing the altimeter's vertical position data specifically for correcting the vertical component of pose estimation. Rather than requiring a complete complex sensor system, the invention selectively applies parameter changes to the vertical axis where IMU drift is most problematic, achieving accuracy improvement with minimal added complexity.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enhances pose estimation accuracy and stability, reducing latency and drift, thereby improving user experience and preventing motion sickness by maintaining stable virtual object placement relative to the real world.
Implementation Method 1
a first altimeter carried by the head-mounted display is configured for outputting a first atmospheric pressure data indicative of an absolute elevation of the head-mounted display; a second altimeter carried by the hand-held control is configured for outputting a second atmospheric pressure data indicative of an absolute elevation of the hand-held control
Implementation Method 2
determine a relative elevation between the head-mounted display and the hand-held control based on a difference between the first atmospheric pressure data and the second atmospheric pressure data
Data Source
AI summary
An extended reality (XR) system, comprises a head-mounted display (HMD) configured for displaying virtual content to a user, a first altimeter carried by the HMD, a hand-held control, and a second altimeter carried by the hand-held control. The first altimeter configured for outputting first atmospheric pressure data indicative of an elevation of the HMD, while the second altimeter is configured for outputting second atmospheric pressure data indicative of an elevation of the hand-held control. The XR system further comprises at least one processor configured for determining a relative elevation between the first altimeter and the second altimeter based on the first atmospheric pressure data and the second atmospheric pressure data.