Dead Reckoning Positional Prediction for AR VR Applications
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In augmented reality (AR) and virtual reality (VR) applications, predicting the position of virtual objects in real-time as the user moves can be challenging, especially when rapid movements occur, leading to artifacts like 'jumping' or 'swimming' of virtual characters due to the lag in visual sensing techniques.
Innovation Solution
The implementation of dead reckoning positional prediction using data from Inertial Measurement Units (IMUs) to anticipate the user's device movement and adjust the virtual camera view accordingly, smoothing the display of virtual objects in relation to the user's physical position and orientation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If visual sensing techniques are used to track user position in AR/VR applications, then the system can capture real-world environment data, but lag occurs during rapid movements causing virtual objects to jump or swim
Solution Approach 1:
The system performs preliminary position prediction using dead reckoning calculations based on IMU data before the actual visual sensing update occurs. By anticipating the user's position change based on acceleration and velocity data, the system prepares the virtual camera view in advance, eliminating the lag that would otherwise cause jumping or swimming artifacts during rapid movements.
2Measurement precision
If the system waits for visual sensing data to determine position, then measurement accuracy is maintained, but real-time responsiveness deteriorates during rapid movement
Solution Approach 1:
The system introduces dead reckoning calculation as an intermediary mechanism between IMU data and visual sensing data. The dead reckoning algorithm processes acceleration and velocity data to predict position changes, providing a real-time responsive estimate that bridges the gap between rapid movement detection and slower visual sensing updates, thereby maintaining both accuracy and responsiveness.
3Speed
If dead reckoning calculation is used to predict position based on IMU data, then real-time position prediction is achieved, but calculation complexity increases
Solution Approach 1:
The system uses the IMU data that is already being collected for other purposes (such as motion detection and orientation tracking) to perform dead reckoning calculations. By reusing existing sensor data and integrating it with the visual sensing pipeline, the system achieves real-time position prediction without requiring additional sensors or excessively complex computational algorithms.
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 effectively reduces or eliminates the artifacts of rapid movement, providing a smoother and more accurate alignment of virtual objects with the physical environment, enhancing the user experience by maintaining a consistent and realistic virtual camera view.
Implementation Method 1
predicting a second change in position of the user device over a second time period, using a dead reckoning calculation and based on the first change in position and data received from an Inertial Measurement Unit (IMU) associated with the user device
Implementation Method 2
data received from an Inertial Measurement Unit (IMU) associated with the user device
Data Source
AI summary
Techniques for predicting a virtual camera view in an augmented reality (AR) or virtual reality (VR) application. A first change in position of a user device over a first time period is determined based on analyzing a plurality of frames of image data related to an AR or VR application. A dead reckoning calculation is used to predict a second change in position of the user device over a second time period, based on the first change in position and data received from an Inertial Measurement Unit (IMU) associated with the user device. A plurality of frames of image data are generated for display in the AR or VR application, based on the predicted second change in position of the user device.