Augmented Diffractive Optical Element for Tileable Structured Light Projection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current structured light projection technologies for wide field-of-view (FOV) depth sensing face challenges such as high zero-order diffraction values, pincushion distortion, increased complexity, and higher costs due to complex diffractive optical element (DOE) designs, which lead to inefficient use of projector power and increased algorithmic complexity.
Innovation Solution
The use of augmented diffractive optical elements (ADOEs) that are designed to prevent projection of light outside a tiling boundary, allowing for efficient tiling of structured light patterns without distortion, thereby achieving a wide FOV while reducing zero-order values and simplifying fabrication processes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If a single wide FOV diffractive optical element (DOE) is used for structured light projection, then the field of view is widened, but the zero-order diffraction value increases significantly
Solution Approach 1:
The patent divides the wide FOV projection into multiple smaller FOV projections by using multiple DOEs, each covering a specific angular range. This segmentation prevents the zero-order value from accumulating across the entire wide FOV, as each individual DOE operates within a limited angular range where zero-order effects are manageable.
Solution Approach 2:
Each DOE in the array is designed with specific local optical properties optimized for its particular angular range. The diffraction patterns and efficiencies are tailored to local requirements, ensuring that zero-order values remain controlled in each local region while collectively covering a wide FOV.
2Object-generated harmful factors
If multiple DOEs are tiled to achieve wide FOV illumination, then the zero-order value is reduced, but pincushion distortion increases and causes pattern overlapping and voids
Solution Approach 1:
The patent applies distortion pre-compensation to the design patterns before they are projected through the DOEs. By pre-distorting the patterns in the opposite direction of the expected pincushion distortion, the final projected patterns remain undistorted and tile seamlessly without overlapping or voids.
Solution Approach 2:
The patent adjusts diffraction parameters and grating designs of individual DOEs to compensate for distortion effects. By changing parameters such as grating spacing, depth, and orientation, the system optimizes projection accuracy across the wide FOV while maintaining tileability.
3Manufacturing precision
If distortion pre-compensation is applied to achieve rectangular projection shapes, then projection accuracy is improved, but fabrication complexity and cost increase
Solution Approach 1:
The patent applies distortion pre-compensation selectively to critical regions of the projection pattern where accuracy is most important, rather than uniformly across the entire pattern. This partial application maintains sufficient accuracy while reducing fabrication complexity compared to full pre-compensation.
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 enables efficient wide FOV illumination with reduced distortion, lower zero-order values, and simplified manufacturing, leading to improved depth sensing accuracy and power usage without the need for advanced fabrication processes.
Implementation Method 1
The illumination sources of the DCA emit light that are diffracted by one or more augmented diffractive optical elements (ADOEs) to form corresponding SL patterns
Data Source
AI summary
A depth camera assembly (DCA) includes a projector, a detector and a controller. The projector emits a tiled structured light (SL) pattern onto a local area. Each illumination source of the projector includes one or more light emitters and an augmented diffractive optical element (ADOE) designed with a pattern mask. The ADOE diffracts at least a portion of light beams emitted from the light emitters to form a first SL pattern projection having a field-of-view corresponding to a first tileable boundary. The pattern mask prevents projection of light that would otherwise be diffracted outside the first tileable boundary. The first SL pattern projection is combined with at least a second SL pattern projection into the tiled SL pattern illuminating objects in the local area. The detector captures images of the objects illuminated by the SL pattern. The controller determines depth information for the objects using the captured images.