Eye-Tracking Drift Correction in Head-Mounted Displays
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Head-mounted displays (HMDs) face challenges in accurately processing motion data due to sensor limitations, leading to drift errors and inaccuracies when in motion, which are not effectively corrected by existing algorithms, especially in scenarios like traveling by train where lateral shifts and rotations occur.
Innovation Solution
The integration of an eye-tracking system that determines a gaze axis and adjusts virtual images to move a target object towards the central axis of the display, using infrared light and cameras to collect data and override motion sensor errors, thereby reducing drift and unintentional movement issues.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If motion sensors are used to track HMD movement, then the system can provide dynamic image adjustment, but drift errors and inaccuracies occur during motion
Solution Approach 1:
The patent introduces an eye-tracking system as an intermediary to override motion sensor errors. The eye tracker independently determines gaze axis and target object position, serving as a mediator that corrects drift errors in motion sensor data without requiring complex sensor fusion algorithms
Solution Approach 2:
The system continuously monitors eye gaze position and provides feedback to adjust the displayed images. By comparing the gaze axis with the central axis and adjusting the tracking rate based on the difference, the system creates a closed-loop feedback mechanism that eliminates drift errors in real-time
2Reliability
If existing correction algorithms are applied to sensor data, then some drift errors can be reduced, but inaccuracies persist especially during lateral shifts and rotations
Solution Approach 1:
The patent replaces the mechanical sensor-based motion tracking system with an optical eye-tracking system. Instead of relying on accelerometers and gyroscopes that suffer from drift, the system uses infrared illumination and camera-based eye tracking to directly measure gaze direction, fundamentally substituting the measurement mechanism
3Stability of the object's composition
If virtual images are anchored to the central axis, then stability is improved, but the system cannot follow user gaze movements
Solution Approach 1:
The system dynamically adjusts the tracking rate based on the difference between the gaze axis and central axis. When the gaze deviates from center, the target object moves toward the gaze position at an adjusted speed, creating a dynamic balance between stability and responsiveness to user attention
Solution Approach 2:
The patent changes the tracking rate parameter based on the angular difference between gaze axis and central axis. By dynamically modifying this parameter, the system achieves both stability (when gaze is near center) and adaptability (when gaze moves to peripheral regions)
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
This approach significantly reduces errors caused by imperfect sensor data and unintended HMD movement, providing a more accurate and stable augmented reality experience by anchoring virtual images to the user's gaze, even during motion.
Implementation Method 1
an infrared light source configured to illuminate the viewing location with infrared light such that infrared light is reflected from the viewing location as reflected infrared light
Implementation Method 2
infrared light is reflected from the viewing location as reflected infrared light. The HMD further includes a camera configured to image the viewing location by collecting the reflected infrared light
Data Source
Figure 1
Figure 2
Figure 3A
AI summary
A head-mounted display (HMD) may include an eye-tracking system, an HMD- tracking system and a display configured to display virtual images. The virtual images may present an augmented reality to a wearer of the HMD and the virtual images may adjust dynamically based on HMD-tracking data. However, position and orientation sensor errors may introduce drift into the displayed virtual images. By incorporating eye-tracking data, the drift of virtual images may be reduced. In one embodiment, the eye-tracking data could be used to determine a gaze axis and a target object in the displayed virtual images. The HMD may then move the target object towards a central axis. The HMD may also record data based on the gaze axis, central axis and target object to determine a user interface preference. The user interface preference could be used to adjust similar interactions with the HMD.